Rajnarayan Chandavarkar - Imperial Power and Popular Politics - Class, Resistance and The State in India, 1850-1950-Cambridge University Press (1998)

Page i
Imperial Power and Popular Politics

Page ii
In this series of interconnected studies Rajnarayan Chandavarkar offers a powerful revisionist analysis of the relationship between class and politics in India between
the Mutiny and Independence. Dr Chandavarkar rejects the 'Orientalist' view of Indian social and economic development as unique and exceptional, which calls for
explanations specific to its culture, and reasserts the critical role of the working classes in shaping the pattern of Indian capitalist development. He demonstrates the
inadequacy of 'culture' as a dominant tool of historical analysis, especially as manifested in those recent subaltern studies which have focused upon colonial discourse to
the almost complete exclusion of the material things. An underlying and recurrent theme of the book is how perceptions of power shaped alignments of class and
influenced changing definitions of social identity. The book ranges widely across the social and political history of the working classes in India, examining the character
of trade unions, the political culture of the working class neighbourhoods, the nature of violence and policing, popular responses to the moral panic of the plague
epidemic and the Gandhian inflection of nationalist rhetoric. Dr Chandavarkar's analysis of political discourse, community structure and class relations in industrializing
India has major implications, and Imperial power and popular politics offers one of the most sustained and sophisticated critiques yet made of both Marxist and
functionalist narratives of industrialization. In their stead Dr Chandavarkar emphasizes the fluidity and flexibility of the relationships between discourse and power,
language and political practice, and in the work's concluding chapter he offers an alternative schematic view of the process of class formation in India, within the
context of 'world capitalism'.
Sustained in argument and elegant in exposition, this book represents a major contribution not only to the history of the Indian working classes, but to the history of
industrial capitalism and colonialism as a whole. Imperial power and popular politics will be essential reading for all scholars and students of recent political,
economic and social history, social theory, and cultural and colonial studies.

Page iii
Imperial Power and Popular Politics
Class, Resistance and the State in India, c. 1850–1950
Rajnarayan Chandavarkar
Trinity College, Cambridge

Page iv
PUBLISHED BY CAMBRIDGE UNIVERSITY PRESS (VIRTUAL PUBLISHING) FOR AND ON BEHALF OF THE PRESS SYNDICATE OF THE
UNIVERSITY OF CAMBRIDGE
© Cambridge University Press 1998
This edition © Cambridge University Press (Virtual Publishing) 2001
PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE
The Pitt Building, Trumpington Street, Cambridge CB2 1RP, United Kingdom
CAMBRIDGE UNIVERSITY PRESS
The Edinburgh Building, Cambridge CB2 2RU, United Kingdom
40 West 20th Street, New York, NY 10011 – 4211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia
© in this collection Cambridge University Press 1998
This book 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 1998
Printed in the United Kingdom at the University Press, Cambridge
Typeset in Plantin 10/12 pt [VN]
A catalogue record for this book is available from the British Library
Library of Congress cataloguing in publication data
Chandavarkar, Rajnarayan.
Imperial power and popular politics: Class, resistance
and the state in India, c. 1850–1950/Rajnarayan Shamrao Chandavarkar.
p. cm.
Includes bibliographical references and index.
ISBN 0 521 59234 8 (hb). – ISBN 0 521 59692 0 (pb)
1. Capitalism – India – History – 19th century. 2. Capitalism –
India – History – 20th century. 3. Working class – India – History –
19th century. 4. Working class – India – History – 20th century.
5. Social classes – India – History – 19th century. 6. Social classes –
India – History – 20th century. 7. Imperialism – History. I. Title.
HC433.C445 1998
330.12'2'0954 – dc21 97–8911 CIP
ISBN 0 521 59234 8 hardback
ISBN 0 521 59692 0 paperback
eISBN 0511004214 virtual (netLibrary Edition)

Page v
For Jennifer

Page vii
Contents
Acknowledgements viii
List of Abbreviations x
Glossary xi
1 1
Introduction
2 30
Industrialization in India before 1947: Conventional Approaches and Alternative
Perspectives
3 74
Workers, Trade Unions and the State in Colonial India
4 100
Workers' Politics and the Mill Districts in Bombay between the Wars
5 143
Workers, Violence and the Colonial State: Representation, Repression and
Resistance
6 180
Police and Public Order in Bombay, 1880–1947
7 234
Plague Panic and Epidemic Politics in India, 1896–1914
8 266
Indian Nationalism, 1914–1947: Gandhian Rhetoric, the Congress and the
Working Classes
9 327
South Asia and World Capitalism: Towards a Social History of Labour
Bibliography 351
Index 377

Page viii
Acknowledgements
This book is part of a wider enquiry into the social history of capitalism in India with which I have been engaged. It develops suggestions offered in The Origins of
Industrial Capitalism in India: Business Strategies and the Working Classes in Bombay, 1900–1940 (Cambridge University Press, 1994), seeks to open up
fresh lines of enquiry and raises questions, some of which I hope to pursue in a further study of workingclass politics in Bombay in the early twentieth century.
The research for this book has been generously supported by the Master and Fellows of Trinity College, Cambridge; the Managers of the Smuts Memorial Fund; and
the General Board of the University of Cambridge. For facilitating this research, I am indebted to the custodians, archivists, librarians and staff of various institutions: of
the Maharashtra State Archives, the Mumbai Marathi Grantha Sangrahalaya and the Bhau Daji Lad Museum in Bombay; of the record rooms of the Bombay
Millowners' Association, the Commissioner of Police, Bombay, and the Deputy InspectorGeneral, Criminal Investigation Department, Maharashtra; the Gokhale
Institute of Politics and Economics, Pune; the National Archives of India and the Nehru Memorial Museum and Library in New Delhi; the Asiatic Society of Calcutta;
the Oriental and India Office Collections at the British Library in London; and in Cambridge, the University Library and the Centre of South Asian Studies, whose
SecretaryLibrarian, Lionel Carter, has once again been unfailingly helpful.
Chapters 2, 4 and 7 have been previously published: chapter 2 appeared in Modern Asian Studies, 19:3 (July 1985) in a special issue which reviewed the
Cambridge Economic History of India; chapter 4 in Modern Asian Studies, 15:3 (July 1981), in a special issue, entitled Power, Profit and Politics: Essays on
Imperialism, Nationalism and Change in Twentieth Century India, 18701940, edited by C. J. Baker, G. Johnson and A. Seal; and chapter 7 in P. Slack and T.
Ranger (eds.), Epidemic and Ideas: Essays on the Historical Perceptions of Pestilence (Cambridge University Press,

Page ix
1994). I thank the editors and publishers for their permission to reproduce those chapters.
Most of the chapters of this book have been presented, in part or whole, at seminars in Cambridge, Oxford, Hull, London, Amsterdam, Boston, and at Dartmouth
College, New Hampshire, Trinity College, Hartford, Connecticut, the Nehru Memorial Museum and Library in New Delhi, the Department of History, Calcutta
University, and the Centre for Studies in Social Sciences in Calcutta. I am grateful to those who participated and thus forced me to rethink and recast my ideas. Tony
Giddens, Tapan Raychaudhuri, Gareth Stedman Jones and John Saville read an earlier draft of this book and provided comments and encouragement. Many friends,
students and colleagues have, through their scholarship, or in conversation and argument, helped to shape this work. It is a particular pleasure to thank for their
friendship and support in numberless ways Arup Banerji, Subho Basu, Chris Bayly, Joya Chatterji, Orlando Figes, Nandini Gooptu, Douglas Haynes, Gordon
Johnson, Anil Seal, Samita Sen, Sunita Thakur, Hari Vasudevan and David Washbrook. There are few ideas in the following pages which I have not rehearsed with
Jennifer Davis, whose insight, patience and understanding has made this book, and much else besides, possible.

Page x
Abbreviations
AICC AllIndia Congress Committee
AITUC AllIndia Trade Union Congress
ATLA Ahmedabad Textile Labour Association
BDEC Bombay Disturbances Enquiry Committee
BMOA Bombay Millowners' Association
BPP SAI Bombay Presidency Police, Secret Abstracts of Intelligence
BRIC Bombay Riots Inquiry Committee
BSEC Bombay Strike Enquiry Committee
BTLU Bombay Textile Labour Union
CEHI The Cambridge Economic History of India
CPI Communist Party of India
CWMG The Collected Works of Mahatma Gandhi
GOB Government of Bombay
GOI Government of India
IESHR Indian Economic and Social History Review
ILO International Labour Organization
ITB Indian Tariff Board
JAS Journal of Asian Studies
MAS Modern Asian Studies
MCC Proceedings of the Meerut Conspiracy Case
MSA Maharashtra State Archives, Bombay
NAI National Archives of India, New Delhi
NMML Nehru Memorial Museum and Library, New Delhi
OIOC Oriental and India Office Collections, British Library, London
PP Parliamentary Papers
PRO Public Record Office, London
PWD Public Works Department
RCLI Royal Commission on Labour in India
TLIC Textile Labour Inquiry Committee

Page xi
Glossary
adivasi—original inhabitants; denotes 'scheduled tribes' under the Indian constitution
ahimsa—nonviolence
akhada—gymnasium; meeting place; residence of religious mendicants
badmash—hooligan
bania—trader, moneylender, grain dealer; also a caste name
bhadralok—gentlefolk; refers mainly to higher caste and literate groups in Bengal
charkha—spinning wheel
chawl—tenement
dacoit—bandit; robber who uses violence and operates in gangs
dada—literally, 'elder brother'; used to describe a neighbourhood tough
dalit—oppressed; 'untouchable' caste
ganja—cannabis
Ganpati—another name for the Hindu god Ganesh
garibi hatao—abolish poverty; Congress Party slogan in the early 1970s
goonda—hooligan, thug
hakim—physician
halalkhore—sweeper who removed refuse and excreta from houses and streets
hartal—strike, cessation of work or trade dispute
havildar—constable
jamat—council of a caste, neighbourhood or sect
khaddar—coarse, homespun cloth
kotwal—police or legal official
lathi—stick, bludgeon
mahajan—merchants' guild, assembly, association
majur—labourer
malik—patron, proprietor, boss

Page xii
maulvi—Muslim priest or learned man
mela—festival; company of dancers taking part in a festival
mofussil—the provinces or hinterland
moholla—neighbourhood, quarter of a town
panchayat—council or tribunal, typically consisting of five people
patel—village official, headman
sadhu—Hindu ascetic; holy man with saintly qualities
sangh—association, organization
saraf, shroff—banker, moneychanger
satyagraha—truthforce; a term used by Gandhi for his technique of passive resistance
seth—wealthy financier; merchant; head of trade guild
shuddhi—ritual of purification, movement for conversion of Muslims to Hinduism
sirdar, sardar—foreman, labour contractor, jobber
swadeshi—product of the nation; goods made in India
swaraj—selfrule
tabut—shrine; a model of the tomb of Husain at Kerbala which is carried in procession during a festival
talimkhana—gymnasium
taluka—an administrative unit below the level of the district
taluqdar—landlord, especially in north India
tamasha—show or entertainment; folk theatre
tanzeem—organization; specifically organization for Muslim unity
toli—gang of men who levied contributions from shopkeepers for Mohurran or other festivals
ugarani—the collection of money which is considered to be due
vaidya—physician, especially in the Ayurvedic tradition
waaz—sermon or discourse preached in a mosque
wadi—quarter or neighbourhood of a town
zamindar—landowner

Page 1
1—
Introduction
This book investigates the interplay between class relations and political discourse in late nineteenth and early twentiethcentury India. It seeks to develop and
elaborate lines of enquiry which I had begun to pursue in earlier work on the social history of capitalism in India.1 This work had suggested that the history of capitalism
and of the working class in India could most fruitfully be investigated, not as an exception – neither as a case of 'precapitalist' development nor as a product of a
peculiar and unique 'Indian culture'2 – but firmly in relation to what are deemed to be the 'rules' or expectations of sociological discourse. An old, persistent and
frequently reactivated 'Orientalist' tradition has long encouraged historians to deem Indian society an exception to every rule of social (and historical) explanation. In
no aspect of the study of Indian society has this assumption of exceptionalism been more resolutely embedded and more subtly manifested than the investigation of the
working classes. Yet to assume the exceptionalism of Indian society is to obscure and distort its character and to deepen its more intractable conundrums. Opened to
the logics of explanation deployed elsewhere, its history offers a significant vantage point from which the theoretical apparatus of social analysis, conventional
explanations and analytical expectations – developed primarily in relation to the unique historical experience of Britain – may be interrogated, reassessed and revised.
This book proceeds, therefore, in a spirit contrary to Kipling's maxim.
The principal aim of this book is to ground the investigation of social conflict and power relations in late nineteenth and early twentiethcentury India more firmly in
their political (and intellectual) context.
1 Especially in Rajnarayan Chandavarkar, 'Workers' Politics and the Mill Districts in Bombay Between the Wars', MAS, 15:3 (1981), special issue, Profit, Power and Politics:
Imperialism, Nationalism and Change in India, 1870–1940, edited by Christopher Baker, Gordon Johnson and Anil Seal (reprinted below as ch. 4), and The Origins of Industrial
Capitalism in India: Business Strategies and the Working Classes in Bombay, 1900–1940 (Cambridge, 1994).
2 This argument has been most explicitly restated by Dipesh Chakrabarty, Rethinking Working Class History: Bengal, 1890–1940 (Delhi, 1989).

Page 2
While the following chapters traverse the quotidian realms of society and culture, their aim is to retrieve the significance of politics from the solvents of 'popular culture'
and 'everyday life' within which it has too often been submerged. The contention of this book is that the form and meaning of social relations, and not least the forms of
knowledge, or the discourses by which they are constituted, are determined by political conflict and defined by the outcome of power relations. It is concerned,
therefore, with how forces of resistance perceive dominant groups and institutions and how agencies of control imagine, define and sometimes actively respond to
problems of order, how perceptions of interests and identities at particular moments can facilitate alignments of power which cut across class, and how political conflict
and debate can lead, within a particular historical conjuncture, to the definition of class.
Furthermore, an attempt is made in this book to scrutinize the nature and meaning of certain concepts, which are central and recurrent in the subject – most extensively,
'industrialization', 'violence', 'crime', 'nationalism', 'world capitalism', and inevitably, of course, 'class' – and, in particular, to examine their construction within a given
historical context. This scrutiny is primarily historical, not philosophical, in approach. The nature of these concepts is examined in their engagement with the historical
evidence, frequently the awkwardness of the evidence with which scholarship, built on the conventional discourse of the social sciences, is copiously presented in the
Indian archives and libraries. The conceptual and interpretative problems addressed in this book are not exhaustive and no attempt is made to provide or disinter a list
of 'keywords'. Nor is this discussion offered as a methodological exercise, which has almost always given pleasure to those who attempt it strictly in proportion to the
pain it causes to those upon whom it is inflicted.
However, close attention of this sort to the conceptual vocabulary of the historian, and its relationship to the meaning of these concepts in contemporary politics, does,
indeed, serve a methodological purpose. It is, of course, a commonplace, often repeated, that the history of the powerless and the poor comes to us primarily in the
voice, and perceptions, of their rulers and exploiters. Some historians have suggested that a path may be cut through such difficulties by reading the sources 'against the
grain'. Others have taken the implication of this dictum a step further and recommended that we should read the sources for their 'silences', by shifting the emphasis
from the text to its marginalia or even its omissions.3
3 In Indian historiography, these exhortations have most frequently emanated from the socalled 'subaltern' historians. R. Guha (ed.), Subaltern Studies, vols. I–VI (Delhi, 1982–89);
P. Chatterjee and G. Pandey (eds.), Subaltern Studies vol. VII (Delhi, 1992); D. Arnold and D. Hardiman (eds.), Subaltern Studies, vol. VIII (Delhi, 1994); S. Amin and D.
Chakrabarty (eds.), Subaltern Studies, vol. XI (Delhi, 1996).

Page 3
Of course, to some extent, and with varying degrees of success, these 'methods' have long been deployed by even the most conventional historians. But the effects of
spelling out the habit of an art into a methodological doctrine have not always been fruitful, least of all, ironically, in the hands of its advocates. To effectively transcend
the discourse of the colonial rulers and the dominant classes, historians would not only have to read their sources 'against the grain' but also subject their own texts to
the same vigorous and selfconscious process of deconstruction. Otherwise, despite their intention to provide a subversive reading, historians were liable, especially,
for instance, in their analysis of violence, crime and popular culture, to replicate the assumptions and prejudices of the dominant discourse. Moreover, the strategy of
reading the sources for their 'silences' and 'absences', of pouring over the record for what it does not contain, has sometimes served to blunt the sensitivity of historians
to, perhaps dull their curiosity for, what they actually can and do tell us and to distract them from fully exploring the uses to which they may be put.
In earlier work, I had suggested that the pattern of capitalist development in India had been shaped largely by the role of the working classes. Business strategies and
entrepreneurial choice in early twentiethcentury India had been determined by the struggle to control and discipline labour. The social responses and political action of
workers, both within and beyond the workplace, often defined the options available to capital.4 Although historians and anthropologists have paid close attention, in
studies of Indian capitalism, to trade and commerce, merchants and entrepreneurship, burghers and bazaars, they have continued to neglect the history of the working
classes. In the light of the historiographical traditions and practices which have dominated the study of colonial India, this is not, perhaps, wholly surprising. It was not
simply that, until recently, historians have largely focused their attention on governance and upon elites, but, more crucially, that the contours of the subject came to be
mapped in relation to the process of imperial conquest and consolidation. The historical narrative of Indian society in the colonial period has turned upon the
relationships and events which facilitated or impeded the entrenchment of British rule: conquest and mutiny, revenue extraction and judicial administration, rural
organization and village society, indigenous tradition and customary practice, collaboration and nationalism, changing strategies to retain the empire and sacral conflicts
which were deemed to have driven the partition of India. If Indian historiography could not fully escape the rigours of colonial knowledge, it was at least in part
because the dominant categories of historical investi
4 Chandavarkar, The Origins of Industrial Capitalism.

Page 4
gation emerged as an adjunct of the colonial task of managing and controlling Indian society.
In this historical narrative, the problem of labour, and the social formation of the working classes, remained no more than a minor, indeed, marginal motif. The clash
between imperialism and nationalism provided the organizing principle of research in Indian history. Questions which could not be contained within these polarities
were simply excluded. Even economic historians primarily debated the counterfactual question of whether British rule had modernized or retarded the Indian
economy. Social history appeared to consist less of the study of the history of Indian society than of evaluating the extent of the 'Westernization' and 'modernization' of
'indigenous' society under the impact of colonialism. In this investigation of 'indigenous' tradition, scholars concentrated upon what were seen as its unique and
exceptional features and embarked, accordingly, upon a cultural anthropology of the past. At a more general level, historians investigating particular social groups often
took their unitary character for granted. Castes, religious and regional groups, occupations and even loosely banded social classes were readily hypostatized.
Moreover, since Indian society was deemed to be essentially 'agrarian' and 'traditional', their attention was directed towards land revenue policies, 'the mode of
production' in Indian agriculture or the rise of the rich peasant. Such a historiographical context, marked by its emphasis on magnates and elites, its consideration of
governance at the expense of social process, its Orientalist attraction to the uniqueness of 'Indian culture', was scarcely conducive to the study of the urban working
classes or, indeed, labour in general.
Yet when historians addressed the subject of the working classes, they did not always serve to liberate it from the deadweight of traditional historiographical practice.
Until recently, the focus of enquiry has rested upon the functionalist question of the role of labour in economic development: whether it could be recruited, trained and
organized sufficiently to serve the needs of Indian industry. The recruitment and adaptation of workers, and the mechanisms for their control and discipline in the
factories, provided the core around which research in Indian (and perhaps, much colonial) labour history has developed.5 Similarly, studies of
5 C. A. Myers, Labour Problems in the Industrialization of India (Cambridge, Mass., 1958), M. D. Morris, The Emergence of an Industrial Labour Force in India: A Study of the
Bombay Cotton Mills, 1854–1947 (Berkeley and Los Angeles, 1965); D. Mazumdar, 'Labour Supply in Early Industrialization: The Case of the Bombay Textile Industry', Economic
History Review, second series, 26:3 (August 1973), 477–96; B. Misra, 'Factory Labour During the Early Years of Industrialisation: An Appraisal in the Light of the Indian Factory
Commission 1890', IESHR, 12:3 (1975) 203–28; C. P. Simmons, 'Recruiting and Organizing an Industrial Labour Force in Colonial India: The Case of the Coal
(footnote continued on next page)

Page 5
the labour movement have, in different ways, turned upon how far and with what consequences, modern trade unions (and modern methods of management and
collective bargaining) could replace the traditional leaders and practices of the working classes.6 These lines of enquiry emanated from and embraced diverse traditions
of scholarship, including those which were hostile to their functionalist inheritance. Frequently, the working classes have been represented in terms of 'cultural'
characteristics attributed to particular economic categories or occupational states of being – 'migrant workers', 'the urban poor', 'landless labour' or 'factory operatives'
– which in turn were overlaid with expectations produced by 'cultural' definition – of caste and religion, for instance – and generated by the discourse within which
Indian society as a whole was objectified and characterized. In other words, historians 'read off' the attitudes, mentalities and social consciousness of these social
groups from the categories within which they lumped them. In this way, the working classes entered Indian history overburdened with the historians' or social
scientists' expectations and, indeed, their tautologies. This has only made it seem more crucial to scrutinize both the language of historical analysis as well as the
vocabulary of contemporary political discourse.
If historians of the Indian working classes thus approached their inquiries within the limits imposed either by the universalist teleologies of the
(footnote continued from previous page)
Mining Industry c. 1880–1939', IESHR, 13:4 (1976), 455–85; R. Das Gupta, 'Factory Labour in Eastern India: Sources of Supply, 1855–1946: Some Preliminary Findings', IESHR, 13:3
(1976), 277–328; L. Chakravarthy, 'Emergence of an Industrial Labour Force in a Dual Economy – British India 1880–1920', IESHR 15:3 (1978), 249–327; R. Newman, 'Social Factors in
the Recruitment of the Bombay Millhands', in K. N. Chaudhuri and C. J. Dewey (eds.), Economy and Society: Essays in Indian Economy and Society (Delhi, 1979); P. S. Gupta,
'Notes on the Origin and Structuring of the Industrial Labour Force in India – 1880 to 1920', in R. S. Sharma, (ed. with V. Jha), Indian Society: Historical Probings: In Memory of D.
D. Kosambi (New Delhi, 1979), pp. 414–34; R. Newman, Workers and Unions in Bombay 1918–1929: A Study of Organisation in the Cotton Mills (Canberra, 1981); Chitra Joshi,
'Kanpur Textile Labour: Some Structural Features of Formative years', Economic and Political Weekly, 16:44–6, (November 1981), special issue, 1823–38; Chitra Joshi, 'Bonds of
Community, Ties of Religion: Kanpur Textile Workers in the Early Twentieth Century', IESHR, 22:3 (1985), 251–80; D. Arnold, 'Industrial Violence in Colonial India', Comparative
Studies in Society and History, 22:2 (1980), 234–55; D. Chakrabarty, 'On Deifying and Defying Authority: Managers and Workers in the Jute Mills of Bengal, circa 1890–1940', Past
and Present, 100 (August 1983), 124–46; Chakrabarty, Rethinking WorkingClass History; Marina Carter, Servants, Sirdars and Settlers: Indians in Mauritius, 1834–1874 (Delhi,
1995); I. J. Kerr, Building the Railways of the Raj, 1850–1900 (Delhi, 1995), especially chs. 4–6.
6 Morris, The Emergence of an Industrial Labour Force; Newman, Workers and Unions in Bombay; D. Kooiman, 'Jobbers and the Emergence of Trade Unions in Bombay City',
International Review of Social History, 22:3 (1977), 313–28; E. A. Ramaswamy, The Worker and His Union: A Study in South India (New Delhi, 1977); Arnold, 'Industrial Violence in
Colonial India', 234–55; E. D. Murphy, Unions in Conflict: A Comparative Study of Four South Indian Textile Centres 1918–1939 (New Delhi, 1981); Chakrabarty, Rethinking
Working Class History.

Page 6
social sciences, derived from both functionalism and Marxism, or alternatively, by the exceptionalism generated by an Orientalist and colonial discourse, they were also
influenced by trends in British and European historiography. Here, the history of the working classes had long been studied as the history of the labour movement, and
then, increasingly, the social conditions and social and cultural practices of its constituent groups. From nineteenthcentury socialism, historians of the 'Western'
working classes inherited the assumption that economic development determined the character of labour, its social organization and political consciousness. The pattern
and pace of industrialization, it was supposed, shaped the character of the social struggle and its political forms. The notion of the stages of industrialization yielded a
matching evolutionary scheme of the stages of class consciousness. The history and development of the labour movement was, thus, assumed to constitute the prelude
to the rise and triumph of socialism. In this way, the history of the working class came to be studied as an inseparable part of the rise of socialism. For some, the point
of interest in studying the history of the labour movement was to investigate how its historic mission, the achievement of socialism, could be advanced or even realized
and to track its progress. For others, its interest lay in discovering how the antagonisms and conflicts embodied by the labour movement could be managed, contained
and defused. More significant than these differences, which sometimes generated fierce debate, was the fact that their shared teleology defined the scope of historical
inquiry.
The effect of this teleology upon the interpretations of the working classes in general have been severely limiting, but in the case of India, in particular, they proved
especially damaging. Economic backwardness, in this reasoning, made the very notion of a working class unthinkable, just as the peculiar cultural institutions of India
seemed to place it in a special category of its own. The nature of the labour force, shaped by a low level of industrialization, could not be expected to develop a class
consciousness or a socialist politics. Moreover, if the weight of the factory proletariat in a predominantly peasant economy was small, and its political potential
correspondingly weak, the history of the working class appeared unimportant, or at any rate it became increasingly difficult to identify and assess its significance. So if
the Indian working classes constituted either the antecedents of Western society or a special and unique case, its history was unlikely to release materials for thinking
about class formation and class consciousness more generally.7 Most crucially, this teleology im
7 It is ironical, therefore, that at least one recent attempt at 'rethinking workingclass history' on the strength of the Indian 'case' has been predicated wholly and insistently on its
cultural exceptionalism (and economic backwardness). Chakrabarty, Rethinking Working Class History.

Page 7
posed upon the working classes an arbitrary and misleadingly narrow definition as an industrial labour force. In this sense, the industrial labour force was abstracted
from its connections with other categories of labour who were proletarianized in nineteenth and twentiethcentury India by similar social processes and who were
increasingly being subjected to the dominance of mercantile and industrial capital.
It was also implicit in this teleological view that industrialization constituted a universal, serial, technological process through which every society passed at an
appropriate level of development. The nature of the Indian working class, it seemed, could scarcely be understood outside the context of the fact that the Indian
economy was passing through an 'early stage' of industrialization. By making Indian society coeval with Britain (or Europe) in a previous and not its contemporaneous
epoch,8 scholars took the first major steps towards entrenching the myth of Indian exceptionalism or, alternatively, the parallel notion that the Indian case was a
defective variant of the West.
In The Origins of Industrial Capitalism, I had sought to retrieve and delineate the wider context within which the working classes formed in India. The links between
factory workers, casual labour and various streams of rural migrants, I had argued, were sufficiently intimate to strain the rigid definition of the 'working classes' as an
'industrial labour force' to the point of obliteration. This book seeks to develop the argument that the history of the working classes can only be grasped fully in the light
of the powerful connections, which were established in the process of their social formation, between factory proletarians, casual workers, rural migrants, agrarian
labour, artisans, 'tribals' and dalits. The Indian working class was largely composed of rural migrants who retained close connections with their village base. Indeed,
they often migrated to earn the wages which would enable them to retain their village holdings, and as a result, they were frequently more militant in defending their jobs
and earnings in the workplace. If the working classes were thus firmly rooted in the countryside, there was no clear line of demarcation between factory hands and the
casual poor. No clear distinction can be sustained between them in their attitudes to work and politics, their lifestyles and job preferences, their response to the law
and their propensity to violence. Yet, tautologically, these imputed behavioural characteristics have often provided the means of distinguishing between 'the casual poor'
and 'the respectable working classes'. As an analytical concept and a heuristic category, the working class were never very clearly demarcated from 'the peasant' or
'the casual poor'.9 Indeed, recruits to 'the working class' in India encompassed very diverse social formations, as tribals were inden
8 J. Fabian, Time and the Other: How Anthropology Makes Its Object (New York, 1983).
9 Chandavarkar, The Origins of Industrial Capitalism, chs. 3, 4 and 5.

Page 8
tured for the tea gardens or recruited for the coal mines, and dalits and landless peasants sought work as field labourers at harvest time or migrated to nearby towns for
employment in the trades or the 'service sector'. Factory hands, who were deemed to constitute the working class in the narrowest and most traditional sense, could
necessarily be described by several other labels. The conventional definition of 'the working class' in terms of an urban factory proletariat appeared particularly limited
when it was measured against the historical evidence of labour force formation in late nineteenth and twentiethcentury India. Ironically, the commonplace that the
working class derived its identity primarily from the relations of production – more specifically, from the factory and the production line – has held firm at the very
moment at which 'the factory system' itself, in its most developed Fordist form, was finally being dissolved.
If, on conventional expectations, the very notion of a working class in India appeared unthinkable, a detailed investigation of workers' politics in Bombay between the
wars suggested that these conventions may have to be revised. For Bombay not only witnessed a scale of industrial and political action which was rarely replicated in
conditions of 'advanced capitalism' but workers' struggles, at particular historical conjunctures, disclosed a fiercely held 'class consciousness'. This exploration of the
nature of workers' politics in Bombay, originally published in 1981, is reprinted below as chapter 4. The remaining chapters in this volume address problems and seek
to develop perspectives which had first emerged in that essay.
The more closely 'the Indian case' was examined the more it cast into doubt the evolutionary assumptions which had held the subject together. Significantly, it became
apparent that it was impossible to understand the central issues of labour force formation in the cotton mills without grasping its intricate, if inextricable, connections
with the social organization of the neighbourhood. These connections between workplace and neighbourhood were vital not only to the recruitment and control of
labour, but, as the mill committees of 1928–29 showed, also to the organization and conduct of collective action. The investigation of the history of the working class, it
seemed, had been unduly restricted by its narrow and exclusive focus upon the workplace.
The investigation of the interplay between the social relations of the workplace and the social organization of the neighbourhood had numerous implications for our
understanding of class formation. First, it was customarily supposed that it was the caste, kinship and religious loyalties of workers which constituted an insuperable
obstacle to the growth of class consciousness. Until further industrialization dissolved these 'primordial ties', class consciousness would not develop. Yet it was
precisely within the social organization of the neighbourhood that the solidarities

Page 9
of collective action were forged and their informing ideologies shaped. Second, the neighbourhood was not only an arena in which the political solidarities of the
working classes were forged, but also the site in which its differences and divisions were seen to become manifest. From the vantage point of the neighbourhood, the
segmentation of the labour market as well as the rivalries of the workplace were laid bare. The aggregation of workers within the factory or the mill did not necessarily
sharpen their sense of common interest; indeed, it could serve to expose the whole spectrum of their differences. Third, as the workers' politics were defined in terms
of the interrelationship between the spheres of workplace and neighbourhood, so their development could be more fully explained in terms of the playing out of
diverse sets of power relations rather than simply as an effect of their relationship to the means of production. Social consciousness and social being were shaped in a
political domain, extending from the daily social relations of workplace and neighbourhood to more public conflicts and confrontations with the state. The political
consciousness of the working classes appeared to be shaped crucially by their experience of, and their relationship with, the state. Their solidarities were not the natural
outcome of popular culture or a reflex of the specific character of production relations, but, rather, they were politically constituted, and as such they were contingent,
sometimes transient and even evanescent. The politics of the working classes had therefore to be situated in the wider context of the social and political alignments
which shaped them. In studies of the bubonic plague epidemic and the discourse of violence in workers' politics, strategies of policing and law enforcement and the
appeal of the Congress among the urban working classes, I have sought, therefore, to investigate the power relations, entailed, described and determined by these
alignments, and the distorting prisms through which the state and the dominant classes and those who challenged and resisted them viewed each other.
From this perspective, the working class could not be grasped as a unitary formation, with a real essence and a single homogeneous identity. Its social formation was
the outcome of processes and relationships which were contingent; it encompassed identities which were varied, conflicting and labile. The social groups which
comprised the working class could be disassembled by diverse and contradictory criteria into numerous fragments, variously constituted. Their broadly similar
relationship to the means of production did not yield objective interests in common, overriding their sectionalism. If sectional difference, based on gender and
occupation, caste and kinship, religion and ideology, came to be negotiated into alliances and unities at specific historical conjunctures, then it became yet more
pressing to explain why these solidarities could be forged at all. Perceptions of mutual interests, indeed the language for their

Page 10
description, were the product of a specific intellectual and political context. Similarly, the interests of these elements which made up the working class depended upon,
and were defined within, specific historical circumstances, which were themselves constantly in flux. As they changed, they served to redefine the interests of their
constituent social groups and to reconstitute their social identities. The sectionalism of the working classes was integral to the process of its social formation; it was not
the product of its unique culture or its backward economy. As historians of the Western working classes paid closer attention to the competing and conflicting identities
of ethnicity and religion, kinship and gender, neighbourhood and nation, they began to focus attention upon the very issues which had led them previously to regard the
Indian case as unique and exceptional in the first place. Yet to describe these divisions within the working class as a process of sectionalism is to lend a certain
coherence to its 'sections' and the interests which made them up, as if they were impervious to change and resistant to further decomposition and reconstitution in a
different form under changed circumstances. In fact, if the formation of the working classes is seen as primarily a political process, contingent upon given historical
conditions, it should follow that its constituent elements could equally plausibly be defined in terms of several other social identities.
Once the significance of the relationship between workplace and neighbourhood for class formation and the economy of labour had been admitted, it severely
undermined the commonplace that the working class discovered its identity exclusively, or even primarily, at the point of production. Nor could the connection
between the level of economic development and the nature of social conflict and the forms of their political expression be readily taken for granted. Yet it was precisely
the belief that there was an immediate and intimate analytical connection between the economic history of the labour movement and its political expressions which had
long underpinned Whiggish and diffusionist notions of industrialization.
If these evolutionary assumptions about class formation could not easily be sustained, it became clear that conventional models of industrialization as a social process
would have to be reexamined. The definition of the working class in terms of a factory proletariat seemed intimately related to an understanding of industrialization as a
serial, technologically determined process which lay beyond the realm of social choice or political conflict. Such a definition could only be justified by, and, indeed, it
may be reduced to a usage of, modernization theory. To challenge this usage is also to demand the rethinking of the social meaning of its informing process:
'industrialization'. Chapter 2 – first published as an essay in

Page 11
1985 – represented an initial attempt to undertake this task. It challenged the assumption, shared by Marxists and functionalists alike, that industrialization constituted a
unilinear process, which shaped society inevitably in a single direction. Indeed, it seemed as if scholarly disputations between Marxists and functionalists about the
nature of industrialization and its social consequences had served to bring them closer together and to consolidate and entrench the assumptions they shared.
Accordingly, the starting point of chapter 2 was a growing scepticism about the received wisdom that industrial development in India was the outcome of a process of
technological diffusion set in motion by the industrial revolution in Western Europe and, indeed, that economic (or industrial) development would lead inevitably to
largescale factory production.
It had become increasingly apparent that evolutionary assumptions about the nature of industrialization had entailed a significant interpretative cost for Indian economic
history. For these assumptions were derived from and served to entrench a normative ideal of the nature and course of industrialization. In turn, it led historians to
abstract largescale industry from its wider economic and political context. It drew us towards a heroic view of entrepreneurs, as if they carried singlehandedly the
burdens of risk and the challenge of innovation, and struggled to raise the economy above the inescapable constraints which held it captive. Most crucially, perhaps,
the implicit assumption that technology constituted a neutral force blinded historians to the extent to which workers turned it into the site of conflicts over its use and,
correspondingly, to the impact of labour upon the pattern of industrial development.
Yet, on closer scrutiny, industrialization did not follow a unilinear path. Its outcome was neither inevitable nor inscribed into its origins. The development of largescale
industry in India was neither simply the result of the visionary genius of entrepreneurs driven to innovate and invest in the newest and best technology, nor were its
limitations the consequence of their 'mercantile' and speculative attitudes or rentseeking motives. Nor could it be supposed that largescale industry alone pointed the
way to the future or represented the only source of dynamism in the economy. As Christopher Baker's study of Tamil Nadu had shown, the handloomweaving
industry had disclosed a remarkable ability to adapt and survive, expand and innovate.10 But this did not represent an exceptional case.
10
C. J. Baker, An Indian Rural Economy: The Tamil Nad Countryside 1880–1950 (Oxford, 1984), ch. 5. Later work has strengthened this picture. See Tirthankar Roy, Artisans and
Industrialization: Indian Weaving in the Twentieth Century (Delhi, 1993); D. Haynes, 'The Dynamics of Continuity in Indian Domestic Industry: Jari Manufacture in Surat, 1900–
1947', IESHR, 23:2 (1986), 128–49 and D. Haynes, 'From Merchant Capital to Weavers' Capital: The Slow Transformation of Artisanal Production in the Bombay Presidency, 1900–
1950', unpublished paper.

Page 12
Moreover, industrial expansion was sometimes facilitated by improvisation with old, even obsolete, machinery, the effective manipulation of raw materials and the
intensive exploitation of cheap labour. Capitalist strategies were frequently and in crucial ways determined by workers' initiatives to defend their jobs and wage levels
and to retain some measure of control over their own labour. The path of industrialization in India was strewn with conflicts and contradictions which often altered its
course and determined its patterns of development.
To a large extent, diffusionist assumptions about the nature of industrialization and evolutionary schemata of its course had held together the conventional narrative
structure of labour history. If, in a radical interpretation, the labour movement signified the emergence of workingclass consciousness, whose further development was
manifested in the rise of socialism, in a more conservative, functionalist variant, more industrialization provided the scope, sometimes even the incentive, for labour to
adapt to its imperatives and abandon spontaneous resistance for formal collective bargaining. In chapter 3, this book examines the nature of the labour movement while
seeking to divest it of such evolutionary and diffusionist assumptions about the character of industrialization. It argues that the nature of trade union organization and the
pattern of industrial action were neither the product of an 'early stage of industrialization' nor a reflex of the 'preindustrial' or 'precapitalist' mentalities of the workers.
Of course, the sectionalism of the working class often made it difficult to form trade unions or to forge wider political solidarities. But this sectionalism was neither
unique to the Indian working class, nor simply generated by its cultural particularities. It was frequently developed by industrialization. It emanated from the division of
labour and the organization of work. It remained, indeed, inherent to the process of labour force formation. Politicians, employers and the state often played upon
these sectional differences and served at times to accentuate them. We cannot take for granted that the aggregation of workers in factories led them naturally to
perceive their interests as uniform.
To recognize that the solidarity of the working class was by no means immanent in the unfolding of industrialization or capitalism was to direct attention to the question
of why trade unions formed in the first place. Workers' resistance did not necessarily lead to its consolidation in trade union organization. Trade unions often appeared
to be no more than a lofty title for a strike committee. When they formed, they were often ephemeral to the dynamic of workingclass action. Conversely, trade union
weakness coexisted alongside the solidarities which workers sometimes forged and the determined struggles they waged to defend their collective interests. No simple
connections can be made between social

Page 13
consciousness, trade union organization and political action. The formation and growth of trade unions did not necessarily reflect the nature of workingclass
consciousness. On the contrary, the characteristics of the labour movement, often represented as symptomatic of a preindustrial society, were the product of the
political context in which they took shape. How the nature of trade union organization and the forms of industrial action were shaped by this political context, fashioned
by the policies of employers and the state, provides the substance of this chapter.
This argument has a wider relevance for explanations of the origins and growth of labour movements. If the weakness of trade unions reflected not the 'preindustrial'
nature of the workforce but the intolerance of the employers, the growth of a labour movement may be best explained by identifying the conditions in which, as the
outcome of the complex relations between workers, trade unions and governments, the employers and the state manifest a willingness to tolerate workers'
combinations. The readiness of capitalists to tolerate workers' combinations has often depended upon political pressure exerted by workers' struggles at the level of
the state and the wider perception of the attendant threat to the social order. If labour movements constituted neither the natural corollary to capitalist development nor
the expression of an immanent social consciousness, their formation calls for a political explanation and, in particular, the examination of their changing relation to the
state.
The implications of this hypothesis are here sketched schematically and cannot be developed in a study of this kind. They call for a more comprehensive and
comparative treatment of labour movements. However, chapters 5 to 8 follow some of its suggestions of an approach to the analysis of class and power. In 'Workers'
politics', I had suggested that the social and political alignments of the workingclass neighbourhoods reached out to wider arenas of politics, encompassing the colonial
state and political organization at various levels. The workingclass neighbourhoods of Bombay did not yield a natural propensity to class solidarity. Nonetheless, the
working classes were able at times to forge impressive solidarities out of a social context riven by conflicts and antagonisms, competing and diverse interests and
cultural and economic sectionalism. But these solidarities could not be reduced to a common denominator of popular culture or interpreted as the realization of class
interests. Within the workingclass neighbourhoods, political organization and action was constituted by the play of power relations and their reciprocities, between
jobbers and workers, landlords and tenants, creditors and debtors, all manner of patrons and clients, as much as between the state, employers and the working
classes. The political consciousness of the working classes formed in relation to the state and the public discourse stimulated

Page 14
by its interventions and the political negotiation around them. It is to this political discourse and the play of power relations in shaping the alignments of class, its broken
solidarities as well as its fundamental differences, that this book turns in chapters 5 to 8, addressing the discourse and practice of violence, methods of policing and law
enforcement, the plague panic of the 1890s and 1900s and the response of the working classes to the protean languages of Indian nationalism.
The central concern of these chapters is to examine the relationship between discourse and power, language and political practice. They examine how particular social
groups were defined and categorized in political rhetoric and public discourse. The fixity which these definitions were attributed in public discourse was frequently
belied by the flexibility of their usage, which could change according to their context. Nonetheless, the definition of social groups, and their accompanying
characterization, sometimes had large consequences – which the colonial state developed with energy and enthusiasm – for the policies of government. Similarly, the
colonial state and its agents played a prominent role in shaping this rhetoric, especially in relation to violence, crime and disease. But the purposes of the colonial state
would not have been served if they had not also been embedded in, and even influenced by, a wider political discourse in which diverse elements of Indian society
shared. The process by which social groups were so defined and characterized in public discourse also created and developed political alignments and shaped social
antagonisms which cut across class and caste, gender and community, religion and nation and sometimes reconstituted these old principles of social division.
Chapter 5 elucidates and analyses a colonial discourse about workingclass violence. But this was a discourse in which Indian elites also participated; at particular
junctures, it could encompass diverse elements of Indian society, including some workingclass groups as well. Later historians, sometimes in quest of 'the mind' of the
working classes, have often unwittingly taken over and replicated this discourse and sometimes even elevated it into grand explanatory frameworks and transcendent
sociological truths.11 Contemporary observers, like the scholars who followed them, attributed a particular propensity to violence to specific groups within the working
class: the casual poor, the rural migrant, the insufficiently proletarianized. Where contemporaries sometimes spoke from fear and sought to employ characterizations of
violence to justify particu
11 D. Arnold, 'Industrial Violence in Colonial India'; Chakrabarty, 'On Deifying and Defying Authority', 127; Chakrabarty, Rethinking Working Class History; Joshi, 'Bonds of
Community, Ties of Religion', 261–3; V. Das (ed.), Mirrors of Violence: Communities, Riots and Survivors in South Asia (Delhi, 1990).

Page 15
lar lines of policy or forms of state action, some historians have tended to valorize subaltern violence. Ironically, historians who valorized the violence of the urban poor
have often replicated the colonial construction of the roughness of the lower orders.
Chapter 5 sets the spurious analytical precision with which scholars have conceptualized violence in relation to the ambiguities of its classification and its social
meanings. The colonial state and its agents were often able to exploit the ambiguities inherent in the very concept of violence to create a wider consensus for some of
their harshest and most repressive policies. How this was effected in practice may be seen in the way in which measures were put together to repress the communist
leaders of the Bombay Girni Kamgar Union after the general strike of 1928, culminating in the tragic, if somewhat laborious, drama of the Meerut Conspiracy Case.
Like its republican successor, the colonial state was often able to evoke widespread support for its acquisition of repressive powers when it was able to demonstrate
the need to use them against particular and 'deviant' social groups which it defined into existence: most commonly, 'goondas', the 'floating population' of large cities, the
casual poor, or 'antisocial elements' in the current phrase. It is particularly striking that those who styled themselves as 'respectable' frequently ascribed the blame for
communal riots – and readily continue so to do – to those whom they perceived as 'rough'. Such representations of the urban poor may be taken historically to reflect
the aspiration of Indian elites to subordinate and control labour and, in the long run, to cheapen its cost. Colonial rule enabled Indian elites to realize this aspiration; half
a century of independence has enlarged their freedom and power to consolidate and extend their control yet further.
In the following chapter, this book turns to the most obvious, ubiquitous and public instrument of this control. Chapter 6 investigates the mechanisms and techniques of
urban policing in Bombay in the late nineteenth and early twentieth centuries. In particular, it seek to situate the police and their methods of operation more firmly within
the context of the social organization of the neighbourhood. Administrative and institutional histories of the police, however, have often tended to present it as a
monolithic force and to exaggerate its discipline and its responsiveness to central command.12 As a result, they have tended to overlook the extent to which the police
in their quotidian functions were exposed to
12 Percival Griffiths, To Guard My People: The History of the Indian Police (London, 1971); A. Gupta, The Police in British India, 1861–1947 (Delhi, 1979); D. Arnold, Police
Power and Colonial Rule: Madras, 1859–1947 (Delhi, 1986); P. Robb, 'The Ordering of Rural India: The Policing of Nineteenth Century Bengal and Bihar', in D. M. Anderson and
D. Killingray (eds.), Policing the Empire: Government, Authority and Control, 1830–1940 (Manchester, 1991), pp. 126–50.

Page 16
the negotiation of power relations in the street and the neighbourhood. The police often became a valuable resource in the earthy battles fought for control and
dominance within the neighbourhood. Moreover, individual connections between policemen and residents, sometimes based on caste, kinship and village, allowed the
working classes to enlist their help and their protection. Around the police, their strategies and their interventions in the neighbourhood, shifting patterns of alliance and
antagonisms between various contending elements formed and fractured.
Attention to the quotidian dimension of policing illuminates a paradox which lies at the heart of the colonial autocracy. British rule in India was an oriental despotism
which rested on force. The British developed through the police and the army, what often appeared to be a formidable apparatus of control and repression; yet they
sometimes appeared to deploy this sanction of physical force with a constraint and economy which was not always evident in Britain.13 Conversely, while the police
often appeared to be constrained by significant organizational and structural weaknesses, they also sometimes revealed an ability to respond to the threat of disorder
with a ruthless brutality which would not always have been contemplated elsewhere. Chapter 6 tries to resolve these conundrums by examining strategies for the
maintenance of order from the perspective of everyday policing. It analyses the assumptions of the agencies of the colonial state which informed their perception of the
threat of violence and drove them at times to respond on a scale which rendered the distinction between 'minimum' and 'maximum' force meaningless. As a result, the
colonial state seemed to threaten its subjects less but killed them more often in India than it did in its own heartland in Britain. As the colonial state passed under the
greater influence of Indian elites in the 1930s, however, it acquired an increasingly repressive demeanour. As they gained leverage upon the instruments of repression,
Indian propertied elites saw no reason to desist from using them to control and discipline labour more effectively, whether in the towns or the countryside. The
Congress ministries which took office under provincial autonomy between 1937 and 1939, erstwhile protagonists of nonviolent satyagraha and civil disobedience,
sometimes swiftly and enthusiastically slid their mailed fists into the velvet gloves of their British predecessors. In this sense, the growing willingness of the Indian state
since Independence to appropriate large executive powers, repress its opponents and under
13 V. A. C. Gatrell, 'Crime, Authority and the Policeman State, 1750–1950', in F. M. L. Thompson (ed.), The Cambridge Social History of Britain, 1750–1950 (Cambridge, 1988), pp.
243–310; K. Jeffrey and P. Hennessey, States of Emergency: British Governments and Strikebreaking since 1919 (London, 1983); Jane Morgan, Conflict and Order: The Police
and Labour Disputes in England and Wales, 1900–1939 (Oxford, 1987).

Page 17
mine civil liberties should not surprise us. It was firmly embedded in the habits of statecraft which the republican regime inherited – more than democracy, a British
legacy for which Indian elites had quickly shown enthusiasm in the colonial period.
From everyday policing, this book turns to the plague epidemic of the 1890s. Chapter 7 takes a further step away from the conventional arenas of politics to examine
the interplay between power and discourse, class and resistance in the construction of the bubonic plague epidemic of the 1890s and 1900s. The case of the plague
epidemic provides further illustration of the process by which different elements in Indian society sought to define themselves within public discourse and to negotiate
alignments and accommodations with each other, sometimes cutting across divisions of caste, religion and class.
Popular responses to the epidemic have sometimes been interpreted in strongly Orientalist terms, as the inevitable clash between Western rationality and Eastern
superstition, and as an expression of an indigenous popular culture.14 Yet these interpretations, or preconceptions, are inadequately supported by the evidence. We
can neither discern a single homogeneous popular response to the plague, nor can the popular reaction be clearly distinguished from the official and the elite. Rather,
the diagnosis of bubonic plague unleashed a moral panic which engulfed medical experts and research scientists, colonial officials and Indian elites, plague
administrators and the victims of the plague 'measures' alike. To investigate how the epidemic came to be constructed seemed to allow a way around the sterile
dichotomies of East and West, science and superstition, rationality and rumour which had enveloped the subject. Moreover, it was clear that the epidemic could not be
seen as a single, integrated phenomenon, which was experienced in the same way by all, but signified different things to different people. For this reason, it seemed
necessary to examine the specific political conjuncture in which the epidemic was constructed. The political construction of the epidemic in the late 1890s was the
outcome of the interaction between its disparate and numerous constituent elements: for instance, colonial perceptions of Indian society, medical and scientific rivalries,
the policies devised to combat the plague and the manner of their implementation, and the response and resistance of its victims. The popular response to the epidemic
was neither culturally specific nor irrational and prescientific.
14 D. Arnold, 'Touching the Body: Perspectives on the Indian Plague, 1896–1900', in Guha (ed.), Subaltern Studies, vol. V, pp. 55–90; reprinted in D. Arnold, Colonizing the Body:
State Medicine and Epidemic Disease in Nineteenth Century India (Berkeley and Los Angeles, 1993), ch. 5; Ira Klein, 'Plague, Policy and Popular Unrest in British India', MAS,
22:4 (1988), 723–55.

Page 18
Rather, it reflected the reluctance of the oppressed to place their faith fully either in the draconian, panicstricken measures or to believe the assurances of their
authoritarian oppressors. Indeed, in the manner of its construction, the plague epidemic of the 1890s in India disclosed remarkable similarities with the onset of the
AIDS epidemic in San Francisco in the 1980s.15
Chapter 8 examines how public discourse could serve not merely to constitute social groups within a particular historical conjuncture, but also to facilitate political
alliances between them. Since these 'social groups', socalled, could scarcely be dissolved into irreducible units, their constitution may be taken to reflect the outcome
of the process by which alliances were negotiated and took shape around a wider political discourse. The focus of this chapter rests upon the popular appeal of
nationalism in its 'mass phase' in the final decades of British rule. Its aim is to explore, and to seek to explain, the nature of the popular response to the Congress.
Conventionally, historians have either treated popular participation in nationalist agitations as a function of elite rivalries or of a diffusionist process of mobilization by
which the 'masses', more or less undifferentiated, are brought into the political domain by their leaders seeking to extend the basis of their support. Alternatively, they
have stressed the autonomy of subaltern politics. In either case, these approaches have made it difficult to grasp the relationship between the Congress and its
fluctuating and fractious following.
In chapter 8, it is argued that an understanding of the nature of the popular following evoked by the Congress, which neither wholly displaces it onto elite rivalries nor
collapses into the pitfalls of mobilization, would have to turn upon the extent to which its political programme provided diverse groups with the means of
comprehending their own social and political situation as well as suggesting realistic means by which they could transform it.16 To this end, the chapter focuses upon
Gandhi's rhetoric about the working classes and sets it in relation to their social and political struggles, particularly in Ahmedabad in the late 1910s and early 1920s. It
is neither intended to suggest that the working classes may be attributed a fixed social and political identity nor that the Congress espoused a single, consistent political
programme, let alone that they articulated it in a univocal rhetoric. Nonetheless, Gandhi's distinctive language increasingly inflected the rhetoric and programme of the
Con
15 Randy Shilts, And the Band Played On. Politics, People and the Aids Epidemic (London, 1988).
16 It draws here upon the seminal essay by G. Stedman Jones, 'Rethinking Chartism', in G. Stedman Jones, The Languages of Class: Studies in English Working Class History, 1832–
1982 (Cambridge, 1983), pp. 90–178.

Page 19
gress. For this reason, I seek to examine the imaginative possibilities which Gandhi's rhetoric offered workers' politics and the limitations which it imposed upon them.
In addition, the chapter argues that the popular interpretation of the rhetoric and programme of the Congress would also depend upon how its often varied and
inconsistent political interventions were perceived by the working classes. Accordingly, this chapter investigates how the changing character of the Congress in the last
decades of colonial rule affected its ability and its willingness to represent the working classes and their discontents. Conversely, it examines the constraints within
which workers' combinations struggled to emerge and the relationships which they sought to forge with politicians and publicists. It is in this context that the often
guarded response of the Congress to the motley and evanescent trade unions which formed and collapsed in this period should be situated. For the Congress, the task
of reconciling the differences among its diverse constituents posed insuperable problems to identifying themselves as the champions of the workers' (or any sectional)
interests. Finally, this chapter examines the nature of workingclass nationalism and, more generally, sets the rhetoric of the Congress in relation to the sectionalism,
whether of class or caste, language or religion, gender or region, of its putative followers.
Chapters 5 to 8, therefore, pick up in various ways and develop a critique of the concept of popular culture and its usage by historians of the subcontinent and, indeed,
elsewhere, which was suggested by the investigation of the workingclass neighbourhoods in chapter 4. Notions of workingclass culture emerged from attempts by
social historians to retrieve from the vitality of civil society the oppositional practices and traditions of resistance among the working classes. For those who concluded
that class and class consciousness could only be found in their real form in advanced capitalist societies, 'culture', especially 'popular culture' provided an alternative to
'class consciousness' and offered a looser category for the discussion of the ideologies and political actions of workers in the context of economic backwardness. Not
only did the concept take for granted, thereby, the evolutionary assumptions which had circumscribed the scope of Indian workingclass history but, more crucially,
perhaps, the use of the term has been riddled with contradictions and conceptual confusions. The concept has enabled historians to attend to the agency and initiative
of the working classes, both within the political and the social sphere. However, assumptions about the cultural characteristics of the working classes have often
informed conclusions about their attitudes to work and politics, as if their choices were governed by a logic consistently beyond their control. While it may appear that
culture would enable historians to identify diversity, it has far more often

Page 20
nudged us towards an assumption of social homogeneity. For if we acknowledge the diversity of customs, values and social practices of a given social group, how
could they be reduced to the single common denominator of culture? In view of the variations of local context, and the tensions and antagonisms by which it is
constituted, it is often merely the assumption of the existence of a 'popular culture' which enables historians to pull across the often fundamental differences of
perception, social being and discourse, the unity and homogeneity of a 'popular culture'. In addition, the use of the term has often concealed, or at any rate failed
adequately to address, significant methodological difficulties. For the delineation of 'popular culture', historians have often depended, not surprisingly in the Indian case,
upon the perceptions and characterizations of elite observers, frequently of functionaries, publicists and journalists. Moreover, historians who have employed the term
have sometimes eschewed social and economic explanation, sometimes even inquiry into the material world, as unduly deterministic; yet, they have often inferred the
values and attitudes of social groups from their modes of existence. The notion of a 'precapitalist' or, indeed, a 'bourgeois' culture remains inescapably deterministic.17
Indeed, such inferences overlook the obvious cultural fact that the circumstances are rare in which most people, and especially subordinate groups, can give free
expression to their cultural preferences. The notion of culture was the product of imperial expansion which brought Europeans into contact with exotic peoples. It has
subsequently provided a method of ordering, disciplining and bringing within a rational discourse these perceptions of the exotic. The working classes and the poor,
whether in the 'North' or the 'South', it would appear, have often remained as exotic to the scholar in his attic in either hemisphere as the European conquistadors and
seafarers found the people they encountered and colonized.
The final chapter seeks to bring together the leading themes and arguments of this book. It offers a schematic view of the historical and social formation of the working
classes in India and seeks to situate it within the context of 'world capitalism'. At the same time, it interrogates the concept of 'the capitalist world economy' and the
explanatory framework of dependency theory from the historical perspective of the Indian subcontinent. In so doing, it stresses the argument that the formation and
reproduction of an industrial labour force was integral to the historical processes which shaped diverse forms of labour use in the Indian economy in the nineteenth and
twentieth centuries. It also attempts to situate the history of indigenous capitalism in its global and colonial setting, not
17 Notably, for instance, in the highly charged polemic against determinism offered by Chakrabarty, Rethinking Working Class History.

Page 21
to track its steady 'incorporation' into the capitalist world economy, but rather to recover more fully the subcontinental context and history of capitalist development.
The subordination of indigenous to international and metropolitan capitalists, mediated and directed, as it was, by colonial power, shaped both the patterns of capitalist
and industrial development in India as well as the forms of labour exploitation.
For historians of India, over the past decade, the tidal waves of intellectual fashion (slightly delayed in this case) have threatened to sweep away the study of the
material world and to leave in its wake a bleaker landscape of deconstruction and textual exegesis. Historians have been led on this terrain towards a preoccupation
with a close reading of texts, with the construction of colonial policy, with the often Eurocentric question of colonial discourse which is taken to have influenced and
encompassed Indian representations of themselves and, inevitably perhaps, given the human condition, with gazing into the mirror of historiography. This is a landscape
which has been increasingly denuded of the investigation of the social and material world, and especially of the struggles waged by the subaltern classes.18 Its
preoccupation with colonial discourse, having promised to expose the Eurocentricity of postcolonial scholarship, has often served to entrench it. Those who focus
upon colonial discourse have been led to practise a 'Eurocentricity' of their own. Close attention to how colonial society constructed Indian society has encouraged a
preoccupation with the intellectual foundations of colonialism while the history of Indian society is faded into the background. Overly concerned with how colonial
discourse and its hegemonized agents represented their subjects, and seeking to liberate themselves from this discursive trap, scholars, rather like colonial ideologues,
have increasingly assumed the mantle of representing the native.
Few today would deny the intimate connections which bind knowledge and power in a single nexus. Nor would many scholars be astonished any longer by the
suggestion that knowledge about Indian society and its past was produced in relation to the harsh facts of imperial dominance. The recognition that the truth about the
past did not lie in a finite archival record waiting to be extracted by scholars, able to discover it by the use of more objective and 'scientific' methods, somehow
untrammelled by 'political' bias, may have appeared as a revelation to literary critics and social scientists, but it was unlikely, even before Said or even Foucault, to
surprise historians, who had necessarily been forced to grapple with the
18 This trend is most readily visible in the recent volumes of Subaltern Studies (especially in vols. V–VII), the fruits of a project which dedicated itself to the study of the 'subaltern
classes', to restoring them to the pages of history, and therein retrieving them from the margins and the footnotes to the main text.

Page 22
unremitting awkwardness of archival evidence. The appreciation that knowledge, the definition of its scope and manner of its creation, was imbricated in the
imperatives of its political context was integral to the historian's craft. Even the most positivistically minded historians – and their tribe has been populous –
acknowledged that their findings were contingent and that their interpretations, in offering no more than one version of the past, only served to locate them in relation to
their sources, the historiography in general and its wider intellectual context.
Nevertheless, with the systematic and lucid exposition of Orientalism, historians of India lost their nerve and found a theory. If colonial discourse was constituted by,
and provided an expression of, imperial domination and control, Indian society and its past appeared simply as its product and its consequence. The motor of history
appeared to lie in the hegemonic colonial discourse and its construction of India. Accordingly, historians now concentrated their attention upon the study of colonial
discourse, the processes by which it was constituted, its manifold constructions of India – its people, its social institutions, its body politic and its intellectual and cultural
history – as the object of its knowledge and its power. Textual exegesis cast its lengthening shadow over social being. Human agency seemed to be yet another
delusion fostered by the Enlightenment. Authors, like actors, were necessarily marginalized within, sometimes vaporized from, the historical account. The material
world was merely the final distortion reflected by the hall of mirrors which historical research seemed to represent. Necessarily, then, scholars have tended to examine
the mirrors minutely and ignore the object whose image it refracts.
The effects and implication of this reasoning have, of course, been deeply conservative. If we refuse to acknowledge the materiality of the social world, we could not
possibly change it. Its deficiencies, injustices and oppressions must seem quite as illusory as the social world and the process of its refraction. To acknowledge the
intimate connections between power and knowledge, and their effect in delineating, even defining, the materiality of the social world is not to deny its existence. More
than some of his disciples and imitators, Edward Said allowed for the distinction between the material world and its representation in public discourse. For 'the lives,
histories and customs' of the Orient, he noted, 'have a brute reality obviously greater than anything that can be said about them in the West'.19 The historian's task in
studying the material world is perhaps to create and enter the space between this level of 'brute reality' and the discourse which is generated by, and relates to it,
whether in India or the West.
19 E. Said, Orientalism (New York, 1978), p. 5.

Page 23
By averting their gaze from the material world, while focusing upon its representation and, more generally, its configuration as knowledge, historians have often served
to replicate colonial representations of the material world and the assumptions upon which they were built. Nowhere has this more consistently been the case than in
the investigation of the working classes, or indeed, peasants and the rural poor. For colonial discourse in India, as elsewhere, was largely a discourse about labour.
The British in India were concerned to put Indians to work in the interests of their larger imperial purposes. The colonial project was characterized, primarily, by the
struggle to acquire closer control over labour, to cheapen its costs and to subordinate it more fully to the disciplines of capitalism. In important ways, the Orientalist
construction of India described the displacement of the colonial construction of its labour problem onto the field its discourse defined as culture.
Colonial ideologues and entrepreneurs built their image of labour upon the assumption that Indian society was traditional, static and timeless. To a large extent, what
colonial discourse identified as traditional about Indian society in the nineteenth century was its own creation as it sought to harness local resources to its global imperial
purposes. In this discourse, Indian tradition was constituted by the village community, the caste system and the agrarian character of the economy. The village
community was portrayed as immutable and harmonious, composed of primarily subsistence, yet interdependent, cultivators who understood and observed their
mutual obligations.20 Intimately associated with the village community was the representation of 'the caste system' as an integral aspect of religion rather than as a
mechanism by which rulers could obtain and exercise control over the whole being of the labourer, not simply the fruits of his labour.21 In colonial discourse, the caste
20 H. S. Maine, Village Communities in the East and West (London, 1871); B. H. BadenPowell, The Indian Village Community (London, 1896). On the impact of the
characterization of the village community on land revenue policies and land tenure, see E. Stokes, The English Utilitarians and India (Oxford, 1959). See also, Louis Dumont, 'The
Village Community from Munro to Maine', Contributions to Indian Sociology, 9 (December, 1965), 67–89. On colonial constructions of the village community, see Clive Dewey,
'Images of the Village Community: A Study in AngloIndian Ideology' MAS, 6:3 (1972), 291–328; for a more systematic analysis of its relationship to a wider Orientalist discourse,
see R. Inden, Imagining India (Oxford, 1991).
21 J. Breman, Beyond Patronage and Exploitation: Changing Agrarian Relations in South Gujarat (Delhi, 1993); G. Prakash, Bonded Histories: Genealogies of Labor Servitude in
Colonial India (Cambridge, 1990); N. Bhattacharya, 'Agricultural Labour and Production: Central and SouthEast Punjab, 1870–1940', in K. N. Raj, N. Bhattacharya, S. Guha and S.
Padhi (eds.), Essays on the Commercialization of Indian Agriculture (Delhi, 1985); J. Banaji, 'Capitalist Domination and the Small Peasantry: Deccan Districts in the Late Nineteenth
Century', Economic and Political Weekly, 12: 33–4 (1977), special number, 1375–1404. For an argument in favour of a political interpretation of caste, see N. Dirks, The Hollow Crown:
Ethnohistory of an Indian Little Kingdom (Cambridge, 1987); see also Inden, Imagining India, ch. 2.

Page 24
system, perceived as the defining moral code of Indian society, and the village community, served to explain why the essence of its civilization might cohere and survive
unchanged in the face of political turmoil and revolution. What underpinned the immutable village community and the caste system in colonial discourse was a stagnant
agricultural economy, characterized by smallholding, selfsufficient cultivators, highly resistant to the rationality of the market. Of course, colonial discourse concealed,
at any rate it did not register, its own active role in creating the traditional and stagnant agricultural economy which it was able to identify by the end of the nineteenth
century.22
The colonial discourse about labour came to be shaped by this representation of the traditional character of Indian society. It sought to deny labour's modernity. Within
its terms, the traditional trappings of caste, religion and village community rendered labour immune, or even antithetical, to capitalist rationality. This was, however, a
discourse riddled with contradictions. For instance, it took for granted that, even as the structural stagnation of the agrarian economy made a growing number of
smallholding families dependent upon wages, it would be difficult to loosen labour from the moorings of the village community.23 If labour was
22 E. Stokes, The Peasant and the Raj: Studies in Agrarian Society and Peasant Rebellion in Colonial India (Cambridge, 1978); D. A. Washbrook, 'Law, State and Agrarian
Society in Colonial India', MAS, 15:3 (1981), special issue, Profit, Power and Politics: Essays on Imperialism, Nationalism and Change in Twentieth Century India, edited by C.
Baker, G. Johnson and A. Seal, pp. 649–721; Banaji, 'Capitalist Domination'; N. Bhattacharya, 'Lenders and Debtors: Punjab Countryside, 1880–1940', Studies in History, new series,
1:2 (1985), 305–42; S. Guha, The Agrarian Economy of the Bombay Deccan, 1818–1941 (Delhi, 1986).
23 Hence, the volume of complaints from entrepreneurs and some officials about the scarcity (and high cost) of labour, especially until the first world war. Statements about the scarcity
of labour are to be found extensively in the annual reports of various employers' associations – especially the Bombay Millowners' Association (BMOA) and the Indian Jute Mill
Employers' Association – trade journals, like the Indian Textile Journal, and various public enquiries into the conditions of labour since the late nineteenth century, especially The
Report of the Indian Factory Labour Commission (Simla, 1908), vol. I – but they were disputed by B. Foley, Report on Labour in Bengal (Calcutta, 1906). See also Report of the
Indian Factory Commission, Appointed in September 1890, under the Orders of His Excellency, the GovernorGeneralinCouncil, with Proceedings and Appendices (Calcutta,
1890). They were also expressed in the official histories of the jute and cotton textile industries; see D. R. Wallace, The Romance of Jute (Calcutta, 1909) and S. D. Mehta, The Cotton
Mills of India, 1854–1954 (Bombay, 1954). The evidence on labour supply is closely examined in Morris, The Emergence of an Industrial Labour Force, ch. 4 and in Das Gupta,
'Factory Labour in Eastern India'. See also Simmons, 'Recruiting and Organizing an Industrial Labour Force'. I have examined the implications of labour migration and the nature of the
rural connections of urban workers in Chandavarkar, The Origins of Industrial Capitalism, ch. 4. For a valuable study of the implications of labour migration in Bengal, particularly for
gender relations, see Samita Sen, 'Women Workers in the Bengal Jute Industry, 1890–1940: Migration, Motherhood and Militancy', unpublished Ph.D. thesis, University of Cambridge,
1992, chs. 1 and 2.

Page 25
24
difficult to obtain, colonial capitalists and officials recognized that it would have to be bought out. But to pay them more than skill and aptitude deserved was
incompatible with the need for cheap labour so crucial to the gimcrack production regimes which colonial expansion brought to India and whose development it
stimulated.25 Yet if labour was not easily mobilized, colonial officials and entrepreneurs nonetheless feared that, once it was set free, it would prove restless and
unrestrained, elusive of control and impossible to discipline. Indentured servitude, 'a new system of slavery' developed under the colonial aegis to replace the old after
its abolition in 1834,26 was, like the jobber system, devised to prise labour out of the village community as well as to pin it down to the mines, plantations and
factories.27 But by the 1920s, as employers grew increasingly resentful of the extent to which they had been cut out of their own systems of labour control and
recruitment, colonial discourse, adopting the high moral ground, came to focus upon the abuses and evils of the jobber system.28
It was assumed that workers had to be pinned down in this way because
24 This argument was explicitly made in Mazumdar, 'Labour Supply in Early Industrialization'. For a critique, see Newman, 'Social Factors in the Recruitment of the Bombay
Millhands', and Chandavarkar, The Origins of Industrial Capitalism, pp. 307–26. Employers, of course, frequently complained about high wages while officials sometimes
expressed concern at the intense exploitation of labour, not least because it gave Indian employers an advantage over metropolitan rivals. This latter concern provided an
important motive for passing factory legislation to regulate the hours and conditions of work. For an account of the official discourse, see D. Chakrabarty, 'Conditions for
Knowledge of WorkingClass Conditions: Employers, Government and the Jute Workers of Calcutta' in R. Guha (ed.), Subaltern Studies, vol. II (Delhi, 1983), pp. 259–310.
25 Especially in the extractive and plantation industries of the nineteenth century: indigo, tea and coal. This style of frontier capitalism which emerged in the wake of colonial expansion
conditioned entrepreneurial attitudes, business strategies and patterns of labour deployment and discipline in most other industries in the twentieth century in significant ways. For an
overview of the evolution of this style of frontier capitalism, see C. J. Baker, 'Economic Reorganization and the Slump in South and Southeast Asia', Comparative Studies in Society
and History, 23:3 (1981), 325–49.
26 H. Tinker, A New System of Slavery: The Export of Indian Labour Overseas, 1830–1920 (Oxford, 1974); Brij Lal, 'Kunti's Cry: Indentured Women on Fiji's Plantations', Indian
Economic and Social History Review, 22:2 (1985), 55–71; E. Valentine Daniel, H. Bernstein and T. Brass (eds.), Plantations, Proletarians and Peasants in Colonial Asia (London,
1992); Carter, Servants, Sirdars and Settlers; J. Breman, Taming the Coolie Beast: Plantation Society and the Colonial Order in Southeast Asia (Delhi, 1989); P. Ramasamy,
Plantation Labour, Unions, Capital and the State in Peninsular Malaysia (Kuala Lumpur, 1994).
27 Recruitment, like the question of labour supply, has provided the organizing principle of research in Indian (perhaps much colonial) labour history. In view of its significance in
colonial discourse, this is perhaps not surprising; but it may be taken as yet another indication of the extent to which colonial discourse has been unwittingly replicated in Indian
historiography.
28 Chandavarkar, The Origins of Industrial Capitalism, pp. 99–109, 195–200, 295–307.

Page 26
they would not honour their contracts. Labour's failure to honour its contracts, indeed its inability to understand the very concept, was inextricably related to the fact it
had to be mobilized out of a society which inhabited a previous, not the contemporary bourgeois epoch. From this characterization of Indian labour as premodern
followed the conclusion that it lacked skill, except, that is, skills which were hereditary and genealogical and associated with caste, but which were not always or easily
transposed to the industrial setting. The notion that skill was relative, contingent and often politically defined was not entertained in this discourse about labour. Nor
was the considerable ingenuity and constant improvisation demanded of workers, and frequently observed, allowed to qualify the axiom that Indian workers possessed
low absolute levels of skill.29
Colonial discourse represented workers as primarily 'agriculturists at heart', despite the fact that, as landless labourers, dalits, tribals, artisans and so on, they were
drawn from diverse occupational strata. Their peasant character was invoked to explain what was perceived as their casual attitude to work, their restless search for
casual and temporary employment wherever it may be found and their lack of commitment to the industrial setting.30 It was supposed, nonetheless, that workers were
greedy for cash, motivated by the desire to earn as much as possible as quickly as they could, in order to return to the land. This supposition could scarcely be
reconciled, however, with the notion that they were unable to hold down a job for any length of time and that they moved restlessly in search of casual employment.31
While this portrayal of workers betrayed colonial capitalist anxieties about wage competition and
29 Ironically, this dependence upon skilful improvisation was necessitated by low levels of investment, especially in technological change. Historians have largely neglected the
detailed investigation of the labour process and the social relations of the workplace but this has not prevented them adopting the characterizations of the colonial discourse.
Assumptions both about 'the precapitalist culture' of the Indian labour force and its lack of skill are central to Dipesh Chakrabarty's portrayal of jute mill labour in Bengal; see
Chakrabarty, Rethinking WorkingClass History. For an alternative view of Bengal jute mill labour, see Sen, 'Women Workers'; S. Basu, 'Workers' Politics in Bengal, 1890–1929:
MillTowns, Strikes and Nationalist Agitations', unpublished Ph.D. thesis, University of Cambridge, 1994.
30 See Report of the Indian Factory Labour Commission, vol. I; Report of the RCLI (London, 1931), which both dismissed the view of workers as peasants in proletarian guise and
then espoused it. For a discussion of the representation of migrant labour, see Chandavarkar, The Origins of Industrial Capitalism, ch. 4.
31 The detailed examination of the urban labour market does not bear out the notions that casual labour sought out temporary work wherever it was to be found or that a clear
distinction can be made between the attitudes to work of casual labour and factory proletarians. See Chandavarkar, The Origins of Industrial Capitalism, ch. 3. On Bengal, see Basu,
'Workers' Politics in Bengal'.

Page 27
32
labour mobility, it also had the selfserving virtue of implying that they were paid enough to accumulate and repatriate substantial savings to their village base.
On the other hand, a recurrent theme of colonial discourse, running in an opposite direction, was that workers did not respond to incentives, whether in the form of
increased wages or improved conditions. Higher wages, it was assumed, would only induce workers to withdraw their labour and retire, clutching their savings, into the
traditional recesses of the village community. This argument, arising out of the assumption of the premodern and precapitalist culture of Indian workers, had been
most explicitly set out by Max Weber in his treatise on The Religion of India:
The workers want to earn some money quickly in order to establish themselves independently. An increase in wage rate does not mean for them an incentive for more work or a
higher standard of living, but the reverse. They then take longer holidays because they can afford to do so, or their wives decorate themselves with ornaments. To stay away from
work as one pleases is recognized as a matter of course, and the worker retires with his meagre savings to his home town as soon as possible. 'Discipline' in the European sense is
an unknown idea to him.33
This ignorance of, or aversion to, discipline in 'the European sense' suggested, in this discourse, why labour could not easily be organized either for work or for politics.
Their peasant character, forged by the hierarchies of caste, rendered them tolerant and acquiescent, but it also accounted for their propensity to violence, lightning
strikes and spasmodic protests.34 Employers and managers held fast to the belief that their workers were ordinarily content and had no 'real' grievances. Labour unrest
was inevitably the result of the baneful influence of outside agitators, although why contented workers should be seduced by the blandishments of troublemakers was
never fully explained. If workers lacked the discipline to adapt to the rhythms of industrial capitalism, they could scarcely acquire it swiftly for the purposes of collective
bargaining or trade union organization.35 Indeed, it seemed obvious to conclude that workers who belonged to an older and not necessarily better age and for whom,
therefore, the concept of contract remained alien and mysterious, could scarcely be expected to negotiate collectively, formulate strictly
32 Employers and colonial officials, thus, paid particular attention to the extent to which workers remitted their savings to their villages and historians have similarly made much of
the limited evidence so far available on the issue.
33 Max Weber, The Religion of India, translated by H. H. Gerth and D. Martindale (Glencoe, Ill., 1958), p. 114.
34 See ch. 5 below.
35 These issues are examined in greater detail in ch. 3 below.

Page 28
36
reasonable demands or display the discipline to stick to bargains struck with their employers.
It was precisely because these supposedly preindustrial workers, rural migrants, seemingly halfpeasant and halfproletarian, could not be relied upon to operate
within the domain of reason that they were perceived, in this view, to be quite so threatening. Not only could they be easily worked up into 'a state of excitement' about
wages and working conditions, but their material grievances were readily turned against government. Similarly, it was thought that their resentments and grievances
could find frenzied expression in their caste prejudices and religious fanaticism, always liable to result, or so it seemed, in violence and bloodshed. Moreover, colonial
officials took it for granted that unless these discontents were contained or resolved immediately they could in an instant develop into a massive political conflagration
beyond the ability of the state to control. Since their political action could rarely be traced back to, let alone contained within, the domain of reason, workers
appeared, in this discourse, to inhabit a very narrow spectrum between passivity and fanatical fury. The imperative for immediate and effective action to contain their
threat seemed at times, therefore, to legitimize measures of repression, whose brutality was disproportionate to the threat they sought to contain, just as an acute
awareness of the vulnerability of the colonial state, when faced with mass revolt, and the futility of ruling India by force, sometimes imposed constraints upon its
uninhibited use.
For all the contradictions and confusions which marked the colonial construction of its 'labour problem', the latter's significance for the colonial project was enormous.
Consequently, colonial discourse about labour extended naturally towards, and encompassed, just as it was informed by, its characterization of 'traditional Indian
culture'. Following in the footsteps of Foucault and Said, often with a rather heavy tread, ponderously in quest of theory, historians increasingly concerned with colonial
discourse and its production of Indian society have concentrated singlemindedly upon what it constituted as 'culture'. When they have sought to investigate the history
of Indian society, they have increasingly focused upon colonial discourse to the exclusion of the material world which it represented, and as a result, perhaps, tended to
replicate the axioms and premises of the colonial discourse sometimes as readily as an innocent, even uncritical, reading of the archival record might produce. The
contradictions which proliferate in the colonial discourse suggest not only the difficulties encountered by colonial rulers and the dominant
36 The lack of a notion of contract in the precapitalist culture of the working classes has recently been offered as one reason for the weakness of trade unions by Chakrabarty,
Rethinking Working Class History.

Page 29
classes in maintaining control over labour but also those entailed in grasping and reconstituting the social world which it inhabits. The injunction 'to read the sources
against the grain' has been so often repeated, not least because of its poetic and mystical ring, that it has come to be something of a cliché. To read against the grain the
sources predominantly produced by colonial rulers and the dominant classes is to read the discourse for its contradictions and through its contradictions to interrogate
its construction of the material world. The following chapters traverse the discursive space between the material world and its varied representations, not merely to
register the available evidence about the former, but also to strip it of at least some of the mythologies narrated by the colonial discourse.

Page 30
2—
Industrialization in India before 1947:
Conventional Approaches and Alternative Perspectives
Models of industrialization and social change, whether Marxist or functionalist, have been derived largely from the historical experience of Western Europe and,
especially, of Britain. Social theories came to be constructed upon a specific reading of a particular, and in some respects, unique, historical development. These
theories or models, now deepseated in our historiographical consciousness, increasingly offer yardsticks against which industrial development elsewhere in the world is
measured. On closer examination, universal postulates thus derived have appeared to generate a large number of special cases. Vast expanses of the globe are
seemingly littered with cases of arrested development or examples of frustrated bourgeois revolutions.
Since the study of the rise of industrialism, the central problem of sociology, has hinged on so specific a historical example, it is not surprising that in spite of
fundamental differences of intellectual traditions, conceptual frameworks and political values, diverse models of economic development and industrialization have been
built upon essentially similar assumptions. Of course, these models have often been criticized in their parts, but they have scarcely been rejected as a whole and while
historians may disclaim some of the assumptions upon which they are based, these continue to be pervasive in the analysis of economic development both in the West
and the Third World. This essay sets out these shared assumptions, common to Marxism and functionalism alike, about the character of industrialization as a social
process. Since they have left their imprint firmly upon the investigation of Indian industrialization and economic development, this essay will attempt to set these
assumptions against the approaches and arguments which they have generated in the specific case of India. To do so may not only indicate the inadequacy of the latter,
but may also serve to highlight the limitations which these assumptions have imposed upon empirical research and historical inter

Page 31
pretation. Finally, this essay will offer some alternative perspectives on, and draw attention to, some neglected questions in the history of Indian industrialization.
The historical example of India offers a particularly helpful perspective upon prevailing models of industrialization. For one thing, in conventional terms, India was
substantially on the road to industrialism from the midnineteenth century onwards without perhaps ever having quite reached its destination. This would suggest that
either we should understand industrialization in India (and other similar examples) as a special case requiring a special model of its own or that existing models may be
in need of revision. Furthermore, the history of industrialization in the West is taken primarily to mean the evolution of factory from craft industry, generally
presupposing the prior development of a market economy, the social differentiation of the peasantry and changing legal and social structures. In India, all these forces
were working together at the same time; and if some showed signs of acceleration in the late nineteenth century, their development usually had long historical roots,
predating the imposition of colonial rule. In other words, no simple evolutionary schemata of social change and economic development can be readily applied to the
Indian evidence.
For functionalist writers, industrialization constituted a serial process through which society would duly pass at an appropriate stage of development. The roads
towards a fully industrialized society were many, but the outcome was invariably the same: 'industrialism . . . that which the industrialization process inherently tends to
create'.1 This 'process' was guided by an underlying 'logic',2 by the 'imperatives of industrialization' which 'cause the industrializing élites to overcome certain constraints
and to achieve certain objectives which are the same in all societies undergoing transformation'.3 The signposts of this transformation were characteristically an
advanced level of technology, largescale enterprise and the
1 C. Kerr, F. H. Harbison, J. T. Dunlop and C. A. Myers, Industrialism and Industrial Man: The Problems of Labour and Management in Economic Growth (London, 1962), p. 33;
C. Kerr and A. Seigel, 'The Structuring of the Labour Force in Industrial Society', Industrial and Labour Relations Review, 8:2 (1955), 151–68; B. E. Hoselitz and W. E. Moore
(eds.), Industrialization and Society (Paris, 1963); W. E. Moore and A. S. Feldman (eds.), Labour Commitment and Social Change in Developing Areas (New York, 1960); N. J.
Smelser, Social Change and the Industrial Revolution (London, 1959) and Theory of Collective Behaviour (New York, 1963).
2 Kerr et al., Industrialism and Industrial Man, p. 17.
3 C. Kerr, F. H. Harbison, J. T. Dunlop and C. A. Myers, 'Industrialism and Industrial Man', International Labour Review, 82:3 (1960), p. 238.

Page 32
formation of a consensus of values governed by goals unknown in 'traditional society'.
If the direction of change from traditional society to industrialism was universal, its tendency was too strong to be resisted and its outcome inevitable.4 Industrialism
worked ceaselessly upon a static, passive traditional order. The society and its constituent elements, moulded by the inexorable advance of industrialism, was as clay in
its hands:
Industrialization redesigns and restructures its human raw materials, whatever the source. Thus, the development of an industrial work force necessarily involves the destruction
of old ways of life and the acceptance of the new imperatives of the industrial work community.5
Whereas 'dynamic elites' initiated industrialization, chose its route and shaped its forms, workers, like other traditional groups, were 'in the end malleable'. Indeed, 'man
is everywhere adaptable to the industrial system'.6 Of course, the process of 'his metamorphosis gives rise to many forms of protest' but these reflect his maladjustment
and 'partial commitment' to the industrial setting. With further industrial development, the worker is reconciled to his machine, and 'incipient protest is moderated,
channelled and redirected', so as to 'conform and contribute to the strategy of the industrializing elites'. Labour resistance, in this view, is irrational and atavistic: a
doomed attempt to stem the tide of history. For the most part, however, as workers adapt to the industrial setting, protest inherently tends to disappear, or at least to
be harnessed ultimately to the larger imperatives of industrialism.
Similar assumptions informed the perspectives of a number of labour economists, development specialists and Parsonian sociologists. For many economists and
sociologists, writing in the 1950s, the 'developed' countries revealed to the underdeveloped the image of their own possible future: 'social and political models'
characterized by full employment, universal adult suffrage and the welfare state. Higher per capita incomes were recognized as an essential precondition for the
achievement of these goals. This made economic growth measured in terms of national income highly desirable.7
At the same time, historians of the industrial revolution in Britain set
4 Ibid., 241; Kerr et al., Industrialism and Industrial Man, p. 47.
5 Kerr et al., 'Industrialism and Industrial Man', 246.
6 Ibid., 245–6.
7
Beneath this desire lurked a different rationale. Development in this sense possessed other advantages, apart from higher per capita incomes. It might help to thwart the communist
menace. It would prevent potential markets, raw materials and investment opportunities from being dragged behind the Iron Curtain. It would more positively open up these resources
to the West, particularly if per capita incomes rose. For a brief but perceptive discussion, see D. Seers, 'The Birth, Life and Death of Development Economics', Development and
Change, 10:4 (1979), 707–19.

Page 33
8
out to examine the 'historical experience of economic development'. 'The identification of the advantages in any one' of the advanced industrial economies, it was
supposed, could be 'of value to those other countries which still seem bereft of a good hand of cards'.9 If 'the central fact of economic development is rapid capital
accumulation', as W. A. Lewis declared, its 'central problem' was to investigate how a given society raised its level of saving and investment from '4 or 5 per cent of its
national income or less' to '12 to 15 per cent or more'.10 Ironically, 'this hypothesis was derived from the empirical study of twentiethcentury underdeveloped
countries, and especially from the case of India'.11 If the industrial revolution in Britain was to serve increasingly as a model against which industrial development in
India could be counterposed, the fine carvings of that model had originally been made upon Indian stone. When this emphasis upon capital accumulation, largescale
industry, entrepreneurial initiative and transformative lead sectors established a firm grip on Indian economic history, the circle had begun to go back on itself.12
But there was a deeper sense in which the history of the industrial revolution in Britain came to acquire the status of a universal model.13 This related not simply to
identifying the circumstances of 'takeoff', nor even to the analysis of the social and economic preconditions of industrial development, but more fundamentally to a
conception of the underlying social processes of industrialization. Thus, David Landes, in his classic study of European industrialization, defined an industrial revolution
as:
8 H. J. Habakkuk, 'The Historical Experience of Economic Development', in E. A. G. Robinson (ed.), Problems in Economic Development (London, 1965), pp. 112–38. For a recent
historiographical survey of the industrial revolution in Britain which shows how this approach came into vogue among economic historians in the 1950s, see D. Cannadine, 'The
Present and the Past in the English Industrial Revolution, 1880–1980', Past and Present, 103 (May 1984), 131–71.
9 E. L. Jones (ed.), Agricultural and Economic Growth in England 1650–1815 (London, 1967), p. 2. See especially, W. W. Rostow, The Stages of Economic Growth: A Non
Communist Manifesto (Cambridge, 1960), p. 139 and passim.
10 W. A. Lewis, 'Economic Development with Unlimited Supplies of Labour', The Manchester School of Economic and Social Studies, 22:2 (1954), 155.
11 F. Crouzet, 'An Essay in Historiography', in Crouzet, Capital Formation in the Industrial Revolution (London, 1972), p. 11.
12 A. K. Bagchi, Private Investment in India, 1900–1939 (Cambridge, 1972); R. K. Ray, Industrialization in India: Growth and Conflict in the Private Corporate Sector, 1914–1947
(Delhi, 1979); M. D. Morris, 'The Growth of LargeScale Industry', in D. Kumar (ed.), CEHI, vol. II: 1750–c. 1970 (Cambridge, 1983), pp. 553–676.
13 Of course, the industrialization of the 'latecomers' has sometimes acquired the status of a model in its own right. However, in terms of the social process of industrialization, it has
usually borne a strong resemblance to the original British model. For a useful summary, see R. C. Trebilcock, The Industrialization of the Continental Powers, 1750–1914 (New York,
1981), pp. 1–21.

Page 34
that complex of technological innovations which, by substituting machines for human skill and inanimate power for human and animal force, brings about a shift from handicraft to
manufacture, and, by so doing, gives birth to a modern economy. In this sense, the industrial revolution has already transformed a number of countries, though in unequal degree;
other societies are in the throes of change; the turn of still others is yet to come.
It gives rise to novel forms of industrial organization, characteristically the factory. The factory constituted something 'more than just a larger work unit. It was a system
of production', which over time created a 'new breed of worker'.14 In this light, industrialization is taken to constitute an autonomous force, determined by a neutral
technology, whose 'imperatives' and 'inherent tendencies' appear to lie beyond the realm of social choice or political control, and which acts to shape society inevitably
in a single direction. It is not intended to suggest that there were no differences of approach between the numerous writers in various fields who investigated the
'transition to industrialism'.15 Rather, what needs to be emphasized is that in spite of specific differences, these assumptions about the character of industrialization as a
social process came to be shared by a remarkably wide range of scholars.
These assumptions, often implicit, have influenced the study of industrialization in three ways. First, by postulating a unilinear direction of change, they have trained our
sights upon 'largescale industry', as constituting the lead sector of the economy, and have obscured not only our understanding of other centres of dynamism and
stagnation within it, but also of their role in determining the structure of the economy as a whole. It has enabled, indeed sometimes depended upon, arbitrary definitions
of 'largescale industry' or the socalled formal sector. Above all, it has often led to the identification of the general problem of economic backwardness almost
exclusively with the specific question of industrial failure.
Second, the crude distinction made between preindustrial and industrial societies frequently fails to advance our understanding of the former or even its transition to
the latter. Preindustrial societies are often taken to be predominantly agrarian societies in which largescale industry has not been established. They are thus lumped
together irrespective of their levels of technology, economic activity or social organization. In this perspective, the mainsprings of dynamism and change within pre
indus
14 D. S. Landes, The Unbound Prometheus: Technological Change and Industrial Development in Western Europe from 1750 to the Present (Cambridge, 1969), pp. 1–2.
15 For a recent exposition of different sociological traditions, stemming originally from the work of Marx, Durkheim and Weber, see P. Abrams, Historical Sociology (London, 1983), pp.
19–146.

Page 35
trial society are obscured. Similarly, models of industrializing societies often provide no means of differentiating between them, even as they operate at greatly differing
economic levels. Consequently, they have generated endless false analogies between fundamentally different societies and inspired scores of meaningless computations
of industrial achievement. Thus, in 1914, we are told, India boasted 'the world's largest jute manufacturing industry, the fourth and fifth largest cotton textile industry
(depending on what is being measured), and the third largest railway network'.16 By 1945, India was said to be 'the tenth largest producer of manufactured goods in
the world'.17
Third, these models frequently postulate a simple and direct relationship between the levels of industrialization and the pattern of social response. Thus, while the early
stages of industrialization are characterized by spontaneous and volatile workingclass behaviour, further development, it is often supposed, leads to the formation of
trade unions, the emergence of 'group protest' and the more ready acceptance of formal collective bargaining; or alternatively to the 'secular decline of protest', for
more industrialization means less labour resistance.18 Conversely, prior assumptions about the forms of social action supposedly appropriate to a given stage of
industrial development have shaped and perhaps limited historical analysis of the nature of workingclass politics.
To a remarkable extent, various forms of Marxist analysis have borne a strong resemblance to functionalist sociology. Like the functionalist models, Marxist theory
built upon the English experience of industrialization. Classical Marxism, like later functionalism, appeared to take for granted that 'underdeveloped' or 'backward'
societies operated at some original stage of development.19 Of the various modes of production, capitalism alone was expansive, relentlessly progressive and inherently
generative of further economic development. As new regions, 'even the most barbarian nations', in Marx's Darwinian language, are pulled into the orbit of capitalism,
they are liberated from the deadening impulses of precapitalist modes of production. Classical Marxism, like later func
16 Morris, 'LargeScale Industry', 553.
17 B. R. Tomlinson, The Political Economy of the Raj: The Economics of Decolonization in India (London, 1979), p. 31; see also Ray, Industrialization in India, pp. 14–21.
18 Kerr et al., Industrialism and Industrial Man, pp. 193–223; C. Kerr, 'Changing Social Structures', in Moore and Feldman (eds.), Labour Commitment and Social Change, pp. 348–59.
19 For a recent restatement of the classical Marxist case, see Bill Warren, Imperialism – The Pioneer of Capitalism (London, 1980). See also, A. Brewer, Marxist Theories of
Imperialism: A Critical Survey (London, 1980), especially, pp. 27–127, 286–94.

Page 36
20
tionalism, assumed that capitalism, like industrialism, would duly be achieved in underdeveloped countries.
Of course, the thrust of 'dependency theory' postulating that capitalism systematically prevented the development of productive forces in 'satellite' economies has been
precisely to attack some of these evolutionary assumptions.21 Where optimists projected a model in which expanding trade and the growth of a market economy
would produce cumulative economic development, dependency theorists envisaged the same processes 'enforcing the rise of economic backwardness'. In pursuing
their 'intention to negate the optimistic model of economic advance derived from Adam Smith', as Brenner pointed out, 'they have ended up by erecting an alternative
theory of capitalist development which is, in its central aspects, the mirror image of the ''progressist" thesis they wish to surpass'.22
Within the Marxist tradition, the focus has rested on the progression from precapitalist to capitalist modes of production. Anticipating a single direction of change, their
theoretical expectations have sometimes encouraged historians to examine the past for the first signs of capitalism, the development of the factory system or the dawn
of class consciousness. It has often led to the explanation of facts and events which belie theoretical expectations in terms of the survival of old modes of production,
arcane social forms or traditional ideologies. And it has sometimes allowed Marxist historians and sociologists to accept too readily the determining character of
production relations and it has nudged them, even as they asserted the power of human agency, towards an understanding of the economic sphere,
as an alien force existing outside them [individuals], of the origin and ends of which they are ignorant, which they cannot control, which on the contrary, passes
20 However, Marx himself did not rule out the possibility that under certain circumstances, capitalism might not effectively breach precapitalist modes of production. 'The
obstacles presented by the internal solidity and organization of precapitalist national modes of production to the corrosive influences of commerce', he observed, 'are strikingly
illustrated in the intercourse of the English with India and China'. Karl Marx, Capital, vol. III (New York, 1967), pp. 333–4, cited by R. Brenner, 'The Origins of Capitalist
Development: A Critique of NeoSmithian Marxism', New Left Review, 104 (July–August 1977), 26, fn. 2.
21 P. Baran, The Political Economy of Growth (New York, 1957); A. Frank, Capitalism and Underdevelopment in Latin America: Historical Studies of Chile and Brazil (New York,
1967). For a sympathetic critique which developed these arguments in fresh directions, see Bob Sutcliffe, 'Imperialism and Industrialisation in the Third World', in Roger Owen and Bob
Sutcliffe (eds.), Studies in the Theory of Imperialism (London, 1972), pp. 171–92; see also Brewer, Marxist Theories of Imperialism, pp. 158–81, 286–94. For a more searching critique,
see Brenner, 'Origins of Capitalist Development', 25–94.
22 Brenner, 'Origins of Capitalist Development', 27.

Page 37
through a peculiar series of phases and stages independent of the will and action of men.23
Not surprisingly, Marxist theories of industrialization and class formation have often nestled under the carapace of modernization theory. The linear development of the
capitalist mode of production was to be witnessed in the progressive evolution of the market economy, increasing bourgeois dominance, the expropriation and
pauperization of the peasantry, the growth of a proletariat and, finally, the emergence of capitalist industry. The 'factory system' was the characteristic – in some senses,
the defining – feature of the new social order. It was from the emphasis upon the transition from 'manufacture' to 'modern industry', from the 'formal' to the 'real
subsumption of labour to capital' that the development of the factory system acquired its centrality in the Marxian tradition.24 This process also served to create for
Marxists what Landes was to call 'a new breed of worker'.
The development of capitalist industrialization, it was assumed, tends to increase the homogeneity of the working class. By concentrating workers into larger masses,
the factory system helps to weld the working class into a greater unity. Workers increasingly form trade unions, join political parties and finally effect revolutions. The
burden of labour history, as a result, has been to explain, in a number of cases, why sectionalism persisted within the working classes and why they so often remained
such reluctant revolutionaries. Yet if the assumption that the working class contained within itself an innate propensity for unity was set aside, the problem itself would
disappear.
For Marxists, as for functionalists, the nature of social consciousness was directly related to the level of industrialization. Of course, the fundamental difference lay in
the Marxist expectation that protest, far from registering a tendency to decline, would be raised to new unprecedented levels of intensity and action. But in relation to
the nature of industrialization and the forms of social change, which are assumed to cause this protest, Marxists have often arrived, albeit by a different route, at the
same destination as functionalist social scientists with whom they have long been locked in academic and political dispute.
The broad acres of these shared assumptions may have narrowed the
23 K. Marx and F. Engels, The German Ideology (London, 1965 edn), p. 54, cited by Abrams, Historical Sociology, p. 38. Of course, this notion of the autonomy of the relations of
production was least evident, if not at times contradicted, in their historical writings.
24 Etienne Balibar, 'Basic Concepts of Historical Materialism', in L. Althusser and E. Balibar, Reading Capital (London, 1970). For an attempt to develop these concepts in historical
analysis, see G. Stedman Jones, 'Class Struggle and the Industrial Revolution', New Left Review, 90 (1975), 35–69. See also, G. Stedman Jones, The Languages of Class: Studies in
English Working Class History, 1832–1982 (Cambridge, 1983), pp. 1–24.

Page 38
scope of historical enquiry and limited the development of social theory. This essay will next examine in the specific case of Indian industrialization how these
assumptions were translated into the medium of historical research and attempt to explore their interpretative cost.
II
Just as economic historians of the West have made it their task to explain why France and Germany took several decades to emulate the technological innovations of
Britain,25 the history of Indian industrialization is counterposed to the example of Western Europe. Industrialization, in this conception, was a process of technological
diffusion, inaugurated by the invention of the steam engine and the spinning jenny in Britain which spread incompletely in various ways, and at different times, over the
rest of the world. 'Industrial development in India' could thus be seen as 'part of the very broad movement which had its origins in Western Europe'.26 Yet 'before the
more productive technology of the industrializing West could become something other than a casual and accidental feature of the Indian landscape', it was clear, in this
view, that Indian society would have to begin to resemble Western Europe on the verge of industrialization.27
How ready, then, was India to embark upon industrialization? The question, although a nodal point around which historical research is organized, is characteristically
posed with chronological vagueness. In one view, the necessary preconditions simply did not exist. Capital was scarce and immobile. The quality of labour was poor,
even if its quantity was abundant. Technology was simple, static and backward.28 There was no 'formal structure of public and private facilities that minimizes
uncertainty', no futures markets, no 'insurance devices', no dependable 'flows of statistical and other information'.29 Furthermore, 'the culture' was not 'preoccupied in
any systematic way with the increase of man's control
25 'In view of the enormous economic superiority of these innovations', wrote Landes, 'one would expect the rest to have followed automatically. To understand why it did not . . .
is to understand not only a good part of the history of these countries, but also something of the problem of economic development in general'. Landes, Unbound Prometheus, p.
126. But if we did not 'expect' the rest to follow 'automatically', the question itself would not arise. See also S. Pollard, Peaceful Conquest: The Industrialization of Europe, 1760–
1970 (Oxford, 1981).
27 Ibid.
28 Ibid., 558–63. Morris argues that the backward technology of iron manufacture restricted the industry's output in the eighteenth century, but that the civilization, of course,
generally adjusted to the limited supply of iron by using it very sparingly (ibid., 560–1). But the increased demand did not produce technological innovation. In other words, backward
technology limited the supply of iron, which was therefore used sparingly. But technology remained backward because demand was small enough to be absorbed through adjustments
in the application of labour.
29 Ibid., 555–6.

Page 39
30 31
over his material environment', while 'a scientific approach to technology was virtually nonexistent'. Such propositions have been sanctified by repetition but
detailed scrutiny has more often highlighted their weaknesses than their strengths.32 In contrast, others have argued that the appropriate circumstances for industrial
development did not exist because they were systematically destroyed by colonial rule.33 Some recent research on the eighteenth century has served to suggest that
whatever the direction in which the economy had been developing, it was distorted and diverted into less fruitful channels by the impact of colonial rule.34 Whatever the
disagreement, the method of approach to a counterfactual question has necessarily taken on a counterfactual tone.
But if Indian society in the eighteenth and nineteenth centuries lacked 'not only an extensive array of basic social, political and economic preconditions but also the
development of an institutionalized capacity to solve new problems that continually emerged in the process of change',35 it must be presumed that these conditions
existed somewhere else at the inception of industrialization. It might be salutary to ask where this was so. In his classic account of the development of the British
economy, Eric Hobsbawm wrote:36
the technological problems of the early Industrial Revolution were fairly simple. They required no class of men with specialized scientific qualifications . . . Most of the new
technical inventions and productive establishments could be started on a small scale, and expanded piecemeal by successive addition. That is to say, they required little initial
investment, and their expansion could be financed out of accumulated profits. Industrial development was within the capacities of a multiplicity of small entrepreneurs and skilled
traditional artisans.
The factors which are now identified as the 'preconditions' of economic development have more often turned out to be the consequences rather
30 Ibid., 562–3.
31 Ibid., 562.
32 On the initiatives of traditional merchants, see C. A. Bayly, 'Indian Merchants in a ''Traditional" Setting: Benares, 1780–1830', in C. J. Dewey and A. G. Hopkins (eds.), The Imperial
Impact: Studies in the Economic History of Africa and India (London, 1978), pp. 171–93; on technology, see I. Habib, 'The Technology and Economy of Mughal India', IESHR, 17:1
(1980), 1–34; on the dynamism of the commercial economy in eighteenthcentury South India and its subsequent decline, see D. A. Washbrook, 'Some Notes in Market Relations and
the Development of the Economy in South India, c. 1750–1850', paper presented to the Second AngloDutch Workshop on Comparative Colonial History, Leiden, September 1981.
33 The best statement of this case is to be found in Bagchi, Private Investment; see also, A. K. Bagchi, 'Foreign Capital and Economic Development in India: A Schematic View', in K.
Gough and Hari P. Sharma, Imperialism and Revolution in South Asia (New York and London, 1973), pp. 43–76.
34 F. Perlin, 'ProtoIndustrialization in South Asia', Past and Present, 98 (February 1983), 30–95.
36 E. J. Hobsbawm, Industry and Empire (London, 1968), p. 39.

Page 40
37
than the causes of growth. We know most often that the preconditions existed only because we also know that development had subsequently occurred. In any
case, accounts which counterpose the history of Indian industrialization against the historical experience of eighteenthcentury Britain and Western Europe overlook the
crucial fact that the tasks of industrialization were becoming by the late nineteenth century increasingly complex and expensive.
If industrialization was, indeed, a technologically determined process which spread outwards from late eighteenthcentury Europe, how did this 'diffusion' occur? In the
existing literature, it is generally supposed to have been the consequence of India's relationship with the West largely through the medium of trade and fortified
subsequently by formal colonialism. Technological diffusion occurred, in this view, through the 'great presidency towns' or 'colonial port cities': Bombay and Calcutta,
centres of an expanding import and export trade, of finance and banking, and of course, of consumption. From the 1850s onwards, 'the first substantial manifestations
of modern industrialism' became discernible.38 Once set in motion, the movement along the path to industrialism was apparently only forward. Although 'largescale
industry', presumably mass production and the factory system, did not spread easily across the economy as a whole 'the pace of its extension within specific sectors
was reasonably brisk'.39
As soon as entrepreneurs recognized 'the commercial possibilities of local factory production', and 'the opportunities must have seemed very obvious',40 they
embarked upon industrialization. Once begun, the process continued through functional necessity. As their interests came to rest upon largescale industries, so
entrepreneurs were committed to technological advance and optimal efficiency. Faced with market fluctuations or intense competition, entrepreneurs either altered their
product or else attempted 'to diversify and upgrade the quality of their output'.41 Sometimes, they attempted to suppress wage costs. Sometimes, and with varying
degrees of success, they attempted to take cover in sheltered markets. But, in the long run, the only answer lay in 'technical and administrative reorganization'; from this
'purgation', their industries emerged healthier and more efficient. This linear view of industrialization
37 'There is scarcely one of these preconditions', Habakkuk had observed nearly twenty years ago, 'which cannot be shown to have been absent in the case of some
acknowledged case of growth. Indeed, it is not difficult to cite cases where the absence of what is commonly regarded as a precondition proved to be a positive stimulus to
growth.' Habakkuk, 'Historical Experience', pp. 118–19.
39 Ibid., 553.
40 Ibid., 574.
41 Ibid., 617.

Page 41
42
has led to an almost exclusive concern with a few major industries, and among them a concentration upon its most important centres.
In this perspective, entrepreneurs are perceived as the decisive force in the process of industrialization. Marxists have focused upon their agency in the development of
capitalism and upon the frustration of their economic interests as the catalyst in the emergence of nationalist movements. The emphasis in neoclassical and functionalist
arguments has rested variously, in the case of apologists for empire, seeking to minimize the destructive effects of colonial rule upon the Indian economy, upon the
failure of businessmen to take their chances, or alternatively, in a nationalist argument, upon the daring and brilliance with which they manipulated limited resources to
maximum advantage. For Morris, the 'economy' itself appears to have been constituted by 'actual entrepreneurial choices'. The decisions of private businessmen
determined 'the allocation of resources' in the economy as a whole; they also serve to 'explain' why 'the scope of industrial development [was] restricted'.43 But the
effect of this argument is simply to reformulate the question. For if 'the range of alternative profitmaking opportunities . . . explains the rate at which businessmen
invested',44 what determined those opportunities in the first place?
Yet, despite the central role thus allocated to the entrepreneur, there are, apart from various hagiographical lives, few studies of individual firms or of entrepreneurial
development.45 Historians who place the entrepreneur at the heart of their story of the rise of industry have sometimes neglected to examine how businessmen made
their choices or indeed how far their intentions were translated into achievement.
Existing interpretations of the course of industrial development in India, based upon models of apparently successful industrialization elsewhere, have taken over
pervasive assumptions about the nature of industrialization: that it was a technologically determined process beyond the realm of social choice; that it was a serial
process whose imperatives were similar in each case; that it was inevitably and inexorably progressive; that, flowing from the West, it constituted the only dynamic
force acting upon a passive 'indigenous' economy. The result has been the develop
42 In dealing with the development of the cotton textile industry, for instance, Morris focuses exclusively upon Bombay despite the fact that it constituted 'a diminishing part of a
stillexpanding industry' in the interwar period. Ibid., 617, 572–83, 603–5, 616–24.
43 Ibid., 554.
44 Ibid.
45 However, on the early colonial period, see C. A. Bayly, Rulers, Townsmen and Bazaars: North Indian Society in the Age of British Expansion (Cambridge, 1983), pp. 369–426; and
A. Siddiqi, 'The Business World of Jamsetji Jejeebhoy', IESHR, 19:3 and 4 (1982), 301–24; and for the nineteenth and twentieth century, see T. Timberg, The Marwaris: From Traders
to Industrialists (New Delhi, 1978), and D. Tripathi, The Dynamics of a Tradition: Kasturbhai Lalbhai and His Entrepreneurship (New Delhi, 1981).

Page 42
ment of a Whig historiography of industrialization whose unflinching focus is the socalled lead sector of the economy, and which faithfully chronicles the continuing
discovery and application of increasingly efficient, 'rational' and modern methods of production. In this perspective, the role of supposedly backward sectors in
determining the structure of the economy as a whole is obscured. Moreover, in this view, industrialization is ripped out of its historical context. The influence of the
agrarian economy, the role of the colonial state, the effects of international capitalism, or the impact of labour is often neglected; at best their interplay in the process of
economic development broadly defined receives cursory attention.46
These teleological approaches to industrialization have informed and shaped the wider field of Indian economic history. First, it has helped to create a hiatus in the
historiography of the Indian economy in the early colonial period. Whereas historians of the seventeenth and early eighteenth centuries, like those of the later nineteenth
and twentieth centuries, address themselves to substantive economic issues – agriculture, trade and markets; the growth of towns and industries; workers and
capitalists – the investigation of the early colonial economy is largely evacuated in favour of guiding themes such as the expansion of British power and the development
of land revenue systems.47 The economic history of colonial India is thus generally commenced only from the midnineteenth century onwards. It is at this point that
historians discern the first signs of modern industrialism, bringing in its wake a period of quickening social change. 'To describe the processes at work', explained one
historian, speaking for many, 'one turns to words like ''anglicization", "modernization", "secularization".'48 Now the historian can trek along the familiar path to
industrialism. The 'historical experience of economic development' can now be applied to the study of India's past; the way is cleared for the history of Indian industrial
development to be counterposed to the experience of Western Europe. The past appears increasingly in the image of the present.
46 Ray, Industrialization in India. Ray's synthesizing history of industrialization combines an interesting treatment of the colonial state with an almost total neglect of labour or
indeed the wider economic context. Morris, 'LargeScale Industry', touches upon the role of labour but finds its impact minimal.
47 So much so that one of the most valuable studies of 'agrarian relations' and land revenue systems in the early colonial period was able to disclaim 'a full discussion of prices,
productivity, cropping patterns and the like'. T. R. Metcalf, Land, Landlords and the British Raj: Northern India in the Nineteenth Century (Berkeley and Los Angeles, 1979), p. xii.
For a recent attempt at integration, see Bayly, Rulers, Townsmen and Bazaars.
48 R. Jeffrey, The Decline of Nayar Dominance (Brighton, 1976), p. xiv. To this armoury, we might also add the generic 'Westernization'. It is significant that in these approaches social
change is represented by essentialist cultural descriptions.

Page 43
Furthermore, the concern of economic historians, their attention attracted to when and how 'modernization' came to India, has been to explain why industrialization
failed to transform the Indian economy and its corollary, why the socalled preconditions were missing. The result has been a primarily counterfactual history. Since
the possibility and prospects for economic development are so closely associated with the growth of largescale industry, it is not surprising that the general problem of
the backwardness of the Indian economy and the more specific issue of the slow rate of industrialization not only seem to warrant the same explanation but at times
even appear to be identical. The diffusionist approach has contributed to the identification of industrialization as the focal point not only for assessments of the
prospects for economic development in India but also for explanations of its failure in the colonial period. Similar 'causes' are held to account for both
underdevelopment and the limited extent of industrialization. The result is often to perpetuate an inherent tautology in such reasoning. If economic backwardness,
reflected by the absence of 'preconditions', is held to explain the modest scale of industrialization, economic backwardness itself is explained in terms of the failure of
industrialization. Since at least the late nineteenth century, writers on the subject have invited us to choose between colonial rule and the Indian social structure as the
main cause of economic backwardness. These options have provided the organizing principles for the bulk of the research and writing in the economic history of
colonial India. But a specifically historical analysis of social and economic change in India may be better served by declining this choice and exploring more fully the
interplay between them.
III
The teleological perspective has concentrated attention on 'largescale industry' as the lead sector at the expense of other supposedly backward activities, on the
triumphs of enterprise while neglecting the nature of business failures, on the role of entrepreneurs as catalysts in economic development while assuming that labour and
other social groups were passive factors moulded and shaped by the autonomous imperatives of industrialization. By examining the categories thus used in the
investigation of industrialization, it is possible to allow alternative emphases and perspectives to emerge. It is the intention of this essay to suggest that our focus may
usefully be shifted from largescale industry to the economy as a whole, not as an autonomous entity subject to the mysterious workings of the laws of supply and
demand, but as it was constituted by production conditions, by the relationship between town and country, by the agency

Page 44
of social classes, by the political presence of the colonial state, and by the larger context of the world economy. There is no reason to assume that the outcome of
economic development was predetermined or that it would inevitably lead to the factory and the production line. Rather, at every step, it was the interaction of a
whole constellation of social forces, whose tensions and antagonisms and conflicts were articulated within a context conditioned by the world economy as well as by
the political priorities of the colonial state, that determined the sometimes wayward direction of change. Measured against the ideal types of social theory and Whig
history, industrial development and social change, both in the West and the Third World, have registered a remarkable propensity to go their own way.
If largescale industry has been commonly accepted as the apex of universal models of social change, historians have only identified the phenomenon with considerable
unease. Largescale industries often produced the same products as smallscale workshops. Frequently, they followed similar production strategies, regulating their
levels of output according to market fluctuations rather than to stock. Patterns of labour use, intensive, yet flexible and casual in deployment, were common to both the
formal and the informal sectors. The labour process was often minutely subdivided in small workshops. Factories were sometimes no more than ramshackle sheds
cheaply put together, using simple, often handdriven machinery. Some traders combined the factory system with 'putting out' to workers who worked on the
premises.49 The advance of mechanization often took eccentric routes. Whereas electroplating processes were replacing the wood block in many small workshops in
Bombay by the early twentieth century, the large silk mills followed traditional practices of stamping designs by hand.50 The characteristics of the capital market which
restricted businessmen to small industrial units also served to narrow the freedom of action of millowners and factory magnates.51 Despite the confidence which the firm
commanded among investors all over India, Tata Iron and Steel Company could raise capital for its Greater Extensions programme in the 1920s only at the expense of
granting 'their leading creditor, F. E. Dinshaw a permanent share of their managing agency commission'.52 A similar arrangement in the coal mining industry placed N.
C. Sircar on the road to bankruptcy which led finally to Benares.53
49 Chandavarkar, The Origins of Industrial Capitalism, pp. 76–94.
50 C. G. H. Fawcett, A Monograph on Dyes and Dyeing in the Bombay Presidency (Bombay, 1896), p. 38.
51 For the case of the cotton textile industry in Bombay, see Chandavarkar, The Origins of Industrial Capitalism, pp. 44–71, 279–95.
52 Ray, Industrialization in India, pp. 27–8.
53 C. P. Simmons, 'Indigenous Enterprise in the Indian Coal Mining Industry, 1835–1939', IESHR, 13:2 (1976), 193–5.

Page 45
54
Ultimately, it would seem, the only factor which distinguished largescale industries from smallscale industries was their size. It is not surprising that faced with such
tautology recent historians of industrialization have been unsure of which industries to include in their accounts: for instance, Ray's account of 'private investment in
specific industries' left out jute, perhaps the most important of them all, and Morris's chapter in the Cambridge Economic History of India took no cognizance of
coal mining.55
If the destination of such linear development is uncertain, the route itself on closer scrutiny appears meandering. Factory industry was not the only source of dynamism
in the economy. Seemingly archaic forms of enterprise not only survived but sometimes even expanded over considerable periods of time. This point may be illustrated
with reference to handicraft industries.
Most historians broadly agree that handicraft production declined in nineteenthcentury India but that its decline proceeded unevenly, affecting different regions at
different times and to varying degrees.56 Although historians may dispute how far the process of dissolution went, there is widespread agreement about its general
trend. For some historians, this process of dissolution was inevitable, 'a worldwide development affecting different countries at different times' and 'as integral a part of
the Industrial Revolution as the coming of the factory system'.57 For Morris, too, the products of textile mills, foreign and especially Indian, ruined the handloom
weaving industry, but its death, if inevitable, was slow and protracted.58
In the case of cotton, the broad outlines of the story are familiar. The import of Lancashire yarn and cloth undermined handicraft spinning and, to a lesser extent,
weaving in the early nineteenth century. The major
54 In their recent study of the Bombay labour market, Joshi and Joshi discovered that their 'family of criteria' for distinguishing between the organized and unorganized sector could
not 'easily be used for empirical investigation' and resorted instead to 'the size of the establishment' as the only workable criterion. See H. Joshi and V. Joshi, Surplus Labour and
the City: A Study of Bombay (Delhi, 1976), pp. 46–7.
55 Ray, Industrialization in India; Morris, 'LargeScale Industry'.
56 A. K. Bagchi, 'DeIndustrialization in Gangetic Bihar, 1809–1901', in B. De (ed.), Essays in Honour of Professor S. C. Sarkar (Delhi, 1976), pp. 499–522; A. K. Bagchi, 'De
Industrialization in India in the Nineteenth Century: Some Theoretical Implications', Journal of Development Studies, 12:2 (1976), 135–64; Morris, 'LargeScale Industry', 668–76; M. J.
Twomey, 'Employment in Nineteenth Century Indian Textiles', Explorations in Economic History, 20:1 (1983), 37–57; G. Pandey, 'Economic Dislocation in NineteenthCentury Eastern
Uttar Pradesh: Some Implications of the Decline of Artisanal Industry in Colonial India', in P. Robb (ed.), Rural South Asia: Linkages, Change and Development (London, 1983), pp.
89–129; M. Vicziany, 'The DeIndustrialization of India in the Nineteenth Century: A Methodological Critique of Amiya Kumar Bagchi', IESHR, 16:2 (1979), 105–46; A. K. Bagchi, 'A
Reply', IESHR, 16:2 (1979), 147–61.
57 D. Thorner and A. Thorner, Land and Labour in India (Bombay, 1962), p. 70.
58 Morris, 'LargeScale Industry', 670–1.

Page 46
centres of the traditional industry in Bengal and the Gangetic Valley were the worst affected. But by producing durable coarse cloth for local markets and to a lesser
extent fine goods and fancy designs which ordinary power looms could not handle, parts of the handloomweaving industry survived until the late nineteenth century.
Then the expansion of the Indian cotton textile industry finally spelt its doom.
Baker's study of the handloomweaving industry in Tamil Nadu shows how far its development ran counter to the apparent allIndian trend.59 The handloomweaving
industry in Tamil Nadu began to expand in the late nineteenth century, precisely when it should have gone into decline in the face of growing competition from Indian
mills. In fact, the handloom sector grew faster than factory industry and by 1939 supplied 60 per cent of the market for cloth in the Madras Presidency. This expansion
was based in part on the home market but exports to SouthEast Asia and the Persian Gulf quadrupled between the late 1880s and 1920s. Production was organized
through various forms of putting out. While many families looked upon the manufacture of coarse cloth 'as a form of social security', they required working capital and
the problems of marketing cloth were sufficiently complex to require specialized knowledge. Precisely because many families turned to handloomweaving when the
local economy was depressed, there was a continuous risk of overproduction and the demand for and supply of both labour and cloth had to be carefully coordinated.
Moreover, little value was added in the process of production so that those who were stranded with stocks when the price of yarn was falling might easily be ruined.
Between the wars, the internal economy of Tamil Nadu, like other regions, was subject to violent fluctuations of prices and demand arising largely from its relationship
with the international economy. The cost of production for handloom workers, as well as the price of cloth, tied to unstable agricultural prices, were highly vulnerable
to international fluctuations. The effects of these conditions were registered in a growing emphasis among handloom capitalists upon coarse goods, an increasing
concentration of control over working capital in the industry, the development of 'larger and more loosely organized production systems' or the emergence of 'large
puttingout networks', and a shift in the location of the industry towards the towns.
It is not clear how far masterweavers or 'independent' weavers were involved in the marketing of their own goods. But they faced obvious disadvantages in dealing
with such unstable markets. Large merchants,
59 C. J. Baker, An Indian Rural Economy, 1880–1955, pp. 393–413. On the growth of weaving workshops in the Deccan, see N. M. Joshi, Urban Handicrafts of the Bombay
Deccan (Poona, 1936).

Page 47
on the other hand, could simply by cutting off supplies of working capital to their weavers reduce the output of cloth. Since they operated wider marketing and
production networks, they were also better placed to identify and to respond to changes in price or demand. Yet these tendencies towards concentration did not mark
a shift towards the factory system. On the contrary, market fluctuations made capitalists even more reluctant 'to commit themselves to regular working and expenditure
on plant'.60
This instability did, however, encourage the handloom capitalists to diversify their interests, buying land, trading in millmade piecegoods, investing in the film industry.
The decline of the export trade in fine goods had undermined the master weavers, engaged in luxury production, and circumscribed the 'independence' of the skilled
'independent' weavers from the traditional weaving communities. Many of them turned to the coarse goods trade, inflated its labour supply and probably accentuated
its instability. As the number of weavers producing coarse cloth increased, their incomes fell and the extent of regular employment available declined. Many were
forced to seek work elsewhere, turning to agricultural labour or service employment in the towns. Consequently, 'the industry was becoming more parttime than
permanent'. By the 1940s it was already being deserted, and this rather than factory competition alone may account for its decline after 1948.61 But there were other
underlying pressures: especially the interplay between the international and the internal economy, and the political priorities of the colonial state which prevented it from
mediating the impact of international fluctuations upon the internal economy and which in fact led to the adoption of fiscal, monetary, tariff and financial policies which
aggravated their effects.62
The case of handloomweaving in Tamil Nadu suggests that nonfactory forms of production organization were capable not only of adaptation and survival in the face
of factory competition, but also of dynamism, expansion and technical and organizational innovation. Nor can this dynamism be viewed simply as the preliminary stage
to the development of the factory. And it is by no means the only such example. In jute, too, 'the handicraft sector expanded rather substantially between the late
1830s and about 1880'.63 Similarly, for much of the colonial period, Indian capital in the coal mining industry was 'typically confined to
60 Baker, An Indian Rural Economy, p. 404.
61 Ibid., pp. 402–13.
62 Ibid., ch. 2; Tomlinson, Political Economy, ch. 2.
63 Morris, 'LargeScale Industry', 567. For the case of sugar, see S. Amin, Sugarcane and Sugar in Gorakhpur: An Inquiry into Peasant Production for Capitalist Enterprise in
Colonial India (Delhi, 1984).

Page 48
64
smallscale individual or family proprietorships mining second class . . . coal from very shallow depths'. Nevertheless, operating on this basis, several firms survived
with remarkable success over surprisingly long periods.65 The factory system and largescale production were not the invariable outcome of industrial development.
If by focusing upon largescale industry, teleological perspectives have tended to overlook handicraft production and apparently archaic forms of enterprise, they have
also concentrated attention on what are readily perceived as the successes of industrialization and this has led to the neglect of the failures. It is almost customary for
historians to preface their accounts of industrial development with passing reference to the first attempts which failed, before hurrying on to chronicle the triumphs of
those which succeeded. It may appear perverse to suggest that the history of industrialization may be usefully approached through the ventures that failed. Yet business
failures frequently occurred as a consequence of the very constraints which also inhibited the firms that succeeded and which established the limits within which all were
forced to work.
The first jute mill was established in 1855 by George Acland, formerly a coffee planter in Ceylon. Despite his 'considerable experience in South Asia', his 'sense of
foreign markets' and his supposedly special knowledge of 'how to organize Indian labour', 'the firm was never very successful'.66 The other four mills established by
1866, however, were 'supposed to have been very profitable'. But this did not encourage a rush of entries and the industry began to expand significantly only in the
later 1870s.67 In coal mining, too, triumphant Indian entrepreneurs, like business failures, were similarly, though not perhaps equally, starved of capital. In the coal
mining industry, capital was not so much progressively 'Indianized' as 'Europeanized'. Coal mining was initiated by Dwarkanath Tagore in the 1830s but the collapse of
his firm in 1847 marked the beginning of European ascendancy in the industry, and in the decades which followed, Indian entrepreneurs were increasingly squeezed
into the least profitable sectors of the industry.68
Early attempts by the East India Company and by European investors to launch iron and steel production failed largely because they were
64 Simmons, 'Indigenous Enterprise', 200, 189–217.
65 In Japan, handlooms locked the cotton mills out of the domestic market until the 1920s and 1930s. For a general survey of the literature on Japanese industrialization which places it
in a comparative context, see B. R. Tomlinson, 'Writing History Sideways: Lessons for Indian Economic Historians from Meiji Japan', MAS, 19:3 (1985), 669–98.
67 Ibid.
68 Simmons, 'Indigenous Enterprise', 200, 189–217. B. B. Kling, Partner in Empire: Dwarkanath Tagore and the Age of Enterprise in Eastern India (Calcutta, 1981), pp. 73–121.

Page 49
undercapitalized and because they could not effectively adapt technology to existing factor costs and to the size of the market. But the same problems inhibited
subsequent attempts. The Bengal Iron Works Company collapsed through undercapitalization in 1879. The failure of the steel project initiated in 1906 by its
reincarnation, the Bengal Iron and Steel Company, was also in part due to the fact that it was 'grossly undercapitalized'. Even the redoubtable Tata family encountered
serious difficulties in raising capital.69 Their survival has commonly been put down to the business acumen and entrepreneurial skill of J. N. Tata and his descendants.
But other factors also contributed to their success. In particular, the inroads made by Belgian steel into the Indian market as well as the lessons of the first world war –
that it was as well to add an ordnance base to the oriental barrack – increased the readiness of the colonial state, whatever the modesty of its ambitions or the limited
efficacy of its policies, to assist the enterprise.70
The first attempts at establishing cotton mills in Bombay failed because the entrepreneurs had 'difficulty mobilizing capital'.71 Yet maintaining the supply of fixed as well
as working capital remained a perennial problem for the Bombay millowners throughout the history of the industry. Millowners with established reputations, who had
gained the confidence of the public investor and who rarely had serious difficulties in raising capital, were liable to be hoist with their own petard. Their ease of access
to funds encouraged a tendency towards overextension and resulted in some spectacular failures: Dwarkadas Dharamsey in 1909; Greaves Cotton and Company,
which managed the largest group of mills in the city, in 1915; Narottam Morarji in 1929; Currimbhoy Ebrahim, who owned roughly 15 per cent of the industry's
capacity, in 1933. The high cost of fixed capital as well as their dependence upon the changeable money market for their working costs led the millowners who
survived to adopt flexible production strategies, geared to the maintenance of their turnover, averse to holding stocks, deploying sizeable proportions of casual labour
and regulating production according to demand.72
Among the various constraints within which entrepreneurs operated, and which led to the collapse of some, the lack of capital was prominent.
70 Ray, Industrialization, pp. 74–93; Bagchi, Private Investment, pp. 291–331; D. M. Wagle, 'Imperial Preference and the Indian Steel Industry', Economic History Review, 34:1 (1981),
120–31.
71 Morris, 'LargeScale Industry', 574. But Morris suggests a few lines later that the first mills 'were not exceptionally costly ventures by local standards' and fairly easy to establish.
Ibid., 574–5.
72 These arguments are further elaborated in Chandavarkar, The Origins of Industrial Capitalism, pp. 67–71.

Page 50
Yet this did not signify an absolute scarcity of supply. It was rather the case that capital was most easily raised in small pools, by entrepreneurs whose fame and fortune
were already legendary, for whose enterprises the markets were already proven and the risks well known. This is not to return by a different route to old notions about
the shyness of capital. Attention to the nature of business failures should alert us, on the contrary, to the highly adventurous spirit of much investment, and serve to
emphasize the magnitude of risk which it entailed.
The evidence is increasingly conclusive that whereas most early Indian ventures failed, European initiatives succeeded: in the coal mines,73 the jute mills74 and the cotton
textile industry of Madras.75 This was primarily because European entrepreneurs enjoyed easier access to and, therefore, a greater command of capital while their
connections with banks and managing agencies facilitated its mobilization. On the other hand, their Indian counterparts had to rely primarily upon their kinsmen, caste
fellows and acquaintances. Yet by deploying their own connections of kin, caste and friendship, European entrepreneurs were able, under certain conditions, critically
to disadvantage their Indian rivals.76 A closer examination of business failures also provides a context within which the nature and quality of entrepreneurship may be
assessed. If the causes of failures also dogged the survivors in their moments of triumph, there is no reason to identify the collapse of firms necessarily with inferior
entrepreneurial skill. Rather, failures served as a reminder that entrepreneurs did not inevitably overcome the difficulties they faced.
In this perspective, it is possible to assess entrepreneurial responses as well as to establish how they might be compared. Historians have continued to address this
problem primarily in terms of community. Morris, for instance, registers 'the different responses by various Indian groups and of natives in contrast to foreigners'.77
Reluctance to break into largescale industry in a particular instance suggests to him the timidity of capital or the lack of enterprise among local business communities.
Since he finds no 'obvious barriers' to entry by Indians to the jute industry, their absence signifies to Morris 'the passivity of Indian capital'. He concludes that Bengali
businessmen were primarily 'small investors and rentiers' rather than 'aggressive entrepreneurs'.78 By contrast, 'the aggressive and
73 Simmons, 'Indigenous Enterprise'.
74 Bagchi, Private Investment, pp. 157–217, 262–90.
75 Baker, An Indian Rural Economy, pp. 339–42.
76 Bagchi, Private Investment, pp. 165–70. See also, O. Goswami, 'Collaboration and Conflict: European and Indian Capitalists and the Jute Economy of Bengal, 1919–1939'. IESHR, 19:2
(1982), 141–79; B. R. Tomlinson, 'Colonial Firms and the Decline of Colonialism in Eastern India, 1914–1947', MAS, 15:3 (1981), 455–86.
77 Morris 'LargeScale Industry', 557.
78 Ibid., 568–70.

Page 51
79
successful merchants' of Bombay were quick to seize their chances. If 'it cost no more and probably somewhat less to set up a jute mill than to open a cotton mill in
Bombay at the same time . . . why was native capital [in Calcutta] so timid?'80 But if the Bengali entrepreneur was backward in jute, how should we explain his
initiatives in coal, under increasingly adverse conditions?81 And in any case, what are we to make of the early failures to promote iron and steel making? For over one
hundred years, efforts by 'Europeans responding to the needs of Europeans', with the help of a modest quotient of official support and encouragement, ended in
catastrophe. The question of whether European capital, apparently the most dynamic agency in the Indian economy, was timid has scarcely arisen in this line of
argument. Yet in the case of iron and steel one (presumably, crucial) cause of failure was that 'even successful charcoal iron operation required a shift to much larger
scale techniques than were ever contemplated'.82
Morris's own evidence suggests that the problem of entrepreneurial response posed in terms of community is not very helpful. For conversely, the successful
penetration of Marwari entrepreneurs into the jute industry in the 1920s and 1930s, like Parsi enterprise in the founding of the Bombay cotton textile industry, is
attributed by Morris not to any innate business acumen but to their firmly established base as traders which enabled them to invest effectively in industry.83 On such
evidence, differences in entrepreneurial responses, it would seem, are better explained in terms of given economic contexts rather than what are perceived to be the
immutable characteristics of particular business communities. Indeed, these communities themselves are often arbitrarily defined, referring interchangeably to different
categories of identity – to race ('European', 'native'), religion (Parsi, Jain), region (Bengali, Marwari, Gujarati) and caste (bania). Any single capitalist from any one of
these 'communal' groups might also be defined in terms of other particular attributes. To approach the problem of differential entrepreneurial response in terms of
community is to assume that these diverse kinds of communities possessed an internal coherence, and it has usually led no further than the discovery that entrepreneurs
of any single community, however defined, often had as much or as little in common with each other as they had with the members of any other group. A more
promising way forward lies in examining the pattern of entrepreneurial response as a whole, for by establishing what was general to businessmen, operating at different
79 Ibid., 574.
80 Ibid., 570.
81 Simmons, 'Indigenous Enterprise'.
82 Ibid., 585. How far does the notion of 'timidity' explain why European capital, having come to dominate exportoriented industries, remained reluctant to explore the possibilities of
the domestic market?
83 Ibid., 580–1, 615–16.

Page 52
levels, we might be better placed to identify what was specific to any particular community: how religion or caste, race or region influenced their decisions and shaped
their actions. This approach may have a general bearing upon the study of the interactions between different communities in their various and changing forms.
IV
Even if we were to view the process of industrialization through the prism of entrepreneurial choice, an investigation of the conditions under which businessmen made
their decisions, what they perceived to be their options and why they made certain choices rather than others at particular moments of time, may well reveal that the
direction of change, far from being unilinear or uniformly progressive, was determined by ceaseless improvisations, which, in turn, pulled the economy several ways at
once. It was not the case that Indian businessmen, having recognized the commercial possibilities of factory production, relentlessly adopted 'the more productive
technology of the industrializing West'.84 The development of industry can be viewed differently. Whatever the difficulties of raising funds in a highly imperfect capital
market, entrepreneurs who become known and whose firms appeared to prosper swiftly attracted eager investors.85 This was the moment for most businessmen to
diversify. Factory production was one possible outcome of this process of diversification. But far from being either a necessary or an inevitable consequence, it was in
most circumstances an unlikely one. There is little reason either to suppose that its inherent tendency pointed in the direction of continuous technological advance or to
assume that it would reach its culmination in the production line. This process of diversification into factory production and the strategies of entrepreneurs for further
industrial development and technological advance can be illuminated by looking more closely at the case of the cotton textile industry.
Entrepreneurial strategies which apparently led to the establishment of factory industry were not always consistently expansive. The shift from trade to industry has
sometimes been portrayed as an obvious, indeed logical progression. Exposed to the agencies of the colonial trade and international capitalism, it appeared to be
merely a matter of time before cotton merchants, involved in the export of raw materials and the import of manufactured goods, would adopt modern techniques,
import machinery and produce yarn and cloth on their own account.86 It is more

Page 53
likely, however, that investment in industry was the response of some merchants to and, in general, a reflection of, their subordination in the export trade in raw cotton.
As Vicziany has shown, a series of innovations in the organization of the cotton trade in the 1860s reduced the previously generous margin of risk in the export trade,
but because their use presupposed a considerable command of capital they accentuated the differentiation between the largescale exporters, in whose hands business
came to be increasingly concentrated, and the merchants who operated on a small scale.87 These innovations combined with the growing importance of Liverpool and
European markets for Indian raw cotton to work strongly in favour of the large European agency houses. Indian merchants subordinated in the export trade began to
invest their capital in spinning mills. By investing in industry, the large cotton traders gained considerable flexibility in their operations. They could buy raw cotton when
prices were low and sell when prices rose. If the cotton trade remained depressed, they could switch stocks intended for export to the manufacture of yarn. Rather
than build warehouses to store their cotton and await the next failure of the American crop or an increased demand among Chinese spinners, they could deploy their
stock of cotton according to its optimal shortterm use, adjust their operations to the uncertainties of the market and adapt to the severe competition in both trade and
manufacture.
This relationship between the trade and industry which entailed investment in the latter to hedge the uncertainties of the former, was to become an enduring
characteristic within the industry. Cotton brokers and piecegoods dealers, indeed entrepreneurs of all kinds, seeking to diversify their commercial interests, provided
the thrust behind the expansion of the number of entrepreneurs in the industry. Subsequently, spinning mills periodically left stranded by the fluctuations of the yarn
market diversified into weaving to consume the surplus yarn which they could not sell.88 Even the 'aggressive and successful merchants'89 of Bombay were not
motivated by the urge to embark upon a process of cumulative technological development, but rather by an anxiety to hedge their bets, to diversify and survive. The
apparently revolutionary process of industrialization was thus launched with limited liability.
Entrepreneurial initiatives in the Ahmedabad cotton textile industry
87 M. Vicziany, 'Bombay Merchants and Structural Changes in the Export Community, 1850 to 1880', in K. N. Chaudhuri and C. J. Dewey (eds.), Economy and Society: Essays in
Indian Economic and Social History (Delhi, 1979), pp. 163–96.
88 Chandavarkar, The Origins of Industrial Capitalism, pp. 61–71, 244–51.

Page 54
were not nurtured in foreign trade, shaped by European influence or inspired by the English example. Ahmedabad had been an important commercial and
manufacturing centre since the sixteenth century.90 The town's longdeveloping commercial institutions, its merchant mahajans and sarafi networks, provided the basis
for industrial development.91 The severance of their role as bankers to precolonial states and armies, the growing competition of the Bombay firms in the raw cotton
trade, the decline of the opium trade, the advent and competition of the English exchange banks were by the midnineteenth century combining to limit the opportunities
for local traders and bankers. Their response was to seek a share of the profits in textile production. By investing in spinning yarn for the sizeable local handloom
industry, local merchants were simply diversifying their interests and many continued to regard their trading and banking activities as the most vital to the business
concerns of their family. Moreover, the development and structure of the textile industry continued to be intimately connected with the structure of indigenous banking
and commercial mahajans in the town. Some of the first mills were built by the most prominent and prosperous merchants. Not only were millmanaging agencies and
sarafi firms sometimes owned by the same family, but there was also considerable interinvestment between them. Mills invested their surplus funds with shroffs, while
in return for their loans, the latter were often given a share in the managing agency commission. Shroffs organized and often supplied the shortterm deposits upon
which the Ahmedabad mills were dependent for the bulk of their fixed capital.92
These financial practices, which government agencies, academic experts and tariff boards found rather primly 'to be open to grave objection',93 also gave the
Ahmedabad mills a number of advantages. Capital could be mobilized by the Ahmedabad mills at some of the lowest interest
90 D. Tripathi and M. J. Mehta, 'The Nagarsheth of Ahmedabad: The History of an Urban Institution in a Gujarati City', Proceedings of the Indian History Congress (1978?), pp.
481–96; K. Gillion, Ahmedabad: A Study in Indian Urban History (Berkeley and Los Angeles, 1968), pp. 11–36.
91 Gillion, Ahmedabad, pp. 37–104; Tripathi, The Dynamics of a Tradition; M. J. Mehta, The Ahmedabad Cotton Textile Industry: Genesis and Growth (Ahmedabad, 1982). For a
critical appraisal of the books by Tripathi and Mehta, and some valuable comments on the Ahmedabad case, see Rajat Kanta Ray, 'Pedhis and Mills: The Historical Integration of the
Formal and Informal Sectors in the Economy of Ahmedabad', IESHR, 19:3 and 4 (1982), 387–96.
92 In the early 1930s, it was estimated that 39 per cent of the total capital of the Ahmedabad mills was drawn from public deposits. Report of the Indian Central Banking Enquiry
Committee, Majority Report (Calcutta, 1931), vol. I, part i, p. 278; Report of the ITB, 1932 (Calcutta, 1932), pp. 82–4.
93 Report of the ITB, vol. I: Report (Calcutta, 1927), p. 90.

Page 55
94
rates in the country. Whereas most mills in India had to borrow a significant proportion of their working capital from indigenous bankers and the money market, their
special connection with the shroffs often served to protect the Ahmedabad millowners from the vagaries of a volatile capital market. By comparison, the practices of
the jointstock banks remained far too rigid, conservative and inflexible adequately to serve the needs of industry.95 Although, during the depression, capital was more
reluctantly mobilized, and interest rates rose in Ahmedabad in the early 1930s, the industry did not suffer the massive withdrawal of public deposits experienced by
Bombay.96 Thus in 1932, the Tariff Board had to recognize that 'in practice, the Ahmedabad mills have seldom experienced any difficulty during the past 30 or 40
years owing to the bulk of their fixed capital being composed of deposits' and was, therefore, led to admit that, 'it seems to have effectively stood the test of time'.97
Moreover, the Ahmedabad shroffs did not merely act as bankers to the millowners, but frequently had interests in the purchase of raw cotton and the selling of
piecegoods. Piecegoods dealers in the domestic market often encountered problems mobilizing funds, in contrast to the importers of foreign manufactures who were
more freely financed by the exchange banks. It is probable that the interconnections between indigenous bankers and the piecegoods dealers relieved some of these
difficulties and they help to explain in part why the Ahmedabad mills were more oriented towards the domestic market for piecegoods, able to establish a firmer grip
upon it in the 1910s and 1920s, and to remain more alert to the specific patterns of demand within it. More immune to the progressive influences of the West than their
rivals in Bombay, the Ahmedabad millowners, whatever their arcane styles of management or traditional connections, were also considered more progressive and
innovative in the management of their industry: they produced finer varieties of goods, competing with imports and of a quality comparable with the best of Bombay's
efforts; they exploited more skilfully the possibilities of the domestic market; managed their mills more closely and more prudently; and were more ready to renew their
plant and replace old machinery.98
The development of the cotton textile industry provides a useful vantage point from which to observe the options which entrepreneurs perceived and the choices which
they made. Cotton textiles constituted
94 Gazetteer of the Bombay Presidency, vol. IV: Ahmedabad (Bombay, 1879), p. 68.
95 Report of the ITB, 1927, vol. I, pp. 90–2.
96 S. D. Mehta, The Cotton Mills of India, 1854–1954 (Bombay, 1954), p. 178.
97 Report of the ITB, 1932, p. 84.
98 Several official reports and contemporary observers commented on the more progressive and efficient management of the Ahmedabad mills, but it was most extensively treated in
the Report of the ITB, 1927, vol. I, passim.

Page 56
India's most important industry and in conventional terms formed the lead sector of the economy. Superficially, at least, it appeared to illustrate perfectly the thesis that
industrialization was an autonomous process of technological diffusion. In the nineteenth century, Indian cotton mills 'confronted the most important, the most
internationally aggressive and politically most powerful industry in Britain'.99 In the face of this formidable competition, the Indian textile industry expanded. The
Bombay mills, equipped by Platt's machines, built to Lancashire's designs, staffed by British technicians, were able to compete effectively with the metropolitan industry
in particular markets. Subsequently, as these techniques were diffused throughout the subcontinent, upcountry centres of production challenged Bombay's dominance
in the domestic market. As the technological innovations of the industrial revolution spread across the globe, India's turn it would appear had arrived. On a closer
examination, however, a different picture of the development of the cotton textile industry begins to emerge. It was not so much the continuous application of new
techniques which characterized the growth of the industry in India as much as relentless improvisation in the use of old machinery, the manipulation of raw materials and
the exploitation of cheap labour.
Three distinct phases of expansion characterized the development of the cotton textile industry. The first spurt of growth, centred in Bombay, in the 1870s and 1880s,
was based upon the penetration of the Chinese market at the expense of British industry.100 Capital, in retreat from the adverse competitive conditions in the export
trade in raw cotton, now sought to exploit Bombay's comparative advantage in labour costs, its access to abundant sources of shortstapled cotton, and its relative
proximity to the market, in order to produce lowcount yarn for handloomweavers in China more cheaply than Lancashire.101
The later nineteenth and early twentieth centuries witnessed steady, if gradual growth and the dispersal of the industry, diluting its early concentration at Bombay. In
Ahmedabad, capital diverted from the lucrative trading and banking activities of the late eighteenth and early nineteenth centuries was gradually attracted to the spinning
mills to supply the needs of the town's sizeable handloom industry, especially for the lower and medium counts of yarn.102 Capital stepped forward with caution: the
100 Bagchi, Private Investment, pp. 229–37; Mehta, The Cotton Mills of India, pp. 40–63; S. M. Rutnagur (ed.), Bombay Industries: Cotton Textiles (Bombay, 1927).
101 Manchester Chamber of Commerce, Bombay and Lancashire Cotton Spinning Enquiry: Minutes and Evidence (Manchester, 1888); Bagchi, Private Investment, pp. 229–37;
Morris, 'LargeScale Industry', 572–83; Mehta, Cotton Mills, pp. 40–63. D. A. Farnie, The English Cotton Industry and the World Market, 1815–96 (Oxford, 1979), pp. 81–134.
102 Gillion, Ahmedabad, pp. 74–104; Mehta, The Ahmedabad Cotton Textile Industry. In

Page 57
mills in Ahmedabad remained much smaller than those in Bombay and their growth was much slower. Although the first mill was built in 1861 by Ranchodlal Chotalal,
who had lost his job in the government service for allegedly accepting a generous bribe,103 there were by the early 1890s only nine mills working in Ahmedabad. Then,
between 1896 and 1899, in a period of famine, agrarian crisis and considerable difficulty for their Bombay counterparts, the number of mills in Ahmedabad doubled.104
In part, this was because mills which were projected in more prosperous times had been able to commence operations only after the downturn. But their survival in the
late 1890s and early 1900s suggests also that the demand from the local handloom industry was far from saturated105 and, further, that the scarcities of the late
nineteenth century and the late 1890s in particular had increased income disparities in the countryside, thus shoring up the purchasing power of some consumers.106
Between 1906 and 1911, there was a further spurt of growth in the industry when in response to economic recovery and the opportunities provided by the Swadeshi
movement and the boycott of foreign cloth, the weaving capacity of the Ahmedabad mills was doubled.107 This period also witnessed a dramatic expansion of looms in
Bombay, reflecting the efforts of its millowners to adapt to the increasingly effective competition of the Japanese mills and of local spinners in the Chinese market.108
The output of the new looms was biased towards coarse and medium varieties of piecegoods which were less exposed to the competition of Lancashire's imports.
The dispersal of the industry reached far beyond Ahmedabad. The new mills mainly produced lowcount yarn and in a few cases coarse cloth as well, intended largely
for local consumption, using cheap labour and local supplies of raw cotton.109 These factors prompted millbuilding in Sholapur, an important centre of the handloom
industry, situated in proximity to the cotton tracts of the Deccan. Similarly, in Kanpur, investment in industry was a response to the demand for cheap lowcount mill
made yarn among local handloomweavers and was further stimu
1900, there were 10,000 weavers operating in Ahmedabad and as late as 1914 there were said to be 1,000 looms in the town, see Gillion, Ahmedabad, pp. 47–9.
103 S. M. Edwardes, Memoir of Rao Bahadur Ranchodlal Chotalal (Exeter, 1920); Gillion, Ahmedabad, pp. 81–5.
104 Gillion, Ahmedabad, p. 88, fn. 16.
105 Tripathi, Dynamics of a Tradition, pp. 45–50; Gillion, Ahmedabad, pp. 46–50.
106 N. Charlesworth, 'Rich Peasants and Poor Peasants in late Nineteenth Century Maharashtra', in C. J. Dewey and A. G. Hopkins (eds.), The Imperial Impact: Studies in the Economic
History of Africa and India (London, 1978), pp. 97–113; C. N. Bates, 'The Nature of Social Change: The Kheda District, 1818–1918', MAS, 15:4 (1981), 771–821.
107 Bagchi, Private Investment, p. 234, table 7.4.
108 Bombay Millowners' Association, Annual Reports; Bagchi, Private Investment, p. 234, table 7.4.
109 M. M. Mehta, Structure of Indian Industries (Bombay, 1955), pp. 163–73.

Page 58
110
lated by the needs of the British Indian Army stationed in the local cantonment. Such expansion was not simply a function of the restless quest by capital for yet
more productive sources of investment but the outcome of its adaptation to specific types of demand and often fairly localized production conditions. Since labour and
raw materials constituted the largest share of the cost of production, these upcountry mills were able to undercut Bombay and Ahmedabad in the coarser varieties and
poorer qualities especially in the local market. Their increasing competitiveness was not based upon the use of machines newer and better than those used in Bombay
and Ahmedabad but upon secondhand machinery which had sometimes been scrapped by their older rivals.
This pattern of location, geared to sources of cheap raw material and labour and to centres of proven demand, was reflected in the geography of the cotton textile
industry in India. By the mid1920s, cotton mills came to be widely scattered throughout the country, outside Bombay and Ahmedabad. Out of the twentythree Indian
states and British provinces which contained cotton mills eighteen had fewer than ten mills, five had no more than a single mill, and only two – the United Provinces and
the Bombay Presidency – had more than twenty. Half the mills in the UP were concentrated in Kanpur, the rest being widely dispersed, and out of the twentyfour
mills in the Bombay Presidency, outside Bombay and Ahmedabad, fifteen were divided equally between Sholapur, Surat and Broach while the remaining nine were
thinly spread across the rest of the Presidency.111
A third spurt of growth occurred in the 1930s. Indian industry held its own under the initial impact of the depression and then began to expand in the middle years of
the decade.112 Whereas the early growth of the
110 Bayly, Rulers, Townsmen and Bazaars, pp. 440–9; Report of the Indian Industrial Commission, 1916–18 (Calcutta, 1918), pp. 28–9.
111
Report of the ITB, 1927, vol. I, pp. 25–6.
112 Tomlinson, Political Economy, pp. 31–4. Of course, in several respects this expansion betrayed a rather fragile character. It took the form of import substitution in a period in which
the purchasing power of the peasantry was declining. It was not accompanied by rising investment in new plant and machinery. It was facilitated by a range of contingent factors and
special circumstances: tariff protection which, though not wholly effective, helped the growing centres rather more than Bombay for whom in large measure it had been designed; the
disruption of international trade and the decline of Lancashire in particular; and the outbreak of war which rescued the industry when it had appeared to reach the limits of import
substitution by the late 1930s. The difficulties underlying this spurt of growth was also demonstrated in the acute problems faced in the 1950s by many of the new centres of textile
production which emerged in this period. For an insight into the demand for cotton textiles, see Bagchi, Private Investment, pp. 237–53; on machinery imports, ibid., tables 7.10 and
7.11, pp. 258–9, and R. Kirk and C. P. Simmons, 'Lancashire and the Equipping of Indian Cotton Mills: A Study of Textile Machinery and Supply, 1854–1939', in K. Ballhatchet and D.
Taylor (eds.), Changing South Asia: Economy and Society (London, 1984), pp. 169–81; on the relationship between town and

Page 59
industry had been financed by the profits of the longdistance trade, both internal and international, the capital which financed the millbuilding of the 1930s was
primarily in retreat from the countryside, in the shadows of agrarian crises and the depression. Creditors who lent money on the condition that it be repaid in grain had
to recover their loans when prices were falling and from clients who frequently could not afford to pay. As the value of land declined and the price of its products
collapsed, the risks of default rose steeply and the means of hedging them disappeared. Under these conditions, as the terms of trade moved in favour of industry, rural
capital migrated to the towns.113
Moreover, the trend towards the dispersal of the industry away from Bombay and Ahmedabad, which had already manifested itself in the late nineteenth and early
twentieth centuries, was accentuated.114 Mills were built once more in proximity to local markets and to sources of raw materials and to cheap and plentiful supplies of
labour. The viability and competitiveness of these mills depended upon their ability to shelter in local markets for coarse goods where foreign competition was least
effective. Moreover, in these lines of production the advantages of the upcountry mills in cheaper labour and raw material costs could be made to count in competition
with a struggling Bombay industry. Like the new mills of the earlier twentieth century, these mills were often equipped with secondhand plant and frequently with
machinery scrapped by older mills which were either going bankrupt or were sufficiently successful and progressive to be retooling.
Historians are generally agreed and indeed have frequently reiterated that these mills were 'very often badly designed and badly managed' and characterized by
'ignorance of the best technical conditions on the part of entrepreneurs and their managers'.115 Their owners and managers have invariably had a bad press. It is
perhaps not their incompetence but their resourcefulness which deserves emphasis. They were not, it is true, concerned with achieving and maintaining what may
theoretically have
country during depression, Tomlinson, Political Economy, ch. 2 and Baker, An Indian Rural Economy, passim; on tariffs, see B. Chatterji, 'The Political Economy of
''Discriminating Protection": The Case of Textiles in the 1920s', IESHR, 20:3 (1983), 239–75, and Chatterji, 'Business and Politics in the 1930s: Lancashire and the Making of the
IndoBritish Trade Agreement', MAS, 15:3 (1981), 527–73; C. J. Dewey, 'The End of Imperialism of Free Trade: The Eclipse of the Lancashire Lobby and the Concession of Fiscal
Autonomy to India', in Dewey and Hopkins (eds.), The Imperial Impact, pp. 35–67, and I. Drummond, British Economic Policy and the Empire, 1919–1939 (London, 1972), ch. 4.
113 C. J. Baker, 'Debt and Depression in Madras, 1929–1936', in Dewey and Hopkins (eds.), Imperial Impact, pp. 233–42; Baker, An Indian Rural Economy, passim: Tomlinson, Political
Economy, ch. 2.
114 Mehta, Structure of Indian Industries, pp. 164–73; Report of the ITB, 1927, vol. I, pp. 4–28, 100–23.
115 Bagchi, Private Investment, p. 252.

Page 60
constituted optimum levels of efficiency. On the other hand, their aim was simple: to maximize profits within the existing economic context; and in this aim they clearly
succeeded in the short term.
Coimbatore, which expanded dramatically in the mid1930s, symbolized most accurately the character of the industry's expansion in this period.116 The Coimbatore
mills benefited from a number of developments taking place in the regional economy: the expansion of local cotton cultivation, especially in the longstapled Cambodian
varieties, which supplied by the early 1930s about half the raw cotton requirements of South Indian mills, thus considerably cheapening its cost; the expansion of the
local handloom industry which provided a ready market; the establishment of the Pykara hydroelectric project which offered a convenient and relatively cheap source
of electricity; tariff policies designed to alleviate Bombay's ills served centres like Coimbatore perhaps rather better. Moreover, the Coimbatore mills were known to
pay the lowest wages in the subcontinent and later in the decade, as they ran into difficulties, they sought to cut wages further.117 The declining profitability of
moneylending and the trade in agrarian produce released local capital for industrial investment, and the substantial cultivators and rural traders of the region, already
deeply involved in the marketing of cotton, now sought the returns to be offered by the spinning of yarn.118 This capital was used economically to establish small
spinning mills largely with machinery scrapped by the large number of Bombay mills which were closing down in this period and the few who were seeking cautiously
to replace parts of their old equipment. Indeed, by the 1930s, the Sassoon group of mills preferred to destroy rather than sell their discarded machinery for scrap lest
the upstart millowners of Coimbatore bought them up and then used these to jostle Bombay out of the domestic market.119
Each phase of the industry's expansion since its earliest beginnings had been characterized not by the progressive adoption of newer and better techniques but rather
by persistent adaptation to available markets, existing production conditions and changing economic constraints. Just as the Bombay mills had penetrated the Chinese
market in the 1870s and 1880s by exploiting their advantage of cheap labour, proximity to suitable raw
116 Baker, An Indian Rural Economy, pp. 339–72.
117 A. Pearse, The Cotton Industry of India: Being the Report of a Journey to India (Manchester, 1930), pp. 105–13; B. Shiva Rao, The Industrial Worker in India (London, 1939), p.
121; E. A. Ramaswamy, The Worker and His Union: A Study of South India (Delhi, 1977), pp. 17–32; Baker, An Indian Rural Economy, pp. 361 ff.; E. D. Murphy, Unions in Conflict: A
Comparative Study of Four South Indian Textile Centres, 1918–1939 (New Delhi, 1981); E. Perlin, 'Eyes Without Sight: Education and Millworkers in South India, 1939–76', IESHR,
18:3 and 4 (1981), 263–86.
119 Proceedings of the TLIC, Main Inquiry, Oral Evidence, Mr F. Stones, Managing Director, Sassoon Spinning and Weaving Company Limited, File 70, 3450–51, MSA.

Page 61
materials and to the market, economizing as far as possible on the cost of machinery and using labourintensive methods to produce cheap, coarse qualities, so the
newer upcountry centres commanded an increasing share of the domestic market at Bombay's expense largely by the same means.
In the 1920s and 1930s, rationalization became an increasingly prominent issue in the affairs of the cotton textile industry. Because the crisis of the industry in the inter
war period was largely localized in Bombay, the pressure to effect rationalization schemes impinged most directly upon its millowners. By the 1920s, Bombay had
been ousted from the Chinese market, lost its competitive advantage in cheap labour costs to both Indian and Japanese rivals and found itself heavily committed to the
production of lowcount, coarse goods which were rapidly becoming saturated in the domestic market as a result of the expansion of the upcountry mills.120 The
general strikes of 1924 and 1925 showed conclusively that direct wage cuts would be met with bitter and possibly damaging labour resistance and therefore offered no
solution to Bombay's problems.121 Tariff protection appeared to be only a slightly more promising remedy: keeping the Indian market open for Lancashire
manufactures had been one of the most important ground rules of the colonial state and Indian millowners made painfully slow progress against this entrenched
tradition.122 In response to their demand for protection, the Tariff Board lectured the Bombay millowners on the need for them to put their own house in order before
seeking the assistance of the state. The answer lay, they argued, in raising their efficiency and lowering their labour costs, in improved labour discipline and increased
productivity, changed patterns of labour deployment, methods of management and composition of output, the adoption of a more rational financial and marketing
structure, the replacement of old plant and the introduction of new machinery.123 Every subsequent committee of enquiry, which enquired into the conditions of the
industry, whether its brief was tariff protection or labour relations, commented upon the need for rationalization, assessed its progress and urged its speedy
implementation. The millowners' commitment to technological advance was increasingly called into question and the terms on which they would undertake to revamp
their methods and organization were now to be increasingly clarified.
121 Ibid., pp. 133–4; for a narrative of these general strikes, see R. Newman, Workers and Unions in Bombay, 1918–29: A Study of Organization in the Cotton Mills (Canberra, 1981),
pp. 142–8, 153–9.
122 Chatterji, 'The Political Economy of ''Discriminating Protection"'; and Dewey, 'The End of the Imperialism of Free Trade'.

Page 62
The Bombay millowners could claim with some justification to have diversified into higher counts and finer qualities, adjusted their production and sales more closely to
the needs of the domestic market, and streamlined their labour force. But their achievements by the end of the 1930s remained remarkably modest. Rationalization, as
one official report commented in 1940, 'has a wider and a narrower meaning' and 'improvements in labour productivity and efficiency' constituted 'one aspect of the
latter sense'.124 The Bombay millowners did not proceed far beyond this partial aspect of the narrow construction of its scope: the implementation of 'efficiency
schemes' which involved increasing workloads and reducing the strength of the labour force. Weavers who worked two looms were now to be put in charge of three
or four while single siders would mind both sides of the spinning frame. Yet even these efficiency schemes were far from sweeping. In the late 1930s, less than onefifth
of the weavers in the Bombay mills operated more than two looms and less than onehalf of the siders worked two sides.125 The millowners did not respond to their
competitive situation or to official exhortations by consistently rationalizing the structure and organization of their industry or by effecting technological change.
The reasons for the weakness of the millowners' attempts to rationalize were complex and although no comprehensive discussion of them can be undertaken here,126
the direction and emphasis of their response is readily demonstrated. The axiom that Indian entrepreneurs were highly speculative in business and inherently averse to
innovation need not detain us. The argument which stresses the constraints imposed upon businessmen by the absence of a domestic capital goods industry is, if more
prosaic, also more plausible. Of course, businessmen were inhibited by the depressed conditions of the trade. However, the positive changes which the Bombay mills
did effect suggest that their owners did not entirely shrink from taking risks, and some millowners at least would not have found it impossible to raise capital for more
farreaching reform. We need to dig deeper than the absolute scarcity of capital for a more convincing explanation.
The supply of capital to the industry, the nature of its markets and the pattern of labour deployment within it interacted to dampen the millowners' enthusiasm for
change. Most mills regulated their production according to the shortterm fluctuations of the market. This business strategy arose, in part, from the dependence of most
mills upon the money market for their working capital and sometimes a proportion of
124 Report of the TLIC, vol. II: Final Report (Bombay, 1953), p. 183.
125 Ibid., p. 187, table 44.
126 I discuss this issue at greater length in Chandavarkar, The Origins of Industrial Capitalism, pp. 271–77, 335–96.

Page 63
their fixed costs as well. In these conditions, they sought to avoid the accumulation of stocks and to maintain a rapid turnover to pay their debts and to attract
investment. Moreover, the Indian market was reputed to take the greatest variety of counts of yarn and types of piecegoods in the world.127 In response, mills
produced a bewildering range of counts of yarn and varieties of piecegoods, each of diverse qualities. The regulation of production to demand in these circumstances
meant that carding machines, spinning frames and looms allocated to one type and quality of output often had to be turned swiftly to another. As the industry diversified
its range of production, so these tendencies were exacerbated. Yet the costreducing effects of investment in machinery could only be fully realized if the flow and
consistency of orders were maintained. For the millowners to remain alert to market fluctuations and to the varieties and specificities of taste was often incompatible
with a commitment to vast outlays on plant and machinery. Far from responding to the unseen logic of industrialism or longterm technological imperatives, the
millowners attempted to maximize their returns within the constraints which they encountered.
A further consideration which often had a more crucial bearing on the introduction of rationalization and reforms was the response of labour. The brunt of these
changes was after all to be borne by the workforce in the form of greater workloads, higher unemployment and sometimes lower wages. In fact, reviewing the
progress of the previous decade, Maloney, the Secretary of the Millowners' Association, observed that 'the main cause of slow progress' in the implementation of
rationalization was not 'the lack of capital' but 'the opposition of the workers'. And as Fred Stones, Managing Director of the Sassoon group of mills added, 'progress
is hampered very much by the fact that there are so many strikes, and it would be inadvisable to invest money in machinery in the circumstances'.128 Changes in
working conditions or wage rates, even those which resulted from shortterm changes in the character of production, had always been liable to prompt industrial
action. Attempts to increase workloads in 1927, which threatened jobs in general and the status and position of weavers especially, prompted industrial action, which
culminated in the general strikes of 1928 and 1929. The strike of 1928 began in April and ended in October in a temporary truce which brought the millowners to the
negotiating table.129 The following six
127 Report of the ITB, 1927, vol. I, p. 48.
128 Proceedings of the TLIC, Main Inquiry, Oral Evidence, Representatives of the BMOA, File 57–A, p. 63, MSA.
129 Report of the Bombay Strike Enquiry Committee, vol. I (Bombay, 1929); Newman, Workers and Unions, pp. 168–209; see also ch. 4 below.

Page 64
months witnessed over seventy industrial disputes until in April 1929 another general strike was effected. This strike proved more difficult to sustain, but although it
began to weaken in July, it did not end until September.130 The question of rationalization once more had an important bearing upon the general strike of 1934.131 The
disruption of production which followed frequently undermined the competitive position of individual mills and, in the case of general strikes, the industry as a whole.
Better by far in these circumstances, especially when their markets were in a slump, for the millowners to do the best they could within the limitations imposed upon
them.
Industrialization was not simply a function of entrepreneurial choice. The impact of labour frequently deflected the course of industrialization and shaped its character.
The actions of workers, rather than technological imperatives alone, frequently determined the options and thus the strategies of capitalists. Indeed, the response of
capitalists to innovation in machinery or production patterns was guided by the general anxiety that changes in working conditions might provoke labour resistance and
unsettle the delicate balances upon which industrial peace rested. Even the Ahmedabad millowners, widely believed to be more progressive than their rivals in
Bombay, and supposedly blessed with a more docile labour force, were even more cautious in the introduction of rationalization and efficiency schemes. 'In spite of
their eagerness to introduce efficiency measures', it was reported, 'they have not been able to do so on account of the determined opposition of the Textile Labour
Association'.132
How far labour resistance shaped the attitude of the entrepreneurs towards the introduction of new technology was exemplified when the Bombay millowners, pressed
to rationalize their industry, considered the adoption of the automatic loom from the late 1920s onwards. The Northrop loom required a larger capital outlay and
suffered a higher rate of depreciation than the ordinary Lancashire loom. Moreover, stores and spare parts for the automatic loom were both relatively more expensive
and less readily obtainable. This additional cost had to be made up, as the Tariff Board noted, by 'higher production, higher prices or a reduction in
130 Report of the Court of Enquiry into a Trade Dispute between Several Textile Mills and their Workmen (Bombay, 1929); Newman, Workers and Unions, pp. 211–50; see also
ch. 4 below.
131 Labour Office, Bombay, Wages and Unemployment in the Bombay Cotton Textile Industry (Bombay, 1934).
132 Report of the TLIC, vol. II, pp. 188–9. The constraints thus exercised by workers upon entrepreneurial choice have often been perceived as a reflection of their preindustrial
mentalities, dysfunctional to the industrial setting. However, entrepreneurs in Britain, at the height of its industrial 'maturity', similarly showed a propensity to shelter in traditional
markets rather than effect fundamental changes which, while increasing their productivity and their competitiveness, also raised the spectre of industrial action.

Page 65
133
labour costs'. In the late 1920s and 1930s, there was little prospect of rising prices. Trial runs conducted in Bombay revealed that the automatic loom did not yield
an appreciably higher output. Furthermore, the automatic loom required weft yarn of a higher quality and greater strength. In turn, this called for a better quality of raw
cotton, improved mixings and preparation as well as the adoption of different techniques of winding, for the weft yarn had to be put onto special pirns. In 1927, these
adjustments alone were estimated to increase the costs of production by seven pies in the pound, at a time when certain lines of Bombay's output were being undersold
at the cost of production in the domestic market by Japanese mills.134 Clearly, they would also have repercussions on the organization of the entire mill. It was scarcely
economical to 'swingoff' the expensive Northrop loom as most mills did with regard to the ordinary loom. To supply the Northrop looms adequately, each process of
production would have to be suitably adapted. Its adoption would require the lengthening and standardization of production runs throughout the mill, from the
preparatory processes to the weaving sheds. Given their financial structures and their volatile market conditions, the millowners could scarcely transform their
production patterns so thoroughly.
The viability of the automatic loom thus came to turn increasingly upon whether labour costs could be reduced. The millowners were reluctant to promulgate direct
wage cuts which they expected would provoke industrial action. The possibility of reducing labour costs would depend therefore upon whether the number of looms
allocated to each weaver could be increased. It was very apparent at the time, however, that 'it would be difficult to get weavers in Bombay to look after more than
four looms'. Yet, even if the number of automatic looms per weaver was raised to six, it was thought that the ordinary loom would still be more economical.135 In one
of its nicer formulations, therefore, the Tariff Board concluded that 'in present conditions no solution of the problem presented by labour costs in Bombay lies in the
introduction of the Northrop loom'.136 Despite the recovery of the mid1930s, a greater measure of protection and the extent of unemployment which created a
reserve supply of labour, the millowners remained reluctant until the end of the decade to invest in the Northrop loom.137 The future, as the millowners perceived it, lay
with the Lancashire loom; and with perpetuating the existing state of affairs, while if possible and where necessary increasing the number of old looms tended by each
weaver. The case of the automatic loom shows that while capital costs, low prices and
133
Report of the ITB, 1927, vol. I, pp. 143–5.
134 Ibid.
135 Ibid.
136 Ibid., vol. I. p. 144.
137 Proceedings of the TLIC, Main Inquiry, Oral evidence, BMOA, File 57–A, MSA.

Page 66
foreign competition as well as fluctuations of quality and composition of output made for the lack of enthusiasm with which the Northrop was received in Bombay, the
most significant factor was the difficulty of getting weavers to work a larger number of looms.
The assumption that industrialization was determined by its own inherent logic rather than by social choice has often led historians to take it for granted that capitalists
were able consistently to create the labour force they sought. The technological imperatives of industrialization, in this view, determined the characteristics of the
workforce. Labour is thus consigned to a peripheral role in the history of industrialization. But the formation of an industrial labour force, like the process of
industrialization itself, cannot be adequately understood outside the context of the choices and actions of the workers themselves.
The bulk of the industrial labour force was composed of migrant workers recruited from varying distances. Yet crucially most workers attempted to retain their rural
connections. This was in part the consequence of low wages and predominantly casual and uncertain conditions of employment. But it also reflected conscious choice.
Rural smallholders migrated to the towns and factories in order to conserve their position in the village, to pay off their debts, to hold onto their land, and to retain their
crop shares. Urban employment, they often discovered, afforded their families at the village base a readier access to cash, but it did not consistently facilitate social
mobility. The indebtedness of industrial workers suggests that by utilizing their greater access to cash and credit, they simply transferred some part of the burden of
debt borne by their family from the countryside to the towns. The maintenance of a rural base was essential to their strategies for subsistence in the industrial setting,
and conversely, urban wage employment played a vital role in the reproduction of the family economy in the village. The formation of an industrial labour force was not
simply a matter of 'raw novices' from the countryside who 'had to be converted into an army of disciplined mill hands, responsive to the general requirements of
industrial work and to the special needs of the mills'.138 Nor was the migrant status of industrial workers a reflection of the 'early stages of industrialization', or a passing
phase in their progressive 'urbanization'. The formation of the labour force cannot be understood in terms of the smooth progression of peasants into proletarians. In
the case of Bombay, workers have retained their
138 M. D. Morris, The Emergence of an Industrial Labour Force: A Study of the Bombay Cotton Mills, 1854–1947 (Berkeley and Los Angeles, 1965), p. 107.

Page 67
rural connections through several vicissitudes since the beginnings of industrialization, over a century ago, until the present day. Far from being backward or passive or
traditional or conservative in their political consciousness, migrant workers were active in the defence of their jobs, conditions and wage levels. In fact, urban
proletarians, without an alternative source of support, however meagre, were more easily recruited to break strikes than those who possessed a rural base and village
connections.139
Industrial workers not only maintained their rural ties but in many cases forged social networks and institutions of their own in the urban neighbourhoods. Through these
neighbourhood connections, they found work, housing and credit. Landlords, grain dealers and moneylenders might extend credit to local residents in periods of
unemployment, rural distress or indeed industrial action. Social connections of the neighbourhood were used to prevent some men usurping the jobs of others during
strikes, sometimes by influence and sometimes with force. In these ways, the social organization of the neighbourhood had an important bearing upon the affairs of the
workplace. It strengthened their bargaining power and often enabled them to resist the policies of employers more effectively.140
It is often supposed that the role of the jobber was created by employers not only to recruit rural migrants but to facilitate their adaptation to the industrial setting. But
this process of adaptation was shaped to a large extent by the actions and autonomous organizations of rural migrants, not by the agency of the employers alone.
Although the jobber has often been portrayed as a figure of awesome power, and the fulcrum of industrial relations, there were numerous alternative locations of
power and patronage in the workplace and the neighbourhood with which the jobber was forced to negotiate and which severely narrowed his freedom of action.
Subject to the pressures of both the management and the labour force, jobbers frequently found themselves inside the jaws of a nutcracker. It was by no means easy
to maintain an adequate supply of labour in the face of shortterm fluctuations of demand, to hire workers at short notice or lay them off according to need, and at the
same time to prevent wages from rising and workers from combining. Indeed, the development of workingclass politics in Bombay in the 1920s and 1930s led to the
gradual erosion of the jobber's position. The state was now forced to fulfil some of the functions once left to the jobber: it intervened increasingly in the maintenance of
labour discipline especially through a growing legislative machinery for the conciliation and arbitration of disputes and an
139 Chandavarkar, The Origins of Industrial Capitalism, ch. 4.
140 See ch. 4 below.

Page 68
increasingly elaborate legal structure which defined the conditions for the conduct of strikes and the existence of trade unions.
The political organization and action of the working class was neither the natural consequence of their aggregation in large factories and cities, nor a direct reflection of
the level of industrialization. The notion that the development of workingclass consciousness and politics matched some inevitable process of its evolution from an
essentially ruralbased, casually employed immigrant labour force to the formation of a mature, industrial proletariat is not borne out by the Indian evidence. In Bombay
city, workers were active in the defence of their interests from the earliest years of the cotton textile industry. Although no trade unions existed among them, they were
always known by their employers as well as contemporary observers to be capable of combining to defend their wage levels.141 At various times, especially in the early
1890s, 1901 and 1908, strikes affecting very large numbers of workers were sustained for considerable periods of time.142 Although nearly onethird of the labour
force in the cotton mills was employed on a casual basis,143 the textile workers were able to mount eight general strikes between 1919 and 1940, which lasted longer
than a month and in 1928–29 were sustained for the most part of one and a half years.144 Indeed, dock workers who were hired almost exclusively on a daily basis at
the dock gates and were generally more quiescent than textile workers effected a threemonth strike in 1932.145 The possession of a rural base, and the development
and utilization of the social connections of the neighbourhood by the city's workers helps to explain why they were able repeatedly to mount and sustain long periods of
industrial action in the 1920s and 1930s. The institutions which workers created and the more or less informal social connections which they forged remained a
decisive influence on the politics of the workplace and exercised effective constraints on the ability of employers to create the labour force they wanted.
141 RCL, Foreign Report, vol. II, The Colonies and the Indian Empire, Memorandum on the Labour Question in India, Evidence, Mr J. M. Campbell, Collector, Land Revenue,
Excise and Opium, Bombay, PP, 1892, vol. XXXVI, p. 128.
142 V. B. Karnik, Strikes in India (Bombay, 1967), pp. 3–56; Commissioner of Police, Bombay, to Secretary, GOB, Judicial Department, no. 10503–6–R, 27 August 1908, GOB, General
Department, vol. 114 of 1908, reprinted in Source Material for a History of the Freedom Movement in India (Collected from Bombay Government Records), vol. II: 1885–1920
(Bombay, 1958), pp. 256–75.
143 RCL, Evidence, Mr J. M. Campbell, Collector, Land Revenue, Customs, and Opium, Bombay to the Chief Secretary, Bombay, PP, 1892, vol. XXXVI, p. 129; Labour Office, Bombay,
Report on the Wages, Hours of Work and Conditions of Employment in the Textile Industries (Cotton, Silk, Wool and Hosiery) in the Bombay Presidency (including Sind), May
1934, General Wage Census, Part I – Perennial Factories, Third Report (Bombay, 1937), p. 20.
144 See ch. 4 below.
145 Karnik, Strikes, pp. 265–6; R. P. Cholia, Dock Labourers in Bombay (Bombay, 1941).

Page 69
VI
This essay has attempted to sketch an albeit ragged consensus of opinion that conceives industrialization as a neutral, technologically determined process, which,
transcending social choice, shapes society in a single, inevitable direction. In this view, industrialization fashions labour in its own image; for some, even according to its
needs. Its culmination is invariably largescale enterprise, the factory system and the production line. It is characterized by an inherent tendency to increasing technical
sophistication and efficiency. Historically, it is assumed to constitute a process of diffusion which, beginning in Britain, spread to Western Europe and the United States,
and then unevenly and incompletely across the globe. For dependency theorists, the process was essentially the same, even if its effect in large parts of the world was
the opposite: not development, but the development of underdevelopment.
These notions about the nature of industrialization have deeply influenced studies of economic development in India. Economic historians of India have been concerned
to explain why the perceived preconditions for industrialization did not exist in early nineteenthcentury India, and why industrialization subsequently failed to transform
the Indian economy. Explanations have boiled down to a residual choice between the baneful effects of colonial rule or the timeless torpor of Indian society. In one
view, general poverty, the deadweight of tradition and social organization that was weakly geared to the rationality of the market meant that the Indian economy
would have remained torpid but for the benefits of colonial rule.146 But if colonial rule had mobilized the Indian economy to the point where the first weak impulses of
'industrialism' began to manifest themselves, why did it not go further? Thus, in another view, the failure of the Indian economy is ascribed conversely to the nature of
British rule. Both sets of arguments assume that external agencies, colonialism and the world economy constituted the only active element in the Indian situation,
whether their force was exerted to plunder Indian resources or to mobilize them. This debate which originated in the late nineteenth century, has long petrified into
deadlock. Its resolution may lie less in choosing between these two positions than in more positively exploring the interplay between Indian social structures and
colonial rule, between the internal and the external economy, though not as an unproblematic marriage between equals.
The counterfactual nature of this debate has arisen in part from an unilinear and evolutionary schemata of social and economic change
146 For a version of this argument, see Morris, 'LargeScale Industry', especially, 558–66.

Page 70
against which the development of the Indian economy is measured. Recently, Irfan Habib has argued that while technology in Mughal India had been surpassed in
sophistication by Western Europe and indeed was not necessarily following the latter's path of development, it was nevertheless dynamic and open to improvement
and innovation.147 The point may perhaps be generalized to the pattern of economic development as a whole. Typologies of social and economic change have
necessarily drawn attention to sharp discontinuities. Historians who have sought to apply them to the past have had to reconcile the far subtler modulations of change
and the longterm continuities in the social processes they encounter with the dramatic ruptures suggested by social theories and typologies of change. The growth and
spread of commercialization in land, labour and produce, capital and credit, scarcely the creation of colonial rule alone, were slow and gradual processes, but did not
develop in a single, unilinear direction.
In addition, the patterns of economic development at a national level, defined in correspondence with political boundaries and states, are liable to mask fundamentally
diverse changes which manifest themselves at more local or regional levels. Recent studies of land tenure and agrarian social structure demonstrated how greatly
production conditions could vary between localities of a fairly modest scale.148 Areas of 'dynamism' sometimes developed into areas of 'enervation'149 while changing
circumstances, not necessarily directly related to the means of production, might facilitate buoyancy in apparently stagnant localities and the development of certain
regions at the expense of others. Studies of national income have emphasized the 'vast disparities in regional income' in the eighteenth century and suggested that the
fortunes of particular regions altered substantially during the period of colonial rule.150 Similarly, Heston and Summers have identified as an essential problem of 'Indian
economic history in the last century' the need 'to deal with the heterogeneous experience of very large regions with populations in tens of millions that have shown
aggregate growth not so much lower than many individual countries, while at the same time dealing with several relatively stagnant regions'.151 And regional studies of
the Punjab have reinforced this
147 Habib, 'Technology and Economy'.
148 E. T. Stokes, The Peasant and the Raj: Studies in Agrarian Society and Peasant Rebellion in Colonial India (Cambridge, 1978).
149 E. T. Stokes, 'Dynamism and Enervation in North Indian Agriculture: The Historical Dimension', in ibid., pp. 228–42.
150 D. Kumar and J. Krishnamurthy, 'Regional and International Economic Disparities since the Industrial Revolution: The Indian Evidence', in P. Bairoch and M. LevyLeboyer (eds.),
Disparities in Economic Development Since the Industrial Revolution (London, 1981), pp. 361–72.
151 A. Heston and R. Summers, 'Comparative Indian Economic Growth: 1870 to 1970', American Economic Review, 70:2 (1980), 97.

Page 71
152
point. These studies serve to suggest that the nature of economic development will be illuminated by studying more closely the fate of regional economies: the
different paths of economic change followed by them, the relationships between the diverse economic activities and social groups within them, and the effects of
various factors, often contingent and political, in determining, altering and redirecting the lines of economic change.153 This may help to offer a dynamic context of social
and economic change within which industrialization and economic development may be placed.
This paper has attempted to argue that largescale industry was neither the only source of dynamism within the economy nor the natural culmination of economic
development. The path of industrial development was constructed by the interaction of various social forces. Its direction was the outcome of these conflicts and did
not follow predetermined lines or proceed towards a single, ultimate destination. The development of industry in India cannot be adequately understood as the
manifestation of a progressive technological diffusion working outwards from Western Europe. Neither was industrialization the only agency of economic development,
nor need largescale industry be construed as its logical culmination, and, indeed, it is doubtful whether largescale industry can usefully be placed in a distinct
functional category of its own. On the one hand, seemingly traditional forms of industrial organization showed themselves able to survive, adapt and prosper within a
given context. On the other hand, business strategies and methods of labour deployment followed in large factories often resembled the practices of smallscale
enterprises in the informal sector. Finally, it is clear that the process of industrialization cannot be fully explained without reference to the crucial impact of labour upon
its course and character. Notions of modernization have offered convenient categories and narrative schemes for understanding social change; yet far from working
upon a passive social order, what are labelled forces of modernization have more usually been tied down and shaped by existing social institutions and supposedly
traditional agencies.
If the burden of predominant approaches to the economic history of colonial India has been to produce a largely counterfactual problematic, the remedy cannot
simply lie in a plea for a more positive historiography. For this line of advance has tended to offer accounts of the performance of the Indian economy, irrespective of
the political fact of colonialism and without seeking to relate this economic performance to the nature of the
152 C. J. Dewey, 'The Agricultural Output of an Indian Province: The Punjab, 1870–1940'. Paper read to the Economic History Seminar, Institute of Commonwealth Studies, London,
30 April 1973; Dewey, 'Some Consequences of Military Expenditure in British India: The Case of the Upper Sind Sagar Doab, 1849–1947', in C. Dewey (ed.), Arrested Development
in India: The Historical Dimension (Delhi, 1988), pp. 93–169.
153 The 'regional economy' before 1857 is examined in CEHI, vol. II, pp. 242–375.

Page 72
154
colonial state and its wider imperial purposes or indeed to the international economy. Thus, studies of British rule and the Indian economy inform us simply about the
performance of the latter in the period no more than chronologically defined by the former.
To examine the interplay between Indian society and colonial rule, it will be essential to absorb more fully into the existing debate about the imperial impact two lines of
enquiry which have already begun to emerge discretely in recent studies. First, existing arguments about the modernization or retardation of the Indian economy focus
almost exclusively on the period after the midnineteenth century. Yet it is obvious that economic and social structures developing in the precolonial period established
the historical context within which the colonial economy formed and within which its dynamic and expansive components emerged. Recent research has suggested that
the world economy was beginning to impinge upon and influence the internal economic and political structure well before the imposition of formal colonial rule in the
late eighteenth century.155 Moreover, what Britain could achieve in India was determined to a large extent by the changing condition of the internal economy and its
developing class structure.156
Second, it is becoming increasingly possible to relate the political priorities of the colonial state to the changing character of the Indian economy. The most common
arguments about the effects of the colonial state on the Indian economy have usually concerned the way it acted to preserve and protect the economic interests of the
metropolitan ruling class. But the colonial state was also vulnerable to the pressures which dominant classes in India could exert upon it. Moreover, under certain
conditions, it subordinated particular, even powerful metropolitan interests to the claims of specific Indian groups. The balance of political power and the composition
of economic interests within the metropolitan ruling classes was subject to change. In any case, 'intention' and conspiracy have
154 For instance, the imperial dimension is played down in the judicious and positive account offered by N. Charlesworth, British Rule and the Indian Economy, 1800–1914
(London, 1983).
155 C. A. Bayly, 'Putting Together the Eighteenth Century in India: Trade, Money and the ''PreColonial" Political Order', paper presented to the Second AngloDutch Workshop on
Comparative Colonial History, September 1981; I. Habib, 'Monetary Systems and Prices', in T. Raychaudhuri and I. Habib (eds.), CEHI, vol. I (Cambridge, 1982), pp. 360–81; Perlin,
'ProtoIndustrialization'; A. Hasan, 'The Silver Currency Output of the Mughal Empire and Prices in India in Sixteenth and Seventeenth Centuries', IESHR, 6:1 (1969), 85–116; J. F.
Richards, 'Mughal State Finance and the Premodern World Economy', Comparative Studies in Society and History, 23:2 (1981), 285–308; K. N. Chaudhuri, The Trading World of Asia
and the English East India Company, 1660–1760 (London, 1978).
156 D. A. Washbrook, 'Law, State and Agrarian Society in Colonial India', MAS, 15:3 (1981), 649–721.

Page 73
sometimes in this line of argument been difficult to prove. On the other hand, it was also the case that India was ruled in the interests of Britain's global imperial
system.157 This fact alone served to shape the Indian economy in ways which were not conducive to growth: the burden of land revenue, especially in the early
nineteenth century; the alignment of the colonial state with predominantly conservative social elements; the commitment to balanced budgets, to meeting the Home
Charges and to guarantees on interest payments which governed the financial, fiscal and monetary policies of the state and whose effects were at times to squeeze the
internal economy and severely damage commercial activity. The imperial impact was governed less by the specifically economic interests of the metropolitan ruling
classes in Britain alone than by the changing political priorities of the colonial state.
157 In the case of India, this point has been most extensively documented in Tomlinson, Political Economy, and his articles, 'India and the British Empire, 1880–1935', IESHR, 12:4
(1975), 339–80 and 'India and the British Empire, 1935–1947', IESHR, 13:3 (1976), 331–52.

Page 74
3—
Workers, Trade Unions and the State in Colonial India
It is a commonplace that the period between 1917 and 1920 constituted a benchmark in the history of the Indian labour movement. Although more unions were
formed, many of them remained ephemeral to the politics of the workplace. Their connections with the workforce were often tenuous. They usually arrived on the
scene of a strike after it had begun and frequently failed to persuade workers to accept the settlements which they had negotiated. Some trade unions had about as
much life as the letterheads which they printed ostentatiously on their notepaper. Most of them existed as little more than a nucleus around which workers would gather
in the course of a strike. When work was resumed, their membership dwindled rapidly and in many cases disappeared altogether. Trade unions existed often as a
loose superstructure constructed over an active undergrowth of informal organization and seemingly spontaneous industrial action.
Conventionally, historians have understood the weakness of trade union organization and the supposed volatility of the workforce as responses characteristic of the
early stages of industrialization.1 Frequently, historians have postulated too simple and economistic a relationship between industrialization and labour protest.
Industrialization has been widely portrayed as an independent force, determined by a neutral technology which lies beyond the realm of political control or social
choice.2 In this view, the level of industrialization has been closely associated with a particular pattern of workingclass behaviour and consciousness. If the early stages
of industrialization are characterized by volatility, further economic development is supposed to lead to the more ready acceptance
1 The most recent and systematic exposition of this convention is Chakrabarty, Rethinking Working Class History. Most accounts of workers' politics, despite substantial
differences of approach, have subscribed to this convention. For some examples see C. A. Myers, Labour Problems in the Industrialization of India (Cambridge, Mass., 1958);
Morris, The Emergence of an Industrial Labour Force in India; Newman, Workers and Unions in Bombay; Arnold, 'Industrial Violence in Colonial India'; Murphy, Unions in
Conflict; Joshi, 'Bonds of Community, Ties of Religion'.
2 The interpretative costs of these perspectives are examined at length in ch. 2 above.

Page 75
of formal collective bargaining. Within the terms of such reasoning, political maturity should properly be measured in yards of cloth or tons of rolling steel.
It would be tempting to fit the Indian case to this conventional wisdom were it not for what the Industrial Disputes Committee observed as 'the capacity of the
operatives to remain on strike for considerable periods despite the lack of any visible organisation'.3 This was exemplified in Bombay where general strikes occurred in
1908, 1919, 1920 and 1923 without the initiative of any effective trade union; in 1928 when both major unions were opposed to its occurrence; in 1934 and 1940,
when the Girni Kamgar Union was unable to play any significant part in the daily relations of the workplace. All these strikes, except for 1908, lasted for at least a
month; in 1928–29, they lasted for about twelve months in eighteen. These extensive, if informally organized, strikes were not confined to Bombay. They were
witnessed in Ahmedabad in 1918, 1923, 1935 and 1937; in Sholapur in 1920, 1928, 1934 and 1937; in the Calcutta jute mills in 1929 and 1937; on various railways
especially between 1928 and 1930; in Nagpur in 1934 and Coimbatore in 1938.4
Confronted by extensive evidence of sustained and prolonged industrial action, it is difficult to accept the common argument that the nature and dynamics of Indian
trade unions simply reflected the level of consciousness among industrial workers. It is clear that for the beginnings of an explanation we will have to dig at deeper
levels. We will have to avoid the assumption that the realization by workers of the utility and indeed necessity of formal trade union organization led simply to its being
effected in reality or indeed that intention was unproblematically translated into achievement. We will also have to pursue the hypothesis that the terrain upon which the
Indian working class, perhaps any working class, fought its battles was determined by its opponents and, therefore, their forms of action reflected not their level of
consciousness, but the range of options available to them inside a particular economic and political conjuncture.
If, on the one hand, the nature of workingclass politics was not merely a function of its adaptation to the industrial environment, the structural characteristics of the
labour force, on the other hand, were not simply sculptured by neutral, technological forces. The politics of employers or the state often actively encouraged, indeed,
developed, many of the supposedly preindustrial characteristics of the workforce. In Calcutta in
3 Report of the Industrial Disputes Committee, 1922, in Labour Gazette, 1:8 (April 1922), 24.
4 Karnik, Strikes in India.

Page 76
5 6
the 1890s and the 1900s and in Bombay in the 1870s and 1890s employers in the jute and cotton textile industries respectively attempted to diversify the social
composition of the workforce by replacing local, urban labour with rural migrants. Their objective was not merely to meet their expanding labour needs but also to
check the growth and intensification of labour unrest which occurred in the 1890s on a scale never before witnessed.
Similarly, in order to meet the often fluctuating demand for labour, employers devised systems of recruitment based on jobbers or sardars who ensured the presence of
a full contingent of workers and were given considerable disciplinary powers at the workplace. In order to fulfil their diverse managerial tasks, jobbers as well as other
supervisory officials had to operate through their caste, kinship, village and neighbourhood connections. It has commonly been assumed that the role of the jobber was
to temper the cultural diversity of the working class as well as to mediate between managers and men, both linguistically and socially.7 It is more likely that by having to
recruit and more generally to operate along the lines of caste, kinship, neighbourhood and village that supervisors and jobbers accentuated the sectionalism of the
workforce.8
Of course, the structural characteristics of the workforce influenced the nature of industrial action and trade union organization; but their relationship was neither
uniform nor consistent. In Bombay, workers who maintained connections with their villages were often considered the most resilient during industrial disputes and more
active in the defence of their interests than those who had severed their ties with the land.9 Ahmedabad appeared to combine the most urbanized, proletarianized
workforce with the most stable and sophisticated trade union in the country. Although the militant sections of the workforce were also the
5 B. Foley, Report on Labour in Bengal (Calcutta, 1906), para 29; Das Gupta, 'Factory Labour in Eastern India'.
6 Report of the Indian Factory Commission, 1890 (Calcutta, 1890), Annual Report of the BMOA, 1892, pp. 11–12, 43–5, ibid., 1896, pp. 149–60, ibid., 1897, pp. 4–6; Morris, The
Emergence of an Industrial Labour Force, pp. 54–6; Misra, 'Factory Labour During the Early Years of Industrialisation'; Mehta, The Cotton Mills of India.
7 Newman, Workers and Unions, pp. 27–34 and passim; Newman, 'Social Factors in the Recruitment of the Bombay Millhands'; Mazumdar, 'Labour Supply in Early Industrialization';
Morris, Industrial Labour Force, chs. IV, V, VIII and IX; Das Gupta, 'Factory Labour in Eastern India'; Simmons, 'Recruiting and Organizing an Industrial Labour Force in Colonial
India'; Chakrabarty, Rethinking Working Class History, pp. 95–114, and passim, especially chs. 5 and 6; Kooiman, 'Jobbers and the Emergence of Trade Unions in Bombay City'.
8 In contemporary discourse, as well as some recent scholarship, it was assumed that jobbers were the product and expression of the caste, kinship and village ties which were taken to
constitute Indian society and its tradition, as well as its working class.
9 RCLI, Evidence, Bombay Presidency (including Sind) (London, 1931), Mr M. S. Bhumgara, vol. I, part i, p. 499; BPP SAI, 1928, no. 3, 21 January, para 61.

Page 77
10
most urbanized, they remained the least integrated into the Ahmedabad Textile Labour Association. Nagpur which recruited its labour locally was noted for its
quiescence, but Sholapur, which did the same, at times appeared turbulent even insurrectionary in its politics.11
Although employers were to gain considerably at times from the sectionalism of the workforce, from caste and religious mentalities, from village, kinship and
neighbourhood ties, it cannot be assumed that these acted only to restrict the possibilities of workingclass politics. Indeed, caste and communal connections
sometimes provided the basis for combination at the workplace. Thus, one of the earliest organizations of Bengal's jute mill operatives was Kazi Zahiruddin Ahmed's
Mohammedan Association founded at Kankinara in 1895. It combined as its major preoccupations the employment of more Muslims in the jute mills and the
renovation of mosques.12 In Bombay, too, the Kamgar Hitwardhak Sabha which played an active role in the settlement of several industrial disputes in the 1910s and
early 1920s was, primarily a welfare association for nonBrahmin workers.13 Marathas from Ratnagiri were among the most ready participants in industrial action and
their village and caste connections contributed to some of the networks upon which communist organization in Bombay was grounded.14 Muslim weavers in
Ahmedabad remained initially aloof and were increasingly alienated from the Gandhian Textile Labour Association, and by the 1930s they had formed their own
communist Mill Mazdoor Sangh.15 Its popularity among the city's workers was sufficient to cause deep concern in the Ahmedabad Textile Labour Association, and it
is this factor, perhaps, rather than capitalist pressure upon the Congress, which explains the highly repressive policies of the Congress Ministry towards the commu
10 RCLI, Evidence, Bombay Presidency, vol. I, part i, pp. 287–9; S. Patel, The Making of Industrial Relations: The Ahmedabad Textile Industry, 1918–1939 (Delhi, 1987); Mehta,
Cotton Mills, pp. 277–86; Gillion, Ahmedabad – A Study in Indian Urban History; Salim Lakha, Capitalism and Class in Colonial India: The Case of Ahmedabad (New Delhi,
1988).
11 GOB, Home (Special), File 550 (14) of 1933, File 543 (53) A of 1934 and File 550 (25) IVA of 1938, MSA.
12 D. Chakrabarty, 'Communal Riots and Labour: Bengal's Jute Mill Hands in the 1890s', Past and Present, 91 (May 1981), 140–69. For an interesting argument which shows that it
would be folly to conclude that these apparently 'communal' forms of organization reflected a political consciousness governed by 'community' or the strength of traditional sentiments
and primordial loyalties among workers, see Basu, 'Workers' Politics in Bengal, 1890–1929', pp. 117–51.
13 Labour Gazette, 2:7 (March 1923), 26; 'The Kamgar Hitwardhak Sabha: A Brief Sketch', Indian Textile Journal, 29 (July 1919), 177–9 and ibid., 29 (August 1919), 209–10. On the
earlier and interesting case of Lokhande, see the recent study by Manohar Kadam, Narayan Meghaji Lokhande: Bharatiya Kamgar Calvalliche Janak (Bombay, 1995).
14 MCC, Statements Made by the Accused, NonCommunist Series, Examination of S. H. Jhabvala, p. 756.
15 GOB Home (Special), File 550 (25) IIIA of 1938, MSA.

Page 78
16
nist unions during its period in office under provincial autonomy between 1937 and 1939.
In Kanpur, the communist revival of the mid 1930s, masterminded by the organizational genius of P. C. Joshi, stemmed from two sources. First, the existing Kanpur
Mazdoor Sabha (KMS) which had been founded in 1919 was led largely by highcaste Hindus, some of whom had been connected with the Arya Samaj. Logically,
the communists turned their attention to Muslim workers who had remained largely aloof from the KMS. Second, they attempted to breathe life into old mill
committees and neighbourhood committees, and sometimes established new ones. Between 1936 and 1938 the old leadership of the KMS found that their base was
moving beneath them, and following the general strike of 1938 the communist ascendancy in the union was complete.17 In following this strategy, P. C. Joshi and the
other communist trade union leaders in Kanpur embarked more systematically on a course which their counterparts in Bombay had discovered accidentally in the later
1920s. It is not intended to suggest that this style of organizing beyond the workplace was peculiar to the communists, but it is clear that in Bombay, and to a lesser
extent in Sholapur and Kanpur, they performed it most effectively. The Bombay Textile Labour Union did not by any means stay out of this critical arena of labour
politics and N. M. Joshi, who was in official circles for over two decades the most trusted spokesman for the 'labour interest', realized soon enough that his credibility
in the councils of government could not be assured if he ignored, or even neglected, the men who cut a figure on the street corner.18 The Congress organization in the
workingclass neighbourhoods of Bombay, during the nonviolent days of civil disobedience, rested on the broad shoulders of the exjobber and neighbourhood boss,
Keshav dada Borkar.19 Ultimately, the most successful trade unions were those which intervened in the informal organizations of the working class, furrowed beneath
the existing empires of street and neighbourhood and indeed resurrected their own. What needs to be examined are the factors which determined this style of trade
union politics.
Clearly, the divisions of caste and kinship, occupational and neighbourhood cultures intensified tensions, encouraged factions and induced rivalries within unions.
However, historians have frequently viewed the problem of trade union rivalries through the wrong end of the telescope.
16 ATLA Papers, File 3, Part 2, microfilm copy, reel 9, NMML.
17 S. M. Pandey, 'Ideological Conflict in the Kanpur Trade Union Movement: 1934–1945', Indian Journal of Industrial Relations, 3:2 (1967), 243–68.
18 V. B. Karnik, N. M. Joshi: Servant of India (Bombay, 1972).
19 Chandavarkar, The Origins of Industrial Capitalism, pp. 204–11. Also see below, ch. 4.

Page 79
In this view, preindustrial societies appear to be characterized by diverse forms of labour use while industrialization increases the homogeneity of the working classes.
As workers are concentrated into larger masses, their interests are thought to become more uniform and their struggles are thought to acquire a common base.
Historians have thus been constrained to explain in each case why workers did not perceive their interests as uniform and why political differences and rivalries
persisted. In India, and as in other cases cast as defective variants of the Western model, preindustrial survivals in the industrial context appear to explain everything.
At the political level, labour history often takes the form of an epic romance in which heroic deeds are followed closely by internecine jealousy and villainous betrayal.
However, if the tendency to hypostatize the working class or indeed the assumption that it contained within itself an inherent propensity towards unity was set aside, the
problem itself would disappear.
If largescale industrial production brought large masses of workers together, the sophistication of the division of labour meant that the impact of trade fluctuations or
managerial policies were felt differentially throughout the workforce. Industrialization did not always reduce, in fact it often accentuated, competition between
workers.20 Trade union rivalries can only be fully understood in terms of this competition between workers over the supply of raw material or the quality of the
machinery; at the level of the jobber system over the control of jobs, the maintenance of wage levels and the domination of neighbourhoods; and over power, at the
sometimes autonomous level of politics itself. It would appear from this perspective that rivalry and competition rather than homogeneity were integral to the formation
of a labour force and endemic to labour politics. To this extent, the supposedly preindustrial characteristics of the workforce do not in themselves offer a sufficient
explanation for workingclass sectionalism and trade union rivalry. On the contrary, employers could bring cultural differences into the workplace and institutionalize
them along the lines of the economic divisions created by industrialization.21 In other words, these supposedly preindustrial characteristics of the workforce were not
epiphenomenal survivals but essential to the process of industrialization. If they informed the tensions and conflicts within trade unions, and could be manipulated by
employers to circumscribe the bargaining power of their workers, they did not simply impede trade union growth and, indeed, their linkages could sometimes
20
Chandavarkar, The Origins of Industrial Capitalism, chs. 7 and 8.
21 GOB, Home (Special), File 344 of 1929, pp. 113–15, MSA, for an example of attempts to use Muslim workers to break a strike. See also G. D. Birla to P. Thakurdas, 4 May 1929,
Thakurdas Papers, File 81 (II) of 1929, NMML.

Page 80
be harnessed for organization as well as for action. The bargaining power of industrial workers was not determined by the neutral advance of technology alone, but
rather by conflicts waged in order to control its use. These conflicts occurred, however, within an arena which was being defined by the policies of the employers and
the state. This chapter will, therefore, turn to an examination of how these arenas were defined and how operating within them shaped the contours and determined the
style of trade union politics.
In most industries, jobbers and similar supervisory ranks played a critical role in the maintenance of factory discipline. To some extent, disputes were resolved within
the jobber's team or arbitrated upon by head jobbers and other supervisory personnel. When workers attempted to combine, they were as far as possible cajoled,
persuaded, threatened or dismissed at these levels. When disputes escalated beyond these levels, employers could continue to manipulate the wage relationship, with
its ultimate sanction of dismissal, in order to thwart combinations. As long as a significant proportion of jobbers and workers were not connected with a particular
union, there was little need for an employer to deal with it; when it began to acquire any considerable influence among the workers, he could discipline or even dismiss
some as a warning to the rest. If their resistance was too great, he could concede in the short term and then gradually begin to replace his employees. At the same time,
by choosing whom he would negotiate with, the employer could decisively influence the larger alliances of trade union politics. On the other hand, the strength of a
particular trade union in the city as a whole could make it a powerful influence dislocating the sensitive balances upon which labour control rested within an individual
factory. As the communist Girni Kamgar Union began to assert its hegemony over the mill districts of Bombay in 1928–29, the Millowners' Association was forced to
deal with it at the level of the industry while individual owners could not prevent its intervention at the level of the mill.22 In general, however, workers who participated
in trade union activity were likely to be victimized, discriminated against in the distribution of raw materials or allocation of machinery and unlikely to be promoted.23
Employers also sometimes used forms of espionage, systems of political intelligence and even physical force to combat the threat of combination. The Sassoon group
of mills in Bombay had an extensive organization for coercion led by one Milton Kubes who styled himself 'a wellknown boxer' and claimed to have 'my own secret
service'.24 And indeed he did. Babaji Rane, giving evidence before the Court of Enquiry investigating
22 See below, ch. 4.
24 BRIC, 1929, Oral Evidence, Milton Kubes, File 5, pp. 241–3, MSA.

Page 81
the 1929 general strike, described himself as a 'a CID man in the Sassoon mills' and said that Kubes was the head of the CID department. Rane himself was an ex
policeman. His job was to take a squad of ten or twelve others to communist meetings, sometimes disguised he said with false moustaches, and armed with a spring
which he described as 'a small thing when folded but it can be stretched out and can be used for attacking others'. Their job was to listen to the speeches and 'report to
the masters', to bring back any handbills which were issued and 'to create some trouble in the meeting and to see that it was broken up'. Rane was later dismissed by
the Sassoon group because he said having had to attend all their meetings and listen to their speeches, he had decided that 'the Red Flag Union was a very good
union'.25 Physical force was more frequently employed against pickets or ordinary workers, occasionally against union leaders. N. G. Ramaswamy's meteoric career as
the leader of the Congress Socialist Textile Worker's Union in Coimbatore was brought to a swift conclusion when he was crippled by the assaults upon him, and died
as a result of his injuries at the age of thirtyone in 1943.26
In an overstocked labour market, industrial action necessarily placed jobs in jeopardy. Faced with a strike, employers often simply dismissed the strikers. Between the
wars, workers' militancy made this more difficult for employers and managers to effect as a matter of course in Bombay, for instance, than it was in Ahmedabad,
despite its model Textile Labour Association.27 Unless workers effected a strike, it was often the case that their grievances were not seriously considered by the
management; but if they were unable to effect a fairly complete strike, they stood little chance of negotiating their demands with the management, let alone achieving
any concessions. If the mood of the workers was too determined and the strike too complete, and especially if the market was in slump, an employer could lock them
out until they were harassed by their landlords and their creditors to return or, alternatively, until he was able to organize a new set of jobbers and workers to replace
them. When an employer wanted to end a strike quickly, his best bet was to keep production going, essentially by recruiting blacklegs, driving them into the factory
and working as many machines as possible. As soon as some workers returned, the danger of losing their jobs often brought several others back to the workplace.
The effect of this reaction once set in motion could overcome a sizeable section of the workforce. The employer was then placed in a position from which he could
dictate the terms and the time at which the mill would be reopened. At times, and with varying
25 Proceedings of the Court of Enquiry, 1929, File 5, MSA.
26 Ramaswamy, The Worker and His Union, pp. 17–32.
27 Labour Gazette, Monthly Report on Industrial Disputes, passim.

Page 82
degrees of success, employers were able to call upon the assistance of the state – which could extend, for instance, to providing a police escort to help them take
blacklegs into the workplace,28 sometimes to arrest and prosecute 'agitators', most dramatically, of course, in the Meerut Conspiracy Case, and, periodically, 'to
restore order'.
The formidable obstacles to combination are illustrated in a report by the moderate labour leader, B. Shiva Rao, highly respected in official circles, on the difficulties he
faced in forming trade unions at Madurai, Tuticorin and Koilpatti. At Madurai and Tuticorin, he reported, the timekeeper had instructed the jobbers 'to attend the
meeting I addressed and take down the names of those present'. At Tuticorin,
the Police Circle Inspector and SubInspector stood right in front of me while I was speaking, and it was only when I referred to the prominence of the police at the meeting that
they thought it better to move aside and sit down.
Prior to the meeting, a procession in Shiva Rao's 'honour' was organized. It passed the mill gates after 5.30 p.m.; the timekeeper had the gates shut 'until the
procession had passed safely beyond the mill gate' and 'everyone was warned that, if he or she attended the meeting there would be punishment'. In Koilpatti, a large
majority of the workers lived in quarters provided free of rent by the employers. To enter the premises, it was necessary to pass the house of the head clerk
who apparently is a sort of a watch man. So the movements of any outsiders enquiring into the conditions of the workers are noted. The workers are afraid . . . lest they should be
turned out of the mill and simultaneously ejected from their quarters.
His tour of these southern textile towns, not surprisingly, proved a failure. 'The workers are so frightened and coerced,' he wrote plaintively,
that they will not come to a meeting until after dark. I am trying to establish a Trade Union at every centre but the influence of the employer is very strong. The police seem to be
on his side, and labour is so plentiful that the workers are afraid of being penalized.
It is possible that in the smaller towns the employers were better placed to exercise such extreme coercion, and that the police were more nervous of the consequences
of 'agitation'. When employers attempted to establish their presence in the workingclass neighbourhoods of Bombay or the jute mill towns of Calcutta, in this way, as
indeed they did, they met with a more fierce and consistent opposition than might have been possible in Koilpatti.29
28 Times of India, 8 August 1929.
29 'Report by B. Shiva Rao on his visit to Madurai, Tuticorin and Koilpatti to organize Trade Unions', N. M. Joshi Papers, File 36, NMML.

Page 83
It is scarcely surprising that under these conditions employers were able to prevent workers' combinations from gaining a permanent foothold in the workplace. Since
the earliest days of industry, workers had to seek the assistance of agents outside the workplace for the provision of necessary, usually legal, services and, because of
the costs of industrial action, to intervene on their behalf, to seek redress for their grievances and, as a corollary, to negotiate and conduct strikes. There is some
evidence to suggest that operatives of the Baranagar jute factory in the Calcutta area turned to the law courts as early as the 1860s and 1870s to recover overdue
wages and in one case to sue an assistant manager for assault.30 Similar cases were filed by Bombay's workers in the small causes courts and police courts of the city.
By the early twentieth century, pleaders' offices proliferated in the mill districts of the city. Some of them operated under what the commissioner of police described as
'various highsounding titles such as ''Labourers' Union", "Millhands' Association", etc.'.31 They provided the base upon which Tilak built his following in the 1900s and
upon which the Home Rule League arose in the 1910s. In these offices, leave notices were written, grievances received and petitions against unfair dismissal or ill
treatment by superiors were drafted. Occasionally, the most successful pleaders intervened in disputes and attempted to negotiate settlements. Nothing was as good
for business or guaranteed to attract custom as the successful negotiation and settlement of a strike.
Between the wars, the legal and administrative structure which impinged upon the working class became more obtrusive and elaborate, creating new arenas in which
pleaders and politicians were better equipped to act than the jobbers and workers themselves. Of course, this provided a spur to the formation of trade unions.32
However, the professionalization of trade unions did not transform their role in industrial relations. They continued to provide workers with social and legal services and
receive and seek redress for their grievances. Their success continued to depend largely upon their own entrepreneurial initiative. Workers might seek their favours,
often as a last resort. But if they were to swell their clientele or expand their constituencies, they had to find a more effective way of making their presence felt in the
workplace.
By the early 1920s, partly in response to the wave of strikes between 1918 and 1922 which affected most parts of the country, some employers
30 D. Chakrabarty, 'Sasipada Bannerjee: A Study in the Nature of the First Contact of the Bengal Bhadralok with the Working Class of Bengal', Indian Historical Review, 2:2 (1976)
339–64.
31 General Department Order no. 3253/62 – Confidential, 15 May 1917, Bombay Confidential Proceedings, 1917, vol. 25, p. 15, OIOC.
32 RCLI, Evidence, Bombay Presidency, GOB, 'Conditions of Industrial Labour in the Bombay Presidency', vol. I, part i, p. 104.

Page 84
33
and, more especially, colonial officials were beginning to talk about the establishment of what they called 'genuine' and 'healthy' trade unions. Their intention was not
necessarily to float 'company unions'; rather the underlying calculation was that the institutionalization of workers' combinations would constrain their propensity to
effect lightning and spontaneous strikes, which seemed to lack leadership and offered no possibility of mediation until relations had broken down completely. Certainly
from 1920 onwards, the Government of India began to nominate a labour representative to the Central Legislative Assembly. Throughout the interwar period, the
positions within legislative assemblies and government councils, the political rewards available to the representatives of labour within the state, were increasing.
The extension of the franchise made the labour vote increasingly difficult to ignore. The Ahmedabad Textile Labour Association played a critical role in the city's
municipal politics.34 Similarly in Calcutta, whitecollar employees, transport workers and selfemployed groups like rickshaw pullers, blacksmiths and cart and carriage
drivers – workers who were most directly affected by the municipal administration – were among the most formally organized workers in the early 1920s, while
several selfstyled trade union leaders, from C. R. Das to Surendra Nath Mallik, were prominent municipal and provincial politicians.35 From 1928, four seats were
reserved for representatives of labour in the Bombay Municipal Corporation.36 Similarly, more nominated and later elected seats became available to labour leaders in
the provincial legislature.
In themselves, these rewards were not sufficient to support numberless political careers. It is not intended to suggest that these developments alone explain why
publicists grew more interested in the cause of labour.
33 Bombay Confidential Proceedings, Judicial, November 1920, vol. 53, pp. 467–84, OIOC; GOI, Home Poll (B), April 1920, Proc. no. 189, NAI; GOI, Home Poll (B), December 1920,
Proc. no. 264, NAI; GOI, Home Poll (B), December 1920, File 291, NAI; GOI, Home Poll (B), December 1920, File 306, NAI; Report of the Industrial Disputes Committee;
Chakrabarty, 'Conditions for Knowledge of WorkingClass Conditions'.
34 ATLA Papers, File 2, Part 2, microfilm copy, reel 10, NMML. Patel, The Making of Industrial Relations, pp. 86–92; see below, ch.8.
35 R. K. Ray, Urban Roots of Indian Nationalism: Pressure Groups and Conflict of Interests in Calcutta City Politics 1875–1939 (New Delhi, 1979), pp. 83–98. Basu, 'Workers'
Politics in Bengal, 1890–1929', ch. 5; Siddhartha Guha Ray, 'Tramworkers of Calcutta: Some Reflections on their Unionisation and Political Experience', Social Scientist, 156 (May 1986),
15–32; Partho Datta, 'Strikes in the Greater Calcutta Region, 1918–1924', IESHR, 30:1 (1993), 57–84; Ira Mitra, 'Growth of Trade Union Consciousness among Jute Mill Workers, 1920–
40', Economic and Political Weekly, 16:44–6 (November 1981), special number, esp. 1840–1; Sanat Bose, 'Industrial Unrest and the Growth of Labour Unions in Bengal, 1920–1924',
Economic and Political Weekly, 16:44–6 (November 1981), special number, 1849–60.
36 RCLI, Evidence, Bombay Presidency, GOB, Vol. I, part i, 145–6; Bombay 1928–29: A Review of the Administration of the Presidency (Bombay, 1930), pp. 24–5.

Page 85
However, it is also true that their effect was to create the space for some trade unionists to establish a certain legitimacy as respectable spokesmen and representatives
of the labour interest. They might thus be called in by employers to settle disputes; they could hope to be consulted by officials on matters of policy; their influence in
the councils of government and the confidence they commanded with employers made them useful allies for jobbers, workers and the lesser leaders of the
neighbourhood. In practice, this position was difficult to achieve and sometimes impossible to maintain. At times, their credentials were challenged by the employers
and the state. At others, their pretensions were rejected by their constituents. More permanently, they were subjected to the competition of rivals. To establish
themselves as important spokesmen for labour, politicians had to organize and control as many unions as possible. Thus, a few individuals in the labour movement
controlled a vast number of unions. N. M. Joshi was the general secretary of the AllIndia Trade Union Congress until the split in 1929 and was president or treasurer
of twelve other Bombay unions. Until he was imprisoned in the Meerut Conspiracy Case, S. H. Jhabvala was connected with at least twenty unions.37 In the later
1930s, P. K. K. Sarma was an officebearer in thirtyseven unions in Madras.38 The list of multiple officeholding in trade unions could be considerably extended both
in Bombay and elsewhere. The largest unions were often built upon extensive, if sometimes uncertain, alliances. Thus Jhabvala said of N. M. Joshi, 'He appointed me
as an officebearer in one of his unions and I appointed him as one in mine', and of another oligarch of Bombay labour, F. J. Ginwalla (who was himself connected with
thirteen local unions), 'We had a constitutional understanding that we would nominate each other wherever we thought it necessary to do anywhere and at any time.'39
By controlling an extensive network which spread itself across several trades, these men acquired important advantages in the conduct of industrial disputes. They
could prevent casual workers in one trade being hired to break a strike in another and crucially they could command greater financial resources. These advantages
were manifested most often on a localscale, but in the mid1920s, N. M. Joshi was inundated with requests from all over India for a share in the funds, collected by
and donated to the AllIndia Trade Union Congress.40 His response was to
37 RCLI, Evidence, Bombay Presidency, GOB, 'Conditions of Industrial Labour', vol. I, part i, p. 112.
38 D. Arnold, 'Labour Relations in a South India Sugar Factory 1937–39', Social Scientist, 65 (December 1977), 16–33.
39 Proceedings of the MCC, Statements Made by the Accused, NonCommunist Series, S. H. Jhabvala, pp. 731, 788.
40 N. M. Joshi Papers, Files 23, 36, 37, NMML. These files indicate the wide range of unions with which Joshi became involved.

Page 86
send small sums to those applicants who could prove their need. But the bulk of this money appears to have been used to provide rations and relief to the Bombay
millworkers during the 1928 general strike.41
Although a few individuals came to dominate the trade union movement at this level, beneath them lay a closely woven web of alignments which ultimately held them
captive. The number of unions they controlled was no reflection of their influence among the workers. Several trade unionists found it difficult to enlist members, mainly
because of the tenuous connection of such organizations with the daily problems of the workplace. Moreover, attempts to organize workers earned the suspicion of
employers, while at the same time the conduct of industrial action intensified their dilemma. If they supported a strike, they earned the displeasure of the employers and
probably the state. If they remained aloof, their paper empires could be swept aside by the militancy of the workers or the intervention of more adventurous rivals.
Thus, N. M. Joshi's Bombay Textile Labour Union was pulled by the momentum of industrial action into the general strike of 1928. With betterestablished political
connections and more to lose from possible failure, Joshi advocated caution. On the other hand, the isolated communists of the Bombay Labour Group calculated that
to lead a strike when the mood of the millworkers was militant offered an invaluable opportunity to establish contacts and extend organization among them.42 During
the course of the strike, the communists were able to burrow beneath the foundations of the Bombay Textile Labour Union and emerged as the dominant force in the
city's labour movement.
The Ahmedabad Textile Labour Association, usually portrayed as exceptional to the general pattern of trade union development, was bedevilled by rather similar
problems. More strikes occurred without its authority than with its leadership and strikes were frequently accompanied by the dismissal and replacement of strikers. It
exercised little influence with the most skilled and most militant section of the workforce, the weavers, who tended to organize outside the union. However, the
machinery for the arbitration of disputes which Gandhi had persuaded the Ahmedabad millowners to accept in 1923 institutionalized the Textile Labour Association as
the main and most effective channel through which workers could express their grievances.43 But the millowners tended to ignore this machinery and preferred the
freedom to make separate settlements, outside the existing institutions of arbitration.
41 The accounts of the Joint Mill Strike Committee in Bombay during the 1928 general strike are to be found in N. M. Joshi Papers, Files 47 and 48, NMML.
43 Patel, The Making of Industrial Relations, pp. 31–63.

Page 87
Labour policy became the cause of open schism within the Ahmedabad Millowners' Association, while at the same time the arbitration machinery threatened to break
down and take away with it much of the union's organization.44
For those trade unions whose existence was not underwritten by the relative tolerance of the employers or the state, there was little alternative but to generate their
connections through action.45 They were forced to pursue strategies of confrontation. Thus, they had to meddle in the daily disputes of the workplace, be willing to
take up even minor grievances and, above all, to demonstrate their ability to secure favourable settlements for their members. This was by no means easy to achieve.
The costs of industrial action were high and militant workers always a target for retrenchment or dismissal. The wider the union's base, the less individual employers
could isolate it at the workplace; yet such a base could only be built up across an industry or an industrial centre by intervening relentlessly at the level of the individual
workplace. However, at certain times, such a strategy of confrontation enabled trade unions to escalate out of the repressive structure of industrial relations. By
intervening in minor strikes, confined to individual departments and mills, these outriders of the trade union movement were able to generate sufficient momentum for
largescale action. Necessarily, they established themselves most successfully during general strikes. Moreover, in relation to their rivals, they had a further advantage.
Since most workers' combinations were effectively excluded from the workplace, trade unions which intervened energetically in minor disputes and assiduously aired
workers' grievances, were able to forge closer and more direct connections with the workplace. Excluded from the workplace, these unions often paid closer attention
to building up connections in the neighbourhood than some of their counterparts. They were able to use these connections to prevent blacklegs usurping the jobs of
strikes, to persuade grain merchants to extend credit to their members, and in various ways to sustain industrial action despite the overwhelming obstacles it
encountered.
In the aftermath of the 1928 general strike, the Girni Kamgar Union was able to formalize these connections, built up through industrial action, both in the
neighbourhood and in the workplace, through their union centres and mill committees.46 The repressive onslaught upon the union and its allies and supporters,
orchestrated by the employers and the
44 G. L. Nanda to S. G. Banker, 23 March 1933, ATLA Papers, File 40, pp. 77–83, microfilm copy, reel no. 1, NMML. These issues are more extensively treated in ch. 8, below. See
also Patel, The Making of Industrial Relations.
46 Proceedings of the MCC, Statement submitted by S. A. Dange, pp. 2498–557. See ch. 4 below.

Page 88
state, which followed the next general strike in 1929 swept these institutions away. But between 1932 and 1934, the Girni Kamgar Union once more asserted itself
through minor strikes which breathed life into these old, withering linkages and culminated in the general strike of 1934.47 For trade unions like the Girni Kamgar
Union, faced with exclusion at the level of the individual mill, lacking the goodwill of the employers, and sometimes subjected to repression by the state, it was clear
that they could not consolidate through patronage the linkages forged in action. To function as a trade union at all, they had often to adopt a stance of continued
opposition to the employers and the state. In this way, the structure of repression which trade unions were forced to confront impelled them towards increasing and
sometimes frenetic activism simply in order to survive.
Trade union leaders who adopted this interventionist style were in the late 1920s known to the jute millworkers of Bengal as 'the strike babus'.48 Moni Singh, the
veteran communist leader of Bengal, and subsequently chairman of the Communist Party of Bangladesh, has left us an instructive, if somewhat surreal, account of their
political style.49 Having accepted 'the contention' of the time that 'the workers were the most revolutionary class', Moni Singh left for Calcutta in 1926 'to start work
among them'. Almost as if he was enacting the fondest fantasies which intelligence officers entertained about 'Bolsheviks', he set up an office on Clive Street 'with the
name, Oriental Trading', presumably as a front for what he hoped would be his revolutionary activities. His primary 'objective', of course, was 'to find out how one
could join the workingclass movement'. Here he was at first thwarted. Although he met a number of 'communistminded' activists, Moni Singh 'could find no
opportunity to participate in the workers' movement'. Then 'one day in 1928', two whole years later, comrades Gopen Chakraborty and Dharani Goswami came to his
house to announce 'a tremendous opportunity to get involved in the workers' movement'. K. C. Mitra, once a railway clerk, now a sadhu, and known as Jatadhari
Babu in deference to his matted hair, had organized a strike in the Railway workshop at Lilooah. Moni Singh 'rushed' with his comrades to the Lilooah Railway Union
Office and arrived the following day. 'Just then a nonBengali worker came and asked in Hindi' for Jatadhari Babu. The millhands of the Keshoram Cotton Mills in
Metiaburuz had gone on strike and, said the worker, they wanted him to represent them. 'Jatadhari
47 GOB, Home (Special), Files 543 (48), 543 (48) K, 543 (48) L, 543 (48) J, 543 (48) E, MSA.
48 Mitra, ''Trade Union Consciousness among Jute Mill Workers', p. 1841.
49 Moni Singh, Life is a Struggle, tr. Mrs Karuna Banerjee (New Delhi, 1988), pp. 17 ff. I am grateful to Subho Basu for drawing this passage to my attention.

Page 89
was not in the office at the time.' Jatadhari Babu was a 'reformist' and Moni Singh and his comrades were scarcely that. Undaunted and now fretting for action, they
presented themselves to 'the nonBengali' weaver, Rehman, as the sadhu's followers, offered on these false pretences to help in his absence, and accordingly
accompanied him to the mill.
Metiaburuz was quite far from that place. We three arrived there. The workers were very happy to see us. Most of the workers in the Keshoram Mills were nonBengali, people
from Bihar and UP. The workers began to explain to us in chaste Urdu, why they had gone on strike. It was a language we did not know. Neither did we have any experience of
textile mills. Gopen Da had a smattering of Hindi. Even he failed to understand much. Seeing our blank looks the workers realized that we could follow nothing. They said, 'Come to
the Secretary Sahib'. Though called secretary, he was actually the manager. The manager was an educated Marwari gentleman, young in age. We considered it better to see the
manager, because we could not understand the problem at all . . . We told him we were nationalist activists. We had come to find out the cause of the strike in the mill. The
secretary welcomed us and asked us to sit down.
The point he made was that previously the weavers used to be paid by the weight . . . Now they would be paid by the yard. This system is followed all over the world. The Birlas
were noble people. They could not possibly harm the workers. The workers were fools, they had gone on strike without understanding anything. Though we had no knowledge
about these things, we came to a conclusion on the basis of our principle that the bourgeoisie could never do anything good for the workers. On this basis, we argued blindly. For
we did not have the least idea about wages in a mill.50
Subsequently, posing as 'a nationalist activist', Moni Singh sought out the weaving master at the Bangalaxmi Mill at Serampore to seek guidance on the intricacies of
the various methods of wage calculation. Fortified by this knowledge, he returned to Metiaburuz where he set up a union and 'within a few months' had acquired 'a
working knowledge of Hindi and Urdu'.51 Of course, it is not intended to suggest that the radicals were alone in encountering the problems created for publicists by the
social and linguistic gulf which separated them from the workers, but rather it is important to recognize that to secure a foothold at the workplace, when they were
excluded from it, trade unionists often required considerable resources of inventiveness, imagination and ingenuity.
The changing arenas of labour politics also helped to facilitate strategies of confrontation. Increasingly, in the interwar period, employers' associations became
necessary instruments in attempts to influence government fiscal, monetary and financial policies. During the slump of the early 1920s and the early 1930s some
centralized attempts, only partially
50 Ibid., pp. 21–2.
51 Ibid., p. 24.

Page 90
52
successful, were made to control production, maintain price levels and sometimes to reduce competition. However, in the cotton textile industry as much as in jute,
the conditions and character of production as well as the methods and objectives of labour management varied considerably from mill to mill. Necessarily, individual
mill managements remained wary of the centralizing initiatives of their employers' association. It cannot be said that employers' associations successfully centralized the
power to formulate and enforce policies across each industry. However, it is clear that individual mill managements were more closely affected by the policies of their
associations in the interwar period than they were before 1914. As policies were formulated for the entire industry, labour responded at the same level. This facilitated
the generalization of disputes, which in turn opened the door to the more adventurous strategies of some trade unions. Unlike Bombay, Calcutta was relatively free of
general strikes. Yet the general strikes in the jute industry in 1929 and 1937 followed policies regarding increased hours of work in the case of the former, and
rationalization in the case of the latter, formulated not by individual employers but by the Indian Jute Mills' Association for the industry as a whole.53 In Bombay, the
general strike of 1928 became general when it was clear that rationalization schemes introduced in particular mills had begun to spread throughout the industry and
when it was recognized that resistance at the level of the individual mill had failed to halt their progress.54 It was frequently at moments of extensive, industrywide
action that trade unions like the Girni Kamgar Union were able to establish themselves or retrieve their position in labour politics.
This argument which differentiates between two fairly general styles of
52 A. K. Bagchi, Private Investment in India 1900–1939 (Cambridge, 1972), chs. 6, 7 and 8; On the cotton textile industry, see Chandavarkar, The Origins of Industrial Capitalism,
pp. 239–396; on the jute industry, see O. Goswami, Industry, Trade and Peasant Society: The Jute Economy of Eastern India, 1900–1947 (Delhi, 1991); O. Goswami, 'Sahibs,
Babus and Banias: Changes in Industrial Control in Eastern India, 1918–1950', JAS, 48:2 (1989), 289–309; Goswami, 'Collaboration and Conflict'; O. Goswami, 'Then Came the
Marwaris: Some Aspects of the Changes in the Pattern of Industrial Control in Eastern India', IESHR, 22:3 (1985), 225–49; and Chakrabarty, Rethinking Working Class History, ch.
1.
53 P. Saha, History of the Workïing Class Movement in Bengal (New Delhi, 1978), pp. 114–31, 148–70; Karnik, Strikes, pp. 234–48, 284–6; on the 1929 general strike, see Basu, 'Workers'
Politics in Bengal, 1890–1929', ch. 6, esp. pp. 208–21; Chakrabarty, Rethinking Working Class History, pp. 196–205; O. Goswami, 'Multiple Images: Jute Mill Strikes of 1929 and 1937
Seen Through Other's Eyes', MAS, 21:3 (1987), 547–83.
54 Proceedings of the BSEC, vol. I, pp. 71–2, MSA. On the nature of rationalization and attempts to implement it, see Chandavarkar, The Origins of Industrial Capitalism, pp. 335–96;
and Rajnarayan Chandavarkar, 'Workers' Resistance and the Rationalization of Work in Bombay between the Wars', in D. Haynes and G. Prakash (eds.), Contesting Power: Resistance
and Everyday Social Relations in South Asia (Berkeley and Los Angeles, 1991), pp. 109–44.

Page 91
trade union politics needs to be qualified. Of course, no simple distinction between strategies of patronage and confrontation are tenable. Trade unions which restricted
their dealings to the councils of state and the offices of the managers discovered that the promise of future concessions or the lure of patronage were not enough to
hold their organization together, let alone allay immediate grievances. Similarly, the linkages of direct action had to be consolidated. There was little point in effecting
industrial action if it secured no concession or merely invited repression. It was crucial for all unions, excluded from the workplace, to maintain their connections with it.
The crux of the matter lay in getting the minor as well as the major grievances of the workers settled. As one labour leader pointed out, 'If workers find that the Union
does not function properly with the Millowners' Association, and that their legitimate grievances are not redressed then naturally their influence goes down.'55 At some
point or other, company unions had to support a strike and yet try to protect their favoured status, while the most radical unions had to face the dilemma of how to
reflect the militancy of their members without permitting it to go too far. Thus the Girni Kamgar Union, too, floundered at times between conciliation and opposition.
Despite the hostility of the Congress Government of 1937–9, it seemed reluctant to initiate industrial action. After 1942, it performed a major holding operation on
workingclass politics. By 1947, its support had declined as a result and its dominant position in the industry was soon eclipsed.56
If the structure of dominance and control at work decisively influenced the nature of trade union activity, it also shaped the forms of action. Certain issues, under
existing conditions of labour management, could only be confronted through general strikes which encompassed an entire industry and which admirably suited a militant
style of trade union politics. However, historians have sometimes taken the frequency of lightning strikes and what they perceived as the riotous, even 'violent' nature of
industrial action to be characteristic of labour politics in the early stages of industrialization.57 It is difficult to reconcile this picture of primordial consciousness with the
more sophisticated political consciousness manifested in the extended, wellorganized general strikes by the same workers in the same centres of industry. Once again,
the beginnings of a more plausible explanation will rest in the considerable risks which
55 Proceedings of the Court of Enquiry, File 6, R. R. Bakhale, MSA; see also RCLI, Evidence, Bombay Presidency, The Bombay Textile Labour Union, vol. I, part i, pp. 352–56.
56 Fortnightly Reports, 1937–47, L/P + J/5/163, OIOC; GOB Home (Special), File 543 (13) B (3) of 1940, 543 (13) B (4) of 1941–3, 543 (13) B (5) of 1943–5, MSA.
57 For instance, Arnold, 'Industrial Violence'; Chakrabarty, 'On Deifying and Defying Authority'. See ch. 5 below.

Page 92
industrial action entailed. In view of the powerful constraints upon the initiation and conduct of industrial action, for workers to warn their employers in advance that
they would strike was also to allow the latter the time and opportunity to subvert industrial action and dismiss its protagonists. As R. R. Bakhale, a moderate trade
union leader who was never throughout his long and illustrious career willingly involved in leading a strike, informed a Court of Enquiry in 1929, there was often no time
for workers to notify the employers of their intention to strike. 'The conditions in the textile industry are such', said Bakhale, referring to the widespread opposition
among employers to trade union organization, 'that lightning strikes, whatever they may be, are inevitable.'58 Indeed, the definition of lightning strikes – 'whatever they
may be' as Bakhale put it – was often the subject of debate and in a sense the object of struggle between employers, trade unions and workers. Wildcat strikes, then,
were not an expression of uninhibited spontaneity; more frequently, they entailed strategic calculation.
This element of calculation did not relate simply to wages and working conditions. The forms which strikes sometimes took suggest that they were staged as an appeal
to a higher level of authority.59 Lightning strikes, when workers stopped their machines and assembled briefly in the mill compound in protest against their treatment by
a supervisor, could be designed to invite the intervention of the manager; the daily processions and mass rallies during general strikes not only maintained morale but
more widely asserted the importance of their cause; when strikers stoned mill after mill in order to close them down, this was in part a response to blacklegging and the
risks consequent upon a partial stoppage, but it sometimes also drove officials to persuade employers to negotiate. Certainly, the forms of struggle as they developed
and changed deserve investigation. It is possible that forms of collective action manifested in particular social contexts emerged from and evolved out of an older
experience of struggle. In any event, the spectacular, and sometimes riotous, character of industrial action was in part adapted to the audience for which it was
performed. Faced with the determined opposition of factory managers, workers sought thereby to attract the notice of the employer and more usually the state in order
to have their grievances heard and redressed. The colonial state did not by any means constitute only the audience in the theatre of industrial relations. It frequently
intervened as an active participant; occasionally, it performed the leading role. For one thing what began as a dispute between employers and workers could rapidly
acquire a public dimension. When workers left their
58 Proceedings of the Court of Enquiry, 1929, R. R. Bakhale, File 7, MSA.

Page 93
60
factories and impinged upon the social spaces of the street, they were often quickly perceived to represent a considerable threat to the social order.
Since the nineteenth century, the police had intervened and sometimes even arbitrated in industrial disputes.61 Well into the twentieth century, even as negotiations were
conducted at more centralized levels of the administration, the police continued to act at more discrete levels, dissuading workers in their chawls from joining a union,
attending a meeting or striking. They might even attempt to persuade shopkeepers to withhold their credit to strikers.62 More overtly, they intervened only to the
detriment of the workers' interest where earlier they may occasionally have defended it. At times, police action defeated pickets and protected or aided blacklegs and
often contributed to the suppression of their demands and destruction of their organizations. Although police action did not constitute as much of a permanent and
structural obstacle to trade union development as the policies of employers, changes in police practice did affect the way in which the working class came to perceive
the state and decisively influenced the development of their political consciousness.63
The presence of the state also began to be felt in the growing body of legislation which governed industrial relations. Laws were passed to compel the registration of
trade unions, thus opening their books to official scrutiny, to install procedures for the conciliation and arbitration of disputes and to require from workers notice of their
intention to strike, thereby differentiating between legal and illegal stoppages, while providing employers with the opportunity of subverting industrial action before it
was effected. From the mid1930s, a series of measures were taken which were applicable specifically to Bombay and designed to isolate the communist Girni
Kamgar Union further from its members and sympathizers in the mills.64 The Trade Disputes Act of 1934 made provision for the appointment of a government Labour
Officer whose function would be to receive workers' grievances and seek to remedy them and thus, while addressing the avoidable causes of popular discontent, they
could encroach upon, and even, perhaps, usurp the traditional role of the trade union.65 Later, in 1938, the Congress Ministry in Bombay passed further
60 See ch. 5 below.
61
S. M. Edwardes, The Bombay City Police–A Historical Sketch 1672–1916 (London, 1923). See ch. 6 below.
62 RCLI, Evidence, Madras Presidency, B. Shiva Rao, vol. VII, part ii, p. 202.
64 Note by Home Member, 16 May 1934, GOB Home (Special), File 543 (48) L, p. 26, MSA; BMOA Annual Report 1934, pp. 22–4.
65 A. W. Pryde, 'The Work of the Labour Office', in C. Manshardt (ed.), Some Social Services of the Government of Bombay (Bombay, 1937).

Page 94
legislation which made conciliation compulsory and stipulated that trade unions, to be representative, had to count among their members a quarter of the workforce.66
It also laid down elaborate procedures of notice, negotiation, conciliation and arbitration as necessary stages preceding a strike and thus robbed the workers of any
advantage they may have gained from determining the timing of a strike. Further legislation in 1946 was passed to encourage the development of a single union for each
industry, on conditions which favoured the Congress trade union, the Rashtriya Mill Mazdoor Sangh.67 Of course, it is not intended to suggest that legislative fiat alone
could undermine the Girni Kamgar Union after 1938 but there can be little doubt that this spate of measures served to severely narrow its freedom of manoeuvre.
Outside Bombay, industrial disputes legislation was less marked by this heavily repressive character. For the most part, and in Bombay perhaps at least until 1940, and
arguably even later, the manipulation of, as well as the control exercised through, the wage relationship was a more persistent and formidable obstacle to combination
than the laws of the state. Moreover, the colonial state was not committed to the unilateral repression of industrial workers,68 however nervous it may have been of the
development of class struggle within the context of nationalist agitation. On the contrary, there are signs that in the aftermath of the civil disobedience movement, the
Government of India, reassured that Indian business interests would hold Congress radicalism in check, began to court labour's friendship and initiate some protective
legislation, even to the detriment of Indian business interests.69 Labour in its turn could now be encouraged to step forward as colonial collaborators. The legacy of
these initiatives was witnessed in the first phase of provincial autonomy until the Congress began in 1938 to set the process in reverse.
These sometimes insuperable obstacles which workers faced in forming a trade union, establishing its presence at the workplace and coordinating its activities across
an industry even in the same locality provides an indication of the scale of the problems which engulfed attempts to create a national framework for trade union
organization.70 If at the level
66 GOB Home (Special), Files 550 (24) of 1938, 550 (25) III A, MSA; ATLA Papers, File 3, part 2, microfilm copy reel 9, NMML; AITUC Papers, File 59, NMML.
67 Morris, Industrial Labour Force, pp. 193–5; R. C. James, 'Trade Union Democracy: Indian Textiles', The Western Political Quarterly, 11:3 (1958), 566–72.
68 This assumption is widely made: for an example, see Arnold, 'Industrial Violence'.
69 The signs were to be found in the greater willingness of the state to countenance limited welfare legislation, see Labour Gazette, 16:10 (June 1937) 760–3; ibid., 16:11 (July 1937),
846–50; ibid., 16:12 (August, 1937), 926–37. The anxieties of the Bombay millowners in this respect can be followed in the Minute Books of the Committee of the BMOA, 1934–8, Office
of the BMOA.
70 N. M. Joshi, The Trade Union Movement in India (Poona, 1927).

Page 95
of particular industries and cities, trade unions operated at one level removed from the workplace, at the national level they rarely, if ever, became effective bargaining
agents. The AllIndia Trade Union Congress was formed in 1920 in order to elect delegates to represent Indian labour at the International Labour Conference.71 Five
years later, the president of the AITUC described his own role in the trade union movement as 'nugatory'.72 This selfdeprecation was unwarranted since scarcely
anything more positive could be said about the organization over which he presided. In the mid1920s, it functioned around the redoubtable figure of N. M. Joshi,
principally as a treasury to which real unions brought their begging bowls.73 The attempt by the communists to capture the AITUC in the late 1920s enlivened its
politics. Following the collapse of the Bombay Textile Labour Union, N. M. Joshi attempted to come to terms with V. V. Giri and B. Shiva Rao, 'moderates' or
'reformists' concerned largely with South Indian labour, in a bid to outflank the communists in the AITUC.74 The culmination of these manoeuvres was that the AITUC
split at its Nagpur session in 1929, when Joshi, Giri and their friends left to form a rival Indian Trades Union Federation.75 Significantly, the issue which split the
AITUC was to which international federation it should affiliate, despite the fact that its national existence was largely fictional. Thus, in 1928 and 1929, the AITUC
witnessed some excitement. However, even at this period of its greatest activity, its president, Jawaharlal Nehru, complained that the job bored him. He wrote to D. B.
Kulkarni, the rival candidate for the presidency at the previous Trade Union Congress, to tell him that he had not wanted to take up the office. 'Owing to my
absence' [at the Jharia Congress], he confessed, 'I was elected and later I did not know how to get out of it . . . I am glad however that my period in office is coming
to an end.'76 After this brief flurry, the national unions settled down to a further period of inactivity. In 1932, the secretary of the newly formed Trade Union Federation
reported that the organization had not found it possible 'to concentrate much of our time to questions directly connected with the working class', let alone 'to give
71 AITUC – Fifty Years On: Documents, introduction by S. A. Dange (AITUC, New Delhi 1973), vol. I.
72 C. F. Andrews to N. M. Joshi, 18 September 1925, AITUC Papers, File I, NMML.
73 N. M. Joshi Papers, Files 2, 6, 13, 14, 15, 21, 25, 30, 31, 33; AITUC Papers, File I, NMML.
74 B. Shiva Rao to R. R. Bakhale, 10 February 1929, N. M. Joshi Papers, File 36, NMML; see also V. V. Giri, My Life and Times, vol. I (Delhi, 1976), pp. 67–8, 46–75.
75 AICC Papers, File 12 of 1929, NMML; B. Shiva Rao, The Industrial Worker in India (London, 1939) pp. 149–77; C. Revri, The Indian Trade Union Movement: An Outline History
1880–1947 (New Delhi, 1972), pp. 115–71.
76 J. Nehru to D. B. Kulkarni, 10 September 1929, AICC Papers, File 16 of 1929, p. 113, NMML.

Page 96
77
direct help to the affiliated unions when they had been engaged in struggles against their employers'.
By the late 1930s, attempts were being made to patch up the differences of the previous decade.78 To a large extent, these developments at the national level were
responses to local imperatives. If the split in 1929 can be explained largely in terms of the trade union rivalries in Bombay city, the attempts at unity in the later 1930s
were the result of the repression of the communists, the expectations aroused by the prospect of gains to be made from the newly elected Congress Government and,
finally, the wisdom, indeed necessity, of all trade unions in Bombay to close their ranks against the threat of the passing of the 1938 Industrial Disputes Bill. It was only
after the 1950s that trade unions acquired a national framework, but this was in part the outcome of their close connections with political parties and their considerable
importance for elections, for the first time based on universal franchise.
The Congress Party, which may have taken a more active role in creating such a national framework, remained for the most part aloof from trade union politics. It
never regretted its connections with the Ahmedabad Textile Labour Association (ATLA); in fact it guarded these connections jealously, never allowing the ATLA to
join up with any of the national federations.79 Yet having developed strong and intimate links with the ATLA, it neither wanted nor sought any further trade union
affiliations. To associate too closely with a particular class interest was for the Congress to sacrifice other more powerful groups which coalesced within it and to
surrender its claim to solely represent the nation as a whole.80 As Nehru wrote when he presided over the AITUC, 'Of course, everyone knows that the Congress is
not a labour organization . . . To expect it to act as a pure labour organization is a mistake. The National Congress is a large body comprising all manner of people.'81
In the 1930s, the Congress Socialist Party professed a more direct interest in the working class and played a more active role in trade union organization.82 But their
activities were conducted, not as organic parts of the national
77 Quoted by Revri, Indian Trade Union Movement, pp. 195–6.
78 N. M. Joshi Papers, File 75; AICC Papers, File LI/1935 and PL2/1938; AITUC Papers, Files 59 and 60, NMML.
79 Anasuya Sarabhai to General Secretary, Textile Labour Union, Bombay, 21 April 1927, N. M. Joshi Papers, File 32, p. 75, NMML; P. P. Lakshman, Congress and the Labour
Movement in India (Congress Economic and Political Studies, no. 3, Economic and Political Research Department, AllIndia Congress Committee, Allahabad, 1947).
80 See ch. 8 below.
81 J. Nehru to D. B. Kulkarni, 10 September 1929, AICC Papers, File 16 of 1929, pp. 111–13, NMML.
82 GOB Home (Special), Files 800 (75) A to 800 (75) A VII of 1938–42 (7 files); File 800 (75) AB of 1935 and File 800 (75) D(I) of 1938, MSA.

Page 97
body, but, in their private capacity, as one group among the 'all manner of people' which the Congress comprised. Consequently, by the mid1930s, organizing and
representing labour had become a specialist activity for politicians and publicists.83
Labour organizations did not officially participate in the Congress agitations of the early 1920s, early 1930s or early 1940s. Gandhi and the Congress sometimes
actively discouraged workers from entering the lists. But the workers did not always follow their leaders' bidding, least of all in Ahmedabad, and even in Bombay
industrial workers, with varying degrees of enthusiasm at different times, participated in the nationalist agitations.84 The depth and character of workingclass
nationalism is not easily assessed. Its particular social meaning, in the context of the highly localized political experience of workers, is difficult to determine. Clearly,
this nationalism did not arise out of commitment to a territorial principle or even from the concern that the interests of the Indian people were homogeneous. It was
probably fed by the racial conflicts between European or Eurasian foremen and Indian workers on the railways.85 It appears to have been more marked in Bengal
where capital was overwhelmingly dominated by Europeans and the jute mills were both owned and managed by them.86 The increasing realization that the colonial
state in the 1920s and 1930s was hostile and repressive in its attitude to labour politics encouraged workers to identify with movements apparently directed against it.87
Conversely, the increasingly vociferous nationalist campaigns also helped workers to crystallize their antagonism towards the state and thereby to identify their own
solidarities.
The weakness of trade union organization in India has often been attributed to 'the early stages of industrialization' and the 'precapitalist mentalities' of the workers.
But it would be difficult to make sense, within these evolutionary terms, of the impressive solidarities which workers forged at the same time or the fiercely disputed
and protracted strikes
83 See ch. 8 below.
84
Patel, The Making of Industrial Relations; see ch. 8 below, where these issues are addressed more fully.
85 Arnold, 'Industrial Violence'; Lajpat Jagga, 'Colonial Railwaymen and British Rule: A Probe into Railway Labour Agitation in India, 1919–1922', in B. Chandra (ed.), The Indian Left:
Critical Appraisals (New Delhi, 1983), pp. 103–45; Proceedings of the MCC, Defence Statement, K. N. Joglekar, pp. 1766–965.
86 Chakrabarty, Rethinking Working Class History, ch. 5; Bagchi, Private Investment in India, ch. 8; Tomlinson, 'Colonial Firms and the Decline of Colonialism in Eastern India'; Basu,
'Workers' Politics in Bengal, 1890–1929', ch. 3. On Kanpur, Chitra Joshi, 'Worker Protest, Managerial Authority and Labour Organization: Kanpur Textile Industry', NMML, Occasional
Papers on History and Society, no. 27 (1985), pp. 10–11; Zoe Yolland, Boxwallahs: The British in Cawnpore, 1857–1901 (Norwich, 1994).
87 This argument is elaborated in ch. 4 below; see also Chandavarkar, The Origins of Industrial Capitalism, especially ch. 9.

Page 98
which they repeatedly sustained. The nature of the labour movement in India cannot be deduced either by measuring the level of industrialization or by the divination of
a particular social consciousness. Rather, its growth and its limitations were shaped by the political conditions of its development.
The formation of trade unions and their methods of operation can most fully be grasped in relation to the formidable obstacles to combination. The fragility of trade
union organizations, their tendency to collapse and disappear as suddenly as they formed, and their failure to entrench themselves within the framework of industrial
relations were largely the consequence of the sustained attempts by employers to repress them at the point of production. The apparent volatility of strikers, their
seeming propensity to take 'lightning' action and the violence which sometimes occurred between strikers and blacklegs were often symptoms of the repression of
labour organization. It is not intended to suggest that the emergence of trade unions would have directed workers towards apparently more 'responsible' forms of
negotiation and bargaining but rather that the repression by the employers infused the politics of the workplace, in every aspect, with the imminent possibility of
'violence'. As employers sought to repress workers' combination, sometimes with the calculated support and sometimes the unwitting complicity of the state, they
began to define the arenas within which labour resistance took shape. In turn, having to operate within these arenas shaped the nature of trade union organization and
forms of action. Undermined within the workplace by the sustained hostility of the employers, workers' combinations had to draw sustenance from the social relations
of the neighbourhood. Frequently dismissed when they expressed their own grievances, workers had to seek the intervention of 'outsiders', intermediaries and political
agents. Denied the means of negotiating with their employers, workers and their representatives were sometimes forced to adopt a strategy of confrontation which
imposed its own pattern of labour relations and collective bargaining. Indeed, the militancy and volatility of the workforce often became a means of insuring against the
considerable sanctions imposed upon strikers. Conversely, if trade union weakness belied the solidarities which workers forged, their industrial and political action did
not necessarily find enduring expression in trade union organization. The formation of trade unions offers no index of the nature of workingclass consciousness. No
simple, direct and necessary relationship linked consciousness, organization and action.
The history of the labour movement in India has commonly been treated as a special case, which was unlikely, precisely because of its economic backwardness and its
cultural peculiarities, to release materials

Page 99
for thinking more generally about class formation and consciousness. In fact, the development of the labour movement in India suggests that the origins and
development of trade unions in general calls for a political, rather than a specifically social or cultural, explanation. The development of trade unions was largely
conditioned by, perhaps dependent upon, the willingness of employers and the state to tolerate them. It neither signified the development of a particular stage of class
consciousness nor did it anticipate the rise of socialism. The quotient of capitalist tolerance for labour organization has usually been determined by the outcome of the
changing relations between workers, trade unions and the state. Significantly, it has depended upon the political leverage which workers were able to gain upon state
power, and thereby, through political pressure and negotiation, to raise the threshold of capitalist tolerance. States have been opened up to a measure of workingclass
influence when the coalition of interests by which they are constituted has recognized the need to expand and diversify their social bases in order to secure their
dominance, sometimes by widening the franchise and extending the representation they afford to the working classes within the institutions of the state. Of course, it is
the threat of industrial and political action which has, most commonly, forced this recognition upon ruling elites. Conversely, as the working classes have consolidated
their position within the alliances which constituted the state, so workers' struggles and their institutional and political forms have been able to exercise greater leverage
upon it. In this sense, the growth of trade unions and the development of labour movements may be more plausibly explained in terms of the stake which the working
classes were allowed in the nation and the political entitlements which they were able to wrest from the state rather than the languages of capitalism and class alone.

Page 100
4—
Workers' Politics and the Mill Districts in Bombay between the Wars
Between the wars, the development of a labour movement in Bombay reflected a growing polarization in social and political relations in the city. This period, which
saw an intensification of social conflict, also witnessed changes in the character of industrial action. Until 1914, strikes in the cotton industry were largely confined to
particular departments and mills; increasingly, after the war, they were coordinated across the industry as a whole. Rising prices and unprecedented profits which
accompanied the postwar boom led to the demand for higher wages supported by two general strikes. In the mid1920s, as the industry's markets slumped, attempts
to cut wages were once again strongly resisted. With a slight improvement in their fortunes in the later 1920s, the millowners introduced 'rationalization' schemes; for
the workforce this meant more work, less wages and higher chances of unemployment. Between April 1928 and September 1929, two general strikes crippled the
industry for about eleven months, and the extension of these schemes and a further round of wage cuts led to another strike wave in 1933–34. Apart from several one
day closures, eight general strikes occurred in the industry between 1919 and 1940. The impact of this militancy was felt not only in other occupations in Bombay but
also in other industrial centres, such as Sholapur and Ahmedabad. As Bombay became the scene of militant workingclass action in India, its labour movement, under
communist leadership since 1928, acquired an explicitly political direction.
Yet even as strikes were coordinated across several mills, no stable trade union growth occurred until the mid1920s; subsequently, the unions remained weak,
vulnerable and often ineffective. To some contemporaries, this suggested the existence of concealed sources of leadership within the workingclass communities;1 to
most, it indicated the
1 After the 1919 general strike, the Government of Bombay believed that 'while the workers had no accepted leaders' the conduct of the dispute 'appeared to indicate the probability
of some controlling organisation'. J. Crerar, Secretary to GOB, to Secretary to GOI, Home, Delhi, 7/15 February 1919, in Bombay Confidential Proceedings, vol. 46, 1919, OIOC. This
riddle of leadership bemused the Bombay Chronicle, too, in 1924 when it

Page 101
malign intervention of the political agitator. More recently, our historiographical common sense has been overtaken by such notions as the political immaturity and rural
passivity of Bombay's workers. Historians have thus been concerned with the 'survival' of the 'preindustrial' characteristics of the workforce, rather than their
rationality within an industrial context. The traditional loyalties of the working class, in this view, obstructed the development of 'modern' trade unions. These accounts
have assumed that the development of labour politics in Bombay can best be understood in the light of existing models of an 'early' factory labour force. In the context
of the Bombay textile industry, this remains a problematic assumption. It is not satisfactory either to portray a factory labour force which had been in existence for
about half a century by 19182 as if it was in a 'nascent' state of formation, or to analyse its history as if it were in transition towards the product of another historical
experience, or indeed to measure its development against some universal paradigm of 'class'.
It is perhaps by focusing too exclusively on the sphere of the workplace, by confining their model of social consciousness to what was reflected by trade union
development, that historians have overlooked the extent to which workers were active in the making of their own politics. The dynamic of labour politics in the inter
war years, in one view, was the struggle between politicians, attempting to mobilize labour, and their traditional leaders, the jobbers in the cotton mills.3 The motive
force behind labour militancy is thus located outside the realm which workers controlled: their political (and moral) choices, it would appear, were consistently being
made by others. In such a view, the history of the working class becomes interchangeable with the history of their leaders, trade unions and political parties. As a
result, the impact of labour militancy upon the development of labour politics in Bombay between the wars has been neglected; instead, the emphasis has rested upon
the role of the nationalist and communist agitator and the role of the jobber, the agent of labour recruitment and control.
However, the weakness of trade union organizations did not prevent Bombay's workers from mounting an effective and sustained defence of
commented: 'it is absurd to suppose that the men are lacking in leadership . . . it is clear that there is good sound leadership among them somewhere'. Bombay Chronicle, 21
February 1924.
2 The first cotton mill was built in 1856, see Morris, The Emergence of an Industrial Labour Force, p. 17.
3 R. Newman, 'Labour Organisation in the Bombay Cotton Mills, 1918–1929', unpublished D. Phil. thesis, University of Sussex, 1970; Kooiman, 'Jobbers and the Emergence of Trade
Unions'. Morris attributes the growth of labour militancy to 'the role of the middleclass intellectual appearing in his first fullblown opposition to British rule', in Industrial Labour
Force, p. 180.

Page 102
their own interests. To understand the development of the perceptions and actions of Bombay's workers, therefore, we need to examine not only the social
relationships of the workplace but particularly the context in which workers lived outside it. Since the earliest inquiries into the conditions of factory labour in Bombay,
the interconnection between the spheres of workplace and neighbourhood have been frequently mentioned; but its implications for industrial politics have surprisingly
remained neglected.
Customarily, the heterogeneity and cultural sectionalism of the working class is identified with the neighbourhood; yet in Bombay it provided an indispensable base for
industrial action. Far from being herded peacefully by their jobbers and neighbourhood 'leaders', workers often acted to constrain them. The momentum of industrial
action was not merely provided by men of prominence; sometimes it was maintained against them. Without organization and action in the neighbourhood, it is doubtful
whether the general strikes could have been sustained. At the same time, the conduct of industrial action in the public arena of the street and the neighbourhood
necessarily generalized the disputes of the workplace, at times brought workers into conflict with the state and created an explicitly political dimension for their struggle.
While it would be misleading to portray Bombay's workers as a 'revolutionary proletariat' or indeed to play down the important tensions and antagonisms between
them, it is in terms of the political culture of the workingclass neighbourhoods that the scale of industrial action and the ascendancy of the communities can be
explained.
From the late nineteenth century, a distinctly workingclass district began to emerge in Bombay. Already in the 1850s, an official investigator had noticed the growing
social and cultural distance between the mass of the population and 'the educated and more influential classes (whether Native or European) of our community'. 'The
principal acquaintance of these [influential] classes with the Native Town', he wrote, 'is generally formed by traversing the Kalbadevee or Girgaum bazaar roads, in
going from the country to the Fort, or from the Fort into the country; and of all the densely peopled districts lying behind these great thoroughfares, they generally
know as little as they do of the interior of Africa.'4 The inception of industry added a further dimension to the city's social geography. An overwhelming majority of the
common mills came to be situated in the
4 H. Coneybeare, Report on the Sanitary State and Requirements of Bombay, Selections from the records of the Bombay Government, new series, vol. XI, (Bombay, 1855), p. 2.

Page 103
three wards to the north of the old 'native town'. Increasingly the working classes, fairly evenly dispersed in the native town of the midnineteenth century, crowded into
this area. By 1925, 90 per cent of the millworkers lived within fifteen minutes' walking distance of their place of work.5 To its inhabitants, this area came to be known
as Girangaon, literally the mill village. As the labour movement gathered momentum between the wars, Girangaon ceased to be a mere geographical entity; rather it
came to represent an active political terrain.
The physical structure of the workingclass neighbourhoods imparted a certain public quality to its social life. The landscape of the mill district was dominated by
ramshackle, jerrybuilt chawls packed closely into the land between municipal thoroughfares. A survey conducted in 1921 discovered that 27 per cent of the
population in Parel and 33 per cent in Umerkhadi lived in rooms containing six or more persons.6 Another investigation conducted in the mid1930s found over 35 per
cent of families of 'untouchable' workers sharing a single room with at least one other family, while over 63 per cent lived in a single room.7 'Every sixth person in the
city', it was reported in 1939, 'lives in conditions which are prohibited even by the existing antiquated law'.8 The extent of overcrowding brought about by high rents,
housing shortages and low wages meant that the inhabitants of the chawl spilled over into the courtyard of the wadi and the street.
The importance of the street did not derive simply from the fact that men lived on it. Street life imparted its momentum to leisure and politics as well; the working
classes actively organized on the street. Thus, street entertainers or the more 'organized' tamasha players constituted the working man's theatre. The street corner
offered a meeting place. Liquor shops frequently drew their customers and gymnasiums their members from particular neighbourhoods.9 Social investigators continue
to be bemused that, when asked to give 'an account of their leisure time activities', the vast majority of workers 'could not be specific and
5 Labour Gazette, 4: 7 (March 1925), 745–7. This survey was based on a sample of 1,349 male and 715 female millhands.
6 J. Sandilands, 'The Health of the Bombay Workers', Labour Gazette, 1:2 (October 1921), 14–16.
7 G. R. Pradhan, 'The Untouchable Workers of Bombay City,' unpublished M.A. thesis, University of Bombay, 1936.
8 Report of the Rent Enquiry Committee, vol. I (Bombay, 1939), p. 9. The committee noted pertinently that the minimum space required by the Bombay Jail Manual for a prisoner was
double that which was stipulated as permissible under the Bombay Municipal Act of 1888. It also reported that 256,379 people lived in rooms occupied by six or more persons and
15,490 lived in rooms with at least twenty others, see pp. 7–9.
9 For a description of the social life of the mill districts in this period, see Parvatibai Bhor, Eka Rannaraginichi Hakikat, as told to Padmakar Chitale (Bombay, 1977).

Page 104
10
said that they pass time roaming, which they consider a mode of relaxation'.
The pleaders' offices, which proliferated along the streets of the mill district during the 1910s, were focal points of organization in industrial and political action; some
also became important centres of social activity. 'There is a constant stream of millhands to these offices', noted the police commissioner in 1914, 'which in the evening
especially become a regular ''rendezvous". Here the millhand gets in touch with the Brahmans or Marathas, who read the vernacular newspapers to them, and not
infrequently incite them to go on strike.'11 With the 'professionalization' of trade unionism in the 1920s, the methods of recruitment and publicity continued to be
reminiscent of the modes of the street entertainer. S. H. Jhabvala, admittedly one of the most 'professional' publicists of labour's cause, and an official of nearly twenty
unions in 1929, thus described his own recruiting drive:
I would stand at the end of the street when the factories were whistled off and would cry 'Ye who are fallen and miserable, come ye here and I shall help you out of the slough of
distress'. A few letters were scribbled on behalf of the distressed individuals, posted by me to their employers and God helps those who help themselves, strange enough a couple
of them were solved, and the poor illiterate flocks thought that I was a good instrument for the redress of their evil lot . . . Often I ventured to take a yellowrobed saint with me who
attracted a larger crowd. Mr Ginwalla managed to pay him eight annas per day, because he rolled in wealth and had no issue. He [the saint] sang Mahratta songs and I afterwards
gave a dose of unionism . . . The result was that in a short time flocks of people, man [sic], women and children anxiously waited for me to hear some of their grievances and to get
them solved.12
In its contrasting political style, the communist Girni Kamgar Union sustained the political momentum of the workingclass neighbourhoods by holding regular
processions and public meetings – at times, these were an almost daily occurrence. Their public commemorations of notable events in the socialist tradition – from the
birth of Marx to the death of Parashuram Jadhav, a worker killed in police firing during the 1928 strike in Bombay – were sometimes well attended, and at all times
contributed to the pageantry of political activity.
Although these forms of social behaviour can be identified with the neighbourhood, they cannot be considered in isolation from the context of work. The separation of
workplace and neighbourhood was more
10 K. Patel, Rural Labour in Industrial Bombay (Bombay, 1963), p. 150.
11 General Department, Order no. 3253/62–Confl; 15 May 1917, in Bombay Confidential Proceedings (1917), vol. 25, p. 15, OIOC.
12 Proceedings of the MCC, statement by S. H. Jhabvala, vol. II, noncommunist series, pp. 786–7. The fact that the Bible – as Jhabvala told the Meerut court – was 'one of my daily
readings' perhaps explains his prose style.

Page 105
13
evident in the cotton textile industry than, for instance, in the smaller artisanal workshops; yet in the textile industry as well these two social spheres were inextricably
connected. Nowhere is this to be seen more clearly than in the role of the jobber, who straddled the boundaries between workplace and neighbourhood. Usually
promoted from the shopfloor, the jobber was delegated vast powers over the workforce. So as to enable him to discipline labour effectively, management allowed him
considerable discretion in the employment and dismissal of workers – the ultimate weapons of labour control. In return, the millowners expected their jobbers to keep
production going: in other words, to maintain an adequate supply of labour, to resolve disputes between workers and to ensure industrial peace. The execution of these
functions was complicated by the fact that the daytoday demand for labour varied, partly because of absenteeism and partly in response to market fluctuations,
which determined the counts of yarn to be spun or the type of cloth to be woven and thereby governed the amount of labour required by management. Each mill
employed a sizeable proportion of its workers on a casual, daily basis. Across the industry as a whole, this was estimated at 28 per cent of the average daily
employment.14 So every jobber had to maintain connections with potential badli or 'substitute' labourers to meet fluctuations in the daily demand for labour.
It was, therefore, integral to the jobber's managerial functions that he should acquire and maintain connections outside the workplace. To recruit and discipline workers
'with success', recorded the Gazetteer, the jobber is 'bound to have a following of men and boys who usually live in the same neighbourhood and often in the same
chawl as himself'.15 Burnett Hurst, in his study of the condition of wageearners in Bombay in the 1920s, observed that the jobber 'endeavours to acquire an influence
over his friends and acquaintances who live in the same or neighbouring chawls. He lends them money, advises them on family affairs and arbitrates in disputes. When
labour is required, he uses the influence so gained and is generally successful in procuring hands.'16 Later evidence, however, suggests that this picture of close
neighbourhood control must
13 For the organization of the handloomweaving workshops in the city see R. E. Enthoven, The Cotton Fabrics of the Bombay Presidency (Bombay, 1897). The separation of
workplace and neighbourhood in the mill district also had its physical aspect. The mill compounds resembled fortresses in the mill district, protected by high walls, iron gates and
sentries equipped with lathis.
14 Labour Office, Bombay, General Wage Census, Part I; the Perennial Factories: Report on the Wages, Hours of Work, and Conditions of Industry in the Textile Industries (Cotton,
Silk, Wool and Hosiery) in the Bombay Presidency (including Sind), May 1934 (Bombay, 1937), p. 20.
15 Gazetteer of Bombay City and Island, compiled by S. M. Edwardes, 3 vols. (Bombay, 1909), vol. I, p. 493.
16 A. R. BurnettHurst, Labour and Housing in Bombay: A Study in the Economic Conditions of the WageEarning Classes in Bombay (London, 1925), pp. 46–7.

Page 106
17
be modified. Jobbers did not always live in the same chawls as their workers, and workers from a single mill did not usually live together. There can be little doubt,
however, that the jobber's power within the workplace rested upon his connections outside, and that at least some jobbers actively invested in the development of a
following and a network of power and influence. What Ambalal Sarabhai said of the jobber in Ahmedabad could equally apply to Bombay: 'He becomes a jobber if
he has friends and relatives in important positions in the mills and is also a favourite of the head of the department; the chances of his becoming a jobber entirely on his
own merit are very few.'18
Not all jobbers sought to build these connections; but few could ignore them altogether. They attempted to establish themselves at influential points within the material
structure of the neighbourhood. Frequently, they acted as rent collectors, sublessors and occasionally even as landlords. They sometimes helped to organize the
khanavalis or boarding houses which catered specially for groups of single workers. They also lent money on their own account and more often guaranteed loans.
Indeed, loans guaranteed by a jobber could be obtained at discounted rates of interest.19 They sometimes ran liquor shops and gymnasiums, and were often active in
the organization of religious ceremonies and festivals. Their authority at the workplace and the influence they acquired outside made them valuable members of chawl
committees and caste panchayats as well as useful allies for politicians at various levels. These highflying connections, deriving from their position at work, in turn
enhanced their value within the neighbourhood. This range of activities did not, however, simply establish the jobber as a provider. His services to the community as
well as his disciplinary function at the workplace placed him in a situation of potential conflict with the workers.
For the millowners, in turn, the jobber's connections outside the workplace increased his value as an agent of discipline. These connections were usually based on the
caste, kinship and village ties of the jobber. Recruitment through the jobber ensured that the cultural diversity of the workers was brought into the workplace;
consequently, the jobber served as an impressive bulwark against combination and provided
17 BDEC, oral evidence, Dhaku Janu Lad, pp. 103–5; Mathura Kuber, p. 499, Daji Sakharam, p. 507 and several others in GOB, Home (Special), File 550 (25) III B of 1938, MSA. Such
evidence should modify the widely accepted picture of the jobber's awesome personal control, which has tended to neglect the institutional basis of his power.
18 RCLI, Evidence, Bombay Presidency (including Sind), 1929–31 (London, 1931), vol. I, pt I, pt i, written evidence, Seth Ambalal Sarabhai, Ahmedabad Manufacturing and Calico
Printing Co. Ltd, p. 277.
19 Proceedings of the Bombay Provincial Banking Enquiry Committee, 1929–31, File 12 c, Replies to the questionnaire . . . submitted by the Currimbhoy Ebrahim Workmen's Institute,
MSA.

Page 107
a useful mechanism for strikebreaking. Significantly, it was when workingclass militancy began to complicate the jobber's task of disciplining labour that the
millowners grew concerned about the efficiency of his role in production and in the mid1930s took steps to modify their methods of recruitment.
Workplace and neighbourhood were brought into relation with each other not only by the methods of labour recruitment but also by the uncertain conditions of
employment. Periods of unemployment and chronic underemployment were commonly experienced by many millworkers, and even the 'permanent' jobs, which were
held in high esteem, offered little security. Not surprisingly, workers organized outside the workplace to hedge against their narrow and fluctuating margins of survival.
These informal welfare systems, or arrangements for mutual assistance, were based on their immediate social connections. Not only did migration occur within these
connections of caste, kin and village, but workers also relied upon them to find work and housing, and turned to them in periods of distress. For instance, groups of
single male workers, often from the same village, would rent a room together. As residents left, their friends and relatives who had moved to the city were also given a
share.20 This practice has inspired thought about its anthropological significance: the recreation of villages within the city or the recourse to traditional ways of life.
However, it probably bears a simpler explanation: that this was an obvious response to housing shortages and high rents.
In 1936, one social investigator noted 'the fact that distant relations, with a view to finding a job in Bombay, come and live with their relatives here'. But he also
suggested the doublesidedness of this dependency when he reported that workers 'find it very difficult to pay the rent . . . and therefore . . . they keep subtenants . . .
People cannot generally afford to have one room per family'.21 Such arrangements fulfilled a reciprocal need: newlyarrived migrants had a place to stay and contacts
through which to find work; the more established residents were able to meet their living costs, renew their rural ties, fulfil family obligations and even extend their
sphere of influence in the city. The importance of these social arrangements was reflected by the fact that, as far as housing was concerned, 'the neighbourhood of
persons of one's own circle is sought'.22 It is the political consequences of these interconnections between workplace and neighbourhood that the rest of this essay will
explore.
20 See Patel, Rural Labour in Industrial Bombay, p. 72. The most noted example of such organization was the 'clubs' established among the Goanese in Bombay. They were
financed by subscription and operated as a welfare system, giving preference to the unemployed among them, see RCLI, Evidence, vol. I, part i, The Bombay Seamen's Union, p.
293.
21 Pradhan, 'The Untouchable Workers', pp. 7–12.
22 Report of the Rent Enquiry Committee, vol. I, p. 20.

Page 108
II
If the social patterns of the neighbourhood cannot be abstracted from their material context, nor can they be portrayed as if they were devoid of political conflict. Not
only in devising strategies for living but also in industrial and political action, workers had to act across the boundaries of workplace and neighbourhood. As spheres of
social action, workplace and neighbourhood are frequently assumed to exert opposite pressures on the development of workers' perceptions and actions. At the
workplace, it is said, economic factors assert their primacy in the conflict between capital and labour and the lines of class antagonism are clearly drawn. The social
patterns of the neighbourhood, on the other hand, are cast in the image of villages transplanted to the city: here, workers appear to be the prisoners of their traditional
loyalties. Yet time and again the urban neighbourhoods belied this image and the mill district became a militant and, at times, even an insurrectionary centre.
The image of the urban neighbourhood as composed of villages ruled by their headmen derived its plausibility from the informal welfare systems operating in Bombay.
Undoubtedly, these welfare systems created opportunities for some people to establish themselves as patrons and providers. But it would be misleading to portray
their power as if it ran in a single direction. A closer examination of these relationships between neighbourhood 'leaders' and their 'followers' suggests the limits of
political command and indicates the social basis for collective action.
The jobber, the dada or neighbourhood boss, the grain dealer, the landlord, the moneylender, each acquired an impressive degree of influence in the course of their
daily commerce. Yet few neighbourhood patrons were able to escape the constraints imposed upon them by the social and political demands of their clients. Their
continued command of resources depended on their ability to fulfil the moral and material expectations of the neighbourhood. For instance, if the jobber's position at
the workplace was based, as we have seen, on his influence within the political and economic structure of the neighbourhood, he was also constrained by this
interdependence. Since his strength derived from the social and commercial ties he established with his workers, he had to remain receptive to their needs and
responsive to their demands. It was when his patronage was extended to the wider organization of credit, housing and recreation that it was exposed to greater
competition from rival jobbers as well as other neighbourhood patrons. Landlords, moneylenders, brothelkeepers and grain dealers no less than workers could
choose between jobbers. The interdependence of his position within the workplace and the neighbourhood meant that a jobber had to

Page 109
extend as far as possible the ambit of his control, unavoidably weakening his own lines of defence. Like all neighbourhood patrons, he had to compete not only for
clients but also for the favour of those more powerful than himself, from employers and trade unions to politicians and minor officials.23
By virtue of their place within the credit structure of the neighbourhood, shopkeepers and grain dealers also commanded considerable influence and some became
desirable political allies.24 In their case, too, their ability to do favours for people from their neighbourhood was central to their own business interests. Often, they were
pressed to finance various social and political activities from festivals to strikes. The expenditure involved was sometimes considerable. For instance, it was reputed
that, during Mohurram, mohollas spent between Rs 100 and Rs 400 to erect a tabut and carry it out in procession. Every street where a tabut was being prepared
would also arrange for a maulvi to deliver the waaz up to the tenth day of the month. For his description over five nights of the martyrdom of Husain, the maulvi was
paid between Rs 30 and Rs 100. These expenses were met – as was common to all religious observances – by the subscription of local residents. During Mohurram,
it was said, 'youths preceded by drummers and clarionet players, wander through the streets, laying all the shopkeepers under contribution for subscriptions'.25 Often
these shopkeepers were nonMuslims. The shopkeepers had paid their dues – often, no doubt, with reluctance – because it was expected that they would. The
relationship was more one of obligation than of enforcement. It was only when the arrangements of the ugarani, the
23 Of course these relationships were not stagnant. Between the wars, the jobber's authority at the workplace diminished. This was partly because the growth of labour militancy
made it increasingly difficult for him to reconcile the demands of his men with the imperatives of management. As the jobber's influence at work declined, it became more necessary
and, at the same time, more difficult for him to entrench himself within the neighbourhoods. It was probably the case that, by the late 1930s, the jobber's position became less
crucial to political and commercial advance in the neighbourhood. The extent of the jobber's decline should not, however, be exaggerated. In the mid1930s, the Bombay
Millowners' Association, in response to the declining efficacy of the jobber, introduced schemes to revamp the system of labour recruitment and control in the industry. However,
individual mill managements remained the jobber's last defender. At the level of the individual mill, the jobber still retained his uses for management. Ineffective in countering
industrywide action, the jobber attempted to entrench himself in the neighbourhood in order to dominate more completely the politics of the particular mill. At this level, BMOA
schemes to control the jobber met with considerable initial resistance from some of its own members. For a summary of the BMOA schemes to control badli hiring and the jobber
system in general see the Report of the Textile Labour Inquiry Committee, vol. II: Final Report (Bombay, 1953), pp. 337–50; BMOA, Annual Report (1935), pp. 27–9 and BMOA,
Annual Report (1936), pp. 37–40.
24 BDEC, oral evidence, Ravji Devakram in GOB, HD (Special), File 550 (25) IIIB of 1938, pp. 277–9, MSA.
25 Gazetteer of Bombay City and Island, vol. I, p. 185.

Page 110
collection of funds for the tabut levied by each moholla, broke down that its operation became evident to the state. For this reason the violence of the arrangement was
most noticeable to the police commissioner who wrote in 1911 that the money was 'extorted – there is no other word for it – from Marwadi and Bania merchants,
who are threatened with physical injury unless they subscribe liberally'.26
The first decade of the twentieth century was a sensitive period for the conduct of the Mohurram festival.27 As the state intervened in this sphere, local shopkeepers
discovered the language in which they could complain about the payments they had hitherto been obliged to make. The desire of the city police to intervene could find
justification in the 'extortion' of which shopkeepers complained; at the same time, this gave the shopkeepers the means by which they could rid themselves of the
burden imposed by these enforced payments. These complexities – and especially the expectations of the neighbourhood – can be illustrated by the outcome of a
complaint lodged by some Marwadi merchants at Pydhoni police station that they were being harassed and assaulted by Muslims of the Bengalpura Moholla. When
the police warned the 'leaders' of the moholla not to continue these extortions, 'this was treated as a grievance and Latiff himself had the impertinence to come to the
Head Police Office and complain that ''the police were not assisting the collection of funds".'28
A similar picture of service, obligation and reciprocity emerges from the role which shopkeepers played in the conduct of strikes. Without their longterm credit, the
general strikes which lasted between a month and six months would not have been possible. During the 1919 general strike, for instance, even as the millworkers were
out in the streets, most of the shops in the mill areas remained open.29 This was at an early stage of the strike. Prolonged strikes often placed immense pressure on
local credit arrangements. During the general strike of 1928, which lasted six months, workers had to turn to their lenders of last resort, reputed to charge the highest
rates of interest: the Pathans. The Pathans' attempts to recover their loans was one important reason for the communal riots of February 1929.30 During the general
strike of 1940, the Bombay Chronicle
26 Commissioner of Police, Bombay to Secretary, Judicial Department, Bombay, No. 545–C, 20 January 1911, reprinted in S. M. Edwardes, The Bombay City Police, Appendix, p.
198. Edwardes' account of Mohurram related largely to areas of the city outside the mill district. But some of these relationships described for these areas were equally applicable to
the mill district.
27 See J. Masselos, 'Power in the Bombay "Moholla" 1904–1915: An Initial Exploration into the World of the Indian Urban Muslim', South Asia, 6 (1976), 75–95.
28 Edwardes, Bombay City Police, Appendix, p. 198.
29 Bombay Chronicle, 13 January 1919.
30 Police Report on the Riots in Bombay, February 1929 (Bombay, 1929); Report of the Bombay Riots Inquiry Committee (Bombay, 1929).

Page 111
reported that the cheap grain shops, offered to workers when they demanded an increased dearnessoffood allowance following the price rises which accompanied
the outbreak of war, were unacceptable because they 'cut away the credit which workers had so far been enjoying with other grain merchants. In times of disputes
between workers and employers, Bania grain dealers allow credit to workers to the extent of five or six months.'31 These were vital connections; even in times of
industrial and political peace they could not be ignored; in moments of conflict, they were indispensable.
If the local shopkeeper was a figure of considerable importance, it was crucial to cultivate his protection and his patronage. For shopkeepers it was their ability to fulfil
these functions that defined their local importance and drew them into more exalted political connections. During the oneday strike of 7 November 1938, Tukaram
Laxman, determined to go to work, turned to the bidi or tobacco shop, when he was stopped by strikers: 'I requested the bidi shopkeeper to send me to work. I said
''Mama, anyhow see that I get to work. Then the Bidiwalla asked the [presumably his] motor driver who was nearby to take me to my mill".'32 On the same day,
however, several grain shops were 'looted'. Baijnath Bahadur complained that strikers entered the shop in which he worked, removed the gunny cloth covering the
grain, ate the grain and ran away.33 The police commissioner described the looting of a shop near the Worli Chawls in similar terms: 'The shopkeepers were arguing
with these people. The crowd seemed to treat the whole affair as a joke. They would just pick up a handful of grain and throw it.'34 The apparent festivity with which
these shops were looted concealed the underlying tensions in the relationship between shopkeepers and the residents of the neighbourhood.
The data – essentially, a title for a neighbourhood leader – fascinated and repulsed contemporary observers. For the dominant classes of the city the data symbolized
the 'roughness' of industrial politics. Burnett Hurst described the 'dada' as 'a hooligan, who lives by intimidation. He is both lazy and dangerous.'35 In public discourse,
neither the employers' nor the workers' organizations cared to be connected with the world of the dada even though they operated within it. Anticommunists used the
term to describe the following of communist unions; communists used it to signify strikebreakers. During the investigations which followed the
31 Bombay Chronicle, 7 February 1940.
32 BDEC, evidence of Tukaram Laxman, in GOB, Home (Special), File 550 (25) III B of 1938, p. 517, MSA.
33 BDEC, evidence, Baijnath Bahadur in ibid., p. 639.
34 BDEC, evidence, W. R. G. Smith, Commissioner of Police, Bombay, in ibid., p. 1049.
35 BurnettHurst, Labour and Housing in Bombay, p. 49.

Page 112
36
communal riots of 1929, Hindu and Muslim witnesses used the term in connection with the rival, rather than their own community. In fact, 'dada' was a term of
respect. Although, in public, everybody tried to dissociate themselves from 'dadas', as one trade unionist pointed out, 'I know personally that Dadas like to be called
Dadas.'37
The dada was not a special kind of working man. Several workers established themselves as dadas by participating in crucial neighbourhood activities such as the
running of gymnasiums or rent collection. In the course of their activities, the dadas became, as V. H. Joshi, an official of the Girni Kamgar Union put it, 'agents dealing
in working people'.38 The metaphor is instructive of the dada's vulnerability to the ultimate sanctions of neighbourhood politics: social and, in reality, commercial
boycott. If the dada was 'an agent dealing in working people', he could not alienate his clientele. This was why 'the dadas left to themselves cannot harm a mass of
people'.39 For this reason dadas could be engaged against strikers least during periods of solidarity and most when they were in some ways least needed, at times of
workingclass vulnerability.
The scale of a dada's activities was determined by his social connections and the base from which he was able to operate. Some, like Keshav dada Borkar, dominated
the whole area of Ghorapdeo for several decades; others were small men, neither recognized nor respected in the next chawl. In a sense, dada was properly a
reputation rather than a status – a reputation for physical prowess or for getting things done. The dada, said Balubhai Desai, 'is a person who has got this reputation of
controlling the hooligans by rendering services to the hooligans and protecting them, giving grain to them and really of course controlling them . . . some of these Dadas
are rich'.40 Their ability to exert this control depended upon their facility in providing such services. They did not always fight themselves, but they could mobilize men
to do their fighting and in any case their leadership depended upon the belief that they were capable of fighting. In order to protect their followers they had to have the
means to pay surety for those of their men who were arrested 'and help in any other way they can'. It was only 'in that way they collect the hooligans'.41 To build and
maintain a following a dada needed influential friends and patrons; but to catch the eye of the great, let alone achieve a following, he needed to cut a figure on the street
corner and in the chawl. Such prominence was often achieved through the leadership of a gymnasium.
36 BRIC, oral evidence, Balubhai Desai, File 8, and A. R. Dimitimkar and S. Nabiullah, File 7, MSA.
37
BRIC, oral evidence, G. L. Kandalkar and V. H. Joshi, p. 71, MSA.
38 Ibid., p. 69.
39 Ibid., p. 61.
40 BRIC, oral evidence, Balubhai T. Desai, File 8, pp. 29–31, MSA.
41 Ibid., p. 29.

Page 113
These gymnasiums, where wrestling contests were held and where men trained in stick play, proliferated in the mill district. Their cultural and political role will be
examined later in this essay. What must be stressed here is that they formed an important part of the dada's domain. It was here that dadas served their apprenticeship
and it was through these gymnasiums that they often built their reputations. According to V. B. Karnik, the prominent trade union leader of the 1930s,
every gymnasium used to have, say, two dozen or three dozen or sometimes even a much bigger number of students and those students were under the control of the gymnasium
– that is the dada who taught at the gymnasium. And that dada could utilise his students in any way that he liked . . . Every party tried to get the support of one dada or the
other.42
The extent to which workplace and neighbourhood overlapped and the roles of jobber and dada could be combined was indicated by Dhaku Janu Lad, a jobber in
the Bombay Cotton Mill. He had been prevented from going to work during the oneday strike of 7 November 1938. His less prominent brother had, however,
managed to enter the mill. Because his brother had not returned when the first shift should have ended, Dhaku Janu walked to the mill to see whether he needed help.
Crowds of strikers who had failed to stop some workers entering the mills, now decided to prevent them from leaving instead. The police might escort the workers out
of the workplace, but they could not extend this service to their doorsteps. For this reason, it was unsafe for the workers to leave the mill. When Dhaku Janu
approached the mill gates, 'those who were working in the mill went up to the Manager as soon as they recognised me'. The manager sought the help of the police to
escort the workers out of the mill, and Dhaku Janu Lad took two separate groups of workers to their rooms.43 The provision of this kind of service was among the
most crucial demands made upon dadas and jobbers. They acted as informal guardians of a public order and morality which they interpreted, sometimes arbitrarily,
and enforced without an excess of decorum.
The material conditions which made informal welfare organization necessary for most workers also created nodes of power and influence in the neighbourhood. The
struggles waged around the jobber, the grain dealer and the dada indicate the reciprocity of these power relations. As people got together to meet their social needs,
their actions defined the extent, and the limits, of social control. It is important to turn from the institutional basis of dominance in the neighbourhood – arising from its
42 Interview, V. B. Karnik, April 1979.
43 BDEC, evidence, Dhaku Janu Lad, GOB, Home (Special), File 550 (25) III B of 1938, pp. 103–17, MSA.

Page 114
material structure – to the patterns of association which occurred within them. The collectivities fostered by the conduct of religious festivals, especially in the earlier
twentieth century, and the gymnasiums of the mill district reveal how social behaviour itself provided a basis for political mobilization.
The relationship between these collectivities and politics is difficult to determine. Of course, the fact that men were brought together to crack a pot during the
Gokulashtami, carrying Ganpati to the sea or dance with the toli bands at Mohurram did not mean that they could then be frogmarched into politics. The observance of
some religious occasions, such as the Ganpati festival, had an explicitly political content and others, like Mohurram during the early years of the twentieth century,
began to reflect social antagonisms, invited the intervention of the state and were dragged into the public domain.44
The associations which emerged in the conduct of religious observances became the focal points of community sentiment and rivalry. The internal structure and
organization of the melas – the companies of dancers at Gokulashtami – provide further insight into the complex interplay between leaders and followers. Participation
in a mela sometimes depended upon the payment of an entrance fee, a monthly subscription and contributions to the general expenses of the mela. Before being
admitted to the mela, each entrant had to take an oath in which he swore not to divulge its secrets to any other mela and not to join its opposing or rival party even if he
severed his connections with his own. Group loyalty was a central feature of these melas. The leader of the mela was afforded considerable respect, usually being a
man of some local prominence, and it was expected that the members of the mela would remain strictly obedient to him. But the leader had to manage the mela,
protect its interests and was held personally responsible for making all the necessary arrangements on the day. His continued leadership depended upon satisfying his
team.45
In their organization, leadership and group loyalties, these associations resembled street or neighbourhood gangs. As one observer of the toli bands which danced at
Mohurram wrote:
Each street has its own band to parade the various quarters of the city and fight with bands of rival streets. If the rivalry is good humoured, little harm accrues; but if, as is
sometimes the case, feelings of real resentments are cherished, heads are apt to be broken and the leaders find themselves consigned to the care of the police.46
44 See R. I. Cashman, The Myth of the Lokamanya: Tilak and Mass Politics in Maharashtra (Berkeley and Los Angeles, 1975), pp. 75–97; Masselos, 'Power in the Bombay
''Moholla",' 75–95.
45 K. Raghunathji, The Hindu Temples of Bombay (Bombay, 1900).
46 Cited by the Gazetteer of Bombay City and Island, vol. I, pp. 187–8.

Page 115
The dynamic of neighbourhood competition on such occasions lay in the reputation which neighbourhood leaders, especially the dadas who led these gangs, were
seeking to gain or conserve. These rivalries were part of the permanent social relationships of the neighbourhood, not the product of spectacular occasions alone.
Like the great festivals, the gymnasium was an important, albeit less public, focus of workingclass culture. The akhada or gymnasium was not necessarily a place
which the 'respectable' abjured. Sir Purshottamadas Thakurdas announced proudly that he had trained at one in his youth and that he now sent his grandson to an
akhada. Balubhai Desai, the Congress politician, claimed in 1929 that he still attended an akhada. It was, however, he added, an akhada only for 'decently behaving
gentlemen', and he chose it because it was the only gymnasium in Bombay with machines 'for reducing fat which I am taking advantage of'. A more common feature of
gymnasiums, however, was lathiplay. Those who trained in akhadas thus acquired a special skill. Balubhai Desai applauded its use as a form of selfdefence. 'A lathi',
he said, 'can give you protection if you are surrounded even by 50 people and you can escape unscathed.'47 But akhadas were not associated with physical culture or
selfdefence alone. Young men, brought together at a gymnasium, skilled at fighting and trained in the use of lathis, had considerable potential for political mobilization,
and frequently provided a basis for neighbourhood action. As social centres, gymnasiums could also become focal points of political organization. According to the
moderate labour leader, Syed Munawar, 'akhadas and teashops were the rendezvous of riff raffs and hooligans . . . those were the best places for them to meet'.48
During the communal riots of 1929, they were again identified as sources for the organization of violence. Indeed, one witness argued that the Muslims had been put at
a disadvantage in the riots by the decline of the Muslim dada 'since the Mohurram taboot processions in Bombay were stopped more than 15 years ago, and since the
closing of the Muslim talimkhanas'.49
Gymnasiums were also pulled into industrial action, on both sides, by strikers and management alike. Some workers, by virtue of being dadas, could deploy the
gymnasium members in support of a strike, while the management recruited strikebreakers from their ranks. The role of the gymnasiums in political mobilization is
more easily identified than the part they played in industrial action. Political pamphlets and the reported speeches of strike leaders often claimed that gymnasiums were
being used in strikebreaking. But it is extremely difficult to document the relationship between gymnasiums and mill managements. Obviously, strike
47 BRIC, oral evidence, Balubhai T. Desai, File 8, pp. 69–71, MSA.
48 Ibid., oral evidence, Syed Munawar, File 3, p. 279.
49 Ibid., oral evidence, A. R. Dimtimkar and S. Nabiullah, p. 271.

Page 116
breaking could offer gymnasiums a means of earning an income; the greater their income the better equipped they would be in relation to other gymnasiums, the more
effective in attracting members and perhaps the more successful in the contests arranged between them. It is easy to see that strikebreaking could become an activity
essential to the success of some gymnasiums. From the point of view of the jobbers or the management, importing the hired strength of a gymnasium to settle scores on
the shopfloor was not always advisable, nor often necessary. It was only when the employer 'became desperate and wanted to see that the mill started again', when he
felt he had exhausted all other options, according to V. B. Karnik, the Royist labour leader,
that he would get hold of a dada and recruit some strikebreakers . . . it all depended upon the market; if there was demand for cloth then he was anxious to reopen the mill; if there
was no demand for cloth then he was not so keen; if the mill remained closed for a week or ten days or even a month it did not matter to him.50
There is as yet little available evidence on the organization and working of gymnasiums. Such evidence as exists suggests that the organization of some gymnasiums
could be extremely elaborate. For instance, the Hanuman Vyayam Shalla was found in 1912 by a certain Narayan Rao. By 1928 it claimed branches in parts of the
city as dispersed as Vajreshwari, on the outskirts of Bombay in the neighbouring Thana district, and Bhoiwada in the heart of the mill district, apart from its
headquarters in Prabhadevi. In January 1928 it acted as host to a contest between fifty other gymnasiums from all over Bombay. This particular occasion involved over
150 wrestling bouts and the collection amounted to over Rs 2,500. It was likely that a lot of money would pass through gymnasiums; no doubt competition for their
control could be fierce. Elections were held to decide the constitution of the committee. Gymnasiums sometimes even advertised their elections in the Marathi press,
notifying their members of the time and place at which they would be held, and announcing how they could establish their qualification to vote. The candidates were
sometimes men of considerable importance. In the case of the Shri Samarth Vyayam Mandir, the nationalist campaigner, Dr N. D. Savarkar, offered himself as a
candidate.51
It was as much a mark of prestige for gymnasiums, as it was for chawl committees and neighbourhood leaders, to be able to invite eminent people to their great
occasions. When the Hanuman Vyayam Shalla held its contest in January 1928 it invited S. K. Bole, founder of the Kamgar Hitvardhak Sabha and, in 1928, vice
president of the Bombay Textile
50 Interview, V. B. Karnik, April 1979.
51 Nava Kal, 6 January 1928.

Page 117
52
Labour Union, to preside at the function. S. K. Bole, it was reported, gave the gymnasium a handsome donation. Because of their obvious importance in political
mobilization, politicians and trade unionists did not treat such connections lightly. Indeed, their political relevance enabled gymnasiums and their dadas to form alliances
at exalted levels, which, in turn, then became an important factor in their position within the neighbourhood.
The tensions and conflicts within the working class were most obviously manifested in the neighbourhood; but here, too, the solidarities of labour politics were forged.
Political experience in this arena was formed, in part, by the struggle to constrain and at times direct neighbourhood leaders. Power and control in the neighbourhood
entailed a set of shifting relationships in which dominance was achieved and limited through negotiation, manoeuvre and sometimes violence. As the neighbourhood was
increasingly brought into the sphere of industrial and public politics as well, it shaped the development of the political consciousness and political action of the working
class.
III
Social relationships in the neighbourhood increasingly impinged upon industrial politics. This was partly because material conditions limited the possibility of
organization at the workplace. In an overstocked labour market, employers were well placed to defeat workers' combinations and at times even exclude them from the
workplace.53 Consequently, if workers were to demand better conditions, fight wage cuts or protect employment levels, it was imperative that they organize in the
neighbourhood as well. The arcane procedures and legal niceties of collective bargaining were never far removed from the baser negotiations of the street.
In dealing with labour unrest, mill managements employed the usual forms of repression, as well as some novel ones. Workers who participated in trade union activity
were less likely to be promoted to more responsible and lucrative posts. They were obvious candidates for retrenchment after an industrial dispute or during a
recession. They were also vulnerable to discrimination in the allocation of machinery or the distribution of raw materials. As the Social Service League pointed out,
'Complaints about victimization of workmen taking a prominent part in the trade union
52 Ibid.
53 One mill manager told B. Shiva Rao, 'For every one who goes out of this gate there are nine more waiting outside who would be grateful for the wages I am paying.' Shiva Rao, The
Industrial Worker in India, p. 55.

Page 118
54
movement are frequently heard.' Trade unions – particularly those which did not meet with the employers' approval – could neither collect subscriptions nor hold their
meetings in the vicinity of the workplace. By choosing with whom they would negotiate, by choosing between rival unions or factions, employers could deal with their
most favoured workers and thus strengthen the organizations they approved of while attempting to destroy those they considered dangerous. Such action was by no
means confined to the textile industry; however, both within and outside it, these measures were most effective when the conditions of employment were casual and the
level of skill low.
Significantly, although the millowners failed to combine across the industry in order to control production when their markets slumped,55 they were able to coordinate
impressively in dealing with industrial action. As early as 1893, the millowners had circulated the names of strikers among themselves.56 As conflict in the textile
industry intensified between the wars, their efforts grew more vigorous. By the mid1920s the Sassoon group, for instance, was employing agents to spy upon the
meetings and organization of their workers as well as to take down and translate such speeches as were made.57 Each mill had in its Watch and Ward department its
own organized force for coercion. The superintendent of the Watch and Ward department at the Sassoon mills was 'a well known boxer' called Milton Kubes. When
asked how he had collected the speeches he claimed to have done in 1928, Kubes said, 'I have got my own secret service'.58 The millowners were also able to
mobilize their own brigades for political action. In opposition to the Red Flag Union, it was said, the millowners 'post their own pickets, publish leaflets, handbills and
keep watch and ward inspectors'. The object of the pickets 'is to help the loyalist workers to go to work . . . and to see that they are not molested. . They simply move
around in the chawls, post themselves as pickets in front of the mill gates, and advise willing workers to go to work and if any of their workers are molested they go to
their rescue.'59 To organize such pickets, millowners relied upon their jobbers either to mobilize the support of their workers against the strike or to encourage anti
strike alliances in the neighbourhood. By the mid1930s, they had become more systematic in keeping an eye on trade union activities, reporting on workers'
54 RCLI, Evidence, Bombay Presidency, vol. I, part i, written evidence, The Social Service League, p. 445.
55 Dissatisfaction in this regard was often expressed in the speeches of the chairmen of the Bombay Millowners' Association at their annual general meetings; see, for instance,
BMOA, Annual Report, 1934, Chairman's speech, p. ii.
56 BMOA, Annual Report, 1893, p. 16.
57 BRIC, oral evidence, Milton Kubes, File 5, pp. 241–3, MSA.
58 Ibid., p. 201.

Page 119
meetings and sharing information with each other. Indeed, this political intelligence was embodied in the monthly report of the labour officer of the Bombay Millowners'
Association to its committee. It was also made available to the police as well as to official inquiries into strikes, disturbances and seditious conspiracies, and seems to
have been treated largely as unproblematic evidence.60
In addition, the millowners were increasingly able to call upon the assistance of the state. Fearful of the infiltration of class struggle into nationalist agitation and
concerned at the spread of support for the communists among Bombay's workers in the late 1920s, the provincial government grew increasingly ready to intervene in
industrial disputes. From the late 1920s, the government constructed a legal framework for the conduct and settlement of disputes, sent more police to the mill gates
during strikes to restrict picketing and control 'intimidation', and prosecuted the communist leaders of the labour movement more readily for incitement or conspiracy.
The presence of the state was most evident, however, in the form of the police when they supervised pickets or escorted blacklegs to work. Introducing the Prevention
of Intimidation Bill in 1929, the Home Member of the Bombay Government recalled his memories of the general strike of that year for the benefit of the Legislative
Council:
One of the most remarkable sights it has ever been my fortune to view was a long procession headed by mounted police, followed by foot police and then by a hollow square with
women workers in the middle and the workmen around them on all sides. The procession was wound up by more armed police and another party of mounted police. As they
marched along the road, the street corners and points regarded as dangerous were guarded by still more police. Day after day these men and women were thus escorted to their
work and away from it in complete security. These measures continued so long as they were necessary. As the number of men at work increased and the danger of their being
overawed by strikers decreased the police precautions were gradually relaxed.61
These were formidable obstacles against which to conduct a strike; they could scarcely leave the forms of industrial action, let alone its possibility, unaffected.
This structure of dominance within industrial relations, ranging from the economic sanctions available to employers at the workplace to the political means of repression
outside, was often sufficient to smother any sustained resistance from the workers. For one thing, industrial action necessarily placed jobs in jeopardy. Moreover,
unless workers were able
60 BDEC, evidence, extracts from the monthly reports of the Labour Office, BMOA, in GOB Home (Special), File 550 (25) III of 1938, pp. 173–245, MSA.

Page 120
to effect a fairly complete strike, they stood little chance of negotiating their demands with management, let alone achieving any concessions. When the state intervened,
workers were placed under greater pressure to devise means by which they could prevent their jobs being usurped by 'blackleg' labour. It shifted the focus of action to
the neighbourhood where social pressure as well as force could be deployed to maintain an offensive. Workers' combinations, excluded from the workplace, were
forced to act in the social arena outside. The disputes of the workplace were brought into the street. Patterns of association developed in the neighbourhood were
integrated into the conduct of industrial action. Managements were, at times, also active in forging antistrike alliances in the neighbourhood, but unless workers had
been able to constrain and immobilize these alliances, they would have been able to offer little effective resistance. As the neighbourhood itself became an arena of
industrial conflict, workers used their social connection outside the workplace in two ways: first, as a material base and second, for varying degrees of direct action.
Neighbourhood social connections, indispensable to the daily life of workers, influenced the possibilities of their collective action. How long workers could remain on
strike was governed by the extent to which they could draw upon the material resources of the neighbourhood and especially upon the credit they were able to
mobilize. If through participation in a strike a worker risked his job, his willingness to strike would to some extent be influenced by his chances of finding another job,
and for this he depended upon his neighbourhood connections. Industrial action sometimes even brought into play the rural connections of the workers. M. S.
Bhumgara, formerly manager of the Khatau Makanji Mills, explained in 1931 that it was upon workers who had lost all connections with the land that 'the millowners
generally depend to break the strike as these people have no home to return to and hence they are the worst sufferers at such times'.62 Those workers who could fall
back upon their village connections were often the most resilient in industrial action.63 Migrants with strong rural connections were expected to be less concerned,
perhaps even less conscious of their economic interests in the city than urban proletarians with nowhere else to turn. In this case, however, it would appear that
migrants with the strongest rural connections could also be the most conscious of their 'urban' interests and most active in their defence:
62 RCLI, Evidence, Bombay Presidency, vol. I, part i, written evidence, Mr M. S. Bhumgara, p. 499.
63 In January 1928, during the strike wave which finally launched the general strike, the police observed: 'The strikers were determined not to work the new system and are gradually
leaving for the native places by the coasting steamers and trains after receiving their wages'. BPP SAI, 1928, no. 3, 21 January, para. 61.

Page 121
Strikers, trade unions and the political parties also had to rely upon the pressure which they could bring to bear upon the community as a whole in confronting strike
breakers. Their actions were based partly on their own strength of numbers, partly on the alliances which they could effect within the structure of neighbourhood power
and partly on their ability to publicize and thereby discredit workers and jobbers, dadas and gymnasiums involved in strikebreaking. It was sometimes said of the
communistled Girni Kamgar Union that it hired 'mavalis and badmashes', literally 'roughs', to stop workers crossing the picket lines or to 'intimidate' blacklegs in their
chawls.64 But most unions did not have money for such enterprises. They were probably most capable of hiring dadas when their membership figures rose dramatically
and their subscriptions permitted them a few luxuries, as for example during some general strikes. Yet at such times, the militant mood of the workers was often enough
to enable them to dispense with these extravagances. On the other hand, as V. B. Karnik put it, if 'usually it was the strikers themselves who used to take the lead in
organising this type of defence' against organized blacklegs, 'sometimes some of the strikers may themselves be dadas'.65
One of the achievements of organization – especially the extensive organization which the communists were able to build up after 1928 – was that unions could deal
with dadas in an attempt to contain their hostility or negotiate their support. From 1928, the Girni Kamgar Union maintained a list of dadas in the mill district and invited
workers to contribute to it.66 Kranti, the union's official organ, published the names of 'loyal' workers, which meant their jobbers and escorts as well.67 Workers, too,
were involved in making the identities of strikebreakers public, and, indeed, moral outrage was repeatedly expressed at their deeds at meetings and through leaflets.
For instance, the residents of a wadi sometimes held public meetings at which local dadas were forced to explain and justify their actions. Blacklegs were often brought
to strike meetings and humiliated. On 31 May 1928, two blacklegs were arrested by workers and brought, their faces blackened with soot, to Nagu Sayaji's Wadi,
the communist stronghold in Prabhadevi. There, the communist leader, S. A. Dange, lectured them on the treachery which blacklegging involved. Dange was later
arrested for his part in this episode, but was released on bail when the two workers failed to pick him out in an identity parade.68
64 BRIC, oral evidence, S. K. Bole, File 3, p. 217, MSA.
65 Interview, V. B. Karnik, April, 1979.
66 BRIC, oral evidence, G. L. Kandalkar and V. H. Joshi, File 16, p. 71, MSA.
67 Ibid., S. K. Bole, File 3, p. 247.
68 Proceedings of the MCC, statement by S. A. Dange, pp. 2447–8.

Page 122
Often, strikebreakers suffered social boycotts. Their names, particularly those of collaborationist head jobbers, were read out at strike meetings. Indeed, during the
general strike of 1940, these lists of names were sent in with so much enthusiasm that it embarrassed the leadership. The secretary of the Council of Action for the
conduct of the strike, R. S. Nimbkar, had to advise speakers not to read out these names as they were not always correct and 'were sent sometimes on account of
personal grudge'.69 Men and women going to their mills were taunted. Strikers would, it was said, call out to somebody on his way to a mill: 'He is a malik's son, that
is why he is going so faithfully to work.' Such action, said Kandalkar, presenting hostile evidence against the communists who had jockeyed him out of power by the
late 1930s, 'no doubt caused some embarrassment to the workers who were going in for work . . . being put to shame in the presence of their brother workmen
naturally annoyed them.'70 That moral pressure could be effective emphasizes the ambiguity inherent in the behaviour of some 'blacklegs'. Although the effect of
working during a strike was clearly to contribute to its defeat, it would be misleading to assume that when workers crossed the picket lines they simply signified total
opposition to industrial action, or revealed thereby an undeveloped social consciousness. Several contradictory pressures, both moral and material, for as well as
against action, operated throughout the conduct of a strike, and governed workers' options. Indeed, it was for this reason that moral pressure, which often entailed
some degree of physical coercion as well, could be effective at all: it found an ideological resonance in the public morality of the neighbourhood.
At the same time, moral pressure and public embarrassment, however effective, were not always enough. Throughout the 1930s, communist leaflets highlighted the
causes of unemployment and argued the case for an identity of interest in the long term between the jobless and the workers in an attempt to deter 'blacklegs', while
maintaining a steady, moralizing attack against 'blacklegging'.71 Notions of morality and justice – or more clearly injustice – infused the most direct and physical forms
of public pressure. At a meeting called to propagate the oneday strike of 1938, Lalji Pendse said that 'some goondas have beaten our volunteers' and called upon
those children of workers who trained at gymnasiums to 'teach a good lesson to these dadas'.72 Towards the last stages of the 1940
69 Commissioner of Police, Bombay, Daily Report, 6 April 1940, in GOB, Home (Special), File 550 (23) CI of 1940, p. 83, MSA.
70 BDEC, evidence, Girni Kamgar Union, Bombay (Kandalkar) in GOB, Home (Special), File 550 (25) III of 1938, p. 431, MSA.
71 See leaflets collected in GOB, Home (Special), File 543 (46) of 1934 and 543 (46) pt I of 1934, MSA.
72 BDEC, confidential statement submitted by the Bombay Millowners' Association in GOB Home (Special), File 550 III of 1938, p. 315, MSA.

Page 123
general strike, the Council of Action of the Bombay Provincial Trades Union Congress had to deal with the exertions of Mane Master. At a meeting on 31 March
1940 a communist worker, Khaire, said that,
Mane Master was defaming the Marathas and blackening the face of the Great Shivaji by conducting on the one hand the Shivaji Gymnasium at Bhoiwada and on the other hand
trying to break the strike. This Mane Master who was a member of the Maratha League had blackened the face of the Marathas and was himself a blot on Maratha society and as
such they should break his legs.73
It was sometimes necessary as well as possible for strikers actively to picket particular neighbourhoods, road junctions and even inside their chawls. For instance,
during the 1938 strikes Madanpura was picketed so effectively that the Simplex Mill reported that its 'jobbers complained that they were not allowed to leave the
moholla'. The experience of the Simplex Mill was by no means exceptional; workers from the New Great Eastern Mills, who lived in Kamathipura, and from the
Madhavji Dharamsi Mills suffered a similar fate.74 The efficacy of such action depended upon the particular political circumstances of each neighbourhood. As S. K.
Patil, the brain behind Congress organization in Bombay city in the 1930s and general secretary of the Bombay Provincial Congress Committee, explained it, not all
strikes or meetings could be broken:
the breaking activities can succeed only in certain areas. Even in the labour area, there are spheres of influence. If you go to a sphere other than your own, it is easier for them to
break up a meeting, because they have a larger following round about. That is not possible everywhere.75
The fact that several mills of the Sassoon group continued to work on 7 November 1938 was attributed to perhaps the most significant dada in Bombay between the
wars, and a Congressman, Keshav Borkar, 'The peculiarity about those mills', said deputy commissioner of police, U'ren,
is that they are in the area which is looked after by Keshav Borkar. He was naturally against the strike . . . It is quite obvious that by virtue of the fact that he holds sway in that
area, the Red Flag Union did not think that they could get much success there . . . The mere fact that he was the headman of that area, I think, was sufficient for the Red Flag
volunteers not to bother with that area.76
73 Commissioner of Police, Bombay, Daily Reports, 1 April 1940, in GOB, Home (Special), File 550 (23) CI of 1940, p. 21, MSA.
74 BDEC, confidential statement submitted by the BMOA, Annexe BI in GOB, Home (Special), File 550 (25) III of 1938, pp. 317–43, MSA.
75 BDEC, oral evidence, S. K. Patil in GOB, Home (Special), File 550 (25) III B of 1938, p. 401, MSA.
76 BDEC, oral evidence, Mr U'ren, Deputy Police Commissioner in ibid., pp. 681–3.

Page 124
The balance of power in the streets was clearly a crucial factor in determining the geography, and sometimes even the possibility, of political action.
Another common response to the structure of control which workers had to face was to impose pressure at the most vulnerable point of most strikes: the jobber. In
1928, strikebreaking jobbers were hounded out of their neighbourhoods. S. D. Saklatwalla of the Tata group of mills informed the Fawcett Committee that one
jobber had 'to change his place of residence twice because they [workers] once found that he had entered the mill and . . . they were therefore persecuting him. He
said he changed his residence although he had to pay increased rent.'77 In one case reported in 1938, Jaysingrao Bajirao, a head jobber of the winding department
related how during the oneday strike of 7 November, workers waited in batches of ten to twenty until 11 p.m. at night 'in order to assault me if I ventured to go out of
the mill gate'.78
It was because workers were often most effective in political action beyond the workplace that the millowners preferred the state to intervene in the conduct rather
than the settlement of strikes: for instance by deploying the police to prevent picketing not only at the mill gates but also in the neighbourhood.79 Ten years later, the
millowners continued to argue a similar case: but more explicitly and with increased vehemence. During general strikes, 'the collection of crowds in streets and
thoroughfares near the mills should certainly be prevented', urged the Bombay Millowners' Association, 'as otherwise free access by employees to their place of
employment becomes impossible'. Such access was a necessary precondition for taking blacklegs into the workplace and maintaining production. Preferably, they
argued, pickets 'should be confined to peaceful conversational persuasion and they should not be permitted to shout slogans or use abusive language or better still they
should not be allowed to speak at all'. They were particularly emphatic that picketing at the workers' 'place of residence' should be made a criminal offence, for 'it is
precisely this type of picketing that is most desirable to prevent'.80
The intimidation of 'ordinary workers' by 'strikers' often explained to the millowners as well as the Home Department why political agitators and their allies were able
to shut down their mills. Clearly, intimidation by itself did not explain the solidarity of a strike, as, for instance, the
77 Proceedings of the BSEC, 1928–29, vol. I, p. 121, MSA.
78 BDEC, oral evidence, Jaysingrao Bajirao in GOB, Home (Special), File 550 (25) III B of 1938, p. 553; see also the evidence of Dhaku Janu Lad, p. 105, MSA.
79 BMOA, Annual Report (1928), Chairman's speech, AGM, p. iii.
80 BDEC, BMOA answers to the questionnaire in GOB, Home (Special), File 550 (25) III of 1938 pp. 141–5, MSA.

Page 125
81
Bombay Millowners' Association believed it did; at the same time without 'intimidation' it was impossible at times to conduct a strike. In public discourse, intimidation
simply meant that union bullies threatened to beat those who went to work. Undoubtedly, the sanction of physical force lay behind most forms of 'political' pressure in
the neighbourhood. But intimidation was not conducted only by such 'professional' groups. It was more usual for workers who favoured a strike to act in their own
chawls to prevent their fellow residents from going to work. Since their own jobs were in the balance, it is unlikely that their actions needed to be instigated or
organized for them. One jobber described the working methods of those who canvassed for the 1938 strike: 'Usually five or ten men are real workers, they approach
people but these five or ten people are followed by a large crowd.'82 When union bullies acted successfully in their selfconscious role as bullies, they appear to have
done so with the aid and approval of the chawl.
As intimidation became a subject for public debate, workers began to use it to their own advantage. One millowner told the Bombay Riots Inquiry Committee, 'I have
had certain talks with groups of work people, and I have questioned them: ''Why don't you come forward and report these people [who intimidate] to the police?"
They say "if we do so, we are marked men".'83 However, there was an underside to the picture presented by the employers and the state. As N. M. Joshi argued,
workers used intimidation as an excuse to remain on strike. 'It may be that there was intimidation on your part', he told S. D. Saklatwalla, during their negotiations after
the 1928 strike, 'and so the men could not tell you the truth.'84 Similarly, K. F. Nariman, the populist Congress leader, pointed out that the intimidation of which
workers claimed to be the victims was often fictional. 'Sometimes what happens is this', said Nariman,
The millhands do not want to go to work for reasons which they believe exist. When somebody on behalf of the millowners asks them 'Why don't you go?' they have not got the
courage to say that they do not want to come [to work]. They say that they are intimidated and so we [sic] do not come. They narrate their grievances to the Union. If anybody
who commands their confidence asks them the question they would narrate their grievances.85
By pleading intimidation as their excuse for industrial action, workers attempted to establish their bona fides as loyal employees and thus to ensure they were given
back their jobs.
81
BMOA, Annual Report (1928), Chairman's speech, AGM, pp. v–vii; BMOA, Annual Report (1933), Chairman's speech, AGM, p. v.
82 BDEC, oral evidence, Dhaku Janu Lad, in GOB, Home (Special), File 550 (25) III B of 1938, p. 105, MSA.
83 BRIC, oral evidence, J. Addyman, File I, p. 85, MSA.
84 Proceedings of the BSEC, vol. I; p. 122, MSA.
85 BRIC, oral evidence, K. F. Nariman, File 6, p. 89, MSA.

Page 126
We have already seen that workers could exert some pressure on their jobbers in a variety of ways and it was by no means customary for the culmination to be
violent. Although there was no positively definable point at which the jobber's position would be entirely rejected, it was essential for him to bend with the political
temper of the mill district, to know when he should act with the workers and when he should act against them. It was particularly in the face of mass action, effectively
orchestrated by a powerful trade union, and extending to more than a single mill, that the limits of a jobber's power were exposed, and that employers appeared
vulnerable without the physical potency of the state. In periods of workingclass solidarity, the jobber's opposition or his participation in victimization could lead to the
desertion of his men, moral opprobrium from the community and the severance of the social and commercial ties upon which his position rested. At such times,
workingclass action to neutralize hostile neighbourhood alliances of all kinds was most successful. It is not intended to suggest that the ability of Bombay's workers to
resist their employers or shackle their neighbourhood leaders was by any means equal or uniform. Their place within the material as well as the muscular structure of
the neighbourhood registered differences between workers; some were plainly better equipped than others to absorb or counter their antagonists. Nor can it be said
that there was any linear development in the balance of power between the 'forces' for or against the labour movement, let alone that these forces in their entirety
remained consistently on either side of the divide. Clearly, the success of the Girni Kamgar Union enabled it in 1928 and 1934 decisively to alter the existing political
balances of the neighbourhood, and it was probably the case that workingclass action was in general most effective when the union was able both to protect workers
at the mill and coordinate their action in the neighbourhood.
The permanent social relations of the workplace, and of the industry, pushed strikes which began within the limits of the workplace into the wider arena of the
neighbourhood. As workers attempted to cope with the limits which this structure of control imposed upon them, paradoxically their actions acquired an important
political edge. Conventionally, we should consider a strike as a form of industrial or even political action, as an event which related directly to the workplace and
concerned particular groups of workers. However, as industrial action was forced into the public sphere, into the streets and neighbourhoods, the effects of industrial
disputes were generalized. In this wider context, the parochial disputes of a mill or a group of mills were placed before the mill district as a whole. By being placed in
the wider arena of the workingclass neighbourhoods, each individual strike became an essential part of the collective

Page 127
experience of Bombay's workers. As a result, the apparently limited nature of industrial disputes became essential to the process by which the social experience and
the social consciousness of the working class as a whole were forged.
IV
It has already been argued that the social exchange of the neighbourhoods shaped the perceptions of Bombay's workers and influenced the forms of industrial action.
But its ramifications were wider still. It exercised an important influence upon the character of workers' politics in the public domain. From the late 1920s onwards, the
communistled Girni Kamgar Union became the dominant force in the politics of the mill district. Not only was the GKU the only union to achieve a more or less
permanent presence in industrial politics but it also led every general strike in the industry after 1928. Throughout much of this period, it was subjected to considerable
repression by the state. In the early 1930s, the Bombay Millowners' Association withdrew its recognition of the union; in 1934, along with other communist
organizations it was declared illegal. As a result, the Girni Kamgar Union was at times incapacitated. But it was a measure of its achievement that although it was
subjected to severe repression and its members to victimization and disfavour, it was repeatedly able to reassert its ascendancy. 'Had it not been for certain
measures', the police commissioner admitted in 1935, referring to the Meerut arrests and the passing of such repressive legislation as the Criminal Law Amendment
Act and the Bombay Special Emergency Powers Act, 'the communists would no doubt have become a positive danger by this time.' For, although its activities 'have
been paralysed to a great extent by the internment of active communists . . . [and] they have comparatively few leaders and organisers . . . the subterranean activities of
the Communists are not effectively kept in check by the measures adopted by the Government from time to time'.86 As late as 1940, Bombay was still considered the
'nerve centre of Communist agitation in India'.87 Not only did the communists survive this repression, but they also succeeded in creating an active political tradition: in
the 1930s, their office became a landmark in the mill district and rival unions competed to adopt the name of the Girni Kamgar Union.88
The spread of support for the communists reflected changes that were
86 Commissioner of Police, Bombay, to Secretary, GOB, Home (Special), Secret no. 3757 B, 8 August 1935, in GOB, Home (Special), File 543 (77) of 1935, p. 77, MSA.
87 Departmental note in GOB, Home (Special), File 543 (42) of 1940, p. 16, MSA.
88 During the 1930s, three unions adopted this name.

Page 128
occurring within the political culture of the workingclass neighbourhoods, changes which were the outcome of growing conflicts, both in the workplace and outside. In
the process, tensions within the working class and between millworkers were exposed. It is not intended to suggest that the intervention of the communists heralded the
dawn of workingclass unity. However, it is also clear that from the late 1920s onwards, an impressive community of political sentiment formed around the
communists. As Syed Munawar, the 'moderate' trade union leader, said in 1929, 'Communist principles have captured the minds of textile workers to a great extent in
the Parel area.'89 For the millworkers in particular, the Girni Kamgar Union created the possibility of a sustained political expression.
The place which the communists came to occupy in the mill district was partly the result of the nature of their intervention in industrial politics. It has already been
argued that material conditions as well as the employer's policies made workers' combinations vulnerable at the workplace. This meant that trade unions had to
maintain an effective presence in the neighbourhood; at the same time, it also meant that they were excluded from the area of the daily social relations of the workplace
and forced to operate often at a level removed from the thrust of workingclass action.
The tension between trade union organization at the level of the individual mill and at the level of the whole industry was crucial to the determination of the politics of the
textile industry. The system of labour control based upon the jobber worked best at the most parochial level. At this level, what mattered was the extent to which
jobbers, acting within the context of the neighbourhood as a whole, were able to resist or incorporate pressures from the workforce. At this level, too, trade unions
were most easily rendered ineffective. As long as a significant proportion of jobbers and workers were not connected with a particular union, there was little need for
the employer to recognize its existence, and even when a trade union acquired any considerable influence amongst his workers, an employer could discipline or at
worst dismiss some of them as a warning to the rest. It was at a more general level that trade unions operated, making alliances with jobbers and then representing their
case to management, to the Millowners' Association and to government when necessary. Significantly, the first trade unions in Bombay were essentially pleaders'
offices where grievances were heard and services, such as the writing of leave notices and the drafting of petitions, were provided.90 Yet, to operate successfully at this
more general level, it was obviously essential to establish more than an ephemeral presence at the workplace.
89 BRIC, oral evidence, Syed Munawar, File 3, p. 267, MSA.
90 See Confidential Proceedings of the GOB, 1917, vol. 25, pp. 15–19, OIOC.

Page 129
The major problem which trade unions faced was their inability to act at both levels. Most trade unions were constrained by this intermediary position. As
intermediaries who built upon their jobber and neighbourhood connections, they were better placed to mediate in the workers' disputes than to lead them. Trade
unions, like the Kamgar Hitwardhak Sabha, the Social Service League and the Bombay Textile Labour Union (BTLU) never became company unions, but nor were
they free to champion the workers' cause. Following the momentum of workers' action could bring them directly into conflict with jobbers and other neighbourhood
patrons. It could also invoke the displeasure of the state and of the employers, whose benevolence and trust was vital to their political survival. For it was their
influence in ruling circles which made these unions valuable allies for the workers and the lesser leaders of the neighbourhood. 'Had the millowners been a little more
sympathetic towards the Union', the representatives of the BTLU mused upon the fate of their own organization, 'the success it had achieved would have been more
substantial and the Union would not have required to go through the agonies it went through after the 1928 strike.' It was 'only recently', the union argued in the
aftermath of the communistled strikes of 1928 and 1929, 'when an undesirable element has entered the trade union fold that the employers have begun to talk in terms
of sympathy towards the unions'.91 Their political alliances inhibited them in advancing the workers' interests and thereby also restricted their membership. At least, by
force of habit, these unions were better placed to act as advocates when they had no clients than as spokesmen when they had no audience.
Not only their material interests but also their conception of their own role in relation to workers limited the efficacy of their leadership. Organizations like the Kamgar
Hitwardhak Sabha and the Social Service League were concerned mainly with social work and the 'uplift' of the poor. They were, as the Labour Office reported, less
trade unions than 'associations for the welfare of their members'.92 Their aim was to rescue workers from the depths of ignorance. In response to low wages, they
suggested more education; as a solution to bad housing conditions, they tried to teach workers hygiene; faced with poverty they advocated thrift. Their strength lay in
speaking on behalf of the poor; in active struggle, they often disintegrated.
It is against this background that the intervention of the communists in the labour movement in 1927–28 was significant. It marked a radical
91 RCLI, Evidence, vol. I, part i, written evidence, the Bombay Textile Labour Union, p. 353. The outcome of these agonies was that by 1931 the union's membership figure stood at
56 and was to fall further to 20 in 1938; see Labour Gazette, 'Principal Trade Unions in the Bombay Presidency', passim.
92 Labour Gazette, 2:7 (March 1923), p. 26.

Page 130
transformation in the style and content of trade union leadership. The communists entered the labour movement in 1927 through the Girni Kamgar Mahamandal, a
trade union founded and organized by jobbers and mill clerks during the 1924 general strike. Since 1927, rationalization schemes had been introduced into certain
mills. Although their object was efficiency, these schemes increased workloads, created the possibility of greater unemployment and induced among the millworkers 'a
genuine fear of less wages'.93 As strikes followed these changes in work practices from mill to mill, it became apparent that individual resistance, however determined,
was doomed. As N. M. Joshi put it later, 'a strike in one mill does not and will not succeed. If there is discontent on a large scale there must be a general strike. Then
only the grievances have some chance of being redressed.'94 Between August 1927 and April 1928, strikes occurred in twentyfour mills.95 Under the impact of this
political determination among the workers, the leadership of the labour movement vacillated.
The only significant pressure in favour of a general strike came from the communists; but they were still incapable of carrying the Girni Kamgar Mahamandal with them.
To lead a strike when the workers' mood was militant offered the communists an invaluable opportunity of establishing organization among the workers. Moreover,
unlike the other unions, the communists of the Bombay Labour Group had two advantages. First, they attributed a positive value to industrial action, for larger
purposes than the immediate conflict, in developing the political and revolutionary consciousness of the working class. At the same time, their enthusiasm was not as yet
weighed down by neighbourhood or even jobber connections which they would have to defend. The Girni Kamgar Mahamandal and the Bombay Textile Labour
Union, on the other hand, with more established political connections were hesitant to risk their linkages in a strike liable to fail. As the strike wave spread across the
mill district, both groups were faced with the danger of being outflanked by the communists. Tensions between the two rival courses of action dominated the affairs of
both these unions. By March 1928, the Girni Kamgar Mahamandal divided and one of its founders, Mayekar, was expelled in a dispute over the control of funds.96
'What happened in this strike', as Dange said later, 'was that the rank and file was forcing the lead on the organisation.'97 This general strike lasted for six months. The
intensity of
93 Departmental note in GOB, Home (Special), File 543 (10) E Pt D of 1929, p. 25, MSA.
94 Proceedings of the BSEC, vol. I, p. 71, MSA.
95 Proceedings of the MCC, statement submitted by S. A. Dange, pp. 2413–15.
96 Proceedings of the MCC, examination of Arjun A. Alwe, p. 961; BPP SAI, 1928, no. 20, 19 May, para. 793; GOB, Home (Special), File 543 (18) C of 1928, MSA.
97 Proceedings of the MCC, statement submitted by S. A. Dange, p. 2424.

Page 131
class consciousness which was expressed in the period was never perhaps to be repeated.
The linkages which were forged in this strike placed the communists firmly in control of the GKM and enabled them to dominate trade union politics. They were now
forced to confront the problems posed by the structure of industrial relations. The initiative taken by the communist leadership in reflecting workingclass militancy
enabled them to establish their political presence at the level of the industry as a whole. To consolidate this support, it was imperative for the Girni Kamgar Union, as it
was now called, to penetrate the level of the individual mill. This was precisely what occurred in the following months. The general strike of 1928 had ended on the
basis of an agreement that the rationalization schemes would not be extended until the committee of inquiry appointed to investigate the dispute had reported. Between
October 1928 and March 1929, seventyone lightning strikes occurred as millworkers resisted victimization or zealously ensured that the agreement was not breached.
The Girni Kamgar Union's intervention in these disputes had, as Dange put it, a 'magical' effect upon organization.98 On 30 September 1928, the Girni Kamgar Union
had a membership of 324; by the end of that year, they boasted 54,000 members.99 The organizational achievement of this period was the mill committees which
sprang up throughout the industry. Workers from each department elected representatives to the mill committee. At the same time, the Girni Kamgar Union opened
several centres within the mill area for the enrolment of members and the collection of subscriptions, but especially to establish and extend connections with the
workers of their neighbourhoods. These centres supervised the work of mill committees in their area. The members of a mill committee would contact their centre as
soon as a dispute arose in their department or their mill. Each centre elected a committee, which in turn elected a managing committee for the union as a whole and to
whose decisions it remained subordinate. The committee of each centre was elected by the most effective unit of the union machine: the mill committee.100
The representatives elected onto the mill committee were responsible for the organizational tasks of the union in their department. They enrolled members and collected
subscriptions; they acted as watchdogs of the workers' interests; they formulated grievances and approached management to negotiate settlements; and if this brought
them no joy,
98 Ibid., p. 2507.
99 'Report of the Court of Inquiry into a Trade Dispute Between Several Textile Mills and their Workmen', 1929, p. 11, MSA.
100 See BRIC, oral evidence, Milton Kubes, File 5, pp. 209–13, MSA; see also Proceedings of the MCC, statement submitted by S. A. Dange, pp. 2498–537.

Page 132
they approached the union, or more frequently in practice, proceeded to strike. In this way, they brought the union to that microcosmic level of the individual mill and
department from which it had so effectively been excluded in the past. While it was their ability to intervene at this microcosmic level which enabled the Girni Kamgar
Union to gain such formidable support and create such an extensive organization, it was precisely their strength across the industry as a whole which prevented the
millowners from excluding them from the politics of the workplace. 'We are helpless', complained Sir Manmohandas Ramji in 1929. 'If we dismiss a man who is a
member of that union, the question of victimization comes in, and we create a strike. If today my mill is working partially and I suspect a man who belongs to that union
and try to dismiss him, there will be a strike next morning.'101 The strength of the Girni Kamgar Union across the industry enabled it to protect its members as well as
advance their interests at the level of the mill.
The mill committees linked, and operated at the junction of the workplace, the neighbourhood, the mill and the trade union headquarters. But the mill committees of
1928–29 did more than this – they also became 'parallel organs of supervision and control' in rivalry with the jobber and constraining his freedom of action. In 1928–
29, they sometimes seemed to give substance to Dange's claim that the Girni Kamgar Union 'overthrew the power of the jobbers and the head jobbers'.102 Through the
mill committees, workers gained access to the union offices. The result was to give meaning to the union as an alternative source of patronage, extending from the
workplace and the neighbourhood to the union headquarters, which operated at a level well beyond the jobber's reach. Their presence forced jobbers to choose
between making an alliance with the union to preserve their position with the workers and risk managerial disfavour, or else to ally with the management to break the
mill committee and isolate the union. As the union penetrated the workplace, it brought new complexities to bear upon the jobber's function of labour control.
However much the mill committees checked the jobber's power in the short term, it did not, contrary to Dange's claim, overthrow him.103 At times the union leadership
even found itself attempting to defend the jobber against the opposition of workers. At a public meeting to elect the mill committee for the Kohinoor Mill, on 24
November 1928, one section of the workers pressed for the exclusion of the head jobber of the weaving department and his six men. Dange and Alwe advocated
restraint: they argued that it would not be practical to exclude men of
101 BRIC, oral evidence, Sir M. M. Ramji, File 2, p. 367, MSA.
103 Ibid.

Page 133
influence, especially those who had the backing of their department. They suggested a compromise in the form of a resolution to warn that those who opposed the
majority opinion of the mill committee or ignored union policies would be removed.104 To attack the jobber, it was clear, the union would have to proceed with care. In
the short term, it was arguably sufficient and perhaps only possible, to constrain him. The mill committee could not overthrow the jobber; but their failure meant that as
the union's position across the industry weakened, the jobbers were able to reassert themselves.
In 1929, the Girni Kamgar Union, by leading another general strike, built upon the momentum established in the previous year and then exhausted it. Already, the
arrests and imprisonment of its most important leaders created chaos in the union's organization. In 1930, G. L. Kandalkar, its new president, declared his support for
the Congress and carried an important section of the union into the nationalist fold. In the face of growing unemployment in the industry, the millworkers' militancy was
seemingly diluted. As the overarching trade union organization grew weaker, its relationship with the mill committees grew more tenuous. The weakening of the bond
between the mill committee and the union or a decline in the activism of the union made it easier for the jobber to turn the mill committee into yet another institution
around which to consolidate his power. However, where they survived, mill committees formed a core of shopfloor organization, through which the Girni Kamgar
Union could rehabilitate itself.
To some extent, the organizational basis of 1928–29 was revived during the strike wave of 1933–34; moreover, it was diversified and extended more formally to the
neighbourhood. The Millowners' Association emphasized the vital role of the union's chawl committees in the conduct of the general strike of 1934.105 In December
1937, the police noted the fact that the organization of the union integrated both workplace and neighbourhood. 'They have gone to great trouble', it was reported, 'to
establish ''communist cells" in mills and industrial concerns, and in addition they have appointed Chawl Committees to influence the workers still further.'106 Although
the communists could not recreate their achievement of 1928–29, these changes enabled them to absorb the growing repressive pressures at the workplace and, at
the same time, to maintain their presence in the neighbourhood.
104 Proceedings of the MCC, vol. X, Marathi Exhibits, Girni Kamgar Union Minute Book, Public Meetings, pp. 6–7.
105 BDEC, evidence, extract from monthly report of the Labour Officer, BMOA, August 1935, in GOB, Home (Special), File 550 (25) III, of 1938, p. 181, MSA.
106 GOB, Home (Special), File 546 (13) B (1) of 1937–38, p. 7, MSA.

Page 134
The reassertion of the communist ascendancy out of the doldrums of the early 1930s occurred through industrial action, in the strike wave of 1933–34. This militancy
forced choices upon workers, jobbers as well as other trade unions. In 1934, when the communists attempted to persuade the Council of Action, composed of the
representatives of several unions, to support a general strike, 'quasicommunists such as Alwe and Abdul Majid felt they had to come in or be pushed aside' and the
Royists 'in order not to lose such influence with the workers as they had, felt impelled to join in and pose as communists'.107 That the communists were able to exert
such pressure on the other unions in 1934 suggests the extent of their recovery. That they recovered at all was due to the powerful base and the political sympathy they
had created in 1928–29.
It was probably their stance of continued opposition to the employers and the state which established for the communists their place within the political culture of the
neighbourhood. The communists came to be identified as the only political group untainted by their association with the state, for instance by nominations to provincial
and central legislatures, to royal commissions, and even to ILO conferences. This enabled the communists to present themselves as the one political group in the labour
movement which acted in the interests of the working class alone. When asked why the Girni Kamgar Mahamandal had permitted the communists to enter and work in
the union although it had recently rejected the leadership of the 'outsiders' of the Bombay Textile Labour Union, Arjun Alwe, President of the GKM, replied: 'we
believed to be true the fellowfeeling which they exhibited towards the workers'.108 After 1928, the communists' exertions in the workplace and neighbourhood served
to confirm, at least for some workers, this assessment.
Between the wars, the state intervened increasingly in the workingclass neighbourhoods. The effect of this intervention was not universally to antagonize workers.
Legislation was passed to protect trade unions and govern working conditions, to grant maternity benefits and to provide compensation for injuries and even to ensure
the prompt payment of wages. The police were known to arbitrate in labour disputes and occasionally even to ensure the payment of overdue wages.109 In practice,
however, there was little life in the new legislation; and the police, the most immediate point of contact between workers and the state, appeared increasingly as the
most organized of the repressive forces which con
107 W. R. G. Smith, Commissioner of Police, to R. M. Maxwell, Secretary, GOB, Home (Special), no. 3035 L, 20 June 1935, in GOB, Home (Special), File 543 (48) L, pp. 99–101, MSA.
108 Proceedings of the MCC, examination of Arjun A. Alwe, 12 August 1931, p. 972.
109 See, for instance, Edwardes, Bombay City Police, Appendix, p. 197.

Page 135
fronted the working class. Indeed, police action during strikes defeated pickets and aided blacklegs and in the process contributed to the suppression of workers'
demands and the destruction of their organizations.
Moreover, as conflicts between national and imperial interests were increasingly articulated in the political domain, they helped to clarify the relationship between the
workers and the state. During and immediately after the first world war, the living conditions of most workers, characterized by rising prices, high rents and general
scarcities, worsened considerably. Grain prices rose almost immediately after the war began, and the government had to take active measures to prevent food riots.
The opening of labour camps in Dadar in 1917 and the work of military recruiting officers led to considerable tension within the workingclass neighbourhoods of
Bombay.110 The impact of the first world war upon workers was to reveal to them that the Indian economy was 'now influenced by international factors'.111 Imperialism
signified another force which governed their conditions of life but over which they had no control. Several factors clarified these perceptions. First, the millowners were
closely identified with the social rituals of a foreign ruling class. Second, the nationalist campaigns of 1917–22 stirred people's minds and involved a racial self
assertion. It could affect the way in which workers related to their AngloIndian and Parsi supervisors who were closely associated with the British rulers. Third, the
economic campaigns of their employers also sharpened the lines of conflict between Indian workers and an imperial state. Indeed, at certain points, the state appeared
to be the cause of their worsening economic conditions and of their industry's problems.
Although the millowners were perceived by the workers as being socially associated with this ruling imperial culture, their attempts to confront the longterm depression
in the industry's fortunes brought them into conflict with the state. When the Government of India refused to abolish the excise duty on Indian mill production, the
Bombay millowners cut wages by 11.5 per cent. This wage cut led to a general strike. Indeed, when in the face of the threat of prolonged workingclass action, the
Government abolished the excise duty, the millowners rescinded the wage cut. Similar connections between the economic policies of the colonial state and the
worsening conditions of the industry and its workers were made by the capitalist class during its currency campaign of 1927–28. Within the labour movement it
appeared as if both the
110 'Statement relating to the disturbances in the City of Bombay in April 1919', in Bombay Confidential Proceedings, 1920, vol. 53, pp. 13–27, OIOC; 'A Report from the
Commissioner of Police, Bombay to the Government of Bombay Concerning Political Developments before and during 1919', in Curry Papers, Box IV, item nos. 54 and 55, Centre of
South Asian Studies, Cambridge; BPP SAI, 1917, no. 29, 21 July, para. 794.

Page 136
Government of India and the capitalists were arguing opposite cases while professing the interests of the working class, and that both posed as the guardians of labour
in order to promote their own particular interests. It was said that while the government's case for a higher exchange ratio rested upon the contention that a lower ratio
would depreciate wages and lead to serious strikes, the capitalists argued that in order to function at the higher ratio they would be forced to reduce wages.
It was not merely at one remove that workers were forced into confrontation with the state. Their economic struggles also brought them into political arenas. In the
immediate postwar period, a pattern of resistance and surrender to wage demands had established itself. Its consequence was to make the power of combination and
the effectiveness of industrial action increasingly clear. During the 1920s, as the millowners organized across the industry to influence the policies of the state, their
Association began to affect the management of individual mills. As the level of the individual mill and the industry were integrated, workers, too, had to act across the
industry to press their demands. As strike activity occurred on a larger scale, negotiations were conducted at more elevated levels. The mill manager no longer
conducted the case for the management alone; the centralizing initiatives of the Millowners' Association became increasingly important. The state intervened less
through the office of the police commissioner and increasingly from Government House. The mutuality of workers' interests became more evident and their conflicts
with the state occurred at new levels. A general strike, a matter of industrial politics, could entail visible forms of class confrontation: from the police escorting blacklegs
across the picket lines to the work of an arbitration court headed by a High Court judge or a civil servant, whose rulings were perceived to be unjust.
To a large extent, the political experience of the working class was constituted in relation to the state; this relationship in turn influenced the development of their
political consciousness. For instance, the police impinged upon the conduct of a strike in various ways. During the general strike of 1928, police reporters attended
workers' meetings; policemen supervised pickets at the mill gates and attempts were made to restrict their number to two;112 and when picketing was carried into the
neighbourhoods, the police presence extended to the chawls as well. At a meeting at the communist stronghold of Nagu Sayaji's Wadi in Parel on 1 June 1928,
according to the police reporter's account, Dange reminded his audience, 'the police have no right to come to your room without a warrant . . . Even if he comes in the
room with uniform but is not armed
112 Ibid., pp. 2438–9; BRIC, oral evidence, K. F. Nariman, p. 87, MSA.

Page 137
113
with a warrant, you can consider him a thief . . . You must protect your own chawl.' The opinion was being more readily expressed that, in general, the police had
shown greater solicitude for the millowners than for the strikers.114 For many people, the police came to represent not the guardians of the law but the long arm of
tyranny. 'Many things are not reported to the police out of fear', said one observer, who also noted that 'Hindus . . . always avoid to go to the courts and police.'115
The Borkar riot which occurred on 11 December 1928 in support of the communist leaders, two months after the general strike had officially ended, showed how in a
single moment the levels of neighbourhood and industrial and public politics could be combined. The origins of the riot dated back to the split within the old Girni
Kamgar Mahamandal in March 1928, when Mayekar was expelled from the union. Finding his old bases of support being pulled away from under him, Mayekar
came to lean upon his friendship with Keshav dada Borkar, gymnasium owner and neighbourhood boss of Ghorapdeo. Borkar's terrain at Ghorapdeo became
Mayekar's last refuge. Throughout 1928 Mayekar, now isolated within the labour movement, opposed the communists with the help of Borkar, and attempted on
several occasions to break up their meetings. 'For six months and more', reported Horniman's Indian National Herald, 'the leaders of the communistled Girni
Kamgar Union were repeatedly disturbed by his unwelcome presence which at once acted as a disintegrating factor on one section of the workers and an infuriating
phenomenon on the other.'116 The effect of Mayekar's intervention 'through his friend Borkar',117 at several communist meetings was interpreted very differently by left
wing sympathizers and by the police. While the police commissioner reported that frequent complaints were received that they were 'seeking to stir up trouble at the
communist meetings . . . but no serious clash occurred',118 the Indian National Herald's version was that the communist leaders 'went to the length of even dissolving
crowded meetings', to avoid confrontation. Indeed, at practically every meeting the leaders exhorted the men to remain restrained in the face of provocation.119 If the
Mayekar–Borkar alliance had been able to create a riot, this would have provided the police with the kind of opportunity they sought
113 GOB, Home (Special), File 543 (18) C of 1928, MSA.
114 BRIC, oral evidence, W. T. Halai, File 5, p. 81, MSA.
115 BRIC, oral evidence, Dr P. G. Solanki, File 6, p. 165; see also, oral evidence, G. L. Kandalkar and V. H. Joshi of the Girni Kamgar Union, File 16, pp. 65–7, MSA.
116 Indian National Herald, 7 December 1928.
117 Letter, Commissioner of Police, Bombay, to Secretary, GOB, Home, Bombay no. 5395 L, 13 December 1928, in GOB, Home (Poll), File 265 of 1928, MSA.
118 Ibid.
119 Indian National Herald, 7 December 1928. Mayekar, claimed the paper, only 'masquerades as a labour leader and is, in fact, alleged to be an agent of the Criminal Investigation
Department'.

Page 138
to take further repressive action. During the strike, Mirajkar had told a strikers' meeting at Kalachowki on 2 August 1928 that:
our strength lies in unity and peace. On Monday attempts are [sic] made to disturb our peace with the use of lathis. If they use their lathis we can also retaliate in the same way;
but as we want to win the struggle, we must keep peace. If we disturb peace, lathis and guns will be used and under that threat they will try to put men into the mills. But our men
are firm and they already know the knavery of the millowners. They [strikers] have already resolved not to fall prey to the hirelings of the millowners.120
On 11 December 1928, a message was received at the Girni Kamgar Union office, calling for their assistance in connection with a dispute at the David Sassoon
Spinning and Manufacturing Mill at Ghorapdeo – the heart of Borkar's territory. The communist leader R. S. Nimbkar, P. T. Tamhanekar, Govind Kasale and a few
others who went to investigate found the complaint to be false. As they left the mill, Nimbkar and his associates were set upon and attacked by Keshav Borkar and a
gang of about twenty men. Complaints lodged at the local police station, however, 'of course failed to trace the assailants'.121 The following morning workers from the
David Sassoon, Morarji Gokuldas, Moon and Shapurji Broacha mills did not resume work, out of sympathy for their bruised leaders. At a meeting of the Girni
Kamgar Union at Poibavdi that morning, Kasale, who had taken the brunt of the attack, displayed his wounds. Clearly the temper of the meeting was highly charged.
Plainclothes policemen in the crowd were identified and assaulted.122 Within minutes, about 500 workers set off towards Borkar's house, 'with the intention
presumably of settling accounts with him. He got intimation of their advance and left his house.'123 By the time the crowd reached Borkar's house they were estimated
to be more than three thousand strong. The contents of his house were pulled onto the street and a bonfire was lit. His furniture and cooking utensils were damaged;
the house was ransacked, the tiles on his roof were removed and thrown away; his gymnasium was wrecked.124
As the morning wore on, mill after mill was brought out on strike by workers who gathered at the gates and stoned the premises until those who had remained inside
the mill were locked out by the management. The
120 GOB, Home (Special), File 543 (18) C of 1928, MSA.
122 Telegram, police commissioner to Secretary, GOB, Home (Special), no. 5368 L, 12 December 1928, in GOB Home (Poll), File 265 of 1928, MSA.
123 Letter, Police Commissioner to Secretary, GOB, Home, Bombay no. 5395 L, 13 December 1928, in ibid., pp. 41–5.
124 Report of H. C. Stokes, Inspector, Byculla Police Station, D. Division in ibid., pp. 13–15; Times of India, 21 December 1928.

Page 139
police were alerted and brought into action: the result was riot. By the time Inspector Klein of Bhoiwada Police Station met the crowd on Suparibagh Road, they
were, according to him, armed with sticks, bamboos, iron rods, gymnastic paraphernalia and 'obviously bent on mischief'. With a small force and awaiting
reinforcements, the police attempted to stop this crowd by throwing a cordon across the road. The result was that the police were routed. Crowds appeared from
every direction and hemmed the police in on all sides, while 'stones were also being thrown from the rooms and windows of the neighbouring houses'. When Klein
fired with his revolver, the crowd 'held back slightly but came on with renewed vigour'. The constables began to climb into the police lorry and Klein failed to 'force
them to stand fast'.125 His deputy was set upon by the crowd and badly beaten. Klein was forced to take refuge in a nearby building while 'the mob furiously attacked
the house from outside'.126 The police lorry had to be chased by Superintendent Spiers, who noticed it hurtling away from the action.127 Spiers returned to the scene of
action, fired into the crowd, 'drove the rioters helterskelter off the roads', and rescued the brave if battered Klein.128 However, several features of this riot must be
noted. First, the nature of the police intervention had the opposite effect to what was intended: for instance, the attempt to stop the crowd with an ineffective cordon at
first, and later by the show of a pistol, and an attempt by some constables to snatch away the red flag which some workers were carrying aggravated the situation.129
Second, five workers died in the riot, four of them from bullet wounds. 'We must put an end to the idea prevailing in police circles', one newspaper reporter wrote
indignantly, 'that human life is so cheap that it can be wantonly destroyed on the slightest provocation.'130 However, it must be taken as an indication of popular anger
and determination that the crowd withstood considerable police firing and returned to counterattack, 'with renewed vigour'. Third, the 'mob' on the streets was neither
undifferentiated nor, in its response, exceptional to workingclass sentiment. As the police commissioner noted, 'I am told by the officers that the stones were being
hurled not only by the rioters on the road, but by millhands who were in the rooms of houses adjoining the road. These stones could not have been obtained on the
road itself and it appears to me that the strikers were out
125 Report by Inspector Klein, Bhoiwada Police Station to Superintendent of Police, E. Division, Bombay in GOB, Home (Poll), File 265 of 1928, pp. 21–5, MSA.
126 Letter, Commissioner of Police, Bombay, to Secretary, GOB, Home, Bombay no. 5395 L, 13 December 1928, in ibid., pp. 41–5.
127 Report by W. D. R. Spiers, Superintendent, E. Division, in ibid., pp. 27–31.
128 Letter, Commissioner of Police, Bombay, to Secretary, GOB, Home, Bombay no. 5395 L, 13 December 1928 in ibid., pp. 41–5.
129 Indian National Herald, 7 December 1928.
130 Ibid.

Page 140
131
yesterday for mischief and had brought the stones with them.' Yet the scene of action – Suparibagh Road – was one of the main thoroughfares through the mill
district. If we are to believe that 'this strike was mainly brought about by those (communist) leaders and was done very secretly and in a well organized way', we must
also believe that this conspiracy involved a vast proportion of the Bombay working class. Significantly, what began as an expression of protest on an impressive scale
against the antiunion activities of Borkar and his men, culminated in a fullscale battle with the police.
While it was by supporting the militant tendencies within the neighbourhood, and through their apparent refusal to collaborate with employers and the state that the
communists staked their claim to be the party of the working class, this claim was not always accepted. In the early 1930s, for instance, their refusal to associate with
the Congress and their attempts to lead their constituents – exhausted by the strikes of 1928–29 and faced with the threat of unemployment – into battle once more
cost the communists membership, neighbourhood allies as well as political sympathy. The support for the communists was not a simple fusion of shared antagonism
towards the capitalist class and the state. Clearly, the Girni Kamgar Union brought to trade union politics a fresh concept of the conduct of workingclass politics, and
in contrast to the condescension of the vacuous sermons of improvement of the early labour organizations, a new concern with the daily issues of workingclass life.132
These departures in trade union leadership arose, in a sense, from necessity. Faced with exclusion at the level of the individual mill, the communists lacked the
resources – essentially the good will of the state and the employers – to renew their linkages through patronage. To function as a trade union at all, the Girni Kamgar
Union had to intervene energetically in the disputes of individual mills and build up enough support across the industry to prevent the employers from disregarding them.
In one sense, the price of survival itself was militancy. However, this brought them into immediate conflict with the state. The conduct of a strike, which required
organization and action in the neighbourhood, carried workers into public forms of confrontation and shaped their political consciousness.
131 Letter, Commissioner of Police, Bombay, to Secretary, GOB, Home, Bombay no. 5395 L, 13 December 1928, in GOB, Home (Poll), File 265 of 1928, p. 45, MSA.
132 Their leaflets, flysheets and public meetings dealt with such questions as jobber tyranny, methods of wage calculations, the shortcomings of 'efficiency' schemes which increased
workloads without improving machinery, the use of the rotation of shifts to weed out troublesome workers and the causes of unemployment. See, for instance, the communist fly
sheets and leaflets collected in GOB, Home (Special), File 543 (46) of 1934 and File 543 (46) Pt 1 of 1934, MSA.

Page 141
'In recent times', K. F. Nariman was to say in 1929, 'a new spirit of organization and class consciousness has come into existence among our labouring classes.'133 But
it is difficult to estimate the impact of this class consciousness upon other competing social identities amongst Bombay's workers. It would be misleading to suggest that
the response of Bombay's workers to the growth of industrial action and the communist ascendancy in labour politics was in any sense uniform. The possibilities of
action varied with their village connections, their position in the neighbourhood and their bargaining power in the workplace. For instance, weavers, working in the
most profitable and rapidly growing sector of the industry, and protected by their level of skill, formed the most militant section of the workforce; while Mahars who
manned the unskilled jobs in spinning departments or north Indian workers, whose lines of supply from their villages were weak, were more easily contained. The
predominant cultural influence within the labour movement was exercised by Marathas from Ratnagiri and Satara. 'Labour activity in Bombay', Jhabvala was to say,
considering his isolation, 'is largely Mahratta in its nature. The leaders must be conversant firstly with the Mahratta language, secondly with Mahratta habits of life and
with a good deal of social outlook upon life that is Mahratta partly in its character.'134 On the other hand, after 1929, Muslim workers were probably increasingly
alienated from the labour movement in Bombay, partly no doubt as a consequence of increasing communal tension within national politics.
Many of these cultural differences were developed into political conflicts and sectarian rivalries by the actions of the employers and the state. As we have seen, the
jobber system operated along the lines of these cultural divisions; it not only facilitated strikebreaking but also could, if necessary, enable employers to replace one
group of workers with those of another caste or religion. Indeed, the communal riots of 1929 began during a strike when Hindu workers tried to stop Muslims from
going to work.135 It is probable that industrialization, far from dissolving caste, strengthened its bonds. The cotton textile industry did not depend upon the perpetuation
of these bonds, but it profited greatly from their use. Caste should, therefore, be seen less as a cultural condition whose
134 Proceedings of the MCC, statements made by the accused, noncommunist series, examination of S. H. Jhabvala, p. 756. This comment must also be read in the light that Jhabvala
by his own admission 'knew very little Mahratti' and was, when he spoke these words, apparently isolated within the labour movement.
135 Memorandum by Director of Information, GOB, 3 May 1929 in GOB, Home (Poll), File 344 of 1929, pp. 113–15, MSA.

Page 142
primacy was being challenged by the emergence of 'class' than another important tension embedded within a class context.
This essay has attempted to depict a network of social relationships out of which the workingclass experience was formed. It was in the neighbourhood that the
classic picture of the Indian working class, bound immutably by their changeless past, their powerless present and their hopeless future was most apparent. Yet the
neighbourhood, which was integral to the relationships of the workplace, became an important base for industrial and political action. It was here, where tensions
within the working class were played out, that the solidarities of class also received their most public expression.

Page 143
5—
Workers, Violence and the Colonial State:
Representation, Repression and Resistance
That Indian workers were prone to spontaneous and violent action was a commonplace among policemen, civil servants and employers. This perception was not
confined to colonial rulers or Indian elites alone. It expressed and informed the anxieties of wide sections of the population. As rural migrants, illiterate and uneducated,
devoid of factory experience, short of commitment to the industrial setting, apparently lacking in proletarian maturity, Indian workers were deemed to be inherently
rough and volatile. They were believed to be acquiescent and incapable of organization, but easily provoked, liable to violent outbursts and particularly vulnerable to
the blandishments of political agitators. Their volatility was often perceived as a function of their preindustrial character. Such workers, it was supposed, could not
readily form, or sustain, trade unions, and their inability to organize only reinforced, in this view, their tendency to express their grievances through violence.
Observers commented on the frequency of lightning strikes. Trivial grievances, they noted, led to immediate closures. Employers complained that workers often went
on strike without formulating their grievances and after the strike had begun would put forward several and extravagant claims. Strikes, it was said, were maintained by
threat, intimidation and assaults upon 'loyal workers'. Trade unions that formed usually failed to formulate demands or negotiate concessions and, if they did, they could
rarely persuade their members to accept any settlements they might achieve.1 As a result, trade unions proved incapable of channelling workers' grievances into
organized and disciplined expression. In any case, in this view, rural migrants were incapable of organization. The circularity of this reasoning captured one of the
numerous contradictions that beset the colonial discourse about labour: their rural and preindustrial mentalities rendered workers too volatile to form trade unions, but
1 Report of the Industrial Disputes Committee, 1922, in Labour Gazette, 1:8 (April 1922), 24.

Page 144
2
the weakness of trade unions explained why they remained volatile. If the preindustrial culture and rural mentalities of workers, in this discourse, predisposed them to
violence, it was also supposed that their backwardness rendered them the innocent prey of political agitators who, for their own particular purposes, were able to
manipulate, incite and provoke their followers to act in ways which damaged the longterm interests of both labour and capital. This perception of their volatility in
politics was intimately related to the parallel notion of their immunity to factory discipline. Workers who were reluctant to commit themselves to regular employment
could hardly be expected to submit themselves to the constraints of collective bargaining or political organization.
The truisms of contemporaries have now been enshrined as the dogma of historians. If contemporary observers often put forward this picture of spontaneous and
spasmodic actions, later historians have done little to disturb it. An impressively wide range of scholars, employing different methods and pursuing diverse, even
conflicting, objectives, appear to agree that Indian workers were given to sudden outbursts of violence and have taken over intact the contemporary characterizations
of their roughness and brutality. Morris D. Morris explained 'the violent and catastrophic behaviour' of Bombay's millworkers in terms of the fact that their unrest
manifested itself before 'strong trade unionism could give that militancy any sense of direction and discipline'.3 For the most part, 'their sullen frustration could only
express itself in seething discontent and a susceptibility to strike at the slightest provocation. Whenever articulate leadership appeared to kindle the spark, protest could
burst out in an industrywide conflagration'.4 Thus, 'in the tumultuous, fearridden and violent atmosphere of the mills', it was plain to see 'the ease with which the
operatives could fall under the Communist influence and be turned into a revolutionary political force threatening the very fabric of the established social order'.5
According to Richard Newman, 'the typical manifestation of discontent in the Bombay mills was . . . a lightning strike, unpredictable, shortlived and unsuccessful'.6
This pattern of industrial action 'arose out of the millhand's lack of commitment to industrial life', for in view of his attachment to his village base, 'he seldom felt it
necessary to fight for reforms in his urban environment'.7 A. D. D. Gordon found the city's labour force 'highly volatile', 'highly suggestible' and 'subject to periodic
bouts of riots'.8
2 This contradiction in the colonial discourse is unwittingly replicated in Arnold, 'Industrial Violence in Colonial India'.
3 Morris, The Emergence of an Industrial Labour Force in India, pp. 185–6.
4
Ibid., p. 207.
5 Ibid., pp. 185–6.
6 Newman, Workers and Unions in Bombay, p. 68.
7 Ibid., p. 67.
8 A. D. D. Gordon, Businessmen and Politics: Rising Nationalism and a Modernising Economy in Bombay, 1918–1933 (New Delhi, 1978), p. 45.

Page 145
These characterizations of workingclass politics were not confined to Bombay. Dipesh Chakrabarty provided a gloss on the complaints of mill managers, caught in
troubled times, and the strike reports of police commissioners, when he observed that 'workingclass protest in the Calcutta jute mills was frequently marked by a
strong degree of physical violence or personal vengeance'.9 For him, the propensity of workers to violence was both a symptom and a consequence of India's
'hierarchical, precapitalist culture'. On the other hand, communal violence was the hallmark of the rural migrant. In the 'sudden emphasis placed by millhands on
communal issues' in the 1890s, Chakrabarty discerned the influence of 'an ''immigrant mind" at work'.10 In late nineteenthcentury Madras, according to David
Washbrook, 'Factory hands, in the difficult state of assimilation into an urban proletariat . . . could be drawn into violent demonstration, without much difficulty or
particular cause.'11 According to Chitra Joshi, violence and crime were a characteristic of early twentiethcentury Kanpur's 'large population of casual poor', mostly
'migrants from the neighbouring villages', a 'marginally employed subproletariat', a volatile 'class of desperate people with dashed hopes', many of whom 'were
apparently willing to do anything for a few annas'. Although they were to be distinguished from the industrial workers, 'the lives of these people were closely
interwoven with those of the workers' and 'the milieu' of 'the worker immigrants' was 'one where street fights, assaults and murders were familiar occurrences'.12 In a
similar portrayal of the undifferentiated roughness of the migrant poor, Gyanendra Pandey observed that Kanpur, 'the fastestgrowing city in the province and easily the
most overcrowded', substantially 'covered with slums' and inhabited largely by 'recent migrants', was by 1930 'noted for its volatility':
It had more than its fair share of the goonda and badmash groups which abound in any large Indian city today and are often in the pay of wealthy men and willing to perform any
service for the money. Some of them engaged in cocaine smuggling, trade and gambling, and readily took to other criminal activities.13
Similarly, Veena Das has sought to explain collective violence in presentday South Asia in terms of
9 Chakrabarty, 'On Deifying and Defying Authority', 127; Chakrabarty, Rethinking Working Class History, ch. 5.
10 Chakrabarty, 'Communal Riots and Labour', 145, 148–9. For a similar recourse to 'thugs', 'goondas' and 'criminals' in explanation of communal riots, see Das (ed.), Mirrors of Violence.
11 D. A. Washbrook, The Emergence of Provincial Politics: The Madras Presidency, 1870–1920 (Cambridge, 1976), p. 249.
12 Joshi, 'Bonds of Community, Ties of Religion', 261–3.
13 G. Pandey, The Ascendancy of the Congress in Uttar Pradesh, 1926–1934: A Study in Imperfect Mobilization (Delhi, 1978), pp. 131–2.

Page 146
the large number of new migrants who live on the margins of the city . . . Generally, such migrants constitute the floating populations which become potential recruiting ground for
the underworld activities of smuggling, drugpeddling and hired assassination . . . The inhabitants of these slums . . . become a human resource for conducting the underlife of
political parties . . .
. . . the unorganized labour sector [which] provides personnel for smuggling drugs, weapons, [sic] and the crowds and processions for various political parties – is also
responsible for providing the personnel that can be mobilized into a hostile crowd.14
In perhaps the most explicit statement of the case, David Arnold has argued that 'industrial violence' was 'a common and persistent feature of labour relations in
colonial India'. The crucial problem for analysis, he tells us, is what 'disposed workers to express themselves through violent rather than peaceful means'.15 This violent
disposition of Indian workers is attributed to a tradition of violence among the rural poor, which they carried with them into the towns and factories; 'the strength and
persistence of preindustrial forms of protest' among workers at 'an early stage of industrialization'; and the failure of trade unions to develop – in a circular argument –
partly because of the migrant and preindustrial character of the workforce, and partly because of 'the colonial and racial context of industrial labour'.16 Obstructed
from forming trade unions, industrial workers, primarily rural migrants, resorted to the kind of behaviour they innately knew best: 'by default they clung to their violent
traditions'.17 Violence is, thus, taken to be the behavioural attribute of a
14 Veena Das, 'Introduction: Communities, Riots and Survivors: The South Asian Experience', in Das (ed.), Mirrors of Violence, pp. 12–13.
15 Arnold, 'Industrial Violence in Colonial India', 235.
16 Ibid. What Arnold has called 'the peculiar colonial and racial context' of Indian industry is offered by him, along with the migrant character of the labour force and the failure of trade
unions to develop, as three factors which explain workers' disposition to violence. Having first ascribed industrial violence to the nature of the labour force, he asserts later that
'colonialism and racism were the basic [emphasis added] causes of the high level of industrial violence in India'. These are large categories and they proliferate confusion. After 1947,
when managerial and minor supervisory ranks in all industries had been Indianized, there was, according to Arnold, no great diminution of violence. Arnold attempts to resolve this
paradox by arguing that 'the Indian state apparatus, especially the police, and Indian and international managers of the post independence period . . . have made their own attitudes
towards labour that formerly characterised colonial India' (255). So what, we might well ask, was either racial or colonial, about these attitudes towards labour? Similarly, Chakrabarty
has argued that 'the authority of the mill manager [in the Bengal jute mills] was bolstered up by his position as a member of the ruling race'. But, he confesses, 'in some respects one
also cannot help noticing the essentially Indian nature of this authority'. This finally led Chakrabarty to explain that 'the word colonial is meant here to include what was indigenous to
Indian society'. Chakrabarty, 'On Deifying and Defying', 132–3; and Chakrabarty, Rethinking Working Class History, p. 166.
17 Arnold, 'Industrial Violence in Colonial India', 254, 236–42; for a more nuanced statement of a similar position, see Chakrabarty, Rethinking Working Class History and Chakrabarty,
'Communal Riots and Labour'. In any case, the constellation of forces

Page 147
particular social group – in this case, a group as large, diffuse and heterogeneous as the Indian rural poor – and, as a corollary, the cultural inheritance of the urban
working classes. Their propensity to violence, it would seem, was inherent, even natural, to their being.18
These perceptions of the violence of the rural poor, the 'marginallyemployed subproletariat' of the towns and industrial workers at the early stages of industrialization
have been given a transcendent theoretical status by Ranajit Guha who wrote in his manifesto for Subaltern Studies:
Elite mobilization tended to be relatively more legalistic and constitutionalist in orientation, subaltern mobilization relatively more violent. The former was, on the whole, more
cautious and controlled, the latter more spontaneous.19
This uninhibited spontaneity and violence formed, according to Guha, whatever its actual social context or manifestation, 'the paradigm of peasant insurgency'. Colonial
officials would not have joined in praise of the rebellious spirit or the celebration of its propensity to violence but they would scarcely have quarrelled with this portrayal
of the roughness and brutality of the poor and they would almost certainly have shared its condescension.
These characterizations of, and explanations for, workers' violence in India have been inseparably intertwined with assumptions about the nature and consequences of
social transformation and the behavioural attributes of particular social groups. The disposition of Indian workers to industrial violence is thus closely related to their
rural origins, their status as casual labour and their incomplete adaptation to the industrial setting. In some accounts, it is readily, indeed explicitly, extended to a
perception of the ease with which 'marginal' proletarians may be recruited to criminal, communal or political violence.
Such characterizations of workingclass violence have not been limited to the historiography of India. They have been shared by scholars working
which disciplined workers at the point of production devastated their attempts at combination more consistently, if less visibly, than police action against strikers or the
prosecution of trade union activists. It should not be assumed that the struggle to form trade unions was waged solely or even primarily in relation to and against the colonial
state. See ch. 3 above. For the specific case of the Bombay textile industry, see Chandavarkar, The Origins of Industrial Capitalism, chs. 7 and 8.
18 So the 'popular culture' of the rural poor and the urban working classes, of which this violent disposition is deemed to be a part, turns out to be little more than an assumption about
'the nature' of the rural poor who inhabit it. At best, it would appear that the attempt to bury one myth about the Indian rural poor – their passivity – has resulted in the resurrection of
another, culturally specific stereotype of their violence. It is perhaps salutary to recall that images of peasant brutality, like those of rural idiocy, are the product of postindustrial
culture.
19 R. Guha, 'On Some Aspects of the Historiography of Colonial India', in Guha (ed.), Subaltern Studies, vol. I, pp. 4–5.

Page 148
on diverse regions and within divergent and conflicting intellectual traditions. The assumption that the level and incidence of violence in society provides an index for
measuring the extent and direction of social change is rooted in a venerable functionalist tradition. Similar assumptions have informed Marxist accounts of the evolution
of forms of protest from 'primitive rebellion' and 'everyday forms of resistance' to revolutionary action, from primordial, through reformist and trade union, to
revolutionary socialist consciousness.20 Just as workers who are insufficiently assimilated to the factory are seen in functionalist accounts as being the most volatile, so
in the Marxist variant, the perpetrators of violence, disorder and even crime are to be found among the insufficiently proletarianized: the casual poor, the rural migrant,
the lumpenproletariat.21 Not only do these arguments derive changes in political style or even consciousness from changes in the structure of the economy, sometimes,
more crudely, from quantitative shifts in the extent of industrialization, but they also portray the development of workingclass politics in terms of a general progression
from spasmodic and spontaneous violence to organized and disciplined action.22 Sometimes, this shift is portrayed in terms of the advance of civilization. Thus, for
instance, Lawrence Stone discerned in early modern England the 'slow, downward drift of less violent cultural norms' as polite manners gradually diffused from
'intellectuals, lawyers, nobles and bourgeois' to the 'violenceprone poor'.23
20 E. J. Hobsbawm, Primitive Rebels: Studies in Archaic Forms of Social Movement in the 19th and 20th Centuries (Manchester, 1959); James Scott, Weapons of the Weak:
Everyday Forms of Peasant Resistance (New Haven, 1985); James Scott and B. J. T. Kerkvliet (eds.), Everyday Forms of Peasant Resistance in Southeast Asia (London, 1986).
21 S. Hall, C. Critcher, T. Jefferson, J. Clarke and B. Roberts, Policing the Crisis: Mugging, the State and Law and Order (London, 1978).
22 C. Tilly. 'The Changing Place of Collective Violence', in M. Richter (ed.), Essays in Theory and History: Approaches to the Social Sciences (Cambridge, Mass., 1970), pp. 139–64. For
Charles Tilly, for instance, modernization transformed 'local food riots' and other 'scattered' and 'sporadic' protest into 'organised demonstrations, bloody strikes and sophisticated
attempts at revolution' and revealed in 'urban rebellion' 'increasing signs of durable formal, politically active working class organisation'. See also Charles Tilly, Louise Tilly and Richard
Tilly The Rebellious Century, 1830–1930 (London, 1975); a similar approach is followed in Dick Geary, European Labour Protest (London, 1981). It is not surprising, perhaps, that
Arnold was to find it 'possible to see' in early twentiethcentury South India the same transformation which Tilly has described for Europe (Arnold, 'Industrial Violence', 241). The
notion that endemic violence characterizes societies in the process of transition to modern capitalist (or industrial) societies has been so deeply entrenched that one scholar set out to
prove that early modern England was less violent than other contemporary societies were, in order to clinch the argument that it had already, by the seventeenth century, passed
through this period of transition. Had it not done so, so the reasoning goes, seventeenthcentury England would have witnessed endemic violence. A. Macfarlane, The Justice and the
Mare's Ale: Law and Disorder in Seventeenth Century England (Oxford, 1981).
23 L. Stone, 'Interpersonal Violence in English Society, 1300–1980', Past and Present, 101 (1983), 29–30. Stone draws primarily upon T. R. Gurr, 'Historical Trends in Violent

Page 149
Yet, it would seem, this diffusion of manners ground to a halt in the face of Irish immigration to nineteenthcentury England and of AfroCaribbeans in the mid
twentieth.24 In recent times, the investigation of popular politics has been considerably advanced by the scrutiny to which historians have increasingly subjected the
meaning and usages of terms like 'the mob', 'the crowd', 'riot' and more recently, 'crime'. Yet historians have continued to conceptualize 'violence' in unexamined ways.
Contemporary observers, as well as later historians, have often lumped together and conflated diverse forms of behaviour under the category of violence. Yet, as this
chapter will argue, violence does not constitute a single, readily defined phenomenon. Indeed, the category has amounted to little more than a misleading label for
disparate actions and has subsumed forms of behaviour with only the most tenuous similarities. It can scarcely, therefore, be offered as a problem for analysis in itself
or accepted as an organizing concept for social investigation. By failing to differentiate between different kinds of violent acts, and the specific contexts in which they
arose, historians in quest of the 'popular culture' of the urban poor have – like contemporaries, seeking to discipline and control them more effectively – generated a
paradigm of their roughness and brutality and, thereby, often unwittingly, have replicated the colonial discourse. But the problems posed by its conceptualization are
not always made easier by breaking it down into its specific parts. Industrial violence, for instance, has included various and fundamentally different forms of action,
from lightning strikes to industrial sabotage, from situations of potential violence to active clashes between strikers and blacklegs or, indeed, the police; and sometimes,
significantly, the threat or even the perception of the threat of violence. 'Criminal', 'collective' or 'crowd' violence have signified acts as diverse as those lumped together
as industrial violence and have offered categories no more coherent for social analysis. To dispute the validity of the category or the coherence of the phenomenon of
violence is to question the very possi
Crime: A Critical Review of the Evidence', Crime and Justice: An Annual Review of Research, 3 (1981), 295–353. For Gurr, early modern Englishmen 'were easily provoked to violent
anger, and were unrestrained in the brutality with which they attacked their opponents'. Ibid., p. 307, quoted by Stone, 'Interpersonal Violence', 25. See also J. A. Sharpe, 'The
History of Violence in England: Some Observations', and L. Stone, 'A Rejoinder', in Past and Present, 108 (1985), 206–15; and J. M. Beattie, Crime and the Courts in England,
1660–1800 (Oxford, 1986), pp. 132–9, especially pp. 136–8; J. S. Cockburn, 'Patterns of Violence in English Society: Homicide in Kent, 1560–1985', Past and Present, 130 (February
1991), 70–106.
24 For the parallels in the portrayals of the roughness of the casual poor, especially Irish and blacks in London in the nineteenth and twentieth centuries, see Jennifer Davis, 'From
Rookeries to Communities: Race, Poverty and Policing in London, 1850–1985', History Workshop Journal, 27 (spring 1989), 66–85.

Page 150
bility of a general explanation for patterns of behaviour which are identified as violent.
But the argument can be pushed further. The reporting of violence has largely determined how the evidence of its occurrence is recorded and its significance evaluated.
Whether workers were believed to be either peaceful or violent in disposition reflected more accurately the perceptions of the observers than the behaviour of the
actors. These perceptions did not reflect reality; in fact they frequently distorted it. The social phenomenon of violence, and its conceptualization, was a construct of
public discourse and private anxieties. That industrial and political action in India, as elsewhere, sometimes took violent forms is not at issue. Rather, the emphasis in
this chapter will rest upon how and why perceptions of violence came to be constructed. The roughness and violence of the working classes and the urban poor, in the
dominant discourse, was the necessary attribute of a social group, which the ruling classes and the state were concerned to discipline, whose resistance they were
anxious to break and whose political movements they were determined to repress.
This chapter attempts to delineate the ideological and political processes by which a discourse about the roughness and violence of the working classes came to be
constructed and the portentous political consequences to which it gave rise. It first examines more closely the opacity of the term 'violence' and the ambiguities which
were inherent in its use. Their effect has often been to proliferate confusion rather than to facilitate clarity in historical and sociological analysis. It then proceeds to
analyse how a discourse about the violence of the working classes came to be constructed in the context of an increasingly militant labour movement and the particular
social groups which came to be characterized in its terms. Characterizations of workingclass violence reflected widespread anxieties about the social order. In the
context of the 1928 and 1929 general strikes in Bombay, led by the communist Girni Kamgar Union, they enabled the colonial state to legitimize its interventions to
contain and repress the industrial and political action of the working classes and, in particular, facilitated and seemingly justified the use of a criminal conspiracy case to
destroy their leadership. The language of violence was not merely fashioned by the colonial state and ruling elites as a blunt instrument of repression; rather, its political
significance derived from the fact that large sections of the population subscribed to its usages. Crucially, in the late 1920s, this language of violence appeared so
plausible to so many people not simply because of the general strikes or an apprehension of the red menace but specifically because of their experience of communal
riots in the city in February and May 1929.

Page 151
II
The meanings, indeed usages, of violence have never been unproblematic. Although the term implies physical assault and the use of physical force, it is also used in
situations which do not necessarily involve actual violence. It could denote the threat of violence or the awareness, in a particular moment of social action, of the
possibility of its occurrence. At times, it has meant little more than the generalized and random perception of the roughness of the lower orders, or the threat posed by
particular social groups. The criminal charge of personal assault includes the possession of an offensive weapon or simply threatening or disorderly behaviour.
Obviously, the threat of violence might obviate the need to use it. But the threat of violence itself is often no more than the subjective perception of its possible use; and
it is only in the context of specific circumstances that it becomes possible to appreciate and sometimes not even then, the shifting threshold at which violence is deemed
necessary or its threat perceived as such by the actors involved.25 The term 'violence' crept into public discourse in India with the full force of these ambiguities.
The scope of these ambiguities is most fully appreciated in the light of particular situations of conflict. For instance, industrial sabotage did not necessarily mean the
physical destruction of machinery; it was used to describe nonviolent actions by which the workforce impeded the attainment of maximum levels of production.
Attempts by workers to retain some control over their own labour, or at best, their refusal to surrender complete control over it, could from a different vantage point
appear as the unruly defiance of the imperatives of factory discipline. At one level, the perception of industrial violence, like the practice of industrial action, arose out
of the permanent antagonisms of the workplace; at another level, its perception was governed by the confidence of the ruling classes in their dominance and of the
colonial state in the fragile alliances upon which it rested – in other words, in their ability to control and discipline the dangerous 'lower orders'.
For instance, in one account,26 evidence for the 'combustibility' and the violent disposition of Indian industrial workers has been drawn largely from the case of the
South India Railway strike of 1928. In this case, the response of the railway authorities as well as the government to the strike lays bare the ideological and political
processes by which the discourse of industrial violence and its perception as an attribute peculiar to the working class came to be constructed. Since the railways were
deemed a
25 See the illuminating discussion in R. Williams, Keywords: A Vocabulary of Culture and Society (London, 1976), pp. 278–9.
26 Arnold, 'Industrial Violence'.

Page 152
public utility industry, strikes which affected them were always likely to invite stern repressive action. But, in addition, this strike occurred at an extremely sensitive
moment marked by widespread labour unrest, the Bardoli Satyagraha, and the arrival of the Simon Commission which promised the renewal of widespread political
agitation. The Government of India, anxious to keep their lines of communication open, were concerned to bring the strike to a swift and satisfactory conclusion. This
strike was classified under the general rubric of 'unlawful and violent movements', and the district magistrates of Madras Presidency were directed 'to be on the
lookout for inflammatory speeches or other incitements to violence'.27 It should not surprise us that magistrates, policemen and functionaries instructed 'to be on the
lookout' for 'violence', should have reported the strike in these terms. In this way, categories were created, a vocabulary supplied and conditions generated for the
definition, reporting and characterization of industrial action, or the conduct of an ordinary trade dispute, in terms of violence. If the meaning of violence is ambiguous,
its attribution to particular social groups by scholars has, to some extent, arisen from an innocent and surface reading of the available evidence.
An implicit but fundamental assumption in the analysis of violence as a social phenomenon is that the level and incidence of violence in one society at a particular time
was higher than in another. Otherwise, it would scarcely warrant interest or explanation. But the basis for such comparison is beset with numerous methodological
problems. Historians of violence have often had to rely on extremely impressionistic evidence. Significantly, for instance, those who have argued that Indian industrial
workers were disposed to violence have not produced convincing evidence to support their claim. Such evidence is, of course, extremely elusive, not simply because,
as one historian recently complained, 'detailed and quantitative comparisons would be difficult to compile',28 but because once collected they would have limited value
for social and historical analysis. Whether particular incidents were labelled 'violent' would be conditioned by the sensitivity of the state to the problems of public order.
The manner in which they were reported to the police, to magistrates or to labour commissioners would affect the way they were recorded. The perception of the
observer – often in the case of industrial politics, that of policemen and employers – would establish the difference between a 'street brawl' and 'industrial violence',
between 'picketing' and 'criminal intimidation', even perhaps between an 'affray' and a 'riot'.29
27 Fortnightly Report for the second half of July 1928, Madras; Political, Simla to Acting Chief Secretary, GOB, F.8/XX/28, 1 September 1928, in GOB, Home (Special), File 543 (10) D
Part A of 1928, MSA.
28 Arnold, 'Industrial Violence'.
29 For an interesting discussion of the subjective bias which vitiates the quality of criminal

Page 153
For assessments of the nature, extent and frequency of violence, we are still to a large extent dependent upon the observations and perceptions of officials and
policemen, and sometimes, when they were its object, of propertied elites, employers and their agents. As Richard Cobb showed long ago, the reporting of violence
and social protest is often distorted by the specificities of police practice; sometimes, it helped to create the social phenomenon.30 By focusing exclusively upon
violence, abstracted from the specificities of its context, historians have become the prisoners of the assumptions and selfimage of the rulers and their agents.31
Since the meaning of the term embraced such a wide range of events, both violent and nonviolent, from cases of physical assault to prognoses about the possibility of
violence, it signified different things to different people and it was subject to varying interpretations by the same observers under different circumstances. This point is
usefully illustrated by juxtaposing two starkly conflicting descriptions, by comparably placed observers, of the same event: the 'riots' and strikes in Bombay which
followed the trial and conviction of the nationalist leader, Bal Gangadhar Tilak in 1908. According to police commissioner, H. G. Gell, Tilak's 'sympathisers and
friends' attempted 'to stir up the feelings of the people against Government' between 24 June when he was arrested and 22 July when he was convicted. Even before
the commencement of the trial on 13 July, 'one thing' was 'pretty clear' to Gell:
and that was that either at the trial or after it large bodies of millhands would attempt to make demonstrations at or near the High Court, and that, if allowed to assemble in any
great masses, they might become disorderly and cause a great deal of damage not only to property but also to life.32
statistics, see J. Ditton, Controlology: Beyond the New Criminology (London, 1979); and J. Davis, 'The Garrotting Panic of 1862: A Moral Panic and the Creation of a Criminal
Class in Victorian England', in V. A. C. Gatrell, B. Lenman and G. Parker (eds.), Crime and the Law: The Social History of Crime in Western Europe since 1500 (London, 1980), pp.
190–213.
30 Richard Cobb, The Police and the People: French Popular Protest, 1789–1820 (Oxford, 1970).
31 At times the price is the abdication of reason. Thus, one historian, having considered the problems of using police evidence, concluded, 'Though onesided and exaggerated, this
evidence is convincing due to its range, depth and detail'. S. Henningham, 'The Contribution of Limited Violence to the Bihar Civil Disobedience Movement', South Asia, new series, 2:1
and 2 (1979), 60.
32
Commissioner of Police, Bombay to Secretary, GOB, Judicial, 27 August 1908, in GOB, General, vol. CXIV of 1908, MSA. Reprinted in Sources for the History of the Freedom
Movement, vol. II, pp. 256–75. On the political context of the Tilak riots, see Cashman, The Myth of the Lokamanya; I. M. Reisner and N. M. Goldberg (eds.), Tilak and the Struggle for
Indian Freedom (New Delhi, 1966); G. Johnson, Provincial Politics and Indian Nationalism: Bombay and the Indian National Congress, 1880–1915 (Cambridge, 1973); and Shashi
Bhushan Upadhyaya, 'Cotton Mill Workers in Bombay, 1875 to 1918: Conditions of Work and Life', Economic and Political Weekly, Review of Political Economy (28 July 1990), PE87–
PE99.

Page 154
To the police commissioner, the potency of this threat, should it be realized, seemed devastating. He estimated that the eightyfive mills in the city employed some
100,000 hands, 'of which at least 50,000 must be ablebodied'.33 'Anyone able to enlist the sympathy of so large a number of men', he proceeded, 'must occupy a
powerful position and if intent on disorder, can practically set all authority at defiance'. It was true, he observed, that at present the millworkers 'had no organisation, no
leader, no common object and no weapons other than stones'. However, there was little reason to be sanguine about the future.
If a combined movement against Government can ever be effected, then we may expect that there will be organisation, a leader, a common object, and there will be weapons, such
as pickaxe, hatchets, crowbars, bludgeons, etc. . . . The object will be the destruction of Europeans, Government buildings, offices, the Railways, the tramways, the telegraph lines,
etc., looting of shops, European for choice, and possibly the burning of mills belonging to Europeans. The area over which they will operate will be the 23 square miles of Bombay
and the numbers engaged will be 50 or 60 thousand ablebodied millhands plus such of the population as are inimical to British Rule.
The police force, in the mind of its chief, would be 'incapable of resisting and, if possible, nipping in the bud, such an outbreak'. Furthermore, he did not believe 'that it
would be possible to strengthen the police in such a manner as to deal with a combined movement which lasted for any considerable time'. In the face of such an
onslaught, the imperial state would be powerless.34 It was rare, indeed, for the 'thin red line' of the British empire to be so clearly exposed or, indeed, the nature of its
physical presence in India, watchful, weary and for ever on its nerveends, to be so fully articulated.
S. M. Edwardes, civil servant and historian of Bombay city, writing at the same time as Gell and, indeed, soon to succeed him as the city's police commissioner, saw
the same situation in a fundamentally different light. What particularly struck Edwardes about the Tilak riots of 1908 was how
remarkable was the quaint juxtaposition during the height of the riots of seething disorder and the quiet prosecution of their daily avocations by the bulk of the people . . . At
Jacob's Circle there was a great display of military and magisterial strength . . . All the pomp and circumstance of Law and Order were represented there and there could scarcely
have been a greater display of armed force, more secret consultations, more wild dashes hither and thither, more troubled parley
33 It is not clear whether the police commissioner was here counting two hands per able body or providing unwitting comment on the conditions under which the millworkers lived
and laboured.
34 Commissioner of Police, Bombay to Secretary, GOB, Judicial, 27 August 1908, in GOB, General, vol. CXIV of 1908, MSA; Sources for the History of the Freedom Movement, vol. II,
pp. 256–75.

Page 155
ing, if the entire city north of Jacob's Circle had been in flames. And yet behind it and around it the daily life of the people moved forward in its accustomed channels.35
This startling contrast between the description of the same riot by the police commissioner and his immediate successor may serve to highlight the nature of British
paranoia in India but it exemplifies in particular the extent to which the subjective perception of violence determined its definition.
The ambiguities which marked the use of the term arose in part from the failure both among contemporaries and later historians to differentiate between particular cases
of violence. This failure to distinguish between different forms of action involving violence, to specify their protagonists or to delineate the actual context in which they
developed has not only resulted in the elision of qualitatively different kinds of events and social actions but it has also made it extremely difficult to probe their social
meaning or indeed to assess their significance.36 Incidents which entailed some violence in the industrial setting are lumped together to constitute 'industrial violence',
qualitatively different forms of action are conflated and each instance contributes cumulatively to the general picture of workingclass volatility. Yet the more closely
particular cases of industrial violence are examined, the less they appear to have in common with each other. Indeed, the social meaning of discrete acts of violence in
the industrial setting are often too diverse to permit analysis under the same label. The extent to which industrial violence has become a catchall term for that whole
area of industrial politics which is excluded by formal collective bargaining and organized trade union activity is indicated by the fact that even wildcat strikes – a
feature of industrial relations everywhere and not necessarily violent – have been considered in much of the work on Indian labour as integral to the special
phenomenon of industrial violence.37 These ambiguities inherent in the definition and analysis of violence illustrate the fact that in moments of collective action, violent
and nonviolent forms of behaviour are often intimately intermeshed and indeed extremely difficult to distinguish. As a result, forms of
35 S. M. Edwardes, The Byways of Bombay (Bombay, 1912), pp. 135–6.
36 Nor, indeed, have historians offered a satisfactory working definition of the phenomenon which they attempt to investigate. The result, of course, is proliferating confusion. Thus,
one historian defined violence during the civil disobedience movement as 'the use (or threatened use) of physical force against people and property: it incorporated sabotage,
intimidation, physical assault and also social boycott' (see Henningham, 'The Contribution of Limited Violence', p. 60). It is plain to see that many of these forms of action were in fact
nonviolent. It also discloses how difficult it is to establish when 'violence' really is violence.
37 Most explicitly, again, in Arnold, 'Industrial Violence'; see also Morris, Industrial Labour Force, pp. 178–97.

Page 156
collective action which were essentially nonviolent were often represented in public discourse in terms of their supposed violence. If this interchangeability arose from
the ambiguities integral to the discursive character of violence, it has only served to accentuate the problems inherent in its definition.
III
The ambiguity and exceptional elasticity of the term 'violence', which has made it elusive and confusing to scholars prone to invest it with a heuristic status, has
endowed it with a powerful, sometimes explosive significance in public discourse. This elasticity meant that it could be put to varied, diverse and even conflicting uses.
But it would be misleading to construe its use in a purely instrumentalist fashion. Labelling individuals or whole social groups as 'violent' might presuppose the
manipulative or, at least, a partial, reading of the evidence and it could serve to justify measures taken by the state for their closer control. However, this labelling,
indeed the partial reading of the facts, in themselves, reflected and expressed deepseated social anxieties and tensions as much as they were instrumental in the
strategies by which the state, its constitutive elements and its agencies, sought to establish their dominance and control.
To appreciate the role of violence in the language of politics, it is necessary to examine how these characterizations of workingclass violence came to be constructed.
To this end, this chapter will focus upon the particular context of Bombay's working classes and examine the implications of this construction of the threat of working
class violence for the struggles waged by them in the 1920s and 1930s. Necessarily, the nature of workingclass political action varied with the particularities of local
social, economic and political conditions. Nevertheless, Bombay's workers provided the paradigm by which Indian industrial workers came to be judged as typically
spontaneous, volatile and spasmodic in their political response.38
Workingclass violence was frequently portrayed by contemporary observers – policemen, civil servants, employers, journalists – in nonspecific, exaggerated and
generalized terms and frequently they attributed arbitrary causes and random consequences to the phenomenon. As social and political tensions mounted, statements
about the violence and the roughness of the lower orders grew wilder and more irrational. Disparate
38 Ironically, Bombay was also sometimes taken in the colonial discourse (as well as by later scholarship) to approximate to what was deemed, for India, a relatively advanced state
of industrialization, and, therefore, considered more likely to yield formal trade union organization.

Page 157
forms of threatening behaviour came to be inextricably linked. This discourse of violence, the assumptions upon which it was built, and the connections which it made,
reflected not so much real social behaviour as the threat which the ruling elites perceived to their control and dominance, at one level, and to their person and property,
at another. Indeed, how generously the ambit of workingclass politics could be drawn around its supposed violence is easily illustrated. 'The outstanding feature of the
year', the Government of Bombay reported, referring to 1921–22,
was the large increase in serious crime, especially in violent and other crime against the person and property. Among the reasons to which this increase may by attributed were the
general contempt for law and order fostered by the NonCooperation movement, the preoccupations of the police with political agitation and its resulting activities, a general
spread of unrest in the industrial world and the withdrawal of a large number of police from their normal duties in connection with the two Royal Visits.39
The official mind perceived a seamless web of connections inexorably linking industrial unrest with nationalist agitation, which combined to foster 'a general contempt
for law and order' and to create diversionary pressures upon the police, and which was likely to culminate in 'violent and other crime against the person and property'.
Such anxieties about the city's working classes were sharpened by the momentum of industrial action witnessed in Bombay in the 1920s. The general strike of 1928,
which lasted for six months and brought the millowners to the negotiating table on nearly equal terms, was followed by several months of lightning strikes, culminating in
another general strike in April 1929. Significantly, during these strikes the labour movement came to be dominated by the communist Girni Kamgar Union. Now
already deepening anxieties sometimes acquired the dimensions of a panic, especially among the millowners, merchants and professional middle classes as well as
among some officials. These anxieties increasingly focused upon the potential for violence among the working classes. The full depth and force of this panic cannot be
documented here. However, its most tangible results were to be seen in a rash of repressive legislation, in at least three public enquiries which functioned at times like a
public trial of the communist Girni Kamgar Union, and above all in the Meerut Conspiracy Case. The effects of this repression were to be felt in the labour movement
throughout the 1930s and the 1940s.
These repressive measures were motivated by the special terror which the threat of communism held for the official mind. But colonial officials
39 Bombay 1921–22: A Review of the Administration of the Presidency (Bombay, 1923), p. 20.

Page 158
could not always afford to give free expression to their darkest fears. These measures could not have been effected without a more general consensus built around
anxieties about the threat of workingclass violence. But these initiatives had more modest beginnings: in a widespread official anxiety about workingclass militancy in
general and the success of the communistled Girni Kamgar Union in particular. 'The respect for authority' noted a senior civil servant in January 1929,
which has hitherto kept the worker in this country in order, has been so grievously weakened during recent years, that they are only ready to accept the lurid hopes held out by
the communists that, if only they combine to overthrow capitalism and the Government, they must succeed and the result will be a millennium for themselves.40
When another general strike seemed imminent, a few months later, in April 1929, another official advocated the banning of workers' meetings on the grounds that
recent experience had established beyond doubt that 'Millhands' meetings inevitably led to the strike becoming general . . . and the communistic teachings result in
disturbances and murder.'41
But this deepseated trepidation about the 'general contempt for law and order' or the consequences of 'the grievous weakening' of a traditional 'respect for authority'
was not, of course, confined to officials alone. It was shared by large sections of the city's elites. In December 1928, Mr T. Watts, the superintendent of the
Currimbhoy group of mills complained to Dange, then the general secretary of the Girni Kamgar Union, that five workers in the roving and the weaving departments of
the Currimbhoy Mill had been leaving their machines in order to recruit members for the Girni Kamgar Union. They had been, he complained, 'neglecting their own
work and moving about in other departments. I beg to bring to your notice that this attitude of the men causes excitement which leads to lightning strikes, riots and
bloodshed, etc.'42 Barely two weeks later, J. H. Roebuck, an employee of the Kastoorchand Mills, wrote to Duff Cooper, the member of parliament for his native
constituency, Oldham, urging him to use his influence to persuade the colonial authorities to take tougher measures to control the red menace. 'No officer in any mill',
he wrote, 'is sure of coming away without serious injury to himself any day of the week, for the labour [sic] is entirely out of
40 Note by Secretary, GOB, General 23/1/29 in GOB Home (Special), File 543 (18) K of 1929, p. 21, MSA.
41 Note by H. F. Knight, Secretary, GOB, Home (Special) n.d. (probably late April 1929), in GOB, Home (Special), File 543 (10) E Part D of 1929, pp. 15–17, MSA.
42 T. Watts to S. A. Dange, 31 December 1928, Proceedings of the Court of Enquiry into a Trade Dispute between several Textile Mills and their Workmen, 1929, File 23, Exhibit 1/40,
MSA.

Page 159
hand and cannot be controlled . . . The attitude of the workers inside the mill is one of open defiance.' The workers prevented the mill officers from attending to the
maintenance or repair of the machinery, he complained, and when this resulted in a loss of production, they would nonetheless demand full pay,
otherwise they stop the mill and stopping the mill means – a howling mob who in many cases don't hesitate to pick up Iron bars, lathis and various other things with the intention
to smash up something or somebody, & the preaching of the leaders of these men has led them to believe that any capitalist or officer in the employ of the capitalist is a fit and
proper person to vent their indignation on.43
While the millowners were convinced in 1928–29 that they were fighting a lastditch battle to preserve order and discipline in the mills, and exerted considerable
pressure for the state to act forcefully against the labour movement, the official mind had for some time shown an urgency to take decisive measures against the spectre
of communism. In the light of the peculiar intensity of the struggles of 1928–29, the fears and anxieties of the state – its imperative to preserve the thin crust of order so
essential to its rule – now found a wider resonance. These perceptions of the potential of the working class for political violence began to dovetail with more widely
held notions of the roughness of the city's poor. These notions of roughness arose in part from the social distance which separated classes and castes, religious
communities and linguistic groups. But sections of the working classes also, and not only the city's magnates looking down from a great height, subscribed to similar
perceptions of their peers. Hindus and Muslims believed the dadas of the rival communities to be particularly sinister. Marathas feared the Mahars. The Ghatis from
Satara, the Bhayyas, as North Indian migrants were derisively known and Julaha Muslims were believed to be wild, rough, exceptionally strong, with a particular taste
for fighting.44 The workingclass claim to respectability, as Kanhoba, the old stalwart in Mama Varerkar's classic novel of the Bombay cotton mills, Dhavta Dhota,
typified, was grounded firmly upon the perception of the roughness of others.
Particularly crucial to public discourse, and increasingly influential in shaping policy, was the belief held especially firmly by those who did not inhabit them, that the
workingclass neighbourhoods were characterized
43 J. H. Roebuck to Duff Cooper, 18 January 1929, Private Office Papers, L/PO/1/23(i), pp. 84–7, OIOC. The mill officers in Bombay, Roebuck reminded Duff Cooper, were 'practically
all Lancashire and Yorkshire men' (presumably of those who were 'Europeans') and several were from Oldham. Duff Cooper, aware that he would, sooner or later, have to face his
own howling mobs at the hustings, took the matter sufficiently seriously to approach the Secretary of State and the Viceroy.
44 See also G. Pandey, The Construction of Communalism in Colonial North India (Delhi, 1992), ch. 3.

Page 160
by their roughness. Here, it was thought, the badmash and the mavali ruled the streets. These men were typically idle, unemployed and mischievous. They were
believed to hand around the street corners or frequent the akhadas and teashops which, according to one labour leader, 'was the best place for them to meet'.45 They
were said to be as easily recruited to break strikes as to man picket lines. They provided agitational fodder for trade unions as well as the Congress politicians. Their
baneful influence was found to be prominently manifested in the communal riots which sometimes consumed the city. Similarly, to their activities was frequently
attributed the disorder which sometimes erupted in the conduct of religious festivals. According to one newspaper, the toli bands of Mohurram imparted to this
religious observance the character of 'a saturnalia of not only the Muhammadans but of all the roughs of Bombay of all creeds'.46 As a result of their presence,
according to another observer, 'heads are apt to be broken'.47 'The tolis', thundered the police commissioner in 1908,
are irreligious rascality let loose for five days and nights to play intolerable mischief in the streets and terrorize the peaceful householder . . . the only unobjectionable feature of the
ten days of the celebration are the nightly Waaz or religious discourse by chosen preachers. But unfortunately these are little patronised by those to whom they would do most
good, namely the bad characters in the tolis.48
Above all, the most commonly identified characteristics of the roughs were that they were rural migrants and that they were unemployed. 'While I am referring to
hooligans', H. D. Nanavaty explained, 'I do not refer to the mill hands. I am referring to the people who have got no ostensible means of livelihood.'49 The labour
leader, R. S. Asavale, explained that 'bad characters turn out to be bad characters on account of unemployment. When people get no work they resort to a bad life to
maintain themselves.'50 One newspaper editorial argued that the cause of the social and political unrest in Bombay was the existence of a class of
men without any definite occupation who are always ready to join any quarrel in the chance of securing some loot. Going to jail has no terror for them as not a few of them are old
jailbirds. The existence in Bombay of this class is a constant menace to peace and prosperity. Government may well consider the means of reducing to a minimum the number of
such hooligans.51
The connection between unemployment and political action was
45 Proceedings of the BRIC, 1929, oral evidence, Syed Munawar, File 3, p. 279, MSA.
46 Bombay Gazette, 25 February 1907. Quoted in Masselos, 'Power in the Bombay Moholla', 82.
47 Gazetteer of Bombay City and Island (Bombay, 1909), vol. I, pp. 185–6.
48 Edwardes, The Bombay City Police, Appendix.
49 Proceedings of the BRIC, 1929, oral evidence, H. D. Nanavaty, File 6, p. 301, MSA.
50 Ibid., oral evidence, R. S. Asavale, File 3, p. 83.
51 Indian Daily Mail, 7 May 1929.

Page 161
simple enough, not because the condition was a cause for discontent, but because by making men idle, it left them free to get into mischief. According to the leading
merchant and millowner, Lalji Naranji, during periods of public 'disorder', 'hooligans take advantage of such troubles and . . . live upon such troubles'.52 Similarly, to
the labour leader, Syed Munawar, hooligans were mostly 'unemployed and unemployable' – except, presumably, for hooliganism.53 The fact that 'hooliganism' could
thus become a field for entrepreneurial activity and a source of employment only sharpened the concern of the state for the social and political organization which lay
beyond its ken. Behind every form of political activity it was increasingly to see the monstrous hand of the unemployed. Similarly, the view was also aired in official
circles that 'the majority of bad characters' were rural migrants and gave rise to the happy suggestion that 'on the threat or outbreak of disturbances all Badmashes
should be rounded up or deported out of Bombay'.54
In this discourse, the working classes were portrayed in terms of a dichotomy between the respectable and the rough, the conscientious and the criminal, stable, loyal
and settled workers as against the idle and dangerous who were always ready to support lightning strikes, irrespective of whether they had a particular grievance. A
distinctive semicriminal class of hooligans and roughs came increasingly to be linked with the mounting political tensions of the 1920s and 1930s. It was a short step to
heaping upon them the invariable responsibility for political violence. The Commissioner of Police, Bombay, used precisely this argument when he wrote in 1926,
when political, industrial or communal unrest culminates in disorder on a large scale in a big city, it is not generally speaking the ordinary resident with a home and an occupation
who keep the police and the military busy. It is the riffraff, the scum of the city that gives the trouble. The habitual criminals and the casual criminals who are kept in some sort of
control in normal times and can commit crime only by stealth, swarm to the surface and taking advantage of the excitement and panic, openly burn and loot and murder and cause
incalculable damage to the city. These are the people whom it is the duty of the police to get under complete control in normal times, and in proportion as the police are successful
in putting the criminal in jail and ridding the city by other means of undesirables so will their labours be lightened in times of disorder.55
The fact that the police were so easily overstretched was an argument in favour of expanding the force and spending a larger share of the govern
52 Proceedings of the BRIC, 1929, oral evidence, Lalji Naranji, File 6, p. 11, MSA.
54 Secretary, BRIC, GOB, Home to P. E. Percival, Chairman, BRIC, Karachi, 5 June 1929, Proceedings of the BRIC, 1929, File B, MSA.
55 Commissioner of Police, Bombay, to Secretary, GOB, Home, File 12305/7, 9 December 1926; ibid., File 6779/239, 27 June 1929 in Proceedings of the BRIC, 1929, File B, MSA.

Page 162
ment's revenues upon it. If 'the scum of the city', who were ordinarily 'kept in some sort of control in normal times' could be brought 'under complete control' by an
expanded police force better able to identify, jail or deport them, they would not be able to 'swarm to the surface' in times of disorder and 'cause incalculable damage
to the city'.
In practice, of course, these distinctions were impossible to maintain. About onethird of the city's mill workforce, the most organized sector of the labour market,
consisted of casual labour, hired daily at the mill gates. For those sections of the workforce considered 'permanent', conditions of employment were not much more
secure. Managerial practices relating to dismissals were notoriously arbitrary. Rules governing leave and absenteeism were not standardized, sometimes not even
stipulated. The casual and uncertain conditions of employment were integral to the shared experience of Bombay's working class.56
Similarly, the facility with which those who had pretensions to respectability referred to 'bad characters' belied the difficulties of separating them from the good ones.
Indeed, the police maintained a list of 'bad characters' just as they drew up a register of criminal tribes.57 This sometimes led people to believe that most 'bad
characters' were known to the police and to criticize them for not putting their knowledge to good effect. In practice, this supposed knowledge was remarkably thin on
the ground while improved lists of bad characters were by their nature unattainable. Indeed, it was by working from a position not of imperfect knowledge, but relative,
and often absolute ignorance, that the police succeeded in creating, or perhaps, more accurately, in reinforcing their own image of a class of 'bad characters'. How this
occurred was revealed when the police responded to the communal riots of February and May 1929 by what they described as 'rounding up' the bad characters.
Their methods were simple. Sometimes assisted by the army they would surround a particular wadi or neighbourhood, cover possible exits, enter the chawls and arrest
anybody they thought looked suspicious. Those who were arrested were then jailed in a block of unoccupied governmentbuilt chawls at Worli.
At the same time, the Government of Bombay was pressing for the introduction of legislation, modelled on the Goonda Act in Bengal, to enable the police
commissioner to intern or deport 'bad characters' for up to a year. As these hapless 'bad characters' were being rounded up, the Home Secretary suggested that the
police be 'instructed to get information as to the birthplaces of persons in Worli for it could strengthen the case [for a Goonda Act] if it could be shown that the
majority of bad hats
56 Chandavarkar The Origins of Industrial Capitalism, esp. pp. 72–122.
57 See ch. 6 below.

Page 163
58
in Bombay are immigrants'. The statistics which relate to the 'bad characters' reflect nothing more than the random manner of their arrest. As many migrants from
outside the Bombay Presidency as migrants from its constituent districts were found among those arrested. Overall, the ratio of Bombay men to outsiders reflected the
relative proportions in the city as a whole. Nevertheless, the Home Member saw in these figures 'strong evidence' to support the argument that bad characters had
emigrated to Bombay and therefore to justify the legislation he sought to 'extern' them. The general figures for arrest and conviction are similarly revealing. More
prisoners were discharged for want of evidence than were brought before the magistrates. By July, five months after the first riots, only about a third of the prisoners
against whom charges could be made had in fact been tried and almost as many were discharged as were convicted. There was only one instance of a man arrested
and released in February who was arrested again in May.59 No doubt these figures suggested to the police the salutary effect which the rounding up of bad characters
in February had upon them in May and confirmed their prejudice that magistrates were unwilling to convict in cases with political overtones. In an alternative, more
literal, reading, however, they neither lend credence to the belief that there existed in Bombay a distinctive class of habitual 'bad characters' who swarmed to the
surface in periods of disorder, nor justify the confidence of Bombay's respectable citizens that the police knew their roughs. Nevertheless, the rounding up of 'bad
characters' was a dramatic operation. As the testimony of several witnesses before the Bombay Riots Inquiry Committee in 1929 indicated, it conveyed or confirmed
the impression not only that a distinct class of bad characters existed but also that they were already known to the police.60 The notion that there existed a distinct class
of semicriminal 'bad characters' within the working class was sanctified by police measures. This loosely defined and easily generalized notion of the social threat
which it harboured was, in the context of increasing workingclass militancy, to prove increasingly convenient for the colonial state.
IV
The general strikes of 1928–29 established beyond doubt that Bombay's millworkers were capable of concerted and sustained action. It also
58 Note by Secretary, GOB, Home (Special) 7/5/29, in GOB Home (Special), File 543 (10) E Part G of 1929, pp. 12–13, MSA.
59 P. A. Kelly, Commissioner of Police, Bombay to E. W. Trotman, Secretary, GOB Home (Special), File 3968/A/122, 16 July 1929, in GOB Home (Special), File 543 (10) E Part G of 1929, p.
73, MSA.
60 Proceedings of the BRIC, 1929, Files 1–17, MSA.

Page 164
became clear that a coterie of communist leaders had gained a powerful following in the workingclass neighbourhoods. Since 1917, the British had been apprehensive
of the Soviet threat to imperial interests as a whole. By the mid1920s, the Government of India was becoming fearful that 'the immense power of mass action' had
been recognized by 'nationalist intellectuals of advanced views' and that even 'communism had begun to earn appreciative comment in quarters which could not be
dismissed as irresponsible'.61 The events of 1928–29 brought these general anxieties into sharper focus.
There were now several reasons for the government to act swiftly. The communists were recognized as a 'serious menace' in Bombay city. They dominated the city's
Congress organization. Without their efforts, it was readily acknowledged, the Simon Commission boycott in the city would have been a dismal failure. More crucially,
it was feared the recent success of the communists might result in the intensification of agrarian agitation and, as one official noted, 'the danger is really far more serious
among the agriculturists than among the industrial workers'.62 As yet, however, the influence of the communists outside Bombay was insignificant. So if the state was to
intervene at all, it was advisable for it to act immediately – certainly before the communists had established a stronger presence in the countryside – in order to detach
them from the momentum of nationalist politics. It was with an easily comprehended urgency, therefore, that the Home Secretary in Bombay stressed in January 1929
'that all questions regarding the industrial situation in Bombay are now subordinate to the need for stopping the spread of communism'.63
However, to declare war on the communists was not necessarily to acquire the means of waging it. To act against a significant focus of political opposition – and in
Bombay city, the communists commanded considerable popular support – the colonial state required some measure of public sympathy. The blessings of the
millowners for repressive measures against the Girni Kamgar Union were not difficult to obtain. Indeed, it was partly at their insistence that the Government of Bombay
had begun to consider seriously the possibility of a concerted campaign against the communists and to take soundings from New Delhi to assess its reactions. The
provincial government also derived some strength from the widespread prejudice – among the city's professional and commercial
61 Note: 'A Review of Recent Communist Activities in India', in GOI Home 7/7/37 Poll, p. 3, NAI.
62 Note, Secretary, GOB, General, 23/1/29, in GOB Home (Special), File 543 (18) K of 1929, p. 22, MSA.
63 Note, Secretary, GOB Home (Special), 23/1/29, in ibid., p. 17.

Page 165
classes and expressed in several newspaper columns – about the roughness of the workingclass neighbourhoods. But it was, nonetheless, imperative that the
government, in attempting to preserve the public order, should not appear to intervene on one side of what was, after all, a trade dispute. In addition, government
intervention to repress 'industrial unrest fomented by the communists' was bound to be ineffective unless it was accompanied by a strategy to contain their political
challenge. To complicate matters further, as colonial officials in Bombay readily recognized, the millworkers had 'genuine grievances' and unless these were resolved
their discontents were unlikely to evaporate.64 For, as the police commissioner pointed out, 'no legislative measures can prevent discontented workmen from following
extremist leaders'.65
It should not be assumed that the colonial state was monolithic or that it took a single or even always a steady view. Nothing was more likely to open up differences
within it, at various levels of government, than the attempt to undertake an elaborate and concerted campaign of repression. It was not enough that the Government of
Bombay had begun to perceive an opportunity to destroy the communists before their influence acquired more menacing proportions. It was essential for Bombay to
obtain the support and concurrence of the Government of India. This was at times to prove somewhat more difficult. If the provincial government had determined upon
the need for repressive action, this was by no means selfevident to the Government of India. Indeed, it required more vigorous justification in New Delhi, which was
distanced from the complexities of the problems of local control, sensitive to its possible repercussions in national politics and answerable to political criticism in
England. As early as June 1928, the Government of Bombay had begun seeking New Delhi's views on the prosecution of the communists.66 The Government of India
took away some of the provincial government's impetus by suggesting that the best means of dealing with the communists was to institute a conspiracy case against all
the major communist leaders in the country, and not just against the Bombay men. They were advised that a conspiracy case under the Indian Penal Code offered a
better chance of success than either prosecution for incitement to violence, which could not, after all, 'be taken seriously' when the general strike of 1928 had been
64 See, for instance, GOB, Home (Special), Departmental Note, 21 December 1928, in GOB, Home (Special), File 543 (18) G of 1928, MSA.
65 'Notes on a Conversation with Mr Kelly, Commissioner of Police, Bombay, on 4 June 1929', by H. G. Haig, 4 June 1929, in GOI, Home (Poll), 303/1929 & KW's I and II, p. 2, NAI.
66 GOB, Home (Special) File 543 (18) K of 1929, MSA; GOI, Home (Poll) 303/1929, NAI.

Page 166
67
'carried on peacefully' for five and a half months, or the acquisition of special legislative powers by the provincial government, for instance, to deport 'goondas' from
the Presidency and communists from British India and to prevent 'intimidation'. Accordingly, the Government of India deputed a special officer to collect, examine and
report on available material bearing on a conspiracy charge and instructed all provincial governments to 'look up evidence with a view to considering the possibility' of
prosecution.68 As late as January 1929, New Delhi, having taken the matter out of Bombay's hands, was unwilling to commit itself to such a prosecution. As the
communist threat in Bombay seemed more menacing and the provincial government's enquiries grew more urgent, the Government of India prevaricated and vacillated
and its replies became increasingly evasive.
For the colonial state there were several complicating factors to be taken into account. In principle, as one senior official noted in consideration of a bill to deport
communists, the Government of India did not want to be seen to be 'endeavouring to take power over those who may merely hold extreme opinions.'69 There was also
the fear of failure. Government officials were clear that 'no case should be instituted unless we are practically certain of a conviction'.70 The researches of the special
officer had soon yielded the alarming insight that the 'big conspiracy case', as officials now began to refer to it, 'would have to be of a very elaborate and
comprehensive character'. The realization had dawned that 'the amount of material to be collected and tested' before the decision to prosecute could be taken was
'very great'.71 The Government of Bombay had a different explanation for New Delhi's prevarication. At an early stage of the official conspiracy to prosecute the
communists, the Government of India had decided to institute the case either in the United Provinces or in the Punjab, but, at any rate, not in Bombay.72 The 'unofficial
reason' for the delay in instituting the case, the Home Secretary in Bombay informed the police commissioner, was that 'no other
67 Note by Secretary, GOB, Home (Special), 21 December 1928, in GOB, Home (Special), File 543 (18) G of 1928, p. 61, MSA.
68 Note by H. G. Haig, Secretary, GOI, Home, 8 November 1928, in GOI, Home (Poll), File F/18/VII/28, NAI; Secretary GOB, Home (Special), to Commissioner of Police, Bombay, 21
August 1928, in GOB, Home (Special), File 543 (18) A of 1928, MSA.
69 Note by H. G. Haig, Secretary, GOI, Home, 12 July 1928, in GOI, Home (Poll), File F/18/VII/28, NAI.
70 Note by H. G. Haig, Secretary, GOI, Home, 3 October 1928, in GOI, Home (Poll), File F/18/VII/28, and KeepWiths I–IX, NAI.
71 Note by H. G. Haig, Secretary, GOI, Home, 8 November 1928, in GOI, Home (Poll), File F/18/VII/28, NAI.
72 Secretary, GOB, Home (Special), to Commissioner of Police, Bombay, 21 August 1928, in GOB, Home (Special), File 543 (18) A of 1928, p. 5, MSA.

Page 167
provincial Government' – that is other than Bombay – 'is willing for the case to be run in its territory, and such a case if put before a Bombay High Court for trial by
jury would end in acquittal on account of the political leanings of the jury men'.73
In November 1928, New Delhi had decided to keep their special officer at his labours until the end of March 1929 and did not expect 'to be in a position to decide
whether the case should be launched until some time in May'.74 Meanwhile, the situation had, if anything, become more difficult and complicated for the millowners and
the provincial government's anxieties had scarcely been allayed. After the general strike had ended in October 1928, scores of strikes occurred in individual mills over
the following two months. In many cases, the millowners had been forced to accede to the workers' demands. In February 1929, a communal riot, which began with
the millworkers attacking Pathan moneylenders, raged through the city. Some officials in the provincial government grew increasingly impatient with their colleagues in
New Delhi. Cases of assault in the mills – two of which resulted in deaths – quickened the panic, increased their sense of urgency and afforced their desire for swift
and decisive action.
The Bombay Government now considered invoking regulations framed in 1827 which would enable them 'to place under personal restraint individuals against whom
there may not be sufficient grounds to institute any judicial proceedings'.75 It was anticipated, however, that 'the Government of India, having no real idea of the
conditions in Bombay will probably object on the grounds of injury to political or labour susceptibilities', despite the fact that, as the Home Secretary urged, 'the
situation is now too serious for these considerations to have weight'.76 To overcome these scruples about 'political or labour susceptibilities' and, at the same time, to
gain a wider basis of support, it was necessary to demonstrate that the consequences of the communist presence went beyond any simply political question. The role of
the communists in workingclass politics, it was recognized, could best be presented in terms of their incitement of 'overt crimes of violence'. Their objection to the
communists, the Government of Bombay accordingly claimed, was not 'political' but rather that 'their activities endanger the lives of the public'. Both the Government of
India and a wide section of public opinion could be expected to sympathize with arguments about the imminent disinteg
73 Note by Secretary, GOB, Home (Special), 21 December 1928, in GOB, Home (Special), File 543 (18) G of 1928, p. 69, MSA.
74 Note by H. G. Haig, Secretary, GOI, Home, 8 November 1928, in GOI, Home (Poll), File F/18/VII/28, NAI.
75 Home (Special), Note 8/1/29 in GOB Home (Special), File 543 (18) K of 1929, p. 5, MSA.
76 Note by Secretary, GOB, Home (Special), 23/1/29, in ibid., p. 18.

Page 168
ration of the social order and the threat of violence to life and property. Furthermore, the Government of Bombay recognized that, as soon as the anxiety that the city
was being overwhelmed by an atmosphere of 'riot and assassination' lessened, public opinion might even oppose the prosecution of the communist leaders, the
repression of political opponents of the Raj and the infringement of what were, in any case, rather fragile civil liberties. On the other hand, public discussions of
violence, the intimidation of 'loyal workers' by strikers and the communist menace in general provided an excellent basis for the Government of Bombay to claim with
force the need to extend by legislation their own executive and police powers. It was in this context that the belief that the Indian working class was disposed to
violence acquired new and explicit political meaning.
Nevertheless, the Government of Bombay remained consistently short of evidence to substantiate their claim that the millworkers were prone to violence, especially at
the behest of communist leaders. In January 1929, three Pathan watchmen of the New China Mills at Sewri were assaulted and killed. The police made over a
hundred arrests and then released about seventyfive millworkers. The remainder who were charged were members of the Girni Kamgar Union's mill committee. The
Government of Bombay concluded that 'the recent murderous outrages . . . are the clear result . . . of these communist speeches' and contemplated the prosecution of
the leaders for incitement to violence. But they foresaw two difficulties: first, 'that there was practically no violence during the general mill strike' of 1928 and second,
'that there is nothing to show that the present outrages are not disapprobated by the communist leaders'.77 Their legal advice was even more unequivocal. Not only was
there no evidence to connect the speeches of the leaders to the Pathan murders, the legal department doubted 'very much if it will be possible to establish such a
connection'.78 Indeed, the Home Secretary took sufficient note of this advice to inform the Bombay Millowners' Association – in flat contradiction to the spirit of his
own jottings on the file – that 'the fact that the general mill strike endured for 5 1/2 months with only one serious collision between the strikers and the police testifies to
the good sense and the good temper of the majority of the workmen' and declared that he had found 'no sign that speeches, however mischievous in intention, are
bearing fruit in violence'.79
When a kidnapping scare implicating Pathans and fights between
77 Note, Home (Special), 23/1/29, GOB Home (Special), File 543 (18) I of 1929, p. 4, MSA.
78 Note by the Remembrances of Legal Affairs, ibid., p. 11, MSA.
79 Secretary, GOB, Home (Special), to Secretary, Bombay Millowners' Association, 26 January 1929, in GOB, Home (Special), File 543 (18) G of 1928, p. 95, MSA.

Page 169
millworkers and Pathan blacklegs and moneylenders developed into a communal riot, in February 1929, the provincial government informed Delhi that 'it is commonly
believed that the kidnapping scare was deliberately organized by communists to intimidate Pathan workmen and promote disorder' but admitted that 'evidence
sufficient for a court of law' was not available.80 The Home Secretary added in a separate note that
the recent murder of Pathans has indicated to other labour – e.g. Pathan to join up with the Communist unions or be forced out of Bombay . . . Unless prompt steps are taken to
deal with Communist propaganda further labour disturbances in Bombay are inevitable in the near future and these may, as the present riots, develop into serious communal
trouble or may be antiGovernment.81
The Home Secretary's laborious memorandum is another seminal document of official paranoia, but it is by no means an exceptional statement of the case. In the
event, it was not forwarded to New Delhi, partly because the Home Member had been led to warn his secretary that the Government of Bombay would be well
advised 'to mention only briefly the theories and suspicions regarding the communist connection with the recent disturbances, which are when all is said and done,
rather debatable ground'.82 Nonetheless, provincial pressure on Delhi to act was mounting. Certainly, the special officer's researches speeded up and the Government
of India, having decided to proceed with their case, ordered the arrest of thirtysix communist leaders on 20 March 1929 to be tried at Meerut for conspiring to
overthrow the KingEmperor.
The Meerut arrests had dramatic consequences in Bombay. They deepened the sense of grievance, which had been building up among the millworkers since the end of
the general strike in October 1928, that the millowners had been nibbling away at piece rates, altering the conditions of work and victimizing members of the Girni
Kamgar Union, even as negotiations between the joint mill strike committee and the Millowners' Association proceeded before the Fawcett Committee. On 26 April
1929, more than a month after the arrests, the Girni Kamgar Union called upon the workers to strike. Two days earlier, the union had agreed that the workers should
enter the mills as usual and then stop work and leave together during the midday recess. The police commissioner was especially impressed that 'this programme was
carried out in a perfectly peaceful way and with remarkable unanimity, something like 80 per cent of the men after leaving the mills at midday declining to return'.83 If
the
80 Telegram, GOB Home to GOI Home, 6 February 1929 in GOB Home (Special), File 543 (18) K of 1929, p. 27, MSA.
81 Draft letter to GOI, Home, in ibid., p. 35.
82 Note by Home Member, 5 March 1929, ibid., p. 49.
83 'Notes on a Conversation with Mr Kelly, Commissioner of Police, Bombay, 4 June 1929', by H. G. Haig, 4 June 1929, in GOI Home Poll 303/1929, pp. 1–2, NAI.

Page 170
millowners and colonial officials alike were alarmed, they were probably no more surprised than the leaders of the Girni Kamgar Union by this show of solidarity. As
the colonial state grew more concerned to contain the threat of labour, the Millowners' Association determined 'to break the Girni Kamgar Union . . . and they are,
therefore, prepared to fight to a finish'.84 The discourse of violence, and the gathering panic about the consequences of the strike and general disorder, now
strengthened the millowners' position. They argued forcefully, and with steadily growing confidence that they would secure the sympathy of the colonial state, that the
workers' solidarity had been sustained by intimidation and that the government should adopt legislative measures to prevent its occurrence. Colonial officials, always
prepared to take the opportunity to widen the powers at their command, and themselves perturbed by the red menace in their midst, gradually joined in the
cacophonous discourse of violence.
Indeed, how consciously the language of violence could be deployed for political purposes now became increasingly evident. On 4 June 1929, the Commissioner of
Police was sceptical of the millowners' claim that the continuance of the general strike was the result of the intimidation practised by the Girni Kamgar Union.85 He was
not, he declared, 'disposed to think that intimidation plays the important part which the millowners attribute to it'. Indeed, picketing was of 'a perfectly peaceful
character' and was in any case 'rendered ineffective by the counterpicketing introduced by the millowners which has turned picketing rather into a matter of ridicule'.
By the end of May, it had seemed as if the strike was petering out. On 24 May, only seven mills were still closed and four days later, only two mills were not working,
and these for reasons unconnected with the strike. On 6 June, after the workers received their wages, however, the strike intensified: seventeen mills closed and over
30,000 workers stopped work. Within the next fortnight, the Government of Bombay completely reversed its earlier position. Portraying the industrial situation as if
intimidation was practised by the Girni Kamgar Union 'with gradually increasing frequency in the streets and in or near the chawls', the Government of Bombay sought
an ordinance from New Delhi to make 'intimidation' a cognizable offence, to be dealt with by direct executive action. The matter, they argued, was much too urgent for
them to draft and debate a piece of legislation. New Delhi, alarmed at Bombay's volteface, refused the ordinance and insisted that the provin
84
J. F. Gennings, Director of Information and Labour Intelligence, Labour Office, Bombay to Deputy Secretary, GOI, Industries and Labour, Simla, 18 May 1929, in GOI, Home Poll,
10/10/1930 & KW, p. 24, NAI.
85 'Notes on a Conversation with Mr Kelly, Commissioner of Police, Bombay, on 4 June 1929', by H. G. Haig, GOI, Home Poll, File 303/1929, NAI.

Page 171
86
cial government face its legislative council but could scarcely now advise against the measure itself.
To some extent, this radical shift in the provincial government's thinking reflected what the Home Secretary in Delhi had observed as an 'uncertainty of outlook' on the
part of its officers and their inability to take 'a definite and consistent line of policy in dealing with a situation which is admittedly most complicated'. Indeed, colonial
officials in Bombay had divided – and vacillated – between the view that the labour leaders, especially the communists, were to blame for the strikes and that the
millowners had been clumsy and cackhanded in dealing with their workers, thus provoking 'the maximum of opposition'.87 The Government of Bombay had found it
difficult to take a steady view. But now there were also some more substantive considerations to focus the attentions of the official mind. The long period of acute
social conflict which Bombay had witnessed since April 1928 had placed the police force under considerable strain. By the middle of 1929, the police commissioner
regarded 'the prospect of these conditions continuing indefinitely' rather nervously. On the one hand, he feared that 'a large number of [police] men have a good deal of
sympathy with the strikers'; on the other hand, the police were 'constantly being badgered by the millowners and by Bombay public opinion for not stopping conditions
over which they really have no power'.88 Cumulatively, it was feared, this served to undermine the morale of the police force. Moreover, the general strike of 1929
was a divisive conflict, unlike the industrial action of the previous year. In its initial stages, Bombay's Home Secretary had warned: 'So long as the strike is not general
and complete, violence may be expected.'89 This formulation described quite persuasively the character and timing of violence during strikes. Certainly, the divisions
among the workers manifested themselves in a flow of complaints about intimidation. Some of these were lodged with the police; many were reported to the mill
authorities who passed them on to the police. The nature of the complaints reveal that this 'intimidation' was not always violent. Frequently, they consisted of 'threat',
'abuse', 'obstruction' and even 'theft of clothes'. For the most part the alleged perpetrators as well as the complainants were described in collective nouns: for instance,
'mawalis' or 'union volunteers', in the case of the former, or 'the millhands of the Bombay Industrial Mill' in the case
86 GOI Home Poll 303/1929, NAI; Telegram, Home, Simla to Bombay, Poona, 2 July 1929 in GOB, Home (Special), File 543 (10) E Part G of 1929, pp. 105–7, MSA.
87 'Summary', by H. G. Haig, 11 June 1929 in GOI, Home (Poll), File 303/1929 & KW's I and II, pp. 5–6, NAI.
88 'Notes on a Conversation with Mr Kelly', by H. G. Haig, 4 June 1929 in GOI; Home (Poll) File 303/1929, p. 2, NAI.
89 Note by Secretary, GOB, Home (Special), 29 April 1929 in GOB, Home (Special), File 543 (10) E Part D of 1929, p. 28, MSA.

Page 172
90
of the latter. Nonetheless such evidence of 'intimidation', construed synonymously with violence, helped to increase the pressure among diverse groups for repressive
measures by the police.
By June 1929, the colonial state had discovered a growing confidence as it moved against the massed ranks of militant workers. The Meerut arrests had scarcely
raised a howl of protest that the limited freedoms of the Indian people were being trampled and no more than a murmur of complaint from the Congress leaders. On
the contrary, 'there is a very strong and growing feeling in the city' the police commissioner observed, 'that drastic and very early action is necessary to deal with the
communists who are responsible for the mill strike'.91 Without a certain confidence in the strength of this feeling, the colonial officials would have found it much harder
to act against the strikers. What informed this feeling was the generalization of the perception that the strikes, 'fomented' by the communist and other mischievous
labour leaders, had created an atmosphere of violence and disorder. Ironically, it was less the declaration of a second communistled general strike in April 1929 than
the communal riots which occurred first in February and then in May which acted to generalize and entrench this perception.
It is not intended to suggest that the millowners and civil servants orchestrated a discourse of violence simply in order to repress an increasingly militant labour
movement. Rather, the fact that images of the roughness and violence of the working classes were so widely accepted enabled the millowners to impose their particular
interpretation upon the strikes of 1928 and 1929 and, further, to prevail upon the colonial state to act quite so explicitly in their own interests. While subscribing to this
discourse, the millowners as well as the state were able to serve their own different purposes. What led a wide range of people in the city, and outside, to so deeply
fear impending disorder and violence and to acquiesce in the appropriation of large repressive powers by the state was the nature of the communal riots as a social
experience.
Bombay's residents, especially its elites, had prided themselves on the pluralist traditions of the city and the crosscommunal basis of its politics and they were startled
by the scale and ferocity of the riots.92 Both riots
90 'Statement of Cases of Intimidation in Connection with the Mill Strike from 26 April 1929 to 11 July 1929, reported to the Mill Authorities', in GOB, Home (Special), File 543 (10) E
(BB) of 1929, pp. 71–85, MSA.
91 Commissioner of Police, Bombay to Secretary, GOB, Home (Special), 11 June 1929, in GOB, Home (Special), File 543 (10) E(BB) of 1929, p. 3, MSA.
92 No communal riots on this scale had occurred since 1893 and it is probable that only a small minority, in this city of migrants, had experienced them.

Page 173
had been sparked off by conflicts which arose out of labour disputes. In February, Pathans had been hired by the employers to break the strike at the oil refinery at
Sewri. Yet the hiring of Pathans as blacklegs had occurred against the background of widespread antagonism towards, and fear of them in Bombay. During the strike
of 1928 and its aftermath, millworkers had borrowed extensively from Pathan moneylenders, who were alone in their willingness to lend to destitute strikers. For this
service, they charged what were seen as exorbitant rates of interest and they were believed to be exceptionally brutal and ruthless in the recovery of their loans. This
combination of fear and loathing, exclusion and exploitation came together in February to make the Pathans the target of workingclass violence. In May, conflicts
between some, predominantly, Hindu strikers and largely Muslim nonstrikers, and attempts by millowners to recruit Muslims to keep their looms and spindles running,
developed into a widespread communal riot. In both cases, the riots soon spread out of the mill districts and into the central parts of the island and as they did so the
participation of millworkers in them declined. There was some evidence to suggest that the labour leaders and especially the communists exhorted the workers not to
be drawn into the communal violence. But none of this shook the growing conviction that the riots were the inevitable outcome of a process set in motion by the
troublemaking communists. The Bombay Riots Inquiry Committee, which met in June 1929 and took evidence, in its investigation of the causes of the violence,
concluded that the underlying cause was 'the speeches of extremist labour agitators from May 1928'.93 Yet those who read the transcripts of the oral evidence
deposed before the committee are likely to find little to substantiate the claim – apart from the repetition of the claim itself – that communism or the manipulation and
intrigues of its political advocates caused the violence.94 In fact, rather, more powerful and complex factors were at work.
Remarkably, it was these communal riots, rather than prolonged strikes or the apparently sweeping successes of the communists, which persuaded many residents of
Bombay – and not merely its elites – that the colonial government should acquire fresh and sweeping repressive powers and deploy them forcefully against the city's
millworkers. The impact of the communal riots on the public perceptions of violence and the social order is perhaps best explained in terms of their nature, as a period
of social conflict. However potent the threat of workingclass violence had seemed during the general strike of 1928, it had in fact passed without violence. The
communal riots entailed real and murderous violence on an extensive scale and the violence they manifested was
93 Report of the Bombay Riots Inquiry Committee (Bombay, 1929), pp. 4–5, 7.
94 The proceedings of the BRIC are preserved in the MSA, Bombay.

Page 174
95
apparently without economy. In February as well as May, there were relatively few confrontations between hostile crowds. For the most part, the violence consisted
of assaults on passersby who had wandered into a strange neighbourhood. Everybody, it seemed, was at risk and nobody could anticipate who might turn upon
whom. At another level, whatever their aversion to the violence around them, everybody was implicated either as potential victims or as potential aggressors.
Millowners and merchants, landlords and shopkeepers, seeking to protect their premises could only do so by the same means as those who threatened them. The
respectable, concerned to secure their neighbourhoods, had no option but to adopt the methods of the rough. This state of fear and anticipation of violence could in its
own right generate further violence. During the riots, agencies which ordinarily maintained order appeared ineffective or were perceived to be partial and tainted by the
conflict. Above all, few could make sense of the riots. They appeared inexplicable and irrational. Since everyday social and commercial exchange did not disclose
clearly demarcated communities, but close connections across boundaries now so clearly drawn, it was difficult to make sense of the polarization of the city between
putative, if improbable, units of Hindus, on the one side, and Muslims, on the other.
Under these circumstances, anxieties about, 'the other', fortified by social prejudices harboured in ordinary times, now acquired an awesome and menacing shape. The
difficulty was that 'the other' now became everybody else. The rich feared the poor; Hindus and Muslims feared each other; women feared men; everybody feared 'the
hooligan'. Indeed, the responsibility for the violence was displaced onto 'the hooligan'. 'The worst offenders', it was reported, 'have been hooligans who haunt the bye
ways and who stab or hit and then run for safety into a maze of houses and gullies'.96 The hooligan became the universal embodiment of 'the other'. His methods
appeared to encapsulate the random and arbitrary nature of the assaults associated with the riots. His presence allowed those who aspired to respectability to absolve
themselves from responsibility for the violence and to dissociate themselves from its brutality. For those who lived through the riots, the hooligan provided an
explanation for the inexplicable, a rationalization of the seemingly irrational, a semblance of sense where nonsense abounded. If for most people this polarization by
religious community made no sense in the pattern of their daily social relations, blaming those they perceived as marginalized, poorly integrated, semicriminal desperate
groups allowed them to grasp the extraordinary events of the riots, and to reconcile the communalization
95 GOB Home (Special), File 348 Part II of 1929; GOB Home (Poll), File 344 of 1929; Report of the BRIC; Proceedings of the BRIC, MSA.
96 Times of India, 6 May 1929.

Page 175
which accompanied them with the syncretism of daily life. Finally, and quintessentially, 'the hooligan' personified the collapse of order. It was not simply that 'hooligans'
were believed to 'swarm to the surface' during the riots, but more significantly, that the violence could not be confined to the domain of 'the hooligan'. As the riots
spread, 'completely disorganizing business and traffic throughout the city',97 they threatened to sweep up everybody in their embrace. None were spared from the
threat of violence. Bombay's inhabitants tracked 'the steady progress of the city towards chaos' and were bewildered and frightened by 'the atmosphere of anarchy'
which prevailed.98
It was in this 'atmosphere of anarchy' that the need for the forceful intervention of the state seemed imperative in the most unlikely quarters. Thus, even the nationalist
Bombay Chronicle observed that 'it would be disastrous' if the city were to 'become a prey to hooliganism on the slightest of pretexts' and reminded its readers, and
the colonial state, that 'the most important function of a modern Government is to prevent the outbreak of lawlessness'.99 The roots of this state of anarchy could be
traced back to the two general strikes, the preaching of the communists and the spreading contempt for law and order which they were believed to have fostered. To
those who had once been sceptical about the colonial state's tilting at the windmill of the red menace or had cast a knowing eye over the millowners' agenda, the
discourse of violence – and the connections which it forged between disorder, hooliganism, workingclass roughness, strikes, communists and communal riots – began
to appear less improbable, sometimes even rather plausible. Public opinion in India – by no means confined to the propertied elites alone – was willing to endow the
colonial state with repressive powers, which it was at times reluctant to acquire itself. Yet these repressive powers would not only undermine the liberties of its subjects
but, once in place, could thereafter be used against opponents of the state, and, potentially, therefore, under changed circumstances, against those who now pressed
for the adoption of the most draconian measures.
This representation of workingclass politics in terms of its essential roughness and violence served to justify repressive action, whether in the streets, or in policies
formulated in the secretariat or in powers assumed by the state and codified in the legislatures. As this representation of the working class became increasingly
pervasive, various interests sought to deploy this discourse to their advantage. The police argued the case for greater resources. Millowners wanted the state to help
them to discipline their recalcitrant workforce and destroy the influence of the communists.
97 Ibid.
98 Ibid.
99 Bombay Chronicle, 6 May 1929.

Page 176
Labour leaders sought to displace their more radical rivals. Indeed, the Gandhian conquest of violence signified at one level the attempt by Congressmen to control, not
merely to lead, their followers. Colonial officials could, within the terms of this discourse, try to justify the appropriation of greater powers to wield against threats to
the social order and, indeed, their own future. Throughout the 1930s, measures broadly taken against labour organizations were justified in terms of the potential for
workingclass violence. This vocabulary was not specific to the vernacular of the British in India; it was also deployed by the provincial ministries of the Congress in the
late 1930s as well by the governments of independent India after 1947. It reflected social anxieties no less real for being so loosely formulated. Indeed, it was precisely
this looseness of definition that imbued the discourse of violence with considerable political value for the colonial state and for its various shifting groups of
collaborators. The power and resilience of violence in public discourse derived from its lack of definition and especially its elasticity of meaning. It was this elasticity of
meaning which facilitated the creation of a vocabulary which carried a resonance for diverse social groups. The perception of the threat of violence – particularly by the
working classes, or groups among them – could for this reason be widely shared, even among those groups who were vulnerable to being characterized in its terms.
Characterizations of workingclass violence, now so enthusiastically taken over by historians, were integral to the dubiously moral economy of the ruling classes in
colonial India.
VI
The notion that the working classes, or particular groups among them, were inherently predisposed to violence arose from the prejudices of the dominant classes about
the roughness of the lower orders. But it was also integral to a wider discourse. Frequently, characterizations of the violence and roughness of the working classes,
often devised by employers, policemen and civil servants, prepared the ground for pushing forward policies which particular interests desired. Beneath their cognitive
surface lay a crucial, if unwarranted, shift from the observation of particular acts of 'disorder' to the definition of violence as a quality innate to the working class. It is
misleading, therefore, to displace these characterizations onto the 'realities' of workingclass life and workingclass politics.
This discourse of violence tended to generalize rather than to specify its meaning. Thus, the consideration of industrial violence came to be inextricably linked with,
sometimes it was merged into, criminal, revolutionary or even communal violence or, more broadly, a pervasive con

Page 177
tempt for authority. Frequently, the causes of 'industrial violence' came to be identified with, indeed embodied in, the very same groups – marginal elements, casual
labourers, rural migrants – who were perceived to have a particular predisposition to criminal and violent behaviour. The identification of the casual poor or the rural
migrant with criminality and violence has reflected police practice rather more accurately than it has described their behaviour. Because they were perceived as
dangerous and criminal, they – or groups within them – were policed more closely and became more vulnerable to arrest. Once they were drawn into the criminal
justice system, their presence afforced the perception of their charactertype as criminal. In this way, the perception that they were prone to crime and violence
preceded the behaviour of any individuals defined as belonging to this stratum and was then affirmed by selective policing. To complete the circle of perception and
practice, this affirmation of their roughness and criminality then served to legitimize policies directed specifically to control them.
The identification of these social groups with violence has developed from the assumption widely, sometimes implicitly, made in the social sciences that endemic
violence characterizes societies in the process of transition to modern, industrial capitalism. By this ahistorical, normative, evolutionary scheme of social change,
apparently applicable to all societies at all times, it followed that workers in the early stages of industrialization were characteristically volatile; that communalism could
appear as a survival from a previous epoch characterized by a sacral conception of the polity and the public domain; and that the violence and criminality of the urban
poor signified the archaic remnant of prebourgeois attitudes among the working classes. The notion that violence is endemic in societies in the process of transition to
modern industrialism has led to its identification in the Indian case as the characteristic of social groups affected, but also marginalized, by that process of transition:
early industrial workers, inadequately assimilated to the factory; the urban poor; rural migrants in the industrial setting. Yet in India – and, of course, not only in India –
the 'casual poor' and rural migrants were not marginal to the economy or the social order, but, rather, were integral to the processes by which these were constituted
and reproduced.
If violence is taken to be integral to the culture of the poor and the charactertypes of whole social groups, its practice is thus abstracted from the particular social and
political circumstances in which it occurs. The claim that their propensity to violence is inherent to the culture and social character of particular groups suggests that it
exists independently of the wider context, even if that context is invoked to explain it. On the other hand, if violence is to be explained as a function of social
transformation,

Page 178
then its investigation would direct our attention to the political and economic context of change rather than the symptoms to which, it is suggested, it gave rise.
Violence, both actual and potential, has been in all societies a powerful instrument for the negotiation of power relations, whether for the maintenance of control or the
organization and conduct of protest. In India, it acquired a special dimension. The final sanction for the maintenance and perpetuation of imperial rule was necessarily
physical force. But the colonial state could not afford to use it freely. It lacked the financial resources or even the military might to rule so vast and turbulent a land by
force. To attempt to do so was always likely to necessitate the sacrifice of other more crucial imperial priorities.
One solution to this problem was to ignore it or, at any rate, to throw a hastily constructed facade of order across deeper levels of social conflict. The perpetuation of
colonial rule depended to a large extent upon the effectiveness of local structures of control. But there was nothing stagnant about the social relationships underlying
these structures. At these levels, the dadas, the street gangs, the retainers who might throw their holds or wield their lathis in the interests of the local boss could as
easily turn against him. This was why when the existing relationships of power in local society were undermined or existing structures of dominance effectively
challenged, they threatened to take away with them much of the facade of order behind which the Raj nestled. Social control has always been more easily theorized
than enforced.
Unlike the postcolonial state, which has proved more willing than its predecessor to turn its cannons on its opponents, colonial rulers did not possess sufficiently
powerful social and cultural ties by which to negotiate social tensions or maintain order. They could neither easily legitimize the use of force nor justify their own policies
of repression. Yet the weakness of their cultural hegemony and their lack of social influence meant that they were repeatedly thrown back upon the use of force,
repressive laws or even extralegal coercion. The further the colonial state intervened in areas of social conflict in order to maintain intact this facade of order, whether
against nonviolent satyagrahis or more fearsome strikers in the industrial setting, the more fundamentally were its weaknesses exposed. It is hardly surprising,
therefore, that the colonial state remained so sensitive to the threat and the possibilities of violence.
At the same time, in the last decades of British rule, the colonial state was brought increasingly under Indian control. Indeed, Indian elites were now more easily able to
deploy the power of the state to maintain their dominance. As the colonial state became more and more Indian, so it lost more of its scruples about the political
susceptibilities of its subjects and

Page 179
showed a growing willingness to repress strikes with greater ferocity. The political conjuncture of 1928–29 in Bombay illustrates how Indian elites gained from colonial
rule the political power to control and exploit labour more intensively. Colonialism in India might, in this light, be defined, in one of its characteristic aspects, as the
process by which labour was cheapened and more fully subordinated to capital, both indigenous and imperial.

Page 180
6—
Police and Public Order in Bombay, 1880–1947
British rule in India was characteristically autocratic and repressive. It could at times be brutal and violent. The rhetoric of civil liberties and individual freedoms
decorated its claims to legitimacy but, in practice, these considerations occupied a lowly place in its order of priorities. In a history characterized by annexation and
conquest and by the maintenance of dominion by force, it is not surprising that the organization and activities of the police should be taken to reflect the authoritarian
nature of colonial rule. For some, it has appeared to be 'expressive of the very nature of colonial rule in India'.1
By focusing upon the administration and organization of the police, historians have often tended to foster the supposition that it constituted a monolithic force.2
However, this organizational perspective has made it difficult to resolve apparent contradictions in the nature of colonial policing and sometimes served to proliferate
confusion. Thus, in one account, the police appeared to constitute the most effective and powerful instrument of colonial repression, but they were also found to be
'often inadequate to meet the major crises of rural control'. On the one hand, 'the coercive strength and disposition of the colonial police' was said to be nourished by
colonial and racist ideology; on the other hand, we are told that 'India's colonial regime fell short of being a police state' or a 'society ruled through fear' because of
'Britain's own political culture' whose innate liberalism served as a check upon unbridled despotism. Colonial and racist attitudes, presumably not drawn from 'Britain's
own political culture', underlay the violence of the state, yet, largely, 'the police alignment was with the propertied classes and not merely the ruling race'. Thus, the
police have been portrayed as the main force of a colonial state
1 Arnold, Police Power and Colonial Rule, p. 235.
2 Percival Griffiths, To Guard My People: The History of the Indian Police (London, 1971); A. Gupta, The Police in British India, 1861–1947 (New Delhi, 1979); Arnold, Police Power
and Colonial Rule; Robb, 'The Ordering of Rural India'.

Page 181
which was both hegemonic and vulnerable; characterized as coercive yet found to be ineffectual; motivated by racism yet restrained by an inherent metropolitan
liberalism; allied closely with Indian propertied elites and yet the bludgeon of the ruling race.3
This chapter will suggest that generalizations about the nature of policing should rest most securely upon the investigation of its daily operations rather than its
administrative design or its organizational form. Accordingly, it will focus upon what might be described as 'everyday policing'. Its aim is to place police methods of
operation in relation to the social organization of the neighbourhoods in Bombay city in the late nineteenth and early twentieth centuries. The daily operations of the
police were determined by the financial and political constraints within which they developed. At this quotidian level, the police were integral to the processes by which
power relations were negotiated in the street and the neighbourhood, and thus constituted an important element in the formation and consolidation of local power. Their
actions influenced, as, indeed, they came to be moulded by, patterns of local dominance. Furthermore, at lesser levels, albeit less systematically, the working classes,
too, could draw upon, appropriate and deploy their personal and social caste and kinship connections with the police. As the police became embedded within the
social and political networks of the neighbourhood, they operated less as simply an instrument of social control, but proved more responsive to influences which were
relatively autonomous of their own internal structure of command. In this light, and viewed through the prism of their daily operations, the police, far from being
monolithic, appear to have been responsive to varied and often conflicting sets of social pressures.
It is not intended to suggest that we should replace the notion that the police were indiscriminately repressive with the notion that they were uniformly benign. In
Bombay, as elsewhere, the police lacked the political and financial resources to control and discipline the working classes as a whole. The conventional and pragmatic
response of the police has been to proceed selectively. Some elements among the poor and the working classes, singled out for particular attention, must have
experienced the police as a particularly brutal and violent force. But if the police were open to recruitment by diverse groups within the urban neighbourhoods, they
could not fulfil their disciplinary, even coercive, function systematically. On the contrary, their selective interventions were probably often experienced as somewhat
arbitrary, but they would have seemed no less oppressive and no more limited for being less systematic. If the police
3 Arnold, Police Power and Colonial Rule;, pp. 147, 230, 233, 235.

Page 182
lacked the political and financial resources to set out to control and discipline the working classes as a whole, they also found it difficult to proceed more selectively by
identifying particular social groups as the targets of their actions. For the task of establishing a general and uniformly applicable consensus about the particular social
groups to be discriminatingly identified as the proper objects of policing often proved beyond the capacity of the colonial state in India. From the perspective of
everyday policing, it would seem implausible to portray the police as simply and unproblematically the enforcers of social control designed by the colonial state and its
collaborators or to suppose that they were able consistently to give practical effect to the intentions of the official mind. Indeed, frequently, their methods of operation
undermined and contradicted the most carefully devised blueprints of the colonial state.
Two further considerations embedded the police more deeply within the social organization of the neighbourhood. First, in India the police force had military origins. Its
primary concern was to police those who had been subjugated, to facilitate the collection of revenue and to ensure the free movement of goods. After the 1860s, when
a uniform police system was put in place, the police understanding of its own role, and its perception of crime and the social order, was coloured by its military
antecedents. The police were liable, therefore, to understand crime primarily in terms of rebellion and disorder, public and political security, rather than simply in terms
of lawbreaking and the security of property and the person. Of course, no police force could set itself the task of rooting out crime wherever it was to be found, with
any realistic expectation of accomplishing it. On the contrary, police activity, and the operation of the criminal justice system as a whole, often served to define and
indeed, therefore, to create crime. The more conscientiously the police set out to abolish crime, the more likely it was to both become aware of it, generate fresh
categories of offences and criminalize old patterns of behaviour. The operations of the police imparted to the notion of public security an explicitly, and narrowly,
political meaning. For this reason, it neither sought to intervene energetically nor to disturb too greatly the practices by which offences against person and property
were handled within informal social networks or local power structures. Of course, the police everywhere have relied to varying degrees upon the collaboration of
local residents. Its consequence in India was that propertied elites, merchants and industrialists and local magnates, created their own private arrangements for
protection and policing. For this reason, too, the scope and the daily practices of the police, and their apparent ambiguities and contradictions, need to be situated
within the context of the social relations of the neighbourhood.

Page 183
Second, although the colonial state increasingly assumed that the maintenance of public order was ordinarily the duty of the police, troops were frequently summoned
to deal with 'local disturbances'. Significantly, public order policing in India, even after the 1860s, was largely conducted by the army. When the police were called
upon to deal with 'local disturbances', their strategies were largely shaped by their own relationship to the social organization of the neighbourhood. It was at this
quotidian level that the parameters within which the police approached larger and more dramatic problems of public order were defined. Senior police officers
recognized the value of establishing close connections with the informal social networks of the neighbourhood, both to detect crime and to protect the public order. But
such connections only served to heighten their anxieties about the quality of the discipline of their rank and file. Frequently, faced with 'a local disturbance', the police
felt they were unable to cope and that they were in danger of being overrun by the 'mob'. They required the army to stand by and strengthen their nerve and sometimes
to intervene and restore order. The armed force which the colonial state could command was formidable and when it perceived a serious threat to its security, it
unleashed it with a ruthless, sometimes murderous, brutality. Its sanction and its power were essential to the maintenance and perpetuation of British rule. Yet the
British sometimes appeared unable to deploy it to repress their opponents. The British in India did not, they could not, constitute an army of occupation. Had they
attempted to so style themselves, the armed force at their disposal would have seemed woefully inadequate.4 In any case, they did not rule India for its own sake; their
purposes were global and imperial and extended beyond the limits of political dominion. For the historian, therefore, the central problem in the policing of colonial India
is how to match the repressive power of the state to the fragility of its control. It was primarily because the police established these connections within the
neighbourhood, remained open, however unwittingly, to its influences, and sought to maintain public order through these alliances, that they were able to postpone for a
time having to bear the ultimate cost of their most brutal repressive efforts. Thus, the supposedly 'coercive disposition' of the police force and its repressive practices
were inextricably connected to the nature of its daily operations in the street and the neighbourhood. Accordingly, this chapter will first set out the constraints within
which the police operated in Bombay; it will proceed to ask how these constraints
4 Thus, in 1938, there was one British soldier to every 88 square miles of the territory and for every 20,000 inhabitants of the subcontinent. 'The British Element in Internal Security
Troops', Appendix E to Annex 2, CID 198–D, CAB 6/6, PRO, cited by D. Omissi, The Sepoy and the Raj: The Indian Army, 1860–1940 (London, 1994), p. 212.

Page 184
influenced police practice; and, finally, it will examine how this dimension of everyday policing shaped their strategies for the maintenance of public order.
II
If the colonial police were, as they have sometimes been described, the most effective instrument of state repression, it is striking that India was far more sparsely
policed than England. Both in personnel and resources, the thin blue line was very thin indeed. The Bombay Presidency had one of the most centralized administrations
in India and, in per capita terms, its police force was perhaps the strongest in the subcontinent. Although the Presidency, excluding the 'native states', was in land area
more than twice the size of England and Wales, the strength of its police force engaged on what were called 'station duties' in the early twentieth century was slightly
over a third of that deployed in England and Wales. While in England and Wales there was a policeman for every 772 people in the first decade of the twentieth
century, the Bombay Presidency, at the time, raised a ratio of one policeman for 1,360 people. The level of expenditure on the police force presents a similar contrast.
The total cost of the police in England and Wales was estimated at £ 99 9s per man per annum; in Bombay, it amounted to slightly more than £ 12.5 These statistics for
the Presidency as a whole included Sind, the laboratory of Sir Charles Napier's early experiments and the model for police organization in India, where the density of
policemen was the highest in the subcontinent.6 As a result, they also conceal the sparseness of policemen in the districts of the Konkan, South Gujarat and Karnataka,
where their ratio to the population varied between 1,100 and 1,300, without allowing for those who were not engaged on 'station duties'.7
Over large parts of the Indian countryside there was no police presence
5 E. C. Cox, Police and Crime in India (London, 1910), pp. 125–27. These figures according to Cox date from 'three or four years ago': ibid., p. 125.
6 According to Hunter, there was a policeman for every 336 people in Karachi and for every 367 people in Thar and Parkar in Sind in 1890. W. W. Hunter, Bombay, 1885 to 1890: A
Study in Indian Administration (London, 1892), p. 426. By 1930, there was a policeman for every 820 people in the Bombay Presidency while there was one for every 64 people in Sind.
Calculated from J. C. Curry, The Indian Police (London, 1932), Appendix, p. 347.
7 Hunter, Bombay, 1885 to 1890, p. 426. Policemen not engaged on station duties included 'a large number of men employed on the guarding of a multitude of Government treasuries,
the escort of Government treasure, the personal escort of officers, the guarding of jails, and of railway trains, stations, and goodssheds, together with other quasimilitary and
miscellaneous duties none of which are required of the police in England'. Cox, Police and Crime in India, p. 126. These men were excluded from the calculations which Cox made of
the ratio of policemen per head of population in India and England, but Hunter's calculations do not appear to have allowed for them.

Page 185
at all. Villages were omitted from the administrative design of the police force. The basic unit of the District Police was the subdivisional outpost which lightly
supervised the works of the village watch, ostensibly conducted by the hereditary servants of the village community, an institution always more active in the sociological
imagination of the official mind than it was in rural society. When it was effective, the village watch was to a large extent the strong arm of its headmen or, more
broadly, its dominant families. According to Edmund Cox, the village watch in the Bombay Presidency was 'for the most part an elegant fiction'. It usually consisted of
'a rabble of over a dozen nondescripts', who rarely 'use the powers that have been conferred upon them' and 'so far from assisting the more disciplined guardians of
the peace, they are actually at the bottom of much concerted crime'.8
The Bombay City Police was organized separately from the provincial force, with its own commissioner on the model of the Metropolitan Police in London. It might be
supposed that a large and growing commercial and industrial centre like Bombay city, with a significant European presence, would be more closely policed. Yet its
police force was half the strength of Calcutta and considerably weaker than London. In 1885, it was reported, 'the strength of the Police available to meet all demands
is considerably less in number than in 1865', its responsibilities were increasing and their range widening. The result was 'a cruel overworking of the Police'. Policemen
were expected to be on duty for 14 to 17 hours 'on alternate nights and for 2 consecutive months'. In addition, following their spell of duty, they were expected to
attend the courts or the commissioner's office to discharge particular cases. The lowest grade of constable received in pay about twothirds of the average millworker's
wage and perhaps as much as the unskilled and lowest paid occupations in the cotton textile industry. Since it was 'impossible to obtain efficient men in any way suited
to make policemen', the commissioner could not afford to be too particular about whom he recruited. The height for entry was lowered to 5 feet 4 inches and the
minimum chest measurements to increasingly modest levels.9 The situation had not changed appreciably by the early twentieth century. 'I have had at times', confessed
one police officer in 1910, 'to enlist men who had not come up to the standard'.10 The Bombay City Police were not a prepossessing force. However much the
commissioner may have yearned for the fine strapping specimens of the martial races, his men were characteristically diminutive and not infrequently consumptive.
What they lacked in muscle they did not make
8 Cox, Police and Crime in India, pp. 254, 258 and 256.
9 Annual Report on the Police in the Town and Island of Bombay for 1884 (Bombay, 1885), pp. 13–15.
10 Cox, Police and Crime in India, pp. 144–5.

Page 186
11
up with arms. In the 1880s, the police force had forty swords between them while the remainder were given batons. In the following decades, a larger component of
the force was armed, usually with muskets, which were as likely to damage the marksman as his mark. It was only in the 1920s, after the Amritsar massacre and
growing financial constraints forced some reconsideration of the role of the army in quelling civil disorder, that an armed police reserve was developed and it was only
during the second world war that it was significantly expanded.12 The organizational weaknesses of the police should not lead us to the conclusion that the police were
complacent about the task they faced in Bombay. Indeed, as one senior policeman in Bombay put it in the 1930s, the city and the Presidency capitals
act as magnets to attract the most hardened criminals of all the criminal classes of Asia. They have incredible underworlds, sordid, inflammable and incurable . . . The mawalis of
Bombay are men who live by trading in every kind of human vice: chicanery, fraud, drugs, women, murder, the whole gamut of evil.13
These disadvantages were compounded by another: the city's residents showed little enthusiasm for serving in the force. Senior police officers expressed a preference
for recruiting those who were 'natives of the district in which they are required to serve', for 'knowledge of the language, customs and geography of the district' was
'very essential' for the effective discharge of their duties. However, as one officer noted in 1910, 'local recruitment is not . . . always possible'.14 In 1913, the
Government of Bombay complained, 'Great difficulty is reported to have been experienced in finding sufficient men to bring the force up to its sanctioned strength.'15 In
other words, the city police could not attract enough recruits to spend the money they were allocated. 'Were it not for the steady flow of Marathas from Ratnagiri', it
was reported in 1913, 'there would be a constant and serious shortage of constables'. Special recruiting parties sent into the Deccan 'met with little or no success'. The
'Pardeshi' (literally, foreign) Hindus from North India showed a decreasing interest in employment in the police force largely due 'to the prospect of higher wages
without the trouble of discipline' elsewhere. The decrease
11 Annual Report on the Police in the Town and Island of Bombay for 1884, pp. 13–15.
12 Bombay Confidential Proceedings, Home Department, January, 1921, vol. 62, OIOC; on the development of the armed police in Madras, see D. Arnold, 'The Armed Police and
Colonial Rule in South India, 1914–1947', MAS, 11:1 (1977), 101–25; on the expansion of the armed police during the second world war, see Gupta, The Police in British India, p. 547,
see also pp. 492–563.
13 Curry, The Indian Police, p. 49.
14 Cox, Police and Crime in India, p. 143.
15 Judicial Dept. Circular, 14 July 1914 in Annual Report on the Police in the City of Bombay, 1913.

Page 187
in Pardeshi applications was not in itself entirely reason for regret. For, in the view of the authorities, 'the Pardeshi makes a much less satisfactory policeman than the
recruit from the Bombay Presidency, for apart from his other failings he is in the first instance less educated and he exhibits a tendency to become somewhat
invertebrate and melancholic under the depressing effects of the Bombay monsoon'.16 In fact, all over India, 'Pardeshis' entered the police force in profusion. 'They
generally possess considerable physical strength', Edmund Cox declared,
they are proverbially honest and faithful, but their brainpower is very limited . . . they are wellfitted [sic] for guards and escorts. For the work of the unarmed police such men are
entirely unsuitable. They usually know nothing of the language of the countries other than their own. They have no friends or interests among the people and are useless for the
purposes of investigation.17
Conversely, Marathas from Ratnagiri often had too many friends and interests among the people to police them reliably and consistently according to the commands of
their senior officers. Distaste for employment in the police force remained significant in the 1920s. The Deccan Marathas, it was observed, were still 'not so enamoured
of service in the police', while now even Hindu migrants from North India sought out alternative employment when the opportunity arose. Only Muslims from the
Bombay Presidency as well as from North India, along with Marathas from Ratnagiri, were more readily recruited into the police force.
In addition, there was a substantial turnover of policemen every year roughly at the rate of about 10 per cent. This prompted the suspicion that 'agriculturalists from the
Konkan and elsewhere are merely making use of the Police Department for tiding over a few months of the year when they might otherwise be idle'. However, further
investigation satisfied the police that this was not the case because 'enlistments and resignations are distributed fairly equally over different months of the year'.18 In fact,
a large proportion of the city's working class retained their rural connections but sought more or less permanent work in Bombay. However, the maintenance of their
village connections demanded considerable flexibility to move between town and country, sometimes at short notice, often for varying periods of time, while retaining a
lien on their jobs in Bombay.19 Certainly, the strategies to preserve their stake in the village often proved incompatible with the maintenance of police discipline. The
fact that recruitment and resignation were distributed evenly throughout the
16 Annual Report on the Police, 1913, p. 16.
17 Cox, Police and Crime in India, pp. 143–4.
19 See Chandavarkar, The Origins of Industrial Capitalism, ch. 4.

Page 188
year need not necessarily rule out the possibility that the rural interests of policemen propelled the rate of turnover among constables.
The low wages of the lowest ranks of the force did not serve to attract recruits. Further disincentives were 'to be found in the unpleasant conditions in which recruits
have to live owing to shortage of housing accommodation and in the large number of vacancies which results in duties being often imposed on recruits in the first
months of their service from which they should be exempt'.20 Periodically, in the 1920s and to a lesser extent in the 1930s, the police found themselves unable to
operate even at the inadequate strength sanctioned by provincial government budgets.21 The building of 'police lines' and barracks, and the housing of constables in
militarystyle barracks has often been interpreted as a reflection of a colonial style of policing but it was primarily a pragmatic response to the difficulties of recruiting a
police force. The provision of housing in Bombay was a necessary inducement to attract recruits and essential for the survival of poorly paid constables. As long as the
police force was undermanned, and women were not generally recruited to serve in this period, it was difficult to weed out recruits who 'did not show good promise
during their training' or more generally 'to enforce a higher standard of discipline' within the police force as a whole.22
Nor did the police become progressively more effective. The value of stolen property reported to the police amounted to about 4 annas per capita in the 1880s and
rose to under 6 annas in 1908 or roughly the average daily wage of millworkers. By inflating their valuations, the police were able to raise this figure to 12 annas in
1910, but were unable to sustain this figure during the decade and it was not until the early 1920s that their estimates amounted to 2 rupees.23 It would be reasonable
to assume that a significant proportion of this property was probably stolen from the vast majority who paid no taxes and exercised no control over municipal
expenditure. Since lawbreaking cost the city's elites so little, they saw no reason why law enforcement should cost them more.
Significantly, between 1865 and 1907, the Government of Bombay tried to hive off the police charges for the city onto the Municipal Corporation. The distribution of
this expenditure between the local and provincial levels of the state was 'a perennial source of friction between Government and Corporation'.24 The successive
financial crises which
21 P. J. Thomas, The Growth of Federal Finance in India (London, 1939), p. 518.
23 Annual Reports on the Police, passim.
24 R. P. Masani, The Evolution of Local SelfGovernment in Bombay (Oxford, 1929), p. 329.

Page 189
the Government of India faced in the 1860s had encouraged it to devolve responsibility for matters such as the police and sanitation, which were marginal to Britain's
global imperial interests, to municipalities and district councils constituted by the wider representation and election of Indian interests. This was known as local self
government, which in view of its supposed educative value enabled the colonial state to do good by doing well. Local elites could thus carry the political costs and
garner the profits of taxing and spending around the parish pump. In 1865, the whole cost of the city's police was thrown upon the Municipal Corporation and although
the provincial government was persuaded to contribute to these charges later in the decade, it simply and suddenly stopped payment in 1873. Although the principle of
the provincial government's contribution was restored through the arbitration of the Government of India, the burden which each body would bear remained a matter of
dispute until 1907, with the Government of Bombay reducing its investment at every opportunity. The parsimony of the provincial government persisted into the 1920s
and 1930s. Although the total expenditure of the Government of Bombay increased gradually in the 1920s, the budget allocated to the police declined while both fell
together in the 1930s. Expenditure on the police declined from about 13 per cent of the provincial budget in 1921–22 to 11 per cent in 1929–30 and although it
increased in the wake of civil disobedience, it still accounted for 11 per cent in 1939–40.25
Various attempts were made from the late nineteenth century onwards to reorganize the police force and increase its efficiency. But their effects were by no means
dramatic. Among the many shortcomings of the city's police force, the reformers identified 'the absence of any proper record of complaints and investigations' and that
'the station is constantly left for several hours' without officers 'capable of recording a complaint properly and commencing enquiries'.26 In fact, it was said, 'the
GovernorinCouncil has reason to believe that there has been a tendency in the past to discourage the lodging of complaints of petty thefts'.27 In the late nineteenth
century, these conditions had invariably led the Government of Bombay to the happy conclusion that the city was populated by peaceable and lawabiding people. In
1875, the Government of Bombay compared crime statistics for Bombay and Calcutta and without allowing for the more intensive policing of the latter or for
25 Calculated from Thomas, The Growth of Federal Finance, p. 518.
27 Circular, Under Secretary, Judicial, GOB to Commissioner of Police, Bombay etc., 2 September 1910, in ibid., p. 1.

Page 190
28
underreporting, applauded the moral character and peaceloving nature of its inhabitants.
Although they continued to regard the volume of crime as 'a natural consequence of the general weakness of the administration',29 the Judicial Department of the
Government of Bombay anticipated by 1910 that improved methods of reporting or, at least, an end to the practice of actively discouraging people from lodging
complaints, would lead to an increase in crime rates.30 There was, indeed, an immediate increase in recorded crime. This increase, explained the police commissioner,'
was the result of 'various changes which have taken place in the police force' serving to increase facilities for the reporting and detection of crime'. However, 'the rapid
increase in genuine reported crime which had been a marked feature in the previous three years in Bombay City', it was noted in 1914, 'suffered a check' for causes
'the nature of which is obscure'.31 Indeed, whether station officers had continued the old habit of discouraging people from lodging complaints or simply manifested 'the
desire to burke cases in order to show good statistical results'32 was not given public consideration even if it caused private anxiety within the Judicial Department or
the police commissioner's office. Two decades later, the tendency to prevent cases being 'burked' by policemen who were 'unwilling to register crime which [they] had
little or no hope of detecting' had by no means abated.33 Yet victims were no more willing to report crimes than the police appeared ready to register them. One police
officer bravely offered 'very rough estimates based upon experience' of unreported crime: 2–3 per cent of murders and between 10 and 40 per cent of reported
burglaries and thefts were 'concealed'.34 But the 'dark figure' of crime is by its very nature unknowable. Indeed, if the real volume of crime is determined largely by the
way in which the machinery of criminal
28 Statement of the State of Crime in Bombay and Calcutta, 17 June 1875, GOB, Judicial, 1875, vol. CDLXXXVII, p. 117, MSA. In 1884, there was a policeman for every 227 residents
in Calcutta and every 507 residents in Bombay. Annual Report on the Police, 1884, p. 13. In 1930, Calcutta had a policeman for every 188 inhabitants, while Bombay had one for
every 265. Calculated from Curry, The Indian Police, Appendix, p. 347. A more recent attempt to calculate crime rates in Bengal showed that 'major crimes', whether violent or
property offences, declined in the last quarter of the nineteenth century and that their frequency per capita was lower than other provinces. This finding led its author to conclude,
in the spirit of the Government of Bombay in 1875, that 'the population of Bengal was less crimeprone and that the criminal administration in Bengal was more effective in terms of
crime prevention'. A Mukherjee, 'Crime and Criminals in Nineteenth Century Bengal (1861–1904)', IESHR, 21:2 (1984), pp. 166–7, 160–2.
29 Edwardes, Bombay City Police, p. 4.
30 Under Secretary, Judicial, GOB to Commissioner of Police, 2 September 1910, in Annual Report on the Police, 1909, p. 1.
31 Annual Report on the Police, 1913, Judicial Dept. Circular, 14 July 1914, p. 1.
33 Curry, The Indian Police, p. 184.
34 Ibid., pp. 184–5.

Page 191
justice bears down upon social exchange, the 'dark figure' is merely an illusion. The low rates of crime and the modest value of property stolen may suggest that the
policeman's experience cannot serve as a reliable guide to the extent of underreporting. But they also bear testimony to the reluctance of people consistently to seek
redress formally through the police.
III
The shortcomings of the police force, its financial and numerical weakness, its lack of discipline and organization combined with the general reluctance to seek redress
through the formal structure of the police and the courts to ensure that the overwhelming majority resorted most commonly to informal sanctions for the settlement of
disputes. This fact was in turn to shape the methods and strategies of policing, whether in relation to the detection of crime or the maintenance of public order. The
nature of these informal sanctions varied enormously according to the nature of the conflict and its protagonists. The locations of power within the workingclass
neighbourhoods were diffuse, fluid and subject to intense rivalry and conflict. Landlords, moneylenders and grain dealers, jobbers and labour contractors, caste
panchayats and chawl and wadi (residents') committees, each provided a forum around which disputes were conducted, sometimes settled, and social relationships
were negotiated, challenged and defined. Dadas, literally elder brothers, were men who had acquired a particular reputation for toughness, sometimes precisely by
asserting their own public role through the enforcement of justice and the protection of their friends, neighbours and clients.35
Jobbers and dadas, akhadas, the melas and tolis, the 'volunteer' corps and neighbourhood gangs were often lumped together inside a notional culture of roughness. But
informal sanctions were by no means either the preserve of neighbourhood roughs nor the prerogative of neighbourhood magnates. Contemporaries, and later
historians and anthropologists, often perceived the social organization of the urban neighbourhoods in the language of rural description. Dadas or jobbers were
portrayed as village headmen. The social organization of the neighbourhood appeared to be reconstituted forms of caste and kinship ties. But these are highly
misleading counters of description; none of them more so than the insistence upon a clear demarcation between the rough and the respectable. To emphasize the
importance of informal sanctions is not to imply that urban society was cohesive, harmonious or consensual. Nor should the
35 See ch. 4 above.

Page 192
operation of informal sanctions be interpreted as the rule of the patrons and magnates of the urban neighbourhoods or its social organization and institutions as the
instruments of social control and hegemony. In fact, magnates were constrained by the demands of reciprocity; their power was contested and limited by their rivals;
honouring their clients often narrowed the patron's freedom of manoeuvre.36
Colonial officials often tried to invest those they perceived as neighbourhood leaders with power and influence which they never possessed. Sometimes, this was
because they picked their leaders badly. More usually, it represented the forlorn attempt by policemen, magistrates and civil servants to place their faith in any existing
structure of power which might maintain order. If they did not enquire too closely how it was maintained, they might not have to care whether it was preserved.
Nonetheless, the official mind which explained political action in terms of incitement or mobilization by leaders and publicists also put its faith in more responsible
patrons and magnates for the restoration of order. During the Mohurram riots of 1908 between rival sects of Muslims, one observer complained, 'there are no regular
leaders with whom those in authority can communicate, nor any recognized heads who possess a real influence to whom the processionists owe allegiance'.37 Similarly,
during the general strikes of 1919 and the early 1920s, the absence, sometimes the evanescence, of trade unions led officials and other observers to wonder how
largescale action by workers could be coordinated and orchestrated without the existence of 'some controlling organization'38 or some 'good sound leadership among
them somewhere'.39 Leadership within the workingclass neighbourhoods appeared heavily concealed and one of the functions of the CID was increasingly to ferret it
out.
Yet even when apparently accredited leaders were found, their influence seemed to be in doubt. When disputes over cow slaughter during Bakrild appeared to
heighten 'communal' tensions between Hindus and Muslims in 1944, the police commissioner approached the leaders of Madanpura but discovered in time that 'these
socalled leaders have very little standing and less authority over the Muslims who live in this area'. When he called a meeting of 'respectable, influential Muslim
leaders', he found that 'They all with one voice admitted that they had no control and no influence with the Muslims who reside in Nagpada and Madanpura.'40 The
assumption underlying these expectations was that Indian society
36
Ibid.
37 Times of India, 15 February 1908. Masselos, 'Power in the Bombay Moholla 1904–15'.
38 Secretary, GOB, Home to Secretary, GOI, Home, 7/15 February 1919, Bombay Confidential Proceedings, vol. 46, 1919, OIOC.
39 Bombay Chronicle, 21 February 1924.
40 Note by Commissioner of Police, 3 November 1944, GOB, Home (Special), File 1002 (1), 1893–1945, p. 257, MSA.

Page 193
was mediated by culturally specific social forms and institutions which regulated its functions, resolved its disputes and restored its harmony. The task of colonial rulers
was then to identify these locations of power and influence and operate through them.41 In this way, public order was founded upon shifting sands. It was maintained in
spite of, and not because of, the strategies so elaborately and feverishly devised by the official mind for the purpose.
Moreover, the respectable operated through agencies which they readily assigned to a culture of roughness. The magnates who joined in the condemnation of
Mohurram as a 'saturnalia of not only the Mahomedans but of all the roughs of Bombay of all creeds'42 frequently financed the celebrations of the moholla. Similarly
grain dealers, shopkeepers and landlords played a prominent role in the organization of the Ganpati or Gokulashtami melas, themselves often little more than street
gangs embellished in a new guise. Landlords hired rent collectors; creditors sometimes recovered dues by force or the threat of force; lawyers employed touts; mill
managers recruited jobbers and street corner bosses to combat pickets or destroy workers' combinations. When millowners were particularly determined to reopen
their mills during strikes, for instance in times of buoyant demand when orders were flowing in and prices rising, or when they could not afford to have their assets
locked up in protracted disputes, they often turned to akhadas or gymnasiums in order to recruit, escort and protect blacklegs.43
E. D. Sassoon and Company were the managing agents for the largest group of mills in Bombay. They had a reputation for managing the best, the most efficient,
progressive and innovative mills and, indeed, for being considerate and farsighted employers of labour. Yet, as we have seen,44 they appointed as the head of their
'watch and ward department' a Eurasian exboxing champion called Milton Kubes and as the communists gained ascendancy among the millworkers in the late 1920s,
the most prominent anticommunist boss of Bombay, Keshav data Borkar, under the portentous title of 'Superintendent of Labour'. Kubes liked to boast that he
commanded his own squad of spies and enforcers, who attended workers' meetings, took down speeches, collected handbills and strikers' pamphlets and, when it
proved possible, simply disrupted the proceedings. The speeches collected by Kubes were made available to the
41 If there is a colonial dimension to the policing of contemporary Britain, as Brogden has suggested, it may perhaps be located in the hypothesization of 'community leaders' and
the ceaseless quest for 'spokesmen' of the ethnic minorities. Their language and approach may suggest how far the official mind in Britain has imagined the colonization of Brixton,
Brick Lane, Handsworth and Toxteth. See M. Brogden, 'The Emergence of the Police: the Colonial Dimension', British Journal of Criminology (1987), 4–14.
42 Bombay Gazette, 25 February 1907, cited by Masselos, 'Power in the Bombay Moholla', p. 82.
43 See ch. 4 above.
44 See ch. 5 above.

Page 194
police and used as evidence in the prosecutions of trade union activists for incitement and notably of the communist leaders in the Meerut Conspiracy Case. Kubes
attended meetings, 'to get an idea of the atmosphere that is brought about by the speeches' and although, as he confessed, 'I know Marathi but little, I can get a sense
of what they say'.45 His ignorance of the language did not apparently impede his collection of evidence. Similar problems arose in relation to the evidence presented
against the communists by the police in the Meerut Conspiracy Case. The problem in this case arose because police reporters who knew Marathi often did not have a
command of shorthand while those English or Eurasian reporters who learnt shorthand did not know the language.46 This conundrum was finally resolved by the police
and indeed the court by trusting God's Englishmen who having spent sufficient time in India were able to divine more accurately from the intonation and inflections of
speech what was, in fact, being said.
Similarly, retailers and shopkeepers in the cotton, piecegoods and bullion markets often hired Pathan watchmen to guard their shops. Frequently, it was said, a head
Pathan accompanied by ten or fifteen others would go around the shops offering to watch them for two rupees or more a piece. Each gang could keep watch for at
least seventy or seventyfive shops. Bombay experienced a number of moral panics about 'the Pathan menace'. They were classified as 'a criminal tribe' in the early
twentieth century.47 In the city, they were believed to be rapacious moneylenders, charging exorbitant interest rates, recovering their dues through violence, taking
away the wives of their debtors in cases of default or kidnapping children of both sexes for prostitution.48 In 1923, the police assigned one officer and two constables
'for keeping watch over undesirable Pathans in the city' at the cost of over 6,000 rupees.49 But they complained that their actions were neutralized by the kind of
businessman who would write to the press to complain of the 'Pathan menace' while at the same time, employing a Pathan 'to look after his shops merely because he is
afraid that if he does not, the Pathan will loot the shop'.50 But Pathans were not alone in prospering from running such neighbourhood protection rackets. Gurkhas and
'Pardeshi' Hindus also
45 Proceedings of the BRIC, oral evidence, File 10, p. 225, MSA.
46 Proceedings of the MCC, Statement by S. A. Dange, made in the court of R. L. Yorke, Esq., I. C. S., Additional Sessions Judge, Meerut, October 26, 1931, in the Case of King
Emperor versus P. Spratt and Others, pp. 2475, 2484–8.
47 M. Kennedy, Notes on the Criminal Classes in the Bombay Presidency (Bombay, 1909).
48 Proceedings of the Bombay Provincial Banking Enquiry Committee, Replies to the Questionnaire, Currimbhoy Ebrahim Workmen's Institute, File 12–C, MSA; Times of India, 9
February 1929; Proceedings of the BRIC, oral evidence, Lalji Naranji, File 4, p. 121; P. G. Solanki, File 6, p. 127 and sundry others, MSA.
50 Ibid., p. 33.

Page 195
51
attempted to develop their own beat and the competition between them was one of the factors underlying the Hindu–Muslim riots of February 1929.
Similarly, informal tribunals were formed around caste and neighbourhood to resolve disputes and maintain order. They were sometimes constituted out of village
connections. These tribunals frequently sat in judgement over the behaviour of local residents. Dadas and jobbers, strikers and blacklegs, moneylenders and rent
collectors could be called upon to explain their behaviour or justify their actions and, at this level too, sanctions were deployed against those who failed to satisfy the
neighbourhood tribunal, the caste panchayat or the chawl committee.52 These institutions should not be perceived as the spontaneous expression of a popular culture,
but the outcome of alliances and rivalries in the neighbourhood, and integral to the processes by which the distribution of local power was determined. There was
nothing static about these local structures of power. It was within and around them that most of the functions of policing, the maintenance of order and the mediation of
conflicts were conducted. Not only low expectations of the police, but also the vitality of neighbourhood organization explains, to a large extent, why few people
lodged complaints with the police.
Significantly, the police often adopted means and methods which were indistinguishable from these agencies within the urban neighbourhood. In fact, they frequently
depended upon alliances which they could make in the neighbourhoods in order to exert their influence within them. Frequently, too, the police were recruited by
prominent figures in the neighbourhood to serve their parochial interests or to swing the balance of local power in their favour. Policemen thus constituted yet another
resource in the politics of the neighbourhood. The structural weaknesses of police organization confined their activities to particular targets often chosen in consequence
of more widespread anxieties about particular social groups. Similarly, their methods were dictated by contingencies and often by the nature of their alliances in the
neighbourhood.
Paranoia was the hallmark of the colonial imagination. Consequently, great emphasis was laid on disguise and detection, subterfuge and surveillance to distil the popular
mood, uncover the plots and rumours of the bazaar for which colonial officials had a particular weakness, and to discern the concealed sources of power, influence
and community leadership. Hartley Kennedy was a police commissioner with a penchant for disguise and a particular hatred for bagatelle players, beggars and foreign
pimps. He posted detectives at the docks to meet foreign steamships and,
51 Proceedings of the BRIC, 1929, Sir M. M. Ramji, File 2, p. 337; Sir P. Klein, Commissioner of Police, Bombay, File 18, p. 167, MSA.
52 See ch. 4 above.

Page 196
53
according to one of his successors, to mark down 'every Jewish trafficker who showed his nose in Bombay'. Kennedy's preference was to dress himself in purdah
and 'thus altered', he was said 'to have wandered about the city after nightfall in company with one of his agents'. Unfortunately, an ankle injury had left him with a
permanent limp, so that this Englishman in drag, hobbling around the city at night, was readily recognized as 'the lame Kennedy saheb' and was long afterwards
remembered 'affectionately', it is said, by 'the old lawbreakers and disreputables who recollect his efforts to bring them to book'.54
In various ways, the police were drawn into the social and political networks of the neighbourhood and, of course, the greater the emphasis on detection, the more
dependent they became upon them. While operating within these networks, the police often found themselves imprisoned within their confines. The official historian of
the Bombay city police was thus able to name a mere handful of detectives upon whom the success and failure of law enforcement depended. In the late nineteenth
century, it was said, Mir Abdul Ali, 'wielded a degree of control over the badmashes of the City wholly disproportionate to his position as the superintendent of the
safed kapdewale or the plainclothes police'.55 Similarly, in the 1920s and 1930s, Rao Bahadur Sabbaji Rao, Superintendent of the Criminal Investigation Department,
was 'one of a few welleducated men' who had entered the force in the 1890s, had 'risen from the rank of Head Constable' and arrived at his post after more than
three decades of experience with the police. The Rao Bahadur acquired a similarly legendary reputation. His particular gift was 'an almost uncanny sixth sense
concerning receivers who were likely to be approached with property of different types'. Thus, following a major jewellery heist in the Fort in April 1928, the Rao
Bahadur moved into action and 'warned the most likely receivers' that they might be approached with the stolen goods. Indeed, 'such was his reputation' that one of
the receivers, visited by the robbers, 'detained them on some clever pretext and informed the formidable, if not quite clairvoyant, Sabbaji Rao.56 Others appeared to
have the almost magical quality of solving crimes 'to which at first there seemed to be no clue whatever'. Their particular, perhaps enabling gift, was 'to keep closely in
touch by methods of their own with the more disreputable and dangerous section of the urban population'.57
Collaboration of this kind brought fame and increasingly sonorous titles to the policemen; but its fruits must also have been tasted by their agents and partners in the
neighbourhood. Not only were the latter often allowed a free hand but the policeman concerned could be required to
54 Ibid., p. 118.
55 Ibid., pp. 72–3.
56 Curry, The Indian Police, pp. 169–70.
57 Edwardes, Bombay City Police, pp. 72–3.

Page 197
bring official authority and resources to bear decisively on the side of their allies in neighbourhood disputes. In some cases, it was said, the police commissioner used
'private agents of his own' to keep a check on the police and to accumulate supposedly 'an uncomfortably accurate knowledge of what was going on in various
quarters of the city'.58 Yet these same methods required on a more systematic scale, could leave the police in darkness, despite the fact, or perhaps because, they
were so deeply embedded in the neighbourhood. After the strike which followed Tilak's trial and transportation in 1908, C. J. StevensonMoore, the Director of
Criminal Intelligence concluded that 'the City Detective Department knows no more about the inner history of the business than my boot'. In fact, he continued,
I am compelled to say that the ignorance of the Bombay police as to the agency and methods used for engineering the strike is nothing short of appalling . . . Many of the
constables live in the chauks from which some of the worst natives come. Yet no information. They are all Ratnagiri men together and of course in league; 75 per cent of the
Bombay Police are Ratnagiri men.59
The police were often deeply implicated in the social and political arrangements of the neighbourhood. In the early twentieth century, when the 'Pathan menace' had
first begun to grip the minds of the police and the public, the police commissioner identified four 'communities' among them, chose leaders for each and ceremonially
invested them with turbans, swords and shawls as symbols of their authority as the head of their respective jamats. In 1929, when the Pathan riots led to a search for
the redoubtable leaders of the community, only one was to be found: Khan Sahib Samad Khan, the sardar and patel of the Kohat Pathans, and especially those who
were employed in the docks. The Khan Sahib told the Bombay Riot Inquiry Committee that the other three headmen had died, two of them nearly twenty years
previously. They had not been replaced because the jamats 'never took the trouble to inform the commissioner of police of the death of their headmen'.60 Perhaps the
jamats had been dissolved and some Pathans had regrouped along different lines. Certainly, some Pathans had formed their own jamats and elected their own patels
without reference to the government. Nonetheless, Samad Khan lamented the decline of the headman appointed by government, confessed that his authority was no
longer what once it had been
58 Ibid., pp. 113–14.
59 C. J. StevensonMoore, Offg. Director, Criminal Intelligence, to Sir Harold Stuart, Offg. Secretary, GOI, Home, 5 August 1908 in GOI, Home Poll (A), December 1908, nos. 149–69, NAI.
60 Proceedings of the BRIC, 1929, oral evidence, Khan Sahib Samad Khan, File 2, p. 201, MSA.

Page 198
61
and took the opportunity to press for its renewal through another ceremony, presumably with more turbans and shawls. The police commissioner regarded these
importunities rather coolly. This had not prevented him in the intervening period from seeking out putative leaders of the 'Pathan community' in order to gain their
collaboration in allaying fears of their 'menace'. In 1924, the police commissioner had invited them 'to keep a strict check on new arrivals and to give me information of
the presence of bad characters in the city with a view to their deportation. They agreed to do so: it remains to be seen if any good will come out of it.'62 The kidnapping
scare of 1929 and the 'Pathan hunt' by millworkers which followed does not suggest that the panic was contained. Yet in the late 1910s and early 1920s, the arbitrator
in disputes between Pathans, whether between moneylenders, straying into each other's patch, or conflicts between dock workers and their sirdars, and their adviser in
their relations with other groups and especially the police was an inspector in the CID and a fellow Pathan called Ubedullakhan.63
The police also intervened informally in the conduct of industrial disputes and were often drawn into communal conflicts. Sometimes, the police intervened at the
highest levels to mediate labour disputes and effect settlements. Favours gained in one sphere could be encashed in another. About the Julaha weavers, 'an extremely
illiterate and fanatical population', the police commissioner claimed in 1911, 'I have had something to do with them, in the matter of getting them reemployed after a
strike and obtaining their backwages from their employers.' During Mohurram in 1911, he decided to realize 'the gratitude which they professed for this help' by
sending for Badlu 'who lives in Madanpura and controls a tabut supported by the Julahi weavers of that locality' and persuading him to keep out of the way of rival
Muslim moholla processions.64 Similarly, during the 1919 general strike, the police commissioner, Vincent, repeatedly tried to discover spokesmen for the workers and
engage them in negotiations with the employers, but failed to find leaders whose terms the workers were willing to accept.65 'Fatty' Vincent, as he was 'usually known
in the service', had been born and brought up in Bombay, where his father had been a police commissioner before him, and had, according to one of his subordinates,
'an exceptionally good knowledge of Marathi'.
61 Ibid., pp. 187–205.
63 Times of India, 16 July 1917.
64 Commissioner of Police, Bombay, to Secretary, GOB, Judicial, March 1911. Reprinted in Edwardes, Bombay City Police, Appendix A, pp. 196–7.
65 Vincent's role as a mediator in the strike and its limitations are evident in newspaper reports in the Bombay Chronicle and Times of India, especially in the third week of January. For
an account of the strike, see Newman, Workers and Unions in Bombay, pp. 120–30 and R. Kumar,' 'The Bombay Textile Strike, 1919', IESHR, 8:1 (1971), 1–29.

Page 199
To Vincent's 'close association with his Marathispeaking ayah and the Maratha sepoys of the Police', during his childhood, was attributed not only his fluency of
language but more especially
his command of colloquial expressions in common use among such people. The term 'colloquial expressions' embraces a wide variety of jests and other phrases which among us
are regarded as indecent. This command of language was not without a useful effect when dealing with people like the millhands of Bombay.66
For Fatty Vincent, however, swapping smutty jokes, exchanging the choicest abuse or indulging in badinage with the workers, was not quite enough to secure a
settlement of the strike.67
In this way, police commissioners sometimes imagined for themselves the role of the kotwals of the great Mughal cities. At lesser levels of the hierarchy, too, and within
local police stations, deals were struck, favours sought and recovered and disputes settled between rival magnates or between employers and workers. For instance,
in 1937, the police were actively involved in an industrial dispute in the tanneries of Dharavi and particularly in its trade union rivalries. The rivalry among Marava and
Adi Dravida leather workers was not simply confined to the workplace or even the locality. It stretched back to Tinnevely district in Madras Presidency from where
these workers had emigrated to Bombay and its ramifications extended to caste conflicts and personal feuds, which appeared to converge upon the figure of Moses
Nallakhan or M. N. Moses as he styled himself. Moses invited S. V. Parulekar, the trade union leader, to form the Dharavi Tannery and Leather Workers' Union.
Moses' attempt to call in the wider resources of trade unionism to alter the local balance of power in his favour and against his Marava enemies backfired when the
tannery owners used this opportunity to replace his friends with his foes in their labour force and thus neutralize the threat of trade union activity. This only served to
intensify the rivalry. At the outset, however, Moses was called to the police station and 'warned', and he was later charged with assault and intimidation. On the other
hand, the police appeared to refuse to take action against the Maravas. In a petition to the Home Minister, Moses drew the conclusion that 'we are not getting justice
because our opponents have got monetary assistance and money to spend to ruin us whereas we are poor and helpless people'. In February 1938, Moses complained
to the police commissioner that although
66 Curry Diaries, 'The Joy of Working', vol. II, pp. 55–7, Curry Papers, Box 1, Centre of South Asian Studies, Cambridge.
67 However, it is clear that both Vincent and the Governor played an important role in persuading the millowners to make concessions and thus to bring the strike to an end. Bombay
Chronicle, 20 January 1919; Times of India, 20 January 1919; ibid., 21 January 1919; ibid., 22 January 1919.

Page 200
he had been assaulted by Ayyaswami Thevar and three others in the presence of a constable, the officer at Mahim police station disregarded his and the constable's
evidence and in fact refused to record a statement from him. Subsequently, when one tannery owner informed Inspector Raje 'about his intention to effect a reduction
in wages', the latter organized a police guard at the workplace, a response which may provide a clue to his earlier attitude to Moses.68 Until the 1920s, the police
intervened in the conciliation of disputes through the commissioner's office; in the following decades, the growing scale of industrial action drove such state intervention
progressively to higher and more centralized levels within the provincial government. As the role of the police was focused, more formally and so less flexibly, upon the
containment and regulation of industrial action, it came to be seen more clearly as a repressive force in workers' politics.69
Policemen active in the regulation of prostitutes often profited from them and some became involved to varying degrees in the organization and protection of brothels.
To this end, they used their authority as policemen informally to gain the compliance of pimps and prostitutes, tailors and moneylenders and a variety of intermediaries
involved in the trade. Sometimes they cuffed their collaborators into profitable, if unequal, partnerships. It was when the tensions between those involved escalated
beyond the ability of the police officers to manage, contain or resolve them, or indeed, when police officers succeeded in making enemies of all their business partners,
that their profitable rackets of protection and terror were brought to light.
Indeed, it was precisely such a breakdown of business arrangements in 1917 which uncovered a complex web of extortion and corruption in the police force, which
seemed to lead to its highest levels. In April 1917, an internal enquiry into various allegations found that Inspector Favel 'had been guilty of systematically taking illegal
gratifications from prostitutes and other persons with whom he came in contact officially'.70 Barely, a year earlier, Inspector Favel had been awarded the King's Police
Medal.71
68 GOB, Home (Special), File 550 (23) A of 1937, MSA.
69 See above chs. 3 and 4.
70 F. A. M. Vincent, Commissioner of Police, Bombay to Secretary, GOB, Judicial, 6 April, 1917, Bombay Confidential Proceedings, Judicial Department, vol. 25, 1917, p. 95, OIOC.
71
Statement of Mr D. Meyer of 23 Rampart Row before F. C. Griffith, Deputy Commissioner of Police, CID, 29 March 1917 in ibid., p. 100. Indeed, as one Mr D. Meyer, a commission
agent, who had fallen in love with Mary Fooks, a Russian prostitute, who worked under the inspector's protection, deposed before the police enquiry, Favel liked to boast that 'he was
the only Jew in the Police who had won the medal'. In fact, Favel expressed the opinion that 'something ought to be done to celebrate it'. Since Favel had already threatened to deport
Mary Fooks, Meyer recognizing that discretion was the better part of love, paid Rs 600 for a picnic at the Elephanta Caves in the inspector's honour.

Page 201
During the war, Inspector Favel had been assigned the task of 'dealing with the large number of enemy and neutral aliens who have come under notice'. Indeed, the
colonial state had attached considerable, perhaps excessive, importance to these duties. In this role, according to Vincent, the commissioner of police, Favel had
become his 'righthand man'.72 Thus, the commissioner took the firm view that 'it would be highly impolitic publicly to disgrace' such an officer by instituting disciplinary
procedures against him or prosecuting him.73 His recommendation that Favel should be allow to resign and to return his medal was quickly adopted by the Government
of Bombay.74
The police enquiry into Inspector Favel's activities focused primarily on his dealings with 'foreign prostitutes'. Mancharam Pitambar, a tailor, known to his friends as
Barny, had followed his father into the trade as a dressmaker to the European prostitutes in Bombay, and on the strength of some seventeen years' experience was
able to provide a recent history of the 'European' brothels.75 Although his history went back to the turn of the century, it was in about 1909 that the European
prostitutes had been placed 'under the direct supervision of Inspector Favel'.76 At this time, the chief impresario of the white slave trade was a Russian Jew known as
M. S. Toster, who acted as a broker 'between the police and the prostitutes' and as a sort of labour exchange for the women, facilitating their transfer between
brothels and meeting and accommodating 'new arrivals', and helping to 'put them in touch with the Mistresses of the brothels'. Favel allied with his rival Maurice
Finckelstein to ease Toster out of Bombay. Finckelstein now stood 'alone in his glory and wielded enormous power over the women, both girls and Mistresses'. In his
moment of glory, however, Finckelstein 'had a quarrel with Inspector Favel' about 'the division of the money extorted from the prostitutes'. As Barny noted
portentously, 'the bond between them had snapped'. Six
72 F. A. M. Vincent, Commissioner of Police, Bombay to Secretary, GOB, Judicial, 6 April 1917, in ibid., p. 95.
73 Ibid.
74 Secretary, GOB, Judicial, to Secretary, GOI, Home, 8 June 1917, in ibid., p. 103. Indeed the Government of Bombay pointed out to New Delhi that after all 'there was very little
difference between resignation in the circumstances disclosed in the case and dismissal'. It also suggested that a prosecution of Mr Favel was unlikely to stand, because of 'the low
moral character of many of the witnesses' and because the evidence consisted of 'mere statements taken down by the police and the persons who made them, being equally guilty of
giving illegal gratification are not likely to adhere to them, and so admit their guilt, in a witness box'.
75 Statement of Mancharam Pitambar alias 'Barny', in ibid., pp. 95–8. Barny claimed that 'by reason of my business dealings with these women, I got to know all intimately and I made
myself useful to them by doing little odd jobs for them outside my more regular business'. Indeed, these 'business dealings' had turned him into something of a linguist. He had learned
'the languages most commonly spoken by these women, namely Yiddish, Italian, and a little German and also a smattering of French'. Ibid., p. 95.
76 Ibid., p. 96.

Page 202
months later, in about 1911, Finckelstein was deported. The deputy commissioner of police, in the Criminal Investigation Department at the time of his deportation was
none other than Fatty Vincent.77
With the departure of Finckelstein, Favel had effectively become both the middleman and, in his police duties, the final receiver and enforcer in the web of extortion
which had been spun around the brothels. As official anxieties about aliens mounted with the outbreak of war, Favel found that his duties afforded large and growing
opportunities for profit. Favel routinely received presents from the pimps and prostitutes. He was paid for leading the women through the maze of regulations that
governed their entry into and movement within the subcontinent. He expected to be rewarded for settling disputes within and between brothels. He charged 'a heavy
commission', from both buyers and sellers, whenever a brothel changed hands and there appears to have been a rather bullish market for shares in them. The broker in
these transactions was a Mrs Markovitch, who had no papers to prove her nationality,78 and was consequently wholly at Favel's mercy. Favel took a cut of the
proceeds of the annual ball, which Mrs Markovitch organized at the Balcon, the hotel of the European prostitutes, and which, it was said, made 'a good deal of
money'.79 Favel colluded with Mrs Markovitch and his former lover, Fritza, the mistress of No. 392 Falkland Road, to ensure that 'all fresh arrivals' went to the latter's
house. As a result, 'Fritza's brothel flourishes with the number of girls while other brothels are starved'.80 Needless to say, Favel also required free access to the
brothels, although his visits, paid for by the mistress of the house, were, it was said, 'by no means as frequent as they used to be'. Nonetheless, these visits would make
Fritza, who was still 'his very good friend', 'very angry and jealous'.81
The threat of deportation was the most powerful sanction which Favel could apply not only to extort money, but also to tighten his control over the pimps, prostitutes
and mistresses. He liked to order the pimps to leave Bombay 'by the first available steamer' but 'on squaring Mr Favel they were permitted to stay on for a few
weeks'. Pimps were, thus, expected 'to pay for the privilege of extending their stay in Bombay and for protection against executive action'.82 The same sanction was
used to release brothels from the protection of his own rivals, named by one 'mistress' as Mr Sloane and Mr Nolan (presumably police officers), and as Mlle Mina of
No. 6 Grant Road found, to bring them more fully under his sway.83 Similarly, the deportation as an enemy alien of Fritza's partner at No. 392
77 Ibid.
78 Statement of Mml. Margot, mistress of No. 6 Grant Road, in ibid., p. 99.
79
Statement of Mancharam Pitambar, alias Barny, in ibid., p. 97.
81 Statement of Mancharam Pitambar, alias Barny, in ibid., p. 97.
82 Ibid., p. 96.
83 Statement of Mml. Mina, of No. 6 Grant Road, in ibid., p. 101.

Page 203
Falkland Road, Sophie Schlamp, a German subject, enabled Fritza to sell a half share in the brothel, realizing nearly 80 per cent on her original outlay in eight years84
and to share the proceeds with Favel.85 Having ostensibly sold her share in the brothel, Fritza continued to live in it. As the purchaser, Mlle Jennette, complained,
Fritza still 'shares the profits with me',86 and after a decent interval she even began 'trying to sell Sophie's share to me'.87 When Mr D. Meyer, a commission agent, fell
in love with Mary Fooks, 'a Russian girl', who worked at Fritza's brothel and 'desired to rescue her from the life she was leading', he became a soft target for Favel.
The inspector used the threat of deportation to pull Mary Fooks back into the brothel and then over two years to extort some Rs 4,000 from Mr Meyer, until once
again and once too often, Favel 'once more started to squeeze me'. Meyer decided this time that he 'had done quite enough for him and I had a row with him and
broke with him'.88
Once Favel had acquired such extensive, virtually monopolistic powers in the trade, it was always going to require deft and dextrous management to hold his business
interests together. Not surprisingly, Favel soon began to gather foes about him. His business partners had been kept in place almost exclusively by threat, coercion and
humiliation. In time, even they became disgruntled. Before the war, Barny, his agent after Finckelstein's departure, believed that some 80 per cent of Favel's takings
'came through me and the balance directly from the ''giver".'89 But, 'since the outbreak of war' when Favel had acquired greater powers to supervise the movement of
alien subjects, Favel's 'income from the prostitutes had increased considerably' and further, 'he now takes much more money direct than he used to take and less of it
comes through me'. To make matters worse, Fritza added insult to penury: taking advantage of her relationship with Favel, she had 'no compunction in calling me "a
bloody cooly".'90 While Fritza remained loyal to Favel, she sometimes felt scorned and was often racked with jealousy.91 Several mistresses resented the collusion of
Favel, Mrs Markovitch and Fritza which starved them of business.92 Favel's handling of Meyer, indeed, of the opportunity which he provided, suggests that he did not
always know where and how to stop. Too many people, often too close to the police by virtue of their
84 Statement of Mancharam Pitambar, alias Barny, in ibid., p. 97; statement of Mrs Fritza Shalome alias Mme Fritza, in ibid., p. 102.
86 Statement of Jennette, mistress of 392 Falkland Road, in ibid., p. 101. This ideal business arrangement for Fritza would presumably have been impossible to maintain without the
protection of her partner. Statement of Mancharam Pitambar, alias Barny, in ibid., p. 97.
87 Statement of Jennette, mistress of No. 392 Falkland Road, in ibid., p. 101.
88 Statement of Mr D. Meyer of 23 Rampart Row, in ibid., pp. 99–100.
90 Ibid., p. 97.
91 Ibid., p. 96.

Page 204
livelihoods, and frequently dependent upon Favel's patronage, had, over the six years since Finckelstein's departure, accumulated too many scores to settle with Favel
for his empire of extortion not to come unravelled.
In view of the range of their involvement in the social organization and political networks of the neighbourhood, it is not surprising that one of the perennial anxieties of
senior officials in matters of public order policing was the discipline of the police force. 'In times of disturbance', S. M. Edwardes commented in reference to his own
period as commissioner, the police 'often reaped a fair harvest of tips and presents from timorous townspeople who desired protection from mob violence, and also
discovered in the aftermath of rioting an easy means of paying off old scores'.93 These opportunities presented themselves most clearly when 'times of disturbance',
whether communal riots or strikes or civil disobedience campaigns, were sustained over extensive periods. At the level of the neighbourhood, the police often
appeared to operate as protection racketeers like the dadas whom they branded as rough and criminal or the groups of Pathans who watched the shops of the bazaar
for a price and apparently on pain of looting them.
At a formal level, the strategies of the police, indeed the meditations of the official mind, merged law enforcement in relation to crime with the maintenance of public
order. Committed to the theory that political movements were the result of the machinations of agitators misleading, manipulating or mobilizing the illiterate, the fanatical
and, especially, the rough, it was easy to conclude that the stricter control of 'bad characters' or 'any tribe, gang or class addicted to the commission of nonbailable
offences'94 would remove some of the most combustible materials available to political troublemakers. Conversely, periods of public disorder, the police claimed,
diverted their attention and allowed criminals the freedom and the opportunity to practise their craft without restraint.95 Similarly, in 1926, the commissioner of police
deployed the same argument to press the case for greater resources for his force. When there was 'disorder on a large scale', he argued, criminals who were 'kept in
some sort of control in normal times' took 'advantage of the excitement and panic to openly burn, loot and murder and cause incalculable damage to the city'. To the
extent that the police were effectively able to jail or deport 'undesirables' in 'normal times', they would be successful in maintaining the peace in periods of 'disorder'.96
94 Cox, Police and Crime, p. 205.
95 Bombay 1921–22: A Review of the Administration of the Presidency (Bombay, 1923), p. 20.
96 Commissioner of Police, Bombay to Secretary, GOB, Home, 9 December 1926, Copy in BRIC, 1929, File B, MSA.

Page 205
Thus, whenever they wanted to clear the streets, for instance during the royal visit of 1911, the police simply jailed those whom they believed were hooligans or 'bad
characters'. The Criminal Procedure Code allowed magistrates to require a person placed before them on suspicion by the police 'to show cause why he should not
give security for good behaviour for a year'. But 'hardened malefactors' underwent a less stringent procedure and could be jailed if they failed to produce the security.
Yet, as Edmund Cox pointed out, one 'remarkable feature' of the regulations was that a person could be proved to be 'an habitual offender . . . by evidence of general
repute or otherwise'. In fact, 'it is easy in India', he observed, 'to obtain evidence of general repute that any person who has rendered himself in anyway unpopular is an
habitual offender'.97 In Bombay, however, the police placed few such cases before the magistrate, and most of them were either discharged or withdrawn. Only fifty
people were placed before magistrates in Bombay in 1922 as compared with 902 in Calcutta.98 This was attributed sometimes to the difficulty of obtaining evidence
from the public who feared retaliation from 'badmashes' and at others, rather more weakly, to the fact that 'in the city there is not that knowledge of one's neighbour's
affairs that is such a characteristic of village life'.99 But the police did not, on the other hand, find it easy to convince magistrates. Until 1921, therefore, when he wished
to round up hooligans, the commissioner of police, 'exercised his powers as a Magistrate and a Justice of the Peace and . . . used to have accused persons put up
before himself as a Magistrate and to remand them into Jail or Police custody'. But the magistracy objected to this procedure and the Government of Bombay
conceded the point and removed his magisterial powers. 'So now', the police commissioner observed, 'it is not so easy to deal with mawalis and such persons as it was
then'.100 But such legal difficulties did not lead them to abandon this strategy altogether.
The police, as we have noted, maintained a list of 'bad characters' but, in practice, this was hard to achieve. While drawing up a register, it was difficult to know whom
to exclude; once compiled, it was no easy task to decide whom to strike off and when. 'I have often found the registers in the utmost confusion', noted one official, 'and
still containing names of
97 Cox, Police and Crime, pp. 108–9.
98 Annual Report on the Police, 1922, p. 17. See Suranjan Das, 'The Goondas: Towards a Reconstruction of the Calcutta Underworlds through Police Records', Economic and
Political Weekly, 29:44 (29 October 1994), pp. 2877–83.
99 Annual Report on the Police, 1923, pp. 22–3.
100 Proceedings of the BRIC, 1929, P. A. Kelly, Commissioner of Police, File 18, p. 207, MSA.

Page 206
101
persons who have led an honest livelihood for years, or who have disappeared, or have long since died.'
Although public discourse lent a certain authority to the term 'bad character', it was not always easy to identify him. In Bombay, as elsewhere, the police readily
identified crime and disorder with the poor and especially the unemployed and the rural migrants.102 But these categories covered the vast majority of the city's
population. The overwhelming majority of the city's population – and, indeed, a significant proportion of police force – were rural migrants, who held on tenaciously to
their rural base and returned regularly to their villages. Few workers enjoyed secure conditions of employment. Casual labour was used extensively both in the so
called formal as well as the informal sector and periodic unemployment was integral to the shared experience of the working classes. Moreover, identifying bad
characters was often complicated by the confusion of stereotypes in public discourse. Thus, in the early twentieth century, while cotton millworkers from Ratnagiri
were believed to be volatile and averse to punctuality and discipline, dock labourers from the Deccan and employed usually on a daily basis were portrayed as steady
and sober, pliant and docile, loyal and obedient.103
Indeed, when the police decided to round up 'bad characters' during the communal riots of February and May 1929, and then undertook a sociological survey of their
prisoners, their results, as we have seen, revealed nothing more than the random manner of their arrests.104 The social profile of the prisoners bore a remarkable
resemblance to a crosssection of the city's population. A senior police official was later to recall that the quarters which had been 'combed out' were 'notoriously
criminal' and that the police had
arrested everyone known to them as having previous convictions and belonging to the apache type. . . . Most of them had previous convictions, sometimes half a dozen or more.
Many criminals fled the city, and conditions gradually became normal.
During the combingout process many of the hooligans came out of their houses without any trouble, when called out by police officers who knew them; other bolted like rabbits,
only to find their earth stopped.
Indeed, 'one gang which made an ineffectual attempt at resisting' was 'a notorious lot of cutthroats known to the British police sergeants as ''Jenny's gang", owing to
the fact that they were controlled by a wicked old woman whose name sounded something like Jenny'.105 Colourful
101
Cox, Police and Crime, p. 205.
102 See ch. 5 above.
103 H. A. Talcherkar, 'Raghu, the Model Millhand: A Sketch', Indian Social Reformer (March 1908); Chandavarkar, Origins of Industrial Capitalism, ch. 4.
105 Curry, The Indian Police, pp. 180–1.

Page 207
stories, many of them less enthralling than these, were no doubt frequently retold in Bournemouth and Bath by the returning heroes of the empire. But there is no
evidence to support the claim that 'combingout' brought the riots to an end. The police may well have tried to round up the usual suspects, but there was in fact little
evidence to connect them to the riots and most of those arrested were released after trial. Certainly, the police operation failed to lend credibility to the notion that 'bad
characters' formed a discrete social group and belied the claim that the police knew who they were.106 The drama of the police operations, indeed their claim of
maintaining a register of 'bad characters', drew attention to the gulf between their grand claims and modest achievements in controlling crime and maintaining order.
IV
The police deployed the same methods in the political sphere which they developed in relation to crime, and frequently within very similar constraints. To the official
mind, Indian politics often appeared to be indistinguishable from a conspiracy. Political movements were mobilized by agitators who incited the credulous and
manipulated the naive. Colonial officials believed that agitations could be prevented and riots averted in their early stages, but once they gained momentum, they could
quickly spread beyond their control. Preventive measures entailed gaining access to what the official mind imagined were the whispered rumours of the bazaars. To
find clues to the popular mood, to ascertain what was unknowable and often irrational, and to penetrate the political networks of the street and the neighbourhood,
colonial officials believed they had to rely upon spies, informers and detectives. Political surveillance was given considerable importance but it scarcely yielded more
satisfying results. From the early 1890s, the CID maintained a 'secret abstract' of political intelligence in which reports from every district were collected. But large
parts of these abstracts consisted of extracts from newspapers or reports of speeches made at various public meetings and political conferences, including the annual
session of the Indian National Congress, which were, especially at the time, readily available elsewhere. The CID also censored plays and proscribed books and
confiscated those which were subsequently published. Some officers must have become very widely read. In 1910, for instance, they claimed to have scrutinized 297
plays, rejected six outright, but passed a further seven after some un
106 GOB, Home (Special), File 543 (10) E Part G of 1929, MSA.

Page 208
107
doubtedly brisk and forthright alterations. In addition, they kept a watch on political agitators and foreigners arriving and departing at the port. The secret abstracts
testify to the fascinating bar gossip which policemen picked up, but their accuracy cannot invariably be relied upon. They also suggest that the surveillance of political
agitators was far from systematic and hardly a permanent obstacle to political organization and action.
Frequently, the police were surprised by the onset of 'disturbances'. In police reports, 'rioters' often appeared in a flash and disappeared without a trace. Thus, after
the communal riots which occurred over cow slaughter in 1893, the commissioner of police reported that although the police had 'an inkling' of what was to come,
when it happened 'the rioters spread like a flash of lightning and in about an hour the whole of the Native Town was in an uproar and fighting and looting was the order
of the day'.108 It was perhaps an inherent characteristic of communal riots that the patterns of violence, its perpetrators and their targets, often seemed arbitrary and
unpredictable to the police as much as to the city's inhabitants. But for all their efforts at surveillance and gathering intelligence, the police appeared to know as little
about the situations in which public order had appeared to break down and sometimes to understand it less than the residents of the neighbourhood. It was not unusual
for the police to be misled by their own intelligence.109
Following the mill strike of 1908, the CID confessed that they lacked the means to watch and follow political agitators effectively and their budgets were further being
squeezed. Indeed, the commissioner of police declared that he needed to double the strength of the CID to watch 'prominent political agitators'. Agitators revealed a
remarkable tendency to multiply, when they were closely watched, so that the work of the CID had increased 'by several hundred per cent in the last few years'.
Political 'suspects', as the commissioner of police pointed out, could arrive and depart at twentyone railway stations in the city. When they drove off in a victoria
carriage on arrival, the detective found it 'almost impossible to keep them in view'. In addition, 'they invariably put up with friends in large blocks of buildings which
makes it very difficult to locate them for some time'. To make matters worse, agitators who became conscious of being watched by the Clouseaus of the plainclothes
police began to take
107 Annual Report on the Police, 1910, p. 7. The impact of their actions upon the local theatre has sadly passed unrecorded. In addition, the CID also took vigorous action to
confiscate 21, 154 proscribed books in the same year.
108 Acting Commissioner of Police, Bombay, to Secretary, GOB, Judicial, 12 August 1893, in GOB, Judicial, 1893, vol. 194, Compilation no. 948, Part I, p. 147, MSA.
109 See, for instance, the case of the Bandra Observation Camp during the plague panic in 1898, in ch. 7 below, pp. 251–2.

Page 209
110
precautions and thus proved yet more elusive. As the Government of Bombay contemplated the problems of political surveillance, they grew increasingly elaborate
and complicated. The Home Secretary agreed that it was advisable to employ a special force to shadow political agitators. In addition, the district and railway police
should be instructed to exercise 'absolute watchfulness', he suggested, when agitators began their journey and to telegraph the 'known or possible place of destination'
while sufficient men had to be detailed to shadow them 'en route' and after arrival.111 Another senior official noted that it was vital to have enough men 'to keep a watch
over each suspect's house, night and day and to watch the headquarter Railway Station as well as the two stations on either side'. It followed that watchers should be
made 'thoroughly familiar with the personal appearance of all suspects' and it was essential to provide them with enough support so that local policemen were made
available to relieve them at the destination as well as en route, in case 'suspects' made a switch.112 On the verge of tying down the rickety apparatus of the police with
their delusions, the Government of Bombay drew back. As they scrutinized their own plans, they began to realize that it would be impossible to provide enough men to
watch all agitators. Instead, the Home Member suggested that the police ought to obtain 'information of any serious movements or plans' by 'employing private
informers and rewarding them handsomely for all information of value'.113 Needless to say, the police had their own network of informers. It is doubtful whether the
Government of Bombay, if they had sought to formalize it, would have found the information they received about political movements and conspiracies worth the
money they considered lavishing upon them.
Moreover, political agitators raised the same problems of procedure as 'bad characters'. It was difficult to gauge who should be watched and when some could be
simply ignored. The consequences of their constraints were sometimes farcical. Kanchan Kumar, a stump orator during the mill strike of 1908, was kept under
surveillance for eight years thereafter and, especially during the first world war, his movements were laboriously recorded in the Secret Abstracts of Intelligence.
Kumar adopted numerous disguises, took on aliases and travelled sometimes as an ascetic, a holy man, sometimes as a peddler. By 1915, he was reported to be
consuming 1 1/2 tolas of ganja daily. His aliases and increasingly exotic
110 Commissioner of Police, Bombay to Secretary, GOB, Home (Special), 3 November 1908, in GOB, Home (Special), File 71 of 1908, pp. 25–7, MSA.
111 Note by Secretary, GOB, Home (Special), 21 October 1908, in ibid., pp. 32–5.
112 Note by JAG, 6 November 1908, in ibid., pp. 37–9.
113 Home Member's Note, 24 November 1908, in ibid., pp. 43–4.

Page 210
disguises, his continuous flight from the police and his ceaseless movement across the subcontinent, his failure to hold down a steady job, his increasingly voracious
appetite for cannabis only served to confirm the suspicion of police intelligence that he was dangerous and prone to crime and that, indeed, in view of his past, the
possibility that he may be involved with a seditious conspiracy could scarcely be ruled out. When, in 1916, eight years after they first took an interest in him, the police,
bemused by his ingenious disguises, troubled perhaps by the several lines which his aliases now took up in the secret abstract, finally hauled him in for questioning,
Kanchan Kumar complained about being pursued by the police. As a result, he had been unable to secure or hold down a job. For no sooner had he found a job, than
the police would turn up to question his employer and he would be sacked. To get away from the CID, he had changed his name, moved to new places, even adopted
disguises to throw the police off his trail. He could no longer stay in the same town long enough to find employment. His last resort, he said, was cannabis which at
least momentarily allowed him to lose himself.114
Undaunted, the police continued to develop their own methods of surveillance. By 1910, an interprovincial list of agitators was being maintained to enable the CID to
follow those who came to Bombay from elsewhere in the Presidency.115 Three years later, the Inspector General of Police suggested an exchange of CID officers with
other provinces so that each could watch arrivals from their own jurisdiction.116 The response from Bengal provided little encouragement: while they were willing to
send staff to Bombay, they replied, they did not really need any officers in exchange. In any case, Bengal extremists did not seem to communicate with Bombay very
much and 'there is no centre outside Calcutta which the latter visit'.117 By 1920, the commissioner of police seemed less than satisfied with the style of operation of the
CID. Their methods had become outmoded. For, he noted, 'Political agitators nowadays do not hide their light under a bushel. Publicity and advertisement are as the
breath of their nostrils and on them they chiefly rely for their success.' Indeed, nothing showed how inappropriate their procedures had become, he observed, than the
fact that at the height of the noncooperation campaign, neither Gandhi nor Vallabhai Patel appeared on the current list of agitators.118
From 1919, special officers were detailed to report on Bolshevik activ
114
BPP SAI, 1914–16, passim.
115 GOB, Home (Special), File 71 of 1908, MSA.
116 Inspector General of Police, Bombay to Inspectors General of Police, Bengal, Central Provinces and Punjab, 1 July 1913, in ibid., p. 163.
117 J. C. Cumming, Secretary, Government of Bengal, Home to C. C. Watson, Acting Secretary, GOB, Political, 19 August 1913, in ibid., pp. 181–2.
118 Note by Commissioner of Police, Bombay, 24 December 1920, in ibid., p. 241.

Page 211
ities, but there were no communists to be found in India as yet. However, later in the 1920s, the political surveillance of communists proved easier to conduct than
spying upon nationalists. The communists were selfdefining, more exclusive and they had a smaller membership. They placed a considerable weight on propaganda,
used a distinctive language and to a large extent operated openly. Their internal feuds also helped to make information more freely available to the police. The evidence
presented for the prosecution in the Meerut Conspiracy Case conducted in the early 1930s shows how readily the coterie of communists who led the Girni Kamgar
Union in Bombay were observed, how often their letters were intercepted and how easily, if not always reliably, their speeches were reported.
Police action has sometimes been portrayed as the greatest obstacle to the growth of trade unions. Its role in Bombay was rather more limited. In fact, workers'
combinations were more consistently thwarted at the point of production.119 However, the momentum of industrial action, and especially the scale of the general
strikes, invited the increasingly robust intervention of the state and, necessarily, the police, in industrial disputes. From 1929 onwards, they tried to limit picketing at the
mill gates to two people and found support in subsequent legislation for this measure. They sometimes escorted blacklegs to work, thus helping to keep the mills
working when the employers were trying to break the strike. They arrested the leaders of the Girni Kamgar Union for the seditious tone of their speeches. In the
general strike of 1934, police detectives were posted to maintain a watch over propaganda and picketing in the workers' chawls.120 Although the Government of
Bombay proclaimed its lofty intention on the eve of the strike 'to keep the ring clear for both parties', one newspaper quipped the following day, that 'the ring in Parel
was so full of armed policemen that there was hardly any room left for the workers'.121 Four days later, all meetings and processions were prohibited in the mill
districts. In the next three weeks, in two stages, twentyeight communists in the Joint Strike Committee were arrested under emergency powers legislation for 'abusing
the liberty of speech'.122 In the later 1930s, by its legislative initiatives as well as at times by police action, the provincial government continued to attack the communists
of the Girni Kamgar Union.
The colonial state was reluctant to wage and sustain a war of attrition against a significant focus of political opposition. But in the case of the
119 See ch. 3 above; also ch. 5
120 GOB, Home (Special), File 543 (48) of 1934, MSA.
121 Bombay Sentinel, 24 April 1934.
122 GOB, Home (Special), File 543 (48) of 1934, p. 115; GOB, Home (Special), File 543 (48) K of 1934, p. 5, MSA.

Page 212
communists they could draw upon wider public sympathy. The millowners had, of course, urged them to repress the communists since 1928. But the government also
gained from the fear of the threat of communism among larger sections of the professional and commercial classes, local Congressmen, and even the press, all of whom
more readily linked communism with their own perceptions and prejudices about the roughness of workingclass neighbourhoods. If the communists could thus be
isolated, the colonial state could attack them without incurring the risk of debilitating political costs. However, it was still necessary to distinguish between state
intervention to preserve the rule of law and maintain public order, on the one hand, and to favour one side of an industrial dispute, on the other. For, in the long run, the
future of colonialism would depend upon its ability to extend and diversify its social base and seek collaboration within the working class.
As the police strove to maintain public order in the early twentieth century, they encountered political action on an impressive scale. Bombay became the seat of an
increasingly militant labour movement and the centre of mass nationalist agitation, and between 1929 and 1947 sporadic communal riots engulfed the city. In the light of
the organizational weaknesses, the financial and political constraints and the strategic and operational shortcomings of the police, we might suppose that they would find
it impossible to maintain public order under growing political pressure. In fact, the police accumulated and used the means of repression with growing readiness.
Increasingly, in this period, and especially in the 1930s, the willingness of the police to swing their lathis and point their guns at antagonistic or disorderly crowds
quickened. But it should not be supposed that the police were able to function as a consistent, systematic or sustained force for repression. Rather, their coercive
character was unevenly, if sometimes rather violently, manifested.
Certainly, perennial anxieties within the higher echelons of the police about the rank and file surfaced under conditions of potential violence. Faced with the tasks of
crowd control, the police could act with considerable brutality. Sometimes, police violence was indiscriminate and sometimes it emanated not from the humble havildar,
whose discipline so often appeared suspect to his commanding officers, but from his superiors, often British and Eurasian, whom they trusted. Thus, for instance, the
Home Member noted in exasperation that, during the civil disobedience movement, in September 1930, on polling day in the council elections, which were being
boycotted by the Congress, there had been

Page 213
a flood of complaints that the 'police, Sergeants and sepoys alike, failed to distinguish between friend and foe . . . and even persons going to the Town Hall to record
their vote were beaten with lathis'. It would have been reasonable to expect, the Home Member went on, that the police would be able to understand that people
dressed in khaddar were against the government and those in 'European or oldstyle Indian clothes' were not, 'Yet after every fracas we hear of several (I nearly said
''numerous") cases in which friendly persons have been indiscriminately beaten. And the Sergeants who ought to know better are said to be the culprits.'123 The
indiscriminate violence of the sergeants during civil disobedience suggests that the official mind had reason for concern about the brittle discipline of the force when it
encountered protesters.
Within the higher ranks of the force, at least, there was probably considerable sympathy for the police commissioner's view, expressed in the immediate aftermath of
the communal riots of 1929 that 'the only medicine for rioters was the bullet'.124 But the police were not always free to translate these sentiments into action. The police
opened fire more frequently and killed more people in India than would have been politically tenable in Britain, but they were often more reluctant to show force. On a
conservative, if impressionistic, estimate the police and the army killed at least seventyfive people between 1919 and 1942 by shooting in order to disperse crowds of
strikers, nationalist agitators or communal rioters. More than thirty people were killed in the first three days of the Quit India movement in August 1942 alone.125 But it
would be misleading to conclude that these deaths were simply a direct reflection of the colonial and racist view that Indian lives were cheap. In fact, senior police
officers were often called upon to justify their use of force and they sometimes went to desperate, even absurd, extremes to explain the actions of their men. The
absurdity of their explanations reflected less the confidence of colonial officials that whatever excuses they offered would be accepted, but rather anxiety and guilt born
out of a desperate search for selfjustification. During the communal riot of May 1929, soon after which the commissioner of police had expressed his view that rioters
deserved to be shot, he reported an incident in which 500 strikers, supposedly Hindus, were said to have stoned a group of 200 Muslim workers in the Fazulbhoy
Mill, who having been hired as blackleg labour, were being escorted to their homes by the police. Thirteen shots were
123 Note by Home Member, Home Department, 26 September 1930 in GOB, Home (Special), File 750 (26) of 1930, p. 185, MSA.
124 Proceedings of the BRIC, 1929, oral evidence, P. A. Kelly, File 18, MSA.
125 GOB, Home (Special), File 1110 (6) A (1) of 1942, MSA; GOI, Home Poll, File 3/15/43, NAI; Hutchins, India's Revolution, pp. 228–33.

Page 214
fired 'to disperse the crowd and save the Mohammedans'. Two strikers received 'slight' injuries, one of them a bayonet would to the head. But, as the commissioner
pondered, there had been no bayonet charge. 'It appears', he concluded solemnly, 'that he [the striker] must have hit himself against the bayonet while running away'.126
Nearly two years later, during the civil disobedience movement, an elderly woman, who had, along with others, mounted a picket outside the Municipal Corporation
buildings, 'fainted' as the police attempted 'to disperse the crowd'. She later alleged that she had first been dealt a blow with a lathi to the stomach and then had been
pushed also in the stomach with the end of a lathi. The first allegation, declared the commissioner briskly in the secrecy of the Home Department's file, was 'impossible'
because no lathi blows were struck and the second was 'equally unlikely' because 'lathis were used horizontally against the backs of resisting women to push them
along. It is possible that the woman may have in some way pressed herself against the end of one of these.'127 Mercifully, there are no recorded instances in Bombay of
miscreants hurling themselves in the path of a speeding bullet, though some no doubt suffered severe damage while resisting arrest.
When the police opened fire, they frequently justified their action by the claim that they were on the point of being overwhelmed. There was a real and pragmatic side
to this perception and the anxieties it generated. Their numerical and organizational weakness meant that police pickets often found themselves surrounded,
outnumbered and attacked. The police operated on the front line of crowd control. They were fully exposed to the 'mob' and brought directly into confrontation with it.
Whether they patrolled the streets in small groups, formed pickets to defend a street corner or entrenched themselves in a police chowky, they became exposed
targets in hostile situations. Significantly, when troops were called out, the army generally refused to adopt 'police methods' precisely because they would thereby be
left exposed and vulnerable to attack.128
Moreover, their techniques of crowd control were never imaginative and often crude. Between the baton and the bullet, they often appeared to have no intermediate
options available. In the early 1920s, the Bombay police developed a sufficient enthusiasm for technological innovation in
126 Commissioner of Police, Bombay, to Secretary, GOB, Home (Special), Bombay, 15 May 1929, in GOB, Home (Special), File 348 Part II of 1929, p. 77, MSA.
127 Note by Commissioner of Police, Bombay, undated (probably 23 February 1931) in GOB, Home (Special), File 750 (26) of 1930, pp. 217–21, MSA.
128 GOI, Home Poll 33/12/32 and 21/6/33, NAI. See N. Narain, 'Cooption and Control: the Role of the Colonial Army in India, 1918–1947', unpublished Ph.D. thesis, University of
Cambridge, 1993, ch. 6.

Page 215
the form of 'lachrymatory gas bombs' to disperse crowds to seek the advice of Scotland Yard and the Philadelphia police. In 1922, the police officer who had been
the commissioner of police two years earlier recalled that he had favoured their use during the general strike in Bombay but had been thwarted because the government
feared that 'political capital would be made of the use of such a measure'. Their particular value lay in the potential they offered for dispersing crowds when otherwise
the police would have to shoot, and thus risk causing 'irreparable damage to a great deal of very valuable machinery and very possibly the loss of a number of lives'.
Lachrymatory bombs provided a means of dealing with 'rioters' 'more effectively and much more mercifully than any other'.129
However, the Government of Bombay continued to be apprehensive about its political consequences. To the provincial government's fear about 'political capital', New
Delhi pointed out that if 'the use of poison gas at war' was unlawful, they could scarcely contemplate its use in peacetime. 'Although the gas itself is reported to be
harmless', the Government of India noted, 'the nervous shock caused by it might result in death', for instance, they elaborated imaginatively, 'in the case of a man
suffering from advanced heart disease', and then, 'the death would be attributed to the gas'.130 More fundamentally, perhaps, colonial officials feared that the 'mob'
might panic and set off a stampede resulting in casualties which would embarrass the government.131 Indeed, the Bombay Municipal Corporation had at least after the
event taken the view that it would have been preferable to use tear gas to disperse satyagrahis during the civil disobedience agitation than for the police to lay about
them with lathis.132 However, throughout the 1920s and 1930s, colonial officials held fast to the belief that tear gas was likely to induce panic and cause stampedes,
that spectators and not just the troublemakers would be affected, and even that 'the police were too stupid to avoid gassing themselves'.133 The axiom that Indian
mobs were unpredictable and excitable led the colonial state, until the 1940s, to regard firing their weapons as no less destructive and far more appropriate under
Indian conditions. Tear gas was used sporadically in the middle and late 1930s, but it was not until the early 1940s that it was eventually accepted by officials and then
used more extensively during the Quit India movement.134
129 Note by Acting Inspector General of Police, Bombay Presidency, 25 April 1922 in GOB, Home (Special), File 610 of 1922, pp. 25–6, MSA.
130 Secretary, GOI, Home (Political) to Secretary, GOB, Home (Special), 16 June 1922, in GOB, Home (Special), File 610 of 1922, pp. 31–3, MSA.
131 GOI, Home Poll, 8/II/30 and 8/II/31, NAI.
132 Ibid., 8/II/31, NAI.
133 Arnold, 'The Armed Police', p. 116.
134 GOB, Home Department, Resolution No. 3766/2–V, 10 July 1941. in GOB, Home (Special), File 610 of 1922, MSA.

Page 216
The representation of the mob in colonial discourse vitally influenced the response it evoked from the state and, in particular, it helps to explain why, when faced with it,
the police often felt overwhelmed. Colonial discourse developed an elaborate aetiology of popular politics and especially of mass violence.135 It has already been
suggested that this discourse often treated Indian politics as a synonym for, and the outcome of, conspiracy, developed through incitement, rumour, propaganda and
sedition and spread by unscrupulous agitators to credulous, illiterate and volatile masses, whose own responses fell outside the universe of reason. This combination of
unscrupulous and irresponsible agitators and irrational and volatile people signified, in this discourse, the ever present and palpable threat that, in no more than an
instant, tension could flicker into riot, mere discontent could erupt into an epidemic of violence, antagonisms could be inflamed into a major political conflagration. The
inflammability of the people and their politics made it crucial that the colonial government should, for the sake of its own subjects as much as itself, maintain and
strengthen public order and stamp out disturbances at the first sign of their appearance. Imperial prestige and the pride of the ruling race required that the colonial state
should assert its control but, more crucially, officials recognized only too well how easily they could be swamped by the swelling tide of frenzied and fanatical natives.
Their fanaticism was often expressed through violence; but sometimes it appeared to know no bounds. In June 1930, the Government of Bombay observed darkly that
Gujarati traders in the city had been 'worked up', by 'the preaching of Gandhi and his followers', to 'a state of fanatical excitement in which they are prepared to suffer
extreme violence while for the most part remaining nonviolent. Their obstinacy makes them very difficult to deal with, while their nonviolence renders forcible
measures against them extremely unpleasant to those engaged in them.'136
Imperial prestige and the vulnerability of its agents were inextricably linked. This sense of their own vulnerability imposed powerful constraints upon the uninhibited
expression of what the maintenance of their prestige demanded. At other times, imperial prestige spurred on the police to take actions which were often more extreme
than the circumstances warranted. As the commissioner of police informed the Home Department in 1930, 'very restrained action by the police . . . only results in
encouraging the dregs of the population to insult and jeer at the Police . . . I therefore propose to take more vigorous action in the future.'137 If
136 Express Letter, Bombay Special, to Home, Simla, 4 June 1930, in GOB, Home (Special), File 750 (26) of 1930, p. 27, MSA.
137 Commissioner of Police, Bombay, to Secretary, GOB, Home (Special), 1 December 1930, in ibid., p. 215.

Page 217
pride and weakness combined to require 'more vigorous action' by the police, the Bombay government, or New Delhi, or, perhaps especially, London, would sooner
or later have to ask themselves how much vigour they could prudently permit. Vigorous action could create a wider sympathy for its victims and deepen and extend
antagonism against the state. Such calculations continually racked the official mind. For instance, Congress processions during civil disobedience allowed British rule to
be pilloried from the street, but they were nonviolent. Any attempt to prohibit them 'would certainly provoke defiance' and the situation could well run out of control,
although just how completely only the official mind could imagine:
A number of processions would begin in different parts of the City. Before long police would be unable to stop them and the military would have to be called in. Shooting on a
large scale would be inevitable and elements of the population now quiet would join in in sympathy. Peace would only be restored after serious riots.138
The urgency which the colonial state felt about the need to maintain order at all times and to thwart any threat to it at the first opportunity arose from its perception of
its subjects as volatile and fanatical as well as its own awareness of its weakness and vulnerability. These perceptions fed and developed the anxiety of the colonial
state and its agents, especially the police, that they might be on the point of being overwhelmed. It also entrenched an intermittent tendency to restore order with
excessive force. In other words, the weakness of the police arose in part from the fallibility of their assessment of the threat to the social order and of their judgement
about the best means of reasserting control.
This tendency to feel overwhelmed and to restore what it perceived as order with excessive force was inherent to the statecraft of colonial rule and embedded in
strategies of public order policing in India. As late as 1920, the Home Secretary reminded New Delhi that 'India possesses an internal frontier', which was marked by
'the relative instability of the social organism . . . the ease with which on skilful instigation all causes of popular discontent can be directed into courses directly hostile to
the authority of Government [and] the growing tendency for political controversies to assume a racial and nationalist complexion'.139 Similarly, J. C. Curry, a senior
police officer in Bombay, observed how the inherent volatility and barely suppressed turbulence of Indian society threatened to expose the weaknesses of the British
position in India. In India, various influences, 'racial, communal, criminal and traditional', fostered 'condi
138 Express Letter, Bombay Special to Home Simla, 4 June 1930, in ibid., pp. 27–9.
139 Secretary, GOB, Home, to Secretary, GOI, Army, 12 January 1921 in Bombay Confidential Proceedings, January 1921, vol. 62, p. 32, OIOC.

Page 218
140
tions of public instability'. This meant that the police 'must be ever ready to suppress ebullitions of violence'. The cotton mills of Bombay, in his view, 'never fail to
find the crowds which become mobs, and the mobs which turn to frenzied slaying'.141 It was imperative, therefore, for the police to establish control and impose order
upon the crowd at the first opportunity. In India, it seemed, it was essential for the police and the state to act quickly and decisively. Thus, 'the first and last rule of
conduct for all ranks', declared Curry, 'is to go straight for any trouble and deal with it before it becomes more serious. There can be no excuse for failing to learn of a
small cloud low on the horizon which . . . may quickly obscure the sky.'142 Yet dealing with the crowd required specialized knowledge and exceptional and particular
skills in India. For
popular disturbances can often be prevented from coming to a head by tact and foresight on the part of officers who understand the various diverse peoples, and know when to
use persuasion, how to appeal to the sense of honour innate in wellbred Indians, and when to hint at the presence of that power which, in the circumstances of India, must always
be available, even if undisplayed and in reserve.143
The implications of the inherent instability and volatility of Indian society extended beyond police practice to embrace the law and the powers of the state which were
necessary to contain it and, therefore, also circumscribe the freedom and autonomy of civil society. It was, after all, to perform precisely this essential function for its
subjects that Britain ruled India. Indeed, Curry continued:
These circumstances render it compellingly necessary for authority to be armed with wide powers for which no need exists in other more homogeneous countries . . . No
government, indigenous or foreign, autocratic or constitutional, can hope to administer such a country, unless its local officials have extensive legal powers and ultimately the
backing of unlimited force.144
The Amritsar massacre in 1919 showed precisely where this line of reasoning was liable to lead. For General Dyer justified his decision to open fire at Jallianwalla
Bagh in 1919 in terms which were familiar to the point of banality and, therefore, commanded an unquestioning authority within colonial discourse. Under his orders,
1,650 rounds of ammunition were fired for ten minutes by the army at a peaceful meeting, killing by the most modest calculation 379 people and wounding a further
1,200. He had been faced, he later told the Hunter Committee, with only two options: either to fulfil his 'very distasteful and horrible duty' of 'suppressing disorder' or
else risk 'becoming responsible for all future bloodshed'.
140
Curry, The Indian Police, p. 86.
141 Ibid., p. 99
142 Ibid.
143 Ibid., p. 86.
144 Ibid.

Page 219
In suppressing disorder, his task was not a matter of 'merely dispersing the crowd, but one of producing a sufficient moral effect, from a military point of view, not only
on those who were present but more specially throughout the Punjab'. Under these circumstances, 'If I fired, I must fire with good effect, a small amount of firing would
be an act of folly.' In the end, he had employed 'the least amount of firing which would produce the necessary moral and widespread effect it was my duty to produce'.
Therefore, the force he employed could not by its very nature be excessive; it had no economy. As he put it, 'There could be no question of undue severity.'145 In this
way, colonial representations of the 'mob' justified the use of repressive measures which could not be contemplated elsewhere. Yet if colonial discourse facilitated and
legitimized the most brutal repression, this did not mean that the state developed, let alone deployed, the means of coercion it may have deemed necessary. Indeed,
until its demise, the colonial state remained particularly aware of its vulnerability to the threat of being suddenly overwhelmed by mass discontent.
VI
It was, in any case, primarily the army and not the police which provided the muscle to move the coercive apparatus of the colonial state. To a large extent, public
order policing in India was conducted by the army. When officials perceived an imminent breakdown of public order, they summoned the army and deemed the
military presence indispensable. Official discourse had enunciated the principle at least since 1860, affirmed by the Eden Commission in 1879 and repeated ever since
that while 'the preservation of the public peace is the duty of the police', the function of the army was 'to suppress rebellion and to resist invasion'.146 Yet, of course, it
was difficult, if not impossible, to distinguish too neatly between disorder
145 Report of the Committee Appointed by the Government of India to Investigate the Disturbances in the Punjab etc. Parliamentary Papers, 1920, vol. XIV, Cmd. 681, p. 1116;
and Disturbances in the Punjab, Statement by BrigadierGeneral R. E. H. Dyer, PP, 1920, vol. XXXIV, Cmd. 771, especially pp. 686–90; see also D. Sayer, 'British Reactions to the
Amritsar Massacre, 1919–1920', Past and Present, 131 (May 1991), 130–64. Yet, in Britain certainly, the question of 'undue severity' posed intractable problems for the military when
it was called upon to provide aid to the civil authority. According to the Secretary of State for War in 1908, 'The law to my mind is clear that the soldier is in no different position
from anybody else . . . an excess of force and an excess of display ought not to be used. The soldier is guilty of an offence if he uses that excess.' See Report of the Select
Committee on the Employment of the Military in Cases of Disturbances, 1908, Examination of Rt. Hon. R. B. Haldane, PP, 1908, Cmd. 236, vol. VII, p. 387.
146 Secretary, GOI, Army to Chief Secretary, GOB, 19 October 1920, Home Department (Confidential) Proceedings, Bombay Confidential Proceedings, January 1921, vol. 62, p. 19, OIOC.

Page 220
and rebellion, let alone anticipate accurately when the former might develop into the latter. Moreover, for the British in India, the introduction of the police had
occurred at a time when 'the pacification of the country and the introduction of a stable civil order were proceeding concurrently'.147 As a result, no clear demarcation
of function between the army and the police had seemed necessary or, indeed, been effected. At no stage in its development had the police been organized on a scale
which presupposed that 'they should discharge the full extent of their duties',148 as they had been defined by the Eden Commission, 'of preserving order, of protecting
property and of quelling disturbances', let alone in the stricter terms of the requirement set out in the early 1920s, that they 'should be able to deal promptly and
effectively with local disturbances without invoking the aid of the military'.149 Indeed, as one official remarked in 1920, the ordinary, unarmed police were 'practically
useless in the case of really violent disturbance', while the armed police force was 'entirely inadequate to put down serious disturbances'.150 In practice, therefore, as
officials knew only too well, it was upon 'the military authorities that the duty falls of seeing that efficient arrangements are made for internal security' and, as a result, far
too frequently, 'regular troops were used for purely police purpose'.151 Whereas officials believed the use of the army 'to deal with domestic disturbances' was in
Britain and elsewhere in the empire, 'little more than a theoretical consideration', in colonial discourse, 'the case of India is entirely different'. In India, troops were
employed several times a year 'to prevent internal disorder and, if necessary, to quell it'.152
Although the troops were called out when the police felt overwhelmed, there was a general reluctance to use the military for police purposes. The extensive use of
troops was an expensive method of ensuring public order and in the face of tightening budgets, the army resented the fact that such use constituted a charge upon the
military estimates. The civilian employment of troops might diminish 'their general efficiency and fighting value' as a military force. 'It exposes them moreover to the
machinations of agitators', as the Government of India observed. In addition, 'it must be remembered that among British soldiers of today there are many to whom
military interference in trade disputes between employers and
147 Secretary, GOB, Home, to Secretary, GOI, Army, 12 January 1921, in ibid., p. 32.
148 Ibid.
149 Secretary, GOI, Army to Chief Secretary, GOB, 19 October 1920, in ibid., p. 19.
150 P. R. Cadell, Commissioner, Southern Division to Secretary, GOB, Home, 25 November 1920, in ibid., p. 28.
151 Secretary, GOI, Army to Chief Secretary, GOB, 19 October 1920, in ibid., p. 19.
152 Report of the Indian Statutory Commission, vol. I: Survey (London, 1930), p. 95.

Page 221
153 154
employed is particularly repugnant'. Yet, ironically, the Indian army was to become increasingly in the 1920s 'a force of screwguns and mules', now best suited,
perhaps, to the task of combating nonviolent satyagrahis and unarmed strikers. The frequent use of the army for public order policing between 1918 and the early
1920s, and especially perhaps the Amritsar massacre, led to a reevaluation of its role in the maintenance of the Raj. The repeated use of troops against civilians with
such brutal consequences could undermine its discipline and sap its strength.
If the army was to be used more sparingly, the obvious solution, in the early 1920s, was to strengthen the police and, in particular, its armed contingent. If the police
were to be expected to maintain public order without the assistance of the army, however, its size would have to be expanded beyond the capacity of the state's
coffers. The prospect of expanding the armed police brought a whole range of problems in its wake. If large forces of military police had to be stationed at what were
deemed 'troublespots', they would incur enormous cost for the relatively rare occasions on which they would be required. Not only would it be difficult 'to provide
sufficient men without an impossible burden of expenditure',155 but the creation of 'a large idle body like this would also be an administrative difficulty and danger'.156
The question of arming the police inevitably raised anxieties about the discipline of the force. If the mob were to disarm them, they could wreak havoc. The armed
police might prove mutinous or they might be partisan in a dispute. They might take sides in a communal riot; they might sympathize with strikers when they were
expected to discipline them; they might be persuaded by nationalist agitators to turn their guns on the government it was their duty to serve. The British most fully
trusted 'European' officers and the suggestion that they should be marshalled in a mobile force to quell disorder was believed to merit serious consideration. But while
the 'Europeanization' of the police force appealed to the primordial sentiments of some colonial officials, wiser counsels appreciated that it might have the effect of
'accentuating the racial question which is generally incidental to disturbances in present times' or 'of raising strong
153 Secretary, GOI, Army Department to Chief Secretary, GOB, 19 October 1920, Bombay Confidential Proceedings, Home Department, January 1921, vol. 62, OIOC.
154 J. Gallagher and A. Seal, 'Britain and India Between the Wars', MAS, 15:3 (1981), 403.
155 P. R. Cadell, Commissioner, Southern Division to Secretary, GOB, Home, 25 November 1920, in Bombay Confidential Proceedings, Home Department, January 1921, vol. 62, p. 28,
OIOC.
156 J. Ghosal, Acting Commissioner, Northern Division, to Secretary, GOB, Home, 17 November, 1920, in ibid., p. 25.

Page 222
157
popular prejudice against the proposed force'. Nor could a riot squad manned by Europeans be relied upon to be more effective. As one district magistrate
observed, Indian police officers of higher ranks had been 'conspicuous for devotion to duty, courage and coolness'. They were, he added, 'better fitted to deal with
mobs than such Europeans as would replace them because they know the people and their language more intimately and can gauge the consciousness of a situation
more accurately'.158
Some troops were ordinarily stationed in Bombay. Their presence was deemed to be essential 'to prevent the serious consequences which would follow from civil
disturbances' in what was after all a major metropolis and a political and commercial centre of vital imperial importance. These troops were sometimes alerted, even if
they were not always 'called out' when officials perceived the threat of 'disturbances'. While the police and the army were most vociferously pilloried for their brutality
during nationalist agitations, the deployment of the military was most readily justified in the case of communal riots. The often apparently random and unpredictable
violence which characterized periods of communal riots swiftly alarmed wide sections of the city's inhabitants, whatever the degree of their complicity in their
occurrence. They were quick to demand the use of troops to restore order and highly critical of both the government and the police for not calling out the army soon
enough. Troops were called to the aid of the civil authorities most often in the case of communal riots. Indeed, 'the vast majority of the disturbances which call for the
intervention of the military have a communal or religious complexion'.159 For 'police forces, admirably organized as they are, cannot be expected in all cases to cope
with the sudden and violent outburst of a mob driven frantic by religious frenzy'. So, in colonial discourse, the fact that Indians were not deemed to have fully entered
the universe of reason – for 'India is a country in which the wildest and most improbable stories of outrage and insult spread with amazing rapidity' – served to justify
the use of the military in civil disturbances, and more brutal repressive practices in general, which either could not be contemplated elsewhere or, at best, amounted to
'little more than a theoretical consideration'.160
Army units detailed for internal security duties consisted of a significant proportion of British troops. The Indian army in 1930 had a ratio of one British soldier for
every two and a half Indian soldiers, but among
157 Commissioner of Sind to Secretary, GOB, Home, 7 December 1920, in ibid., p. 30.
158 District Magistrate, Hyderabad, to Commissioner of Sind, 19 November 1920, in ibid., p. 31.
159 Report of the Indian Statutory Commission, vol. I, p. 95.
160 Ibid.

Page 223
troops allotted for internal security, 'a majority of British troops is employed', on average in the ratio of eight British soldiers to every seven Indians. In communal riots,
the demand, as the Indian Statutory Commission noted, was overwhelmingly for the use of British troops who could be seen to have 'no bias, real or suspected by
either side'.161 Yet even when they could be seen to be tainted with 'bias', their presence was rather reassuring to colonial officials. Thus, while the Government of
Bombay readily acknowledged in 1920 'the serious disadvantages' entailed in 'the employment of troops in industrial and political disturbances', it was precisely 'the
character of the political agitation at present being carried on, and the rapid development of the labour situation' which determined them to reject any attempt to restrict
their option to invoke military aid when they considered it necessary.162 If, in colonial discourse, no form of political action could be readily distinguished from sedition,
it was folly to assume that labour disputes could simply be taken for what they were – disputes between employers and their employees. 'It is well known', as colonial
officials often gravely told each other, 'how easily a mass of uneducated Indians ordinarily welldisposed towards Government can be excited against the authorities by
a few firebrands.' Thus, colonial officials took it for granted that 'any unrest even of an economic character can be easily directed against the forces of law and order'.
Moreover, in Bombay Presidency, it seemed that 'nearly all local disturbances . . . though often purporting to be connected with questions of pay and employment
have been engineered by outsiders . . . prominently connected with agitation against government'.163 In the context of the political turbulence witnessed in Bombay and
the city's vital strategic, political and economic importance to imperial interests, and by comparison with the brutality of the military repression effected elsewhere, what
is surprising is how rarely, not how extensively, troops were used. Although troops were kept in readiness during each of the general strikes, they were not by any
means called out as a matter of course. They were alerted and mobilized more frequently during the nationalist agitations of the early 1920s and early 1930s, but on
these occasions officials often appeared to invest the 'moral effect' of their proximity with greater significance than their active use. Nonetheless, between the two world
wars, it was reported that 'the use of the Army for the purpose of
161 Ibid.
162 Secretary, GOB, Home to Secretary, GOI, Army, 12 January 1921 in Bombay Confidential Proceedings, vol. 62, pp. 32–4, OIOC.
163 P. R. Cadell, Commissioner, Southern Division, to Secretary, GOB, Home, 25 November, 1920 in ibid., p. 27.

Page 224
164
maintaining or restoring internal order was increasing rather than diminishing'.
To some extent, the increasingly frequent deployment of the army to restore order reflected the growing scale of social conflict and political action. In addition,
organizational changes, arising in part from improved communications and the expansion of the railway network, made it possible to mobilize the army more easily. It
could now be concentrated and garrisoned in larger units rather than dispersed in small outposts and it could, therefore, be deployed more quickly, more efficiently and
even more flexibly. Thus, units of the field army could swiftly reinforce those assigned to internal security and fewer troops had to be tied down to the tasks of watch
and ward.165 Moreover, political reform and the devolution of ministerial power to elected politicians heightened British anxieties and contributed to their readiness to
perceive the collapse of order and to summon the army. Accordingly, during the successive rounds of constitutional reform in the late 1910s and the mid 1930s, the
British sought to maximize their control over the repressive apparatus of the colonial state, even as greater power was devolved to elected representatives.166 The army
remained suspicious, throughout this period of reform, that provincial governments invoked military aid more frequently and perhaps more often than was necessary,
largely because they were reluctant to spend adequately on the maintenance and expansion of their police forces. So it seemed to the generals that while Indian
politicians lost no opportunity to castigate the Government of India for its vast military expenditure, they took advantage of their growing power in the provinces to
divert expenditure away from the vital needs of the police and of public security. But if the police were underresourced, the army would have to be called out more
often. In this way, the actions of Indian elected representatives served to increase the demands on the military estimates which they were themselves pressing to cut. As
the repressive power of the colonial state was expanded, so the ranks of the police and the army not only continued to be staffed by Indians but their officer corps,
especially in the case of the latter, underwent a process, however slow and halting, of Indianization. More crucially, however, increasing state repression coincided with
increasing Indian control over the institutions and agencies of the state. In an important sense, the growing dependence upon the army for public order policing
reflected the tendency of local elites, as they gained in
164 Report of the Indian Statutory Commission, vol. I, p. 95.
165 Omissi, The Sepoy and the Raj, pp. 199–214; Narain, 'The Colonial Army in India, 1918–1947', ch. 6.
166 Arnold, 'The Armed Police'; Omissi, The Sepoy and the Raj, pp. 194–231; Narain, 'The Colonial Army in India', ch. 6; S. P. Cohen, The Indian Army: Its Contribution to the
Development of the Indian Nation (Berkeley, 1971).

Page 225
creasing influence over the local and provincial machinery of the state, to deploy its resources to discipline and control labour, whether in the towns or the countryside.
During the period of provincial autonomy, between 1937 and 1939, the Congress Ministry showed few inhibitions about taking repressive action and, sometimes, a
suspiciously greater willingness to force strikers back to their machines under the muzzle of a gun, in Bombay as well as in Ahmedabad and Sholapur, than the
preceding colonial regimes.167 'We had expected the Bombay Ministry to taboo the old bureaucratic violence', observed the Congress Socialist after the police had
opened fire on the workers during the oneday strike of 7 November 1938, 'Our expectation has not been fulfilled.'168
VII
The system of policing which the British created in the midnineteenth century grew out of the need to separate its function from the revenue administration, on the one
hand, and the military, on the other. The British had become increasingly aware that 'the preservation of the public peace' would require an agency which was less
cumbersome and more flexible than the army. If the British were to assert the legitimacy of their rule, especially after 1857, beyond the claims of conquest and
occupation, this agency would have to be more clearly subject to civil authority. In any case, the extensive and repeated use of the military for public order policing
entailed the risk of overly exposing soldiers, especially Indian troops, to situations of political and social conflict, which could compromise their loyalty and undermine
their discipline. In Britain, the police force had taken its modern shape in the midnineteenth century when the political order had been consolidated and rulingclass
anxieties came to be focused primarily upon the protection of private property.169 For the
167 On the Bombay Trade Disputes Bill sponsored by the Congress, see GOB, Home (Special), File 550 (25) IIIA of 1938; GOB, Home (Special), File 550 (24) of 1938; on the oneday
strike to protest against the bill, see Bombay Sentinel, 7 November 1938 and Times of India, 8 November 1938; Bombay Chronicle, 8 November 1938. See also C. Markovits, Indian
Business and Nationalist Politics, 1931–39 The Indigenous Capitalist Class and the Rise of the Congress Party (Cambridge, 1985), ch. 6; Sujata Patel, The Making of Industrial
Relations: The Ahmedabad Textile Industry, 1918–1939 (Delhi, 1987); on Madras, see E. D. Murphy, Unions in Conflict: A Comparative Study of Four South Indian Textile
Centres, 1918–1939 (New Delhi, 1981), pp. 148–219; on Jamshedpur, D. Simeon, The Politics of Labour under Late Colonialism: Workers, Unions and the State in Chota
Nagpur, 1928–1939 (New Delhi, 1995), chs. 7–9; V. Bahl, 'Attitude of the Indian National Congress Towards the Working Class Struggle in India, 1918–1947', in Kapil Kumar (ed.),
Congress and Classes: Nationalism, Workers and Peasants (Delhi, 1988), pp. 1–33.
168 Congress Socialist, 13 November 1938.
169 C. Emsley, Policing and Its Context, 1750–1870 (London, 1983); R. Reiner, The Politics of the Police (Brighton, 1985), ch. 1.

Page 226
British in India, however, the political order seemed irremediably fragile and its social base inherently volatile. The function of the police was primarily to safeguard
India's 'internal frontier' and to secure its political order rather than simply to detect or prevent crime. Yet ironically the police could not discharge this task without the
assistance of the army, which in turn was organized and deployed to defend 'the internal frontier': 'to hold India against the Indians', as Kitchener discovered in 1905,
rather than 'with regard to possible danger from outside'.170 As a distinct police system was separated out from the army and institutionalized, the role which was
delineated for it ensured that it never fully escaped its military origins.
This preoccupation with public order meant that the police left the prevention and detection of crime largely to be handled within local power structures. It also
determined the definition and ordering of crime. Thus, for instance, it was the British attempt to establish their sway over large tracts of a mobile and strifetorn
countryside which led to the invention of 'thuggee' in the 1830s.171 Similarly, the criminalization of particular tribes and castes arose out of the need to pin down
wandering cultivators, pastoralists, forest dwellers and itinerants, so that they could be taxed and policed more effectively.172 'Dacoity' became another catchall
category within which a wide range of 'crimes' could be lumped and diverse kinds of agrarian conflicts addressed.173 Riots were no more easily definable than dacoity
and no more easily counted. In fact, they were often mistaken for each other and each was sometimes subsumed within the other. Nonetheless, by the late nineteenth
century, in some rural districts, as the magis
170 Kitchener to Roberts, 12 January 1903, Kitchener Papers, 30/57/29, PRO cited by Omissi, The Sepoy and the Raj, p. 206.
171 See the excellent article by Stewart N Gordon, 'Scarf and Sword: Thugs, Marauders and State Formation in 18th Century Malwa', IESHR, 6:4 (1969), 403–29, especially 407–15; see
also Radhika Singha, 'Providential Circumstances: The Thuggee Campaign of the 1830s and Legal Innovation', MAS, 27:1 (1993), 83–146.
172 S. Nigam, 'Disciplining and Policing the Criminals by Birth, Part I: The Making of a Colonial Stereotype – the Criminal Tribes and Castes of North India', IESHR, 27:2 (1990), 131–64
and S. Nigam, 'Disciplining and Policing the Criminals by Birth, Part II: The Development of a Disciplinary System, 1871–1900', ibid., 27:3 (1990), 257–87; M. Radhakrishna, 'The Criminal
Tribes Act in the Madras Presidency: Implications for the Itinerant Trading Communities', ibid., 26:3 (1989), 269–95; and M. Radhakrishna, 'Surveillance and Settlements under the
Criminal Tribes Act in Madras', ibid., 29:2 (1992), 171–98.
173 However, see D. Arnold, 'Dacoity and Rural Crime in Madras, 1860–1940', Journal of Peasant Studies, 6:2 (1979), 140–67. For all the problems of arbitrary classification, Arnold
concludes that dacoity can be taken to constitute a real, discrete and measurable activity rather than a function of the perceptions of the colonial rulers and their agencies of law
enforcement. Thus, he notes, 'Although the crime statistics published by the police from 1863 were doubtless inaccurate . . . there is no reason to believe that the figures fail to show
the general pattern of the incidence of dacoity.' Ibid., p. 143.

Page 227
174
trate of Saran noted, 'the most important offence with which the police have to deal . . . [is] rioting'. In this instance, the police appear to have been managing
disputes over land and rural resources between rival land controllers rather than the actions of the dispossessed against the dominant elites.175 For the colonial state
largely left rural magnates free to discipline and control their dependants and labourers, when they could, even if they were willing to bolster the efforts of rural elites
when this proved necessary. Indeed, by the late nineteenth century, the criminalization of pastoralists, nomads and wandering groups may be seen to have extended to
landholding elites and prosperous tenants the means and opportunities to exert considerably closer control over labour.176
This pattern of policing was produced by a force which, as Christopher Baker has observed, 'was useful in the rural areas only in an emergency and even then with
mixed results'.177 Yet in the cities, and Bombay was no exception, the police lacked resources, discipline and sometimes even manpower. To a very large extent, both
in the towns and the countryside, Indian society relied upon informal systems of policing and conflict resolution. The structural and organizational weaknesses of the
police in Bombay forced them to operate through alliances, networks and connections made within the neighbourhoods. Sometimes, these were alliances forged
between the police and dominant elements within the neighbourhood. Sometimes, they were personal and kinship connections which could be put to use by the humble
and powerless. The working classes could, therefore, sometimes use their connections with the force, however individual and personalized, to obtain protection, secure
advantage or resolve conflicts. At a daily level, therefore, the police did not act monolithically as a force external to the working classes as a whole. Moreover, the
rank and file of the police were sometimes responsive to social pressures within the neighbourhoods, irrespective of the internal structures of command within the force.
It is not surprising, perhaps, that superior officers of the police remained concerned about the state of discipline among the subordinate staff. On the other hand, they
also put their faith devoutly in the value of these street and neighbourhood connections for the detection of crime. In fact, their belief was belied by the
174 'Annual Crime Report for 1882', no. 346, 14 February 1883, Saran Collectorate, Faujdari Basta, cited by A. Yang, 'The Agrarian Origins of Crime: A Study of Riots in Saran
District, India, 1886–1920', Journal of Social History, 13:2 (1979), 292.
175 Yang, 'Agrarian Origins'; see also S. Freitag, 'Crime in the Social Order of Colonial North India', MAS, 25:2 (1991), 227–61; and S. Freitag, 'Collective Crime and Authority in North
India', in A. Yang (ed.), Crime and Criminality in British India (Tucson, Ariz., 1985), pp. 140–63.
176 Freitag, 'Crime in the Social Order'; Nigam, 'Disciplining and Policing the Criminals by Birth', Parts I and II.

Page 228
criminal statistics. For the low rates of recorded crime suggest both a reluctance on the part of residents to report them and on the part of the police to record them.
This reluctance only serves to afforce the suggestion that the prevention and detection of crime, as well as the meting out of punishment, remained throughout this
period largely the preserve of the social networks of the neighbourhood.
The constraints under which the police functioned meant that they were only likely to be effective if they marginalized and concentrated upon selected targets. In this
process of selection, they would necessarily have to rely upon a general consensus about which groups in society were especially prone to criminal activity and might
constitute, therefore, the proper objects of policing. In the context of colonial India, this consensus proved particularly difficult to achieve. Of course, by enacting this
principle of selection, the colonial state was able to create criminal tribes and castes. But while it might police them energetically, the criminal tribes were scarcely, by
the late nineteenth century, a potent threat to the social order. To marginalize more substantial social groups was an intractable and often hazardous proposition. In
particular localities and regions, it was sometimes possible to find a consensus which isolated and marginalized some social groups. But social groups which were
marginalized in one area might be harder to isolate in another; indeed, they might even exercise some influence on the strategies for policing the public order. Among
those who laid claims to respectability and sought to distance themselves from the rough, there were many for whom the experience of selective policing was never too
remote. With the development of nationalist politics, a greater variety of people found a possible consensus about the strategies of policing and the definition of its
proper objects unconvincing and untenable, in the light of their own experience, real or imagined, of the state. The intensification of communal conflict only made this
consensus more difficult to achieve. By generating and developing a political vocabulary of sectarianism, the colonial state in fact foreclosed the possibility of creating a
viable language of order. This was why it often found it difficult to satisfy the logic of its position by freely expressing its coercive disposition. The function of crime for
the capitalist industrial state, we are sometimes told, was to regulate society more closely and to discipline and punish the working classes. The colonial state, however,
was thrown back upon the more slippery alliances and political contingencies, in particular shifting alliances with its collaborators, to underwrite the social order.
The exigencies of public order policing, and especially situations of crowd control, exposed the weaknesses of the police and, indeed, laid bare their own sense of
vulnerability. It was an essential characteristic of

Page 229
the Indian crowd, in colonial discourse, that unless it was quickly dispersed it would rapidly get out of control. Frequently, therefore, they tended to perceive the
potential for crowd violence too readily and invoke military assistance. Yet, for all their sense of weakness, the police were sometimes ruthless and even brutal in their
response to the crowd. In part, this use of excessive force may be seen as a symptom of panic, arising out of their awareness of their own weakness. But it also
reflected what might be described as the statecraft of the police and was integral to their approach to the problems of public order. Assumptions about the cultural
peculiarities of Indian society suggested that it could only properly be managed by methods, codes and knowledge which were specific to its culture. So the detection
of crime acquired, in this context, a new and unusual meaning. Crime became an important part of the endeavour of colonial anthropology. Informers and spies became
essential to unravelling the hidden meanings of the East. Policemen, civil servants and magistrates developed out of this quest for the culturally specific a whole
ethnography of evil. While in reality crime went largely unreported and unrecorded, police reports and memoirs identified and described in painstaking detail crimes of
savage brutality or extraordinary guile and cunning or those which reflected exotic customs and elaborate rituals. Of course, this was particularly the case with 'thuggee'
and the criminal tribes and castes, whose supposed criminality was represented as an inheritance and a profession, inextricably connected to their lineage and
genealogy;178 but this search for the culturally specific infused the tone and language of the police reports in Bombay city as well. The mysteries of the Orient extended
to criminal behaviour and demanded that colonial rulers fashioned culturally specific measures to control them.
Similarly, colonial discourse about popular politics began with the assumption that the populace lived beyond the realm of reason. Rumours, malicious information,
skilful instigation were always liable to push a fanatical people over the edge of violence. This propensity only heightened the need for swift, firm and effective action to
prevent a minor affray turning into a major riot or a local disturbance into widespread and uncontrollable rebellion. The speed of response by the police and the army
always seemed crucial. Moreover, it was deemed that the efficacy of the police and the army in any single situation would determine its 'moral effect' upon potential
'miscreants' elsewhere. Colonial officials were often concerned that without the option of using extralegal compulsion and unlimited force they could not be certain of
maintaining order. Yet the imperative for immediate and effective action to impose control
178 See, especially, Nigam, 'Disciplining and Policing the Criminals by Birth', Parts I and II; Gordon, 'Scarf and Sword'.

Page 230
tended to obscure the distinction between minimum and maximum force. In addition, the social chasm which divided commanding officers of the police and the army
from the crowds they sought to control also contributed towards this tendency to overreact. Frequently, they found it difficult to judge a situation accurately and tended
to perceive the potential for violence rather too hastily.
The notion that 'the mass of uneducated' Indians could be easily worked into 'a state of excitement' which, because it pushed them outside the domain of reason,
rendered them impossible to control, was compounded by another aspect of the way colonial rulers represented themselves and their subjects. The legitimacy of
British rule in India was founded in this discourse upon the unfitness of Indians to govern themselves. While the purpose of their rule was, they claimed, to bring
enlightenment and civilization to areas of darkness and to prepare Indians for eventual selfgovernment, it was also incumbent upon the British, while educating their
subjects, to discipline them with firmness and justice. Paternalist benevolence was inseparable from necessary punishment which, happily, in any case, redounded to
the good of their subjects. Of course, paternalism had to be enlightened and from its lofty standpoint, the authoritarian excesses of colonial agents were harshly judged.
Just as the moral progress of empire could be set against the crass notion of economic imperialism, so moral outrage at the lapses of colonial agents camouflaged the
exploitative purposes of imperialism.179 In this way, colonial discourse insisted not only upon the use of excessive force but deemed it essential for 'moral effect' and
necessary for the good of its subjects. In this light, it becomes possible to see why the police and the army, for all their apparent weaknesses, sometimes responded to
revolting and disorderly subjects with a ruthless and murderous brutality. It also becomes clear that Dyer's actions at Amritsar were not, in Churchill's characterization,
'an episode without precedent or parallel in the modern history of the British Empire . . . a monstrous event, an event which stands in sinister and singular isolation'180
but, on the contrary, it was integral to and the outcome of the colonial conceptualization of and response to the problems of public order.
It is sometimes suggested that the autocratic nature of the colonial state in India was tempered by the liberal and democratic political culture of Britain. This is a
romantic view and should be regarded with scepticism.
179 For an excellent delineation of this discourse, see R. Robinson, 'Oxford in Imperial Historiography', in F. Madden and D. K. Fieldhouse (eds.), Oxford and the Idea of the
Commonwealth: Essays Presented to Sir Edgar Williams (London, 1982), pp. 30–48.
180 Quoted in Omissi, The Sepoy and the Raj, p. 218; see especially, Sayer, 'British Reactions to the Amritsar Massacre'.

Page 231
181
For some social groups in Britain, like the casual poor and the Irish, the state must at times have appeared extremely authoritarian and oppressive. Nor were the
activities of troops and police in the Welsh coal fields in the early twentieth century or the Black and Tans in Ireland redolent of a political culture which honoured civil
liberty above all else.182 Similarly, the British state was able to respond to strikes with a ferocity it did not often attempt in India. When Clydeside workers struck in
January 1919, 'infantry and tanks were rushed in, and machine gun posts set up in the City Chambers'. During the police strike in Liverpool in the same year, 'the army
was sent in with bayonets fixed to keep the peace and a battleship was posted to the Mersey, no doubt to cow the masses'.183 The scale of military mobilization during
the general strike of 1926 in Britain, 'when the government responded with the full panoply of law enforcement, with police and special constables backed up by
battalions of troops, many warships, and marines and thousands of civilian volunteers drawn from the loyalist middle class'184 was never, in the course of a trade
dispute, matched in India.185
The British in India showed no aversion to deploying their armed force with destructive effect when they believed that their position was under threat and when they
feared the repercussions of their repression less than the consequences of disorder. The most violent campaigns waged by the army carried a large punitive, even
retributive, content, effected as an example and a lesson to their subjects: in the aftermath of the Mutiny of 1857;186 Jallianwalla Bagh and the repression which
accompanied the Rowlatt satyagraha in the Punjab;187 the suppression of the 'fanatical'
181 Gatrell, 'Crime, Authority and the Policeman State', pp. 243–310; J. Davis, 'Jennings' Buildings and the Royal Borough: The Construction of an Underclass in midVictorian
England', in D. Feldman and G. StedmanJones (eds.), Metropolis: London, Histories and Representations since 1800 (London, 1989), pp. 21–40.
182 P. Townshend, The British Campaign in Ireland, 1919–1921: The Development of Political and Military Policies (Oxford, 1975).
183 K. Jeffrey and P. Hennessey, States of Emergency: British Governments and Strikebreaking since 1919 (London, 1983), pp. 11–13.
184 Jane Morgan, Conflict and Order: The Police and Labour Disputes in England and Wales, 1900–1939 (Oxford, 1987), p. 25.
185 The British, however, mobilized massive military power during the Quit India movement in 1942.
186 W. H. Russell, My Diary in India in the Year 1858–59, 2 vols. (London, 1860); J. W. Kaye, A History of the Sepoy War in India, 1857–58, 3 vols. (London, 1864, 1870 and 1876); S.
N. Sen. Eighteen Fifty Seven (New Delhi, 1957); P. C. Joshi (ed.), Rebellion – 1857: A Symposium (Delhi, 1957).
187 Report of the Committee Appointed by the Government of India to Investigate the Disturbances in the Punjab etc. PP, 1920, Cmd. 681, vol. XIV; Punjab Disturbances, 1919–20, 2
vols., Report of the Commissioners Appointed by the Punjab SubCommittee of the Indian National Congress to Look into the Jallianwalla Bagh Massacre (Delhi, 1976); Sayer,
'British Reactions to the Amritsar Massacre'; V. N. Datta, Jallianwalla Bagh (Ludhiana, 1969).

Page 232
188
Mapillas in the early 1920s, which may have resulted in some ten thousand deaths; the sustained military initiatives, supposedly against terrorism in Bengal in the
early 1930s;189 and the Quit India movement in 1942, when British military power was at its greatest and its mobilization for the maintenance of internal security was,
indeed, formidable.190 These military campaigns, part punitive, part repressive, were staged largely in areas where government was least firmly entrenched and the
British were least confident of the stability of their collaborative base: Punjab, Malabar, Medinapur, eastern UP and north Bihar. In Bombay city, where British rule
was more securely based, the colonial state did not embark on a sustained and generalized campaign of repression against their opponents in the same fashion.
Nonetheless, it is significant that despite the strength of their collaborative alliances and the entrenchment of their rule, the police and the army often resorted to violence
to preserve the public peace at a considerable cost to human life.
The imperial purpose in India was to deploy its resources in Britain's global interests, neither simply nor necessarily to maintain colonial rule. The time might come when
dominion would have to be sacrificed to these larger imperial priorities. The fact that India was ruled in the global interests of British imperialism imposed a selfdenying
ordinance upon its Oriental Despotism. British rule had to preserve the public peace to enable the mobilization of Indian resources for its imperial purposes. But it
could not secure its rule and mobilize resources without the collaboration of powerful Indian interests. So it was essential for the British to invest in and honour existing
configurations of power in Indian society while trying to control, in the last resort, those domains of power which
188 K. N. Panikkar, Against Lord and State: Religion and Peasant Uprisings in Malabar, 1836–1921 (Delhi, 1989), p. 163, see also pp. 139–90; S. F. Dale, Islamic Society on the
South Asian Frontier: The Mappilas of Malabar, 1498–1922 (Oxford, 1980), pp. 179–218; C. Wood, 'Peasant Revolt: Interpretation of Moplah Violence in the Nineteenth and
Twentieth Centuries', in C. J. Dewey and A. G. Hopkins (eds.), The Imperial Impact: Studies in the Economic History of Africa and India (London, 1978), pp. 132–51.
189 GOI, Terrorism in India, 1917–1936 (Simla, 1937); T. Sarkar, Bengal 1928–34: The Politics of Protest (Delhi, 1987); see also, for the wider political context of Bengal, Joya
Chatterji, Bengal Divided: Hindu Communalism and Partition, 1932–47 (Cambridge, 1994); for details of the military campaign in Bengal, see Omissi, The Sepoy and the Raj, pp. 223–
5; and Narain, 'The Colonial Army in India', ch. 6.
190 GOI, Home Poll, File 3/15/43; GOI, Home (Poll), File 3/52/43, NAI; P. N. Chopra (ed.), Quit India Movement: British Secret Report (Faridabad, 1976); F. Hutchins, India's Revolution:
Gandhi and the Quit India Movement (Cambridge, Mass., 1973), especially, chs. 7 and 9; A. C. Bhuyan, The Quit India Movement: The Second World War and Indian Nationalism
(Delhi, 1975), especially pp. 93–102; Narain, 'The Colonial Army in India', ch. 6; M. Harcourt, 'Kisan Populism and Revolution in Rural India: The 1942 Disturbances in Bihar and East
United Provinces', in D. A. Low (ed.), Congress and the Raj: Facets of the Indian Struggle, 1917–1947 (London, 1977), pp. 315–48; G. Pandey (ed.), The Indian Nation in 1942
(Calcutta, 1988).

Page 233
most directly served these global interests. The pattern of collaboration which allowed the British to build up a formidable armoury of repressive force in India also
prevented them from using it indiscriminately against their political opponents. Moreover, these patterns of collaboration were sufficiently varied and complex that their
apparatus of repression could not simply be laced at the disposal of Indian magnates. If the British were to leave themselves free to deploy Indian resources to their
global and imperial purposes, they could not afford to become the prisoners of their collaborators. The Quit India movement represented the most potent threat to
British rule, at least since 1857. But the British in 1942 commanded sufficient military force in the subcontinent to snuff out the movement in a month. It might appear
that the British were at the zenith of their power in India and their armed forces could be decisively used to secure their rule. However, the prospect of a replay under
the changed conditions after the second world war led the British to the conclusion that they would have to fundamentally alter the basis of their rule, and adjustments
to their position in India after 1945 pointed them in the direction of dismantling their empire.
So here lay one of the abiding contradictions of the colonial state. In the late nineteenth century, the British had recognized that the only solution to the problem of
order was to leave it to the disciplinary mechanisms of local structures of power. Increasingly, after the 1920s, this technique of rule seemed constantly on the verge of
breakdown. Colonial rulers who witnessed the growth of popular politics as the disorderly effusions of the ignorant and the fanatical, easily roused by disaffected
publicists, were increasingly driven by the imperative to restore public order at the earliest opportunity, indeed with immediate and moral effect. Under the weight of
this imperative, the distinction between the use of minimum and maximum force was obliterated. Yet the survival of Britain's autocratic rule in India and the fulfilment of
its imperial purposes imposed upon its guardians a forbidding economy of repression, which was somewhat more likely to be observed in the breach.

Page 234
7—
Plague Panic and Epidemic Politics in India, 1896–1914
Between 1896 and 1914, bubonic plague killed over eight million people,1 a modest estimate which does not allow for cases which were concealed, misdiagnosed or
wrongly classified. Of all the various epidemics which afflicted India in the late nineteenth and early twentieth centuries, a Kaliyuga, a period of very high mortality,
stagnant, even falling population and declining life expectancy,2 the plague was not the most destructive. Malaria and tuberculosis killed more than twice as many
people over a similar period;3 in barely four months, the influenza epidemic of 1918–19 accounted for twice as many;4 smallpox and cholera counted their death toll in
millions.5 Yet no other epidemic evoked the fear and panic generated by the plague.
The plague epidemic prompted massive state intervention to control its spread. It also sometimes provoked fierce resistance, riots, occasionally mob attacks on
Europeans and even the assassination of British officials. The vigorous and energetic intervention of the state, in itself prompted by the general panic, bore no direct
relation to the virulence of the epidemic. The focus of the state's most vigorous measures was Bombay city and its
1 Annual Reports on Sanitary Measures in India, PP, passim; R. Pollitzer, Plague (Geneva, 1954), p. 26; L. Fabian Hirst, The Conquest of Plague. A Study of the Evolution of
Epidemiology (Oxford, 1953), pp. 296–301.
2 The population of Bombay Presidency declined in three of the five decennial periods between 1871 and 1921. See L. Visaria and P. Visaria, 'Population, 1757–1947', in D. Kumar (ed.),
CEHI, vol. XI: 1757–1970 (Cambridge, 1982), pp. 463–532, especially Tables 5.7 and 5.12; Kingsley Davis, The Population of India and Pakistan (Princeton, N.J., 1951), pp. 33–66,
especially, on life expectancy, Tables 16 and 17, p. 63.
3 I. Klein, 'Death in India', JAS, 32:4 (1973), 642–3; I. Klein, 'Malaria and Mortality in Bengal', IESHR, 9:2 (1972), 132–60; Davis, Population, pp. 53–7.
4 I. D. Mills, 'Influenza in India during 1918–19', IESHR, 23:1 (1986), 1–40; Davis, Population, Appendix B, p. 237.
5 D. Arnold, 'Smallpox and Colonial Medicine in Nineteenth Century India', in D. Arnold (ed.), Imperial Medicine and Indigenous Societies (Manchester, 1988), pp. 45–65; D. Arnold,
'Cholera Mortality in British India, 1817–1947', in T. Dyson (ed.), India's Historical Demography. Studies in Famine, Disease and Society (London, 1989), pp. 261–84; D. Arnold,
'Cholera and Colonialism in British India', Past and Present, 113 (1986), 118–51.

Page 235
Presidency between 1896 and about 1902. But plague mortality continued to rise thereafter, reached its peak between 1903 and 1907, exceeding the levels of the late
1890s by twelvefold, and proved far more lethal in the Punjab. Yet neither plague policy nor plague riots in the Punjab appear to have displayed the zeal or acquired
the political prominence they achieved in Bombay.6
Underlying the nature of this response is a further paradox. The late nineteenth century was a period of enormous selfconfidence in medical science and, particularly,
in its newly founded and burgeoning branch of tropical medicine.7 The plague epidemic in India became the occasion for the most intensive international research on
bubonic plague and its eventual findings largely laid the foundations for, indeed established much of, what is now known about the disease.8 Yet the measures adopted
by the colonial state at the turn of the twentieth century remained highly reminiscent of the Black Death or the epidemic in seventeenthcentury England.9
The intensity of the panic which gripped colonial officials and humble subjects alike suggests the need to examine how the epidemic came to be constructed. Studies of
epidemics in India have ranged widely, examining their demographic implications, or tracing their course and their attendant social and political effects. They have
examined the administrative problems posed by epidemics,10 the conflicts and rivalries which they opened up within government,11 how the state and elites exploited
the chaos they caused to extend their control,12 how political factions within
6 Over 354,000 died of the plague between 1896 and 1900. This figure rose to slightly under 4.5 million for the period 1903–7. Between 1903 and 1907, Punjab alone accounted for
between one quarter and one half of the plague deaths each year. Annual Reports on Sanitary Measures in India, PP, passim.
7 M. Warboys, 'The Emergence of Tropical Medicine: A Study in the Establishment of a Scientific Specialty', in G. Lemaine, R. Macleod, M. Mulkay and P. Weigart (eds.), Perspectives
on the Emergence of Scientific Disciplines (The Hague, 1976), pp. 75–98; H. H. Scott, A History of Tropical Medicine, 2 vols. (London, 1939).
8 Hirst, Conquest of Plague; I. J. Catanach, 'Plague and the Tensions of Empire, 1896–1918', in Arnold (ed.), Imperial Medicine, pp. 149–71; Scott, Tropical Medicine, vol. II, pp. 702–
67.
9
P. Slack, 'The Response to Plague in Early Modern England: Public Policies and their Consequences', in J. Walter and R. Schofield (eds.), Famine, Disease and the Social Order in
Early Modern Society (Cambridge, 1989), pp. 167–87.
10 Arnold, 'Cholera and Colonialism'; Arnold, 'Smallpox and Colonial Medicine'; D. Arnold, 'Touching the Body: Perspectives on the Indian Plague, 1896–1900', in R. Guha (ed.),
Subaltern Studies, vol. V, pp. 55–90; I. J. Catanach, 'Plague and the Indian Village, 1896–1914', in P. Robb (ed.), Rural India. Land, Power and Society under British Rule (London,
1983), pp. 216–43; Catanach, 'Plague and the Tensions of Empire'; I. Klein, 'Urban Development and Death: Bombay City, 1870–1914', MAS, 20:4 (1986), 725–54; Klein, 'Plague, Policy
and Popular Unrest in British India', MAS, 22:4 (1988), 723–55.
11 See, especially, Catanach, 'Plague and the Tensions of Empire'.
12 Arnold, 'Touching the Body': Arnold, 'Cholera and Colonialism'.

Page 236
13 14
the Congress arranged themselves around the event and even the 'indigenous' cultural response to Western science and colonial policies. These studies have been
illuminating but they have often proceeded piecemeal. This chapter is predicated on the assumption that epidemics do not represent a single, integrated phenomenon
but signify different things to different people. It argues that their inherent interest lies less in the discrete events which occur in their wake, than in the manner of their
construction. The historical process of their construction not only illuminates wider relationships between social groups and between state and society, but it can also
be argued that the constituent events of an epidemic upon which historians focus might be grasped most firmly when they are acknowledged to be, separately and
discretely, a function of the very process of its construction. This chapter will, therefore, try to explore how the plague epidemic was put together and the interplay
between some of the numerous elements which made it up: colonial perceptions of Indian society, medical and scientific rivalries, the interaction between the plague
administrators and the people upon whom their attentions focused. These perceptions, ideologies and political processes were not created for the first time by the
plague epidemic; they were indeed endemic to colonial India. However, the significance of these relationships, and of the tensions and antagonisms they generated
during the epidemic, lies in the ways in which their interplay helped to construct the panic of the late 1890s.
The policies of the state and the popular response to them have been frequently portrayed in terms of the inexorable conflict between 'Western antiplague measures
and popular culture',15 the unavoidable clash of 'two different, often antagonistic valuesystems, the one Indian, the other European'.16 Indian responses, it is suggested,
served as 'a reminder of the great cultural gulf which divided the colonizers and the colonized'.17 There was, we are told, an 'Indian view of disease and its treatment'
which called for 'family involvement and religious ministration, not secular segregation'.18 I will be arguing, however, that there was neither a uniform nor a
homogeneous, culturally specific Indian response, and further that the response of the populace, like that of the state, was integral to, and the product of, the
generalized panic from which none escaped. While antipathy to the plague measures was often shared by disparate social groups who might have nothing else in
common, it arose
13 I. J. Catanach, 'Poona Politicians and the Plague', in J. Masselos (ed.), Struggling and Ruling. The Indian National Congress, 1885–1985 (New Delhi, 1987), pp. 198–215.
14 Arnold, 'Touching the Body'; Arnold, 'Cholera and Colonialism'; and Klein, 'Plague, Policy and Popular Unrest'.
15 Klein, 'Plague, Policy and Popular Unrest', 739.
16 Arnold, 'Cholera and Colonialism', 119.
17 Ibid., 134.
18 Ibid., 137.

Page 237
less from the stirrings of an autonomous realm of popular culture than from the political conjuncture in which the plague was constructed.
On 23 September 1896, Dr A. C. Viegas, medical practitioner and local politician, declared before the Bombay Municipal Corporation, and thereby to the world, that
bubonic plague had broken out in the city.19 Thus, the plague acquired the character of an epidemic. The colonial state was, at first, reluctant to lend its authority to a
panic which might cripple trade and threaten the social order. The Health Officer conceded that 'the peculiar type of fever referred to by Dr Viegas . . . was of a
suspicious character' and 'appeared to be in some respects of a bubonic character'.20 In the following weeks, as the death toll mounted, officials referred to the disease
as the 'fever plague' or 'bubonic fever' but never 'bubonic plague'.21 It was only in early October when the Viceroy provided London with confirmation of 'true bubonic
plague'22 that the epidemic finally received official sanction.
But the mere suspicion of the plague was always likely to wrest the situation out of their control. Viegas's announcement had set the alarm bells ringing across the
world. The Calcutta Corporation, fearing that Bombay's fate today might be Bengal's tomorrow, began to discuss precautionary measures on the following day.23 By
early October, ports in SouthEast Asia, the Persian Gulf and East Africa had applied plague regulations to all vessels arriving from Bombay.24 The French feared the
prospect of riots at Marseilles25 and pressures mounted for 'restrictive measures to prevent passage of [sic] plague to Europe'.26
Once the colonial state was compelled to act, it initiated a vigorous, indeed draconian, programme of measures. The plague measures, as W. C. Rand described them
from Poona, 'were perhaps the most drastic that had ever been taken to stamp out an epidemic'.27 They were, indeed, to
19 Times of India, 24 September 1896, pp. 4, 5.
20 Ibid., p. 5.
21 Hirst, Conquest of Plague, p. 77.
22 Viceroy to Secretary of State for India, 2 October 1896, 'Papers Relating to the Outbreak of Bubonic Plague in India with Statement Showing the Quarantine and Other Restrictions
Recently Placed upon Indian Trade, up to March 1897', p. 3, PP, 1897, vol. LXIII.
23 Times of India, 25 September 1896, 5.
24 GOB, General, vol. 131, Compilation no. 178, Part I, and vol. 132, Compilation no. 178, Part II, MSA.
25 Catanach, 'Plague and the Tensions of Empire', pp. 151–2.
26 Secretary of State for India to GOB, 18 January 1897, 'Papers Relating to the Outbreak of Bubonic Plague', PP, 1897, vol. LXIII, p. 55.
27 Draft of Report to Government of Bombay by the Late Mr W. C. Rand, I.C.S., Chairman, Poona Plague Committee (n.p., n.d [1897?]), p. 3, MSA.

Page 238
cost Rand his life. Their primary objective was to identify and isolate the sick, remove them swiftly to hospital and segregate their contacts in 'health camps'. Their
houses were to be disinfected, their floors dug up, for they were believed to harbour the offending microbes, and their personal effects fumigated and sometimes burnt.
Physicians and families were required by law to notify all cases of sickness and 'all cases of fever were treated as ''suspects".'28 The sick were to be isolated in
hospitals, where most died, and their relatives segregated in special camps. Corpses had to be compulsorily inspected before disposal and the procedures for the
registration of deaths were tightened. Vigorous steps were taken to disinfect the whole environment. Houses and gullies 'under any suspicion were at once flushed and
disinfected'; buildings were limewashed or cleansed, usually at the owners' expense; and the drains and sewers were flushed every day with 3 million gallons of a
dilution of carbolic acid and sea water. The city was literally drenched in disinfectant solution. Hankin, a bacteriologist with the Plague Research Committee, reported
that he 'had to put up an umbrella before entering some plague houses in order to protect himself against the deluge of carbolic acid solution descending from the upper
stories into which the disinfectant was pumped by a fire engine for spraying on the walls and floors of the premises'.29 In addition, a system of surveillance was
instituted to examine cargo and passengers at the ports,30 and inspection checkpoints and detention camps became a common feature of the railway system as the
movement of people into or out of the city was severely restricted. As plague began to spread, similar measures were extended across a wider area and were
sometimes enforced by British soldiers.
This forceful and aggressive intrusion of the colonial state into the private domain was not simply dramatic and brutal but also novel and unprecedented. Its effect was
often to intensify and quicken the panic occasioned by the disease. As the epidemic spread across the subcontinent, the severity of the plague measures was matched
only by their desperation and it is not surprising that they met with fierce, if sporadic, resistance.
Then, almost as suddenly as the disease had appeared, the stringency of the plague measures was relaxed. In October 1906, the Times of India observed that no
notice was taken of the revival and increase of plague in Poona, 'whereas a few years ago it would have caused a panic'.31 This was
28 The Bombay Plague, Being A History of the Progress of Plague in the Bombay Presidency from September 1896 to June 1899, compiled by Capt. J. K. Condon (Bombay, 1900),
p. 125.
30 Port Health Officer quoted in Condon, The Bombay Plague, p. 136.
31 Times of India, 13 October 1906, quoted in Catanach, 'Plague and the Tensions of Empire', p. 169 fn. 82.

Page 239
not simply a reaction to popular hostility and violence. Although plague riots had often been heeded by the state as a warning that it should proceed with caution, the
recrudescence of the epidemic frequently 'compelled us to attempt more drastic and comprehensive measures'.32 Nor can this change of heart be attributed to the
decreasing virulence of the epidemic since plague mortality rose as the panic subsided.
It was perhaps more important to the changing direction of plague policy that the disease had begun to acquire a specific social character. If, at the outset, Viegas had
failed to find any poor people among the afflicted, it became increasingly obvious, as the epidemic worked its way through the subcontinent, that 'the poorer classes
suffered most severely'.33 In the memorable phrase of SurgeonGeneral Harvey, DirectorGeneral of the Indian Medical Department, bubonic plague was 'a disease of
filth, a disease of dirt, and a disease of poverty'.34 As the disease took on the character of a plague of the poor, it came to be seen as endemic. 'We Europeans are
indifferent', declared one of their number as the plague revived yet again, in its annual cycle, in March 1902: 'for the statistics show that fewer Europeans have died
from plague than die each year from cholera, so we can chance plague as we chance cholera.'35 Bubonic plague had become simply another disease in the formidable
pantheon of plagues which flourished in India's malignant climate and integral to the burden which the white man carried dutifully.
II
The pattern of state intervention in the plague epidemic was unique in the history of colonial India. It cannot be taken to exemplify 'the interventionist ambitions and
capacity of India's mature colonial state'.36 The scale and consistency of state intervention, as it entered homes, meddled with caste and religious practices, regulated
the disposal of the dead and restricted the free movement of people, was unprecedented. And the colonial state would never again orchestrate such a penetrative
programme of government, intrude so remorselessly upon the private domain, or attempt to exert such ambitious and extensive measures of social control. The frenzied
zeal with which the colonial state launched itself at
32 Lord Sandhurst and others, GOB, Judicial, to Her Majesty's Principal Private Secretary of State for India in Council, London, 9 April 1898, in GOB, General (Plague), vol. 389,
Compilation no. 298, 1898, MSA.
33 Condon, The Bombay Plague, p. 132.
34 Quoted in the Report of the Indian Plague Commission, 1898–99, with Appendices and Summary, vol. V, p. 170, PP, 1902, vol. LXXII.
35 Pioneer, 15 March, 1902, quoted in Catanach, 'Plague and the Tensions of Empire', p. 159.
36 Arnold, 'Touching the Body', p. 56.

Page 240
its subject was integral, as both symptom and cause, to the panic which accompanied the epidemic. It played a determining, perhaps creative, role in the political
construction of the plague epidemic.
The severity and desperation of the government's response may be explained partly in terms of folk memories of the Black Death and partly in terms of the possible
imperial consequences of the plague. An international embargo on Indian shipping not only threatened to close an important market and source of raw materials for
Britain but also disturb the intricate system for the multilateral settlement of its balance of payments, in which India played a large and vital part.37 The extensive
commercial and financial connections centred on Bombay made it seem highly improbable that the epidemic could be contained within the city.38 If it spread through
the subcontinent, it might devastate India's social order and economic base, flatten the pivot of empire and undermine the foundation of Britain's influence between the
Yellow and the Red Seas.
These were powerful pressures but they do not provide a sufficient explanation. The European powers were quickly impressed by the stringency of the plague
measures and by 1897 there was little evidence that the scourge was about to engulf the West. Nor was the threat to trade insuperable. The International Sanitary
Convention which met at Venice in 1897 to discuss the plague favoured the medical inspection of people and their personal effects over embargoes on the import or
movement of merchandise.39 Moreover, plague mortality in the first three years of the epidemic did not distort the ordinary death rate for the subcontinent. Although
the threat to trade and empire seemed less grave in the late 1890s, the vigour of the plague administration did not slacken.
So the immoderate, perhaps irrational, severity of the state's response requires that we dig at a deeper level. The official construction of the plague epidemic was
shaped by its assumptions about its own statecraft as well as its perceptions of the governed. In the course of the epidemic, colonial officials not only expressed these
assumptions with unusual freedom, but also developed them to their logical extreme and further were able to manifest them in practice. The plague, it would appear,
became the focus of the most terrible anxieties which India evoked in the British imagination. India appeared to be a land of potential, sometimes hidden, dangers,
political and corporeal, moral and cultural. The defence of the 'thin red line' could not be left to the redcoats alone. The struggle
37 S. B. Saul, Studies in British Overseas Trade (Liverpool, 1960), pp. 188–207; Tomlinson, 'India and the British Empire'.
38 For an account of Bombay's relationship with its hinterland and its role in the Indian economy, see Chandavarkar, The Origins of Industrial Capitalism, ch. 1.
39 Hirst, Conquest of Plague, pp. 389–90.

Page 241
to maintain it had to be waged individually against moral and physical sickness, collectively against intrigue, conspiracy and rebellion.
Among the most awesome and compelling dangers represented by India, because it was both intimate and insidious, was the threat of disease. If, in the British
perception, India was a repository of infectious diseases, there were two possible remedies. Since most diseases were caused by filth, one obvious solution lay in a
massive programme of sanitary measures undertaken by the state. Alternatively, a minimalist answer might be found in segregation and the rigorous maintenance of a
substantial social distance from the native town and its inhabitants.
Neither proved to be practicable. The British in India, as the Sanitary Commissioner reported in 1894, would 'never be safe so long as the native population and its
towns and villages are left uncleansed to act as a reservoir of dirt and disease'.40 But the task of cleaning the subcontinent was too gigantic to contemplate. Despite the
prominence of the threat of disease in British perceptions of India, sanitation and sewers, town planning and public health occupied a low place in the imperial order of
priorities. The colonial state was unwilling to incur the cost and averse to bearing the political risks of sanitizing India. For such a project would require the British to
meddle deeply and dangerously in the habits and customs of the natives. Yet, as they knew only too well, the key to the enjoyment of their political kingdom lay not in
social engineering but in salutary neglect. If the problem of public health was thus conceived in terms which could not possibly allow its resolution, the insanitary and
unhygienic conditions of India's towns and villages, however dangerous, were increasingly portrayed as innate and natural to the subcontinent. Colonial officials, as the
Sanitary Commissioner for the NorthWestern Provinces observed in 1885, did not hold themselves 'to blame for this condition of things. Plainly, indeed, the view was
expressed that if the natives chose to live amidst such insanitary surroundings, it was their own concern. And how they managed to do it without greater penalty of
death than seemed apparent, was a frequent cause of expressed surprise.'41
It was no more viable for the British to segregate themselves in hygienic, nativefree, sanitary enclaves. Indian society was too complex and too turbulent to be
managed from afar. It was unrealistic to expect that the army, which employed the overwhelming majority of British residents in India, would function effectively if it
was segregated along racial lines. Segregation in India remained more a conceptual than a physical reality. Thus, army cantonments were inhabited predominantly by
poor, low
40 Annual Report of the Sanitary Commissioner with the Government of India, 1894 (Calcutta, 1896), p. 27.
41 Eighteenth Annual Report of the Sanitary Commissioner of the NorthWestern Provinces, 1885 (Allahabad, 1886), p. 60.

Page 242
42
caste Indians – prostitutes, hawkers and halalkhores – whose proximity the British so feared and whose habits they so deeply deplored. Similarly, in the larger urban
settlements of Europeans, segregation was even more fervently imagined and even more ineffectually maintained. At various times in the nineteenth century, town
planners in Bombay city envisaged residential segregation as an integral part of its future development. But as an architect told one such committee: 'A good many
middle class Europeans live in Tarwadi and Byculla. Poor Europeans live in the same class of houses as poor Natives.' Moreover, he added, 'the rich and poor have
always lived together – the former in the principal, the latter in the back streets – and always will'.43
As the possibility of treating the cause of their anxieties grew more remote, so these anxieties grew more entrenched and their social and political implications seemed
more menacing. Increasingly, the British became fatalistic, though never sanguine, about the threat of disease. The outbreak of bubonic plague challenged this fatalism.
The plague represented the apotheosis of the threat of disease. Moreover, its threat was highly personalized; it attacked scores of Europeans in the first few months.44
Immediate and summary action now seemed imperative. Policy initiatives which had once seemed impolitic now seemed indispensable; those which seemed to lie
beyond the capacity of the state suddenly fell within its grasp and no effort or expense was to be spared. Intervention in a style which was considered unthinkable
before was now seriously pursued.
Bombay city had grown prodigiously in the second half of the nineteenth century, but it had lacked the most basic infrastructure to accommodate this growth. Its most
densely populated areas in 1900 had only decades earlier been lying beneath the sea. Health Officers in the city had regularly predicted impending doom and
devastating epidemics. In 1875, T. S. Weir, observing 'the low standard of living and the insufficient diet of the majority of the people' and 'their weak and puny
constitutions', predicted the periodic and recurrent 'outbreak of one or other of the epidemic diseases; it may not be Cholera; it may not be Small Pox, but
42 Report of the Sanitary Commissioner for Bombay, 1867 (Bombay, 1868), Report on Colaba, Inspection Report, no. 1, pp. 10–11. See also the excellent article by J. W. Cell,
'AngloIndian Medical Theory and the Origins of Segregation in West Africa', American Historical Review, 91:2 (1986), 307–35.
43 Proceedings of the Committee on the Extension of the City of Bombay, 1887, Evidence, Mr D. Gostling, p. 2. GOB, PWD (General), vol. 1162, Compilation no. 4133 W, 1868–89, MSA.
44 Report by T. S. Weir, Executive Health Officer, Bombay Municipality, in P. C. H. Snow, Report on the Outbreak of Bubonic Plague in Bombay, 1896–97 (Bombay, 1897), pp. 144–5.
Eightyeight Europeans died of the plague between September 1896 and May 1897; many others were attacked by the disease.

Page 243
45
some disease will arise, and sweep off the most effete of the population'. In the early 1890s, the inadequacy of the city's drainage system suggested that 'there will
surely come a time when the population of each district will not be able to live in health'.46 With the plague, it seemed that the time had come. The apocalypse which
the Health Officers had long anticipated was now firmly in their midst. To many colonial officials, the plague appeared not simply the product of ineffectual or
misconceived policies, but, more especially, divine retribution for sanitary neglect.
III
It was from this psychology of guilt and terror that the official mind formulated its strategy to combat the plague. Public health policies had been often conceived as the
application of modern, scientific knowledge among people who were not only ignorant of the principles of hygiene but whose traditions and modes of life were violated
by them. The British in India sought to tame the dangers of the subcontinent and impose order upon its chaos with justice, rationality and science. It was their privileged
access to reason, the superiority of their knowledge and their ability to implement it with an incorruptible justice which legitimized their harshest, most vigorous
measures. Thus, W. L. Reade, who took over the plague administration in Poona after Rand's assassination, proclaimed, 'I consider that plague operations properly
undertaken present some of the best opportunities for riveting our rule in India' and 'also for showing the superiority of our Western science and thoroughness'.47
Sanitary and medical science was as integral and critical to the official perception of their statecraft as education and justice. Yet the formulation of an effective plague
policy was seriously hindered by general ignorance about the causes and transmission of the disease as well as about possible methods of treatment. At the start of the
epidemic, plague authorities still favoured the view that its cause lay in a localized miasma, which could nonetheless be caught through contact with infected persons.48
The identification of the plague bacillus simultaneously by Yersin
45 Tenth Annual Report of the Health Officer, in Annual Report of the Municipal Commissioner of Bombay for the Year 1875 (Bombay, 1876), pp. 148–9.
46 Report of the Health Officer in Administrative Report of the Municipal Commissioner for the City of Bombay for the Year 1892–3 (Bombay, 1893), p. 383.
47 W. L. Reade to Arthur Godley, 3 March 1898, quoted in Catanach, 'Plague and the Tensions of Empire', p. 154.
48 Hirst, Conquest of Plague, p. 41. According to Hirst, this 'miasmaticocontagious point of view' was 'favoured until the end of the nineteenth century by many writers on plague' and
indeed, as late as the 1920s, there was 'a vigorous reaction on the part of an influential group of British epidemiologists against current conceptions of the role of microbes in the
causation of disease': ibid., p. 89.

Page 244
and Kitasato in 1894 had not resolved the problem of causation when the epidemic took root in Bombay two years later. Unless the aetiology of the disease was
known, it was unclear whether the bacillus was indeed its cause or simply its consequence. Thus, with the advent of plague in Bombay, several official teams of
scientists, sponsored by their national governments, 'hastened to the affected city, charged with the task of studying the disease. Probably never before or since has
such an imposing array of epidemiological talent assembled in one place for research into a specific disease.'49 Their findings were to throw as much darkness as light
on the subject while thousands continued to die. At times, their investigations proceeded along lines determined less by scientific evidence than by social assumptions.
The initial consensus which had emerged out of the Hong Kong epidemic was that the plague bacilli entered the body through the alimentary canal. This hypothesis had
been developed by analogy with cholera and other food and waterborne infections, which had recently occupied the attention of bacteriologists and sanitary
commissioners in India. It was swiftly replaced, following feeding experiments in Bombay which produced negative results, by the suggestion that human beings
became infected when cuts and abrasions on their bare feet came into contact with bacilli in excreta of rats, who it was still assumed acquired it through ingestion. This
notion coincided with the recognition that 'patients mostly belonged to the lower classes who commonly go barefooted and barelegged'.50 Circumstantial evidence
that 'halalcores who remove the nightsoil from the houses, and who form probably the dirtiest portion of the population, were notably free from plague' cast some
doubt on this hypothesis. But this difficulty was briskly overcome with the suggestion that 'they are a strong, wellnourished class, of whom only the fittest have
survived, and they are highly paid and live well'.51 Tests on plague patients showed only a tiny proportion of cases in which the bacillus was found in skin abrasions. By
1898, the studies of Hankin and Simond demonstrated that the plague bacillus survived only briefly outside the body and was only rarely, if ever, recovered from
supposedly infected objects, including the surface of the soil, foodstuffs, the floors of houses or other articles in areas of infection.52
The major breakthrough in the aetiology of bubonic plague occurred when the French bacteriologist, P. L. Simond, published a paper in 1898 identifying it as a rat
disease and postulating that its transmission to man occurred through rat fleas. The case was not experimentally proven to the
49 Ibid., p. 105.
50 Condon, The Bombay Plague, p. 104; Hirst, Conquest of Plague, pp. 111–19.

Page 245
satisfaction of his fellow bacteriologists and least of all the Indian Plague Commission. To them, it seemed to rely too heavily upon the imagination. The Plague
Commission preferred the theory of barefoot Indians. It was convinced that if rats were involved in the initial outbreak of the disease, the infection was primarily
spread through human agency. It poured scorn on Simond's hypothesis that transmission was effected by a bloodsucking insect and although it acknowledged that
plague bacilli in rat corpses rapidly died, it adhered to the notion of infection through microbes on the surface of the ground.53 The rat flea theory provoked such
vigorous scepticism primarily because it undermined the assumptions, connecting hypotheses about the nature of the disease to notions of social behaviour and cultural
characteristics in India, upon which epidemiological research had been proceeding. Significantly, it was from Sydney, where the cholera analogy and the barefoot
theory were somewhat less obviously sustained, and where images of proliferating microbes and metaphors of contagion strained credulity, that some of the early
confirmation of Simond's hypothesis came forth.54
In India, medical and official opinion adhered to the view, attractive in this repository of disease, that the plague was spread by contact between human beings.
Consequently, the emphasis of policy was placed primarily on early detection and diagnosis, segregation, hospitalization, disinfection and the close inspection of
travellers and merchandise in transit. These policies had been frequently implemented to control infectious diseases. But in the case of the plague they were carried to
their extreme. Since the 1860s, the response to outbreaks of cholera in the army was to move soldiers out of their barracks and house them in temporary camps;55
now whole villages and even small towns were evacuated. Isolation in a contagious diseases hospital might appear a reasonable precaution; now it was effected under
an armed military guard. Disinfection was another common recourse in times of smallpox or diphtheria; now cities were flooded and incessantly sprayed with mercuric
chloride solution. Fantasies about cleaning up the subcontinent now took on a new and substantive meaning. No form of executive action, it seemed, was too
extravagant in the frenzy and panic of the epidemic.
The medical and scientific experts did not always privilege the claims of evidence over their own preconceptions about Indian society and they were eventually
ensnared within them. Sometimes they overlooked or
53 Report of the Indian Plague Commission, 1898–99, vol. V, pp. 68–71, 75–7, 101–2, 108–11, 122–7, PP, 1902, vol. LXXII; Cell, 'AngloIndian Medical Theory', 326–7; Catanach,
'Plague and the Tensions of Empire', pp. 158–9.
54 Hirst, Conquest of Plague, pp. 160–9, 144–8; Cell, 'AngloIndian Medical Theory', 325–8; Catanach, 'Plague and the Tensions of Empire', pp. 162–3.
55 Cell, 'AngloIndian Medical Theory', 322.

Page 246
neglected the evidence; sometimes, they simply could not read its signs. Thus, it was clear, at the outset of the epidemic, that 'the contagion of the prevailing fever is
very slight',56 and it was soon demonstrated that 'the quickest possible isolation of the sick had no effect on the march of the malady'.57 As early as 1897, 'discerning
observers in Bombay' knew, as Hirst pointed out, that bubonic plague was not 'infectious in the ordinary sense'.58
Nonetheless, policies, formulated on the assumption that the plague was a virulently infectious disease, proved at best oppressive and at worst fatal. Thus, the stringent
inspections along the railway lines and at ports yielded a minute number of plague cases, although many tens of thousands were detained under 'suspicion'.59 When
roofs were removed, floors dug up, houses flooded with disinfectant, the rats simply moved away and spread the infection. By pumping the sewers with disinfectants,
rats were driven into houses and carried the fleas with them to infect the inhabitants.60 Had the scientists and medical experts given the rat flea theory more serious
consideration, they may have resolved some of the conundrums – for instance, the erratic and spasmodic pattern of its dissemination – posed at the time by the
epidemiology of the plague. Moreover, its implications for changing the direction of plague policy, swiftly seen as ineffectual, were substantial. For it suggested the need
to switch the emphasis from the inspection and control of human beings to that of merchandise and from disinfection to disinfestation. As it was, plague policies put
forward as rational measures to control the epidemic among traditional, ignorant people fell prey to the superstitions of science, derived from preconceptions about
Indian society and generated by a wider discourse in which the experts shared.
IV
From the onset of the epidemic, officials readily acknowledged that plague measures were likely to have 'undesirable consequences' but the only alternative was to
endure a growing incidence of sickness and a rising death toll. To them, it seemed imperative to do whatever was necessary to stamp out the plague. Yet it was by no
means clear how far they could go in asserting the virtues of science over the traditional sentiments and religious susceptibilities of the people. The riots which occurred
at the
56 Times of India, 25 September, 1896, p. 4.
58 Ibid., p. 119.
60 A Monograph on Evacuation as a Protective and Combative Plague Measure, Complied Under the Orders of Sir A. Wingate, KCIE, Acting Chief Secretary to Government by Lt.
J. K. Condon, UnderSecretary to Government (Plague Department) for the Use of the Indian Plague Commission, 30 March 1899 (Bombay, 1899), p. 1.

Page 247
Arthur Road Hospital in Bombay as early as October 1896 suggested to the provincial government 'that it was extremely doubtful whether we could enforce in
Bombay City all the measures which high medical authority commended to us as desirable to combat the epidemic'.61 The annual revival of the plague and its
dissemination came increasingly to be explained not in terms of misconceived and misdirected policies but of 'the failure of the people to comprehend its characteristics
and the value of the measures'.62 Popular resistance was perceived as irrational and dysfunctional, a product of the prescientific attitudes, the sacred traditions and
customary prejudices of the Indian people, a view which later historians have sometimes unwittingly endorsed.
Contemporary officials tended to homogenize the Indian response to the epidemic. In fact, it was extremely uneven. Resistance was not the only response to the plague
measures and collaboration was by no means confined to the elites. When the Government of Bombay drew up a list of the city's inhabitants who merited rewards and
commendations for their assistance in carrying out plague measures, Demos was strongly represented, including street corner bosses like Hashim Dada of Nagpada.63
'In every street row', British officials were relieved to discover, 'some were found to stand beside the executive and calm the mob.'64 The epidemic yielded neither a
homogeneous popular response to the epidemic nor a simple and consistent opposition between the colonial state and the Indian people as a whole. Nor should we
postulate a natural affinity between the colonial state and the 'Indian elites', to be neatly distinguished from the mass of the population. To W. C. Rand in Poona, it
appeared, 'some of the most influential men in the City . . . were more likely than not to work against any operations that might be set on foot by Government'.65 In
Hubli, where the local Muslim population had 'given trouble ever since Plague operations began', the Collector of Dharwar reported, 'the list of ringleaders' included 'a
prominent Kazi and an exMunicipal Commissioner', who, he believed, had thereby betrayed a pact he had made with the town's elites.66
Forms of resistance also varied. Perhaps the most common response to the frenzied plague measures, and indeed the scourge itself, was flight.
61 Lord Sandhurst, Governor of Bombay and Others, GOB, Judicial, to Her Majesty's Principal Private Secretary of State for India in Council, London, 9 April 1898, GOB, General
(Plague), vol. 389, Compilation no. 298, MSA.
62
Condon, The Bombay Plague, p. 11.
63 GOB, General (Plague), vol. 8, Compilation no. 712 P/11 Confidential, 1899, MSA.
64 Snow, Report on the Outbreak of Bubonic Plague, p. 19.
65 Draft of Report . . . by the Late Mr W. C. Rand, p. 4, MSA.
66 Collector, Dharwar, to Commissioner, Southern Division, 13 March 1898, GOB, General (Plague), vol. 389 of 1898, Compilation no. 298, MSA.

Page 248
From most large towns, a substantial proportion of the population simply ran away. This was why the evacuation of whole settlements met with the least resistance and
in some places, like Bassein, it was 'in high favour with the people'.67 Concealment and evasion were also extensively practised. 'Incredible shifts were resorted to',
reported Rand from Poona, 'to prevent the authorities from becoming aware of the occurrence of cases. Plague patients were hidden in lofts, cupboards and gardens –
anywhere in fact, where their presence was least likely to be suspected.'68 In Surat city, where 'the virulence of the disease was never great' in 1897 and 1898, patients
were moved from house to house to avoid search parties with the effect it was erroneously supposed, of spreading the germs more quickly.69 Various stratagems were
adopted by travellers to avoid the inspection at ports and along railway lines.70
Popular violence and riot provided the least frequent manifestation of such resistance, but they were also, of course, the most dramatic. In view of the zeal with which
the plague administrators invaded the homes and the physical and social privacy of the people, riots appear to be remarkably rare occurrences. But the colonial state
often took careful note of local skirmishes which it might otherwise have simply overlooked. In part, this was because these 'riots' did sometimes exercise a check on
what the official mind grasped with certitude as necessary measures. In addition, these moments of collective action often exposed the weakness and vulnerability of
local administrations, at a time when they were parading their power and exerting control and in an atmosphere governed by perceptions of physical danger. Finally,
these riots often displayed what colonial officials charmingly described as 'racial characteristics and innate prejudices'.71 Thus, in the Bombay riots of March 1898, the
'mob' chased Inspector Coady shouting 'Mardalo goreku, mardalo goreku' or 'kill the white man'. At the statue of the Standing Parsi, 'it was impossible for any
European to pass with safety. No distinction was made . . . the principle the rioters acted upon was to hit a solah topee wherever they saw it and it mattered not
whether the victim had ever had any participation in the plague administration.'72 Numerous Europeans were attacked in the streets and at least two soldiers were
lynched. The only shop to remain open on Abdul Rahman Street sold revolvers and 'the owner of this establishment was much sought after by Europeans during the
afternoon'.73
It might be supposed that to understand popular hostility to the plague
69
Condon, The Bombay Plague, p. 177.
70 Ibid., pp. 138, 146.
71 Ibid., p. 131.
72 'An Account of the Riots in Bombay on 9 March 1898', in GOB, Judicial, 1898, vol. 217, Compilation no. 669, Part I, MSA.
73 Ibid.

Page 249
measures we need to look no further than their character and the style of their implementation. If the plague measures were harsh, it might appear that popular
resistance to them was wholly rational. It would be tempting to suggest that while the most extravagant fantasy seemed credible and the wildest speculation highly
plausible to the boffins of the government bacteriological laboratories and the learned practitioners of the medical sciences, Demos alone took a steady view. This is a
useful corrective to the supposition that the idiom of popular thinking was based on rumour and religion rather than fact or reason. But, of course, nobody was exempt
from the 'unreasoning panic'. Popular responses, no less than elite reactions, the frenzied exertions of the state or the brittle and sometimes destructive certainties of
medical and scientific experts, were shaped by the political conjuncture of the plague.
The most prominent feature of the plague operations, because of the stress laid on 'detection', was the search parties. In Poona, but less intensively elsewhere, the
chosen agents for search were British soldiers and the searches were conducted like a military operation. The locality was surrounded by cavalry which paraded the
streets, while soldiers moved from house to house, accompanied ideally by 'a native gentleman' and at least a hospital assistant to adjudicate upon suspicious cases.
But inevitably resources were overstretched. There were too few doctors to accompany every search party74 and while there was no shortage of 'native gentlemen',
they were not always willing to assist: some, according to Rand, 'worked steadily and well . . . others irregularly, others not at all'.75
The 'detection' of sickness was a sensitive and intimate operation. 'To eliminate the suspicious cases', as one official report put it, 'a careful individual examination of
each native was necessary' which involved in particular 'a careful exploration of his body for glandular enlargement'.76 Not surprisingly, there were complaints that 'all
the females are compelled to come out of their houses and stand before the public gaze in the open street and be there subjected to inspection by soldiers'.77 Soldiers
were said to 'behave disgracefully with native ladies'78 and the tenor of the official response that they had merely 'joked with a Marathi woman' suggests that sexual
harassment probably did occur.79 Shripat Gopal
74 In Bombay, in 1898–9, only five doctors were available for 'plague duty'. See Condon, The Bombay Plague, pp. 131–2.
75
Note by Chairman, Poona Plague Committee, 7 July 1897, in GOB, General (Plague), vol. I, Compilation no. 70/P, 1897, p. 157, MSA.
77 V. M. Bhide et al. to Chairman, Poona Plague Committee, 7 April 1897, in GOB, General (Plague), vol. I, Compilation no. 70/P, 1897, p. 8, MSA.
78 Bhide et al. to chairman, Poona Plague Committee, 20 April 1897, ibid., p. 15, MSA.
79 Chairman, Poona Plague Committee, to Secretary, GOB, General, 20 July 1897, ibid., p. 179, MSA.

Page 250
Kulkarni, an octogenarian, complained that ten or twelve soldiers had burst into his house, forced him to undress, 'felt . . . the whole of my body and then made me sit
and rise [several times] and, sitting around me, went on clapping their hands and dancing'.80 In addition, there were complaints that while conducting their house
searches, soldiers 'put into their mouths whatever eatables they find', 'insist upon being supplied with milk and other drinks' or 'put into their pockets such things as
come into their fancy'.81 Petitioners were often aggrieved that there was a wanton and indiscriminate destruction of property during searches. Officials knew only too
well that the 'concealment of cases is also practised in order to avoid the inconvenience and petty expenses of disinfection, destruction of clothing etc.'.82
Those who had the misfortune of being placed under suspicion, however wrongly, by any one plague agency, it was said, could suffer the attention of several others.
The suspicion of the search party brought in its train the disinfection gangs and limewashing parties. It could also result in the hospitalization of suspects, the segregation
of their relatives, the loss of their property, perhaps the destruction of their homes.83 Sometimes, it was said, 'their neighbours and in many cases even the passers by
are indiscriminately seized and sent to segregation camp'.84
The most common complaint concerned false diagnosis which resulted in 'perfectly healthy persons' being 'seized and forcibly taken away by the search parties'.85
There was some considerable uncertainty in the diagnosis of the plague, particularly by the search parties. The chairman of the Poona Plague Committee conceded that
the Medical Officer in charge of the Plague Hospital had been 'complaining of the excessive number of cases sent to him which were not plague'.86 It is probable that
before September 1896 bubonic plague had long gone undiagnosed87 but once the panic gathered force, doctors as well as the lay search parties found plague
everywhere.88
80 Petition by Shripat Gopal Kulkarni, ibid., pp. 306–7, MSA.
81 V. M. Bhide, Chairman, Deccan Sabha, Poona, to Chairman, Poona Plague Committee, 20 April 1897, ibid., p. 15, MSA.
82 Condon, The Bombay Plague, pp. 51–2.
83 Petition of BhikubinTatya Shimpi to President, Deccan Sabha, 1 August 1897, GOB, General (Plague), vol. I, Compilation no. 70/P, 1897, pp. 245–7, MSA.
84 V. M. Bhide, Chairman, Deccan Sabha, et al., to chairman, Poona Plague Committee, 20 April 1897, ibid., p. 13, MSA.
85
Ibid, MSA.
86 Note by Chairman, Poona Plague Committee, 7 July 1897, ibid., pp. 154–5, MSA.
87 Snow, Report on the Outbreak of Bubonic Plague, pp. 1–3; see also, Condon, The Bombay Plague, p. 68.
88 A dispute over the misplaced suspicions of a search party culminated in the Bombay riot of March 1898. See GOB, Judicial, vol. 217, Compilation no. 669, Part I, MSA. On Dr W. J.
Simpson's diagnosis of a case of syphilis as the plague, which caused a panic in Calcutta in 1896, see 'Papers Relating to the Outbreak of Bubonic Plague in India . . . up to March 1897',
PP, 1897, vol. LXIII, pp. 41–2; Catanach, 'Plague and the Indian Village', p. 241 fn. 107.

Page 251
The Government of Bombay vociferously denied the allegations made by petitioners and the press and few of them could be proved. Sometimes, the author of the
petition could not be traced which in the prevailing terror is neither surprising nor significant. Of course, the truth of these allegations matters less than the fact that, as
their collector was subsequently to point out, 'they were rife in the City'.89 It is unlikely that the truth would dramatically alter the significance of their narratives.
Historians of India have recently paid considerable attention to rumours about epidemics and especially the plague. In particular, David Arnold has characterized them
'as a form of popular discourse'90 which shows 'a significant divergence of outlook between the middle classes and the subordinate population'.91 There are several
problems with this argument. First, the historical source of rumours lies most frequently in official reports, the memoirs of civil servants and newspapers, which created
as much as they reported them, and which were compiled largely by the middle classes. Rumours may also be taken, therefore, as an elite discourse about popular
attitudes.
Second, the British in India were highly susceptible to rumours. This susceptibility reflected their isolation from, imperfect knowledge of and vulnerability to Indian
society. Falsehoods put about in the bazaar might spread discontent and provoke rebellion, which they might not always be in a position to control. In part, this was
why colonial officials compiled and reported 'rumours' with such exasperation and dread as they occurred, such bemusement, curiosity and fascination after the event.
After the 1898 riots in Bombay, the Commissioner of Police reported, 'Rumours, of course, are flying about in plenty, but I think it is a pity that people are so very
anxious to credit even the most absurd stories, the origins of which it is almost impossible to trace.'92 Sometimes, however, it was his own officers who appeared most
anxious to credit them. In the immediate aftermath of the riot, the Plague Committee peon at Bandra, to the north of the island, reported that some Mohammedan
'roughs' were throwing stones at his shed near Mahim causeway and the superintendent on duty saw 'a large gang of Mahammedan roughs . . . to be collecting near the
Railway in the vicinity of the [Bandra Observation] camp'. Superintendent A. H. Bingley telegraphed for reinforcements, as he later explained, because 'I was aware
that a persistent rumour had been spread abroad for the last few days that a disturbance might be expected on Friday.'93 The
89
Statement by V. M. Bhide, Chairman, Poona Plague Committee, 1 August 1897, GOB, General (Plague), vol. I, Compilation no. 70/P, 1897, p. 108, MSA.
91 Ibid., p. 68.
92 Commissioner of Police, Bombay, to Secretary, GOB, Judicial, 1 April 1898, in GOB, General (Plague), vol. 389, Compilation no. 298, 1898, MSA.
93 Superintendent of Police, Bandra Observation Camp, to Plague Commissioner, Bombay, 25 March 1898, ibid, MSA.

Page 252
police reinforcements were rushed from Thana districts only to discover that 'no stone throwing had taken place at the camp and that there was not the slightest cause
of alarm'.94
Moreover, Indian elites subscribed to rumours about the plague as much as the poor. For instance, one rumour, quoted by Arnold,95 was recorded when 'a well
disposed Brahmin' asked the epidemiologist Hankin, 'in all seriousness . . . whether it was true that an English sahib had put snakevenom into the Bombay water
supply and thus produced the great epidemic in the city'.96 Similarly, P. C. Snow, the Municipal Commissioner for Bombay, reported, in 1896, that of all the plague
measures, segregation and hospitalization 'caused the most alarm' and 'reports were freely circulated that the authorities merely took them there to make a speedy end
of them'.97 But these were measures which 'the whole people, high and low, viewed with the wildest hostility'. Similarly, diverse social groups subscribed, as we have
seen, to the 'stories' about soldiers and segregation, search parties and disinfection gangs, medical diagnoses and hospital treatments contained in the petitions to the
Poona Plague Committee and found them highly convincing.
Finally, there can be only the most slender distinction between the rumours of the populace and the superstitions of science. To attribute the causes of the plague,
against the evidence, to microbes harboured in the earth seemed to many officials and experts to constitute 'in all seriousness' an immutable fact: real knowledge
sanctified by scientific learning. But this was not very far removed from the 'welldisposed Brahmin's' theory of snakevenom.
Rumours about the plague need not be understood in terms which are specific to Indian culture alone. They may be profitably read as the expression of a pervasive
mood of unreality and mortal danger, shared by many across divisions of class, caste and creed, as the plague raged and retreated around them. Rumours were an
earthy, accessible and, in a sense, even tangible way of sustaining hope, expressing anger, of paradoxically keeping in touch with reality. Rumours provided a magical
idiom for discussing the most horrible and menacing realities while sometimes providing a means of liberation from them. Rumours cited by Arnold anticipating the
collapse of British rule or predicting impending doom are examples of the latter. Moreover, sometimes ordinary, daily anxieties could find an exaggerated and exotic
focus. Thus, perhaps, stories of being 'kidnapped' by soldiers, and kept segregated, only to return to discover relatives had died, articulated a deepseated fear of
94 District Superintendent of Police, Thana to Secretary, GOB, General (Plague), 27 March 1898, ibid, MSA.

Page 253
separation and loss, at a time when both abounded. Stories about the behaviour of soldiers may have borne a considerable measure of truth but they also reflected the
nightmarish invasion and violation of privacy – even godrooms and kitchens – by the most frightening, powerful, uniformed and foreign agent of public authority.
Sexual harassment by the soldiers and their 'disgraceful behaviour towards native ladies' almost certainly occurred – and, indeed, physical examination, 'the exploration
of the native's body' in the streets or at railway checkpoints may themselves be regarded precisely as that – but reports of them also served as a metaphor for the
violent eruption of the state into the privacy of people's lives. Rumours about poisoning and snakevenom which attributed the plague to the conspiracy of officials did
not simply represent a popular appreciation of the 'undivided malevolence'98 of the state. It was rather that by blaming official conspirators it was possible to evade
one's own irredeemable vulnerability to the epidemic. For the British, too, it was preferable to blame the spread of the plague on the incomprehension and
obscurantism of the native, rather than on the failure of their hopelessly misconceived policies, or to attribute their own continued danger from the disease as well as
physical assault by riotous mobs to the currency of rumours propagated by troublemakers in the bazaars and the credulity of the simple and illiterate, if sullen and
resentful, poor.
Rumours during the plague, or indeed at other times, should not be interpreted literally, either for their foundation in reality or as a text, to 'discern' the
'preoccupations'99 of the popular mind. Rather, rumours, and stories, even when they circulated fictions more terrifying than the facts, were a means of mediating the
unremitting horrors entailed in the actual circumstances of the plague. Nor can rumours be regarded as the exclusive, or even primary, idiom of the poor. Men of
science and letters, physicians with formidable reputations, the 'most imposing array of epidemiological talent' in the world, and the handpicked brilliance of Oxford
and Cambridge who filled the ranks of the Indian Civil Service, were all similarly susceptible. If the plague epidemic facilitated the expression of the deepest popular
anxieties, it had also become the focus of British anxieties about their inability to control India and their vulnerability to the numerous epidemics, medical as well as
political, which it harboured.
While the prospect of receiving the attention of the search parties, with their militarystyle campaigns of detection, filled people with dread and
99 Ibid., pp. 72, 75–7.

Page 254
provoked them to anger, the threat of hospitalization loomed alarmingly large in public anxiety. Indeed, at times it culminated in riots and mob attacks upon hospitals.
But this should not be taken to signify an innate, indiscriminate and implacable antipathy to Western medicine. Of course, it is not intended to suggest that the Indian
people embraced Western medicine enthusiastically at the first opportunity, but rather that popular attitudes were too complex and too diverse to be reduced to a
simple choice between acceptance and rejection. Indeed, this complexity stemmed precisely from the fact that popular attitudes were not simply and reflexively drawn
as frequently assumed from a reservoir of religious and cultural traditions, but were shaped by historical and political contingencies and, in particular, by the experience
of various medical cures.
In late nineteenthcentury India, the 'Western' medical presence was extremely thin on the ground. In Bombay city, where its density was greatest, the census returned
555 qualified practitioners in 1901,100 while in the countryside, outside the district towns, there was generally 'no skilled medical assistance of any kind'.101 To obtain
medical attention, most people had to travel considerable distances, suffer a loss of work and incur considerable expense. Nonetheless, where they existed, rural
dispensaries were extensively used.102 In Bombay city, too, there is little evidence that 'Western' medical practitioners were abjured. On the contrary, at the very start
of the epidemic, Dr Thomas Blainey, visiting the afflicted district of Mandvi with Dr Viegas, found people eager for medical attention. 'The people around me', he
wrote,
urged me to go and see other similar cases at varying distances and said that many persons in their locality were similarly attacked. Several of them informed their friends that I was
a municipal representative deputed to inquire into the present fever outbreak and that every facility should be afforded me to make the inquiry complete. I have no doubt that Dr.
Viegas corrected their mistake on my official identity . . . The people, though ignorant, are quite alive to the dangerous character of the prevailing fever.103
Many of them were, of course, shortly to die. But their response to Dr Blainey was far removed from the response evinced by the plague authorities as the epidemic
wore on.
Like the colonial officials before them, historians have sometimes too
100 Census of India, 1901, vol. XIA, Bombay Town and Island, Part VI, tables, compiled by S. M. Edwardes (Bombay, 1901), Table XV, p. 138. These were listed as 'practitioners
with diploma'. Another 398 'practitioners without a diploma' were also returned by the census.
101 J. K. N. Kabraji, Second Assistant Collector, to R. B. Stewart, Ag. Collector, Nasik, 16 August 1897, in GOB, Revenue, vol. 19, Compilation no. 67, Part IV, 1898, MSA.
102 R. B. Stewart, Collector, Nasik, to Commissioner, Central Division, 25 September 1897, ibid, MSA.
103 Times of India, 25 September 1896, p. 4.

Page 255
104
readily taken for granted that 'Western medicine outstripped popular disease comprehension'. In fact, 'Western' medicine knew little about the disease and its ability
to learn more was inhibited precisely by its perceptions of Indian culture and society. To the extent that policies were shaped by its wisdom, their effect was often to
intensify and disseminate the epidemic. The plague was notoriously difficult to diagnose. As a result, 'numbers of cases had to be segregated . . . not because the
patients had plague but because they had suspicious symptoms'.105 Depending upon which symptoms predominated in a given case, bubonic plague could resemble
relapsing fever, severe cases of malaria, typhoid, typhus, glandular fever and even drunkenness. Physicians sometimes misread its symptoms for another disease or
treated rather more benign illnesses as if they were the plague. The official history of the plague cited the case of Govind Jeeva who, treated for alcoholic poisoning by
the doctor, was thus speeded on his way to death from bubonic plague.106
Nor were there any known remedies for the disease.107 By 1900, 'Western' medical practitioners had discovered 'no specific remedy' for the plague and none of those
which they tried had influenced 'favourably the mortality among those attacked'. The plague hospitals could offer nothing more than 'hygienic and symptomatic
treatment'. Not knowing what else to do, they tried to tide 'the patient over a certain period' and trusted to 'the natural tendency to recover'.108 Medical practice,
especially in the 'Western' tradition, was seen to be experimental in its procedure and ineffective in its results. The descriptions of their own activities by plague officials
suggest that this perception was on the whole justified. To Bombay's Municipal Commissioner, 'The outbreak of plague in the House of Correction afforded a
particularly good opportunity of watching the effect of M. Haffkine's prophylactic treatment.'109 When Haffkine took a medical team to inoculate the villagers of
Undhera, a village near Baroda, in February 1897, they were able to report their satisfaction that 'the conditions approached very nearly the strictness of a laboratory
experiment'.110 It is doubtful whether the subjects of their experiments shared their satisfaction as the threat of death loomed over them and it is scarcely surprising, in
the
104 Klein, 'Plague, Policy and Popular Unrest', p. 739.
107 For a description of some remedies tried at the start of the epidemic, see Times of India, 25 September 1896, 4.
110 Condon, The Bombay Plague, pp. 43–4; W. M. Haffkine, Experiment on the Effect of Protective Inoculation in the Epidemic of Plague at Undhera Taluka, Baroda, February and
March, 1898 (Bombay, 1898).

Page 256
light of these attitudes, that medical officers found it hard to win public confidence.
The widespread hostility to hospitalization, shared by diverse social groups, was not simply a function of caste and religious sensibilities. Indeed, it extended 'even to
hospitals established and managed by Hindoos for their own caste fellows'.111 Hospitalization represented the culmination of all plague measures, perhaps the most
coercive manifestation of a brutally intrusive state, and the end of a terrifying chain of events which began with the search parties. The case mortality rate of plague
patients entering hospital averaged over 80 per cent at the height of the first epidemic in Bombay.112 The overwhelming majority who were admitted to hospital did not
return alive. The arrival of the ambulance must, therefore, have seemed like the state's death sign on the patient or the suspect. 'People who thought the poor . . . ought
to be happy because they had been born to it', reported T. S. Weir from Bombay, 'almost wept when they saw one of the same poor ill from Bubonic Plague, lifted
into a municipal ambulance.'113 'Peals of screams' from plague patients in the Contagious Diseases Hospital in Bombay 'not only pervaded the whole hospital, but even
attracted the notice of passersby on the road', one of the major thoroughfares through the mill districts.114 Not surprisingly, hospitals came to be perceived in the
public imagination less as a refuge from the ravages of the plague, than as a potent and destructive instrument of terror, as 'places of torture and places intended to
provide material for experiments'.115
If to the Indian mind, as one historian has recently suggested, hospitals were 'a place of pollution, contaminated by blood and faeces', this was often less because they
appeared 'inimical to caste, religion and purdah',116 than because these were the real conditions to be found there.117
111 Condon, The Bombay Plague, p. 126; Snow, Report on the Outbreak of Bubonic Plague, p. 16.
112 Snow, Report on the Outbreak of Bubonic Plague, p. 17; Condon, The Bombay Plague, pp. 132–3.
113 Report by Executive Health Officer, Bombay, in Snow, Report on the Outbreak of Bubonic Plague, p. 81.
114 'Report on Bubonic Plague Cases Treated at the Arthur Road Hospital from September 24, 1896 to February 28, 1897', by Khan Bahadur N. H. Choksi, Extra Assistant Health Officer,
in charge of the Arthur Road Hospital, ibid., p. 237.
115 Report by Executive Health Officer, Bombay, ibid., pp. 73–4.
117 For a description of the plague hospital in Poona by one who was taken there on suspicion and later released, see the letter from Sodaji Pundlik More to Dynanprakash in GOB,
General (Plague), vol. I, Compilation no. 70/P, 1897, pp. 197–238, MSA. On conditions in the Arthur Road Hospital in Bombay, see 'Report on Bubonic Plague Cases Treated at . . .' by
N. H. Choksi, in Snow, Report on the Outbreak of Bubonic Plague, pp. 210–12, 237.

Page 257
But this was not simply an 'indigenous' perception. When his daughter contracted the plague in Poona, Surgeon Major Barry, a British officer of the Indian Medical
Service, recognized that she would have a better chance of recovery at home. Rather than entrust her to what should have appeared to his 'Western eyes' as 'the
sanitized and healing environment of the hospital',118 he conspired with his colleague, Surgeon Major Baker, who examined the patient, to conceal the fact and break
the regulations they were working so vigorously to enforce upon others.119
Nonetheless, popular attitudes to the plague hospitals, and to Western medicine in general, were by no means inflexible. Some caste and communal hospitals, having
created an environment which people felt more able to use, found general favour.120 As Bombay's inhabitants fled the plague in 1897, the Petit Mills, facing an acute
shortage of labour, 'promised to erect a temporary plague hospital' where their workers, 'would be looked after instead of being sent away to the Municipal hospitals'.
As a result, it was said, they 'abandoned the idea of going away'.121 If hospitals were not always hospitable places, they were particularly forbidding to the poor.122
By contrast, the occasional glimpses that we have into the domestic treatment of the sick during the epidemic suggest the enormous care often taken by friends and
relatives. W. D. Shepherd, the Collector of Poona, complained that 'the inability of the family to refrain from attending on sick relatives ensured that a large number
should die where many might have lived'. While they were 'generally very keen to keep plague out of their villages', they refused to allow their own 'separation from
their sick relatives'.123 The official historian of the plague expressed his frustration at those who, sharing rooms with plague patients, contracted the infection from being
in constant attendance upon them. 'The poor and ignorant' were said to be more vulnerable to infection than, for instance, nursing staff, because of 'the common
custom which exists of friends receiving the sputa of the sick in their hands, and using their hands and clothing to wipe away discharges from the patients'
119 W. L. Harvey, Municipal Commissioner, Bombay, to Plague Commissioner, Bombay, 14 March 1900, in GOB, General (Plague), vol. 16, Compilation no. 15, 1900, pp. 129–31, MSA.
120
Report of the Municipal Commissioner on the Plague in Bombay in the Year Ending 31st May, 1899 (Bombay, 1899), p. 318.
121 Ibid., pp. 354–6.
122 This is also suggested by some evidence from the 1920s and 1930s. See, for instance, RCLI, Evidence, Bombay Presidency, 1929–31, Mr Dattatraya Ramchand Mayekar and Mr
Narayanrao Kulkarni representing the Girni Kamgar Mahamandal, vol. I, part ii, p. 387. See also Proceedings of the TLIC, 1938–40, Main Inquiry, Evidence, Cotton Mill Workers,
Spinning Side, File 60 A, pp. 1020–2, MSA.
123 Report from Poona District in Condon, The Bombay Plague, pp. 234–5.

Page 258
124
mouth'. It would be folly to idealize the Indian family; not even among kin might we find a single and undifferentiated response. From Surat district, for instance, it
was reported that 'the relatives and caste fellows of the Hindu patients often shrank from performing the last offices for their own dead', while those who contracted the
disease in private segregation camps 'crawled back to their empty houses to die' and neighbours threw their corpses into the streets before officials marked their
houses for special attention.125
At the height of the panic, Western medicine must have often appeared as nothing less than the theological carapace of an intrusive and oppressive state, and the
hospital its most terrifying institutional embodiment. We might under these circumstances expect most people to have turned their backs upon the whole apparatus of
Western medicine. In fact, the public response to Western medicine – if indeed an 'Indian' response may be inferred from the evidence at all – was more nuanced and
flexible and many showed a willingness to use both doctors and hospitals. During the epidemic, some people strained to hear the 'Western' physician's mantra; others
read the shaman's lips; still others turned to the hakims and vaidyas or the folk remedies which the most wizened village elders could recall. Each was as likely to be as
effective as the other. There is nothing to suggest that those who sought the intercession of Sitaladevi did not also seek the help of the governmentapproved physician.
In fact, popular attitudes to Western medicine in the 1890s, as perhaps more recently, were determined by its efficacy, its accessibility and its cost.126 Popular
resistance to medical intervention and hospitalization during the plague epidemic is more plausibly explained by the terms on which it was offered to those stricken or
threatened by the disease than by some primordial sentiment or cultural essence embedded in the Indian mind.
VI
The plague epidemic did not give rise to a single, homogeneous Indian response. Neither is it possible to identify in it any consistent pattern of social differentiation. The
diverse responses to the epidemic reflected in part the various ways in which it was perceived and experienced. People found themselves pitted against each other in
the panic more often than they were gathered into large social solidarities. Social tension, competi
125 Report from Surat District, ibid., pp. 178–9.
126 Cf. A. R. Beals, 'Strategies of Resort to Curers in South India', in Charles Leslie (ed.), Asian Medical Systems: A Comparative Study (Berkeley and Los Angeles, 1976), pp. 195, 192–
4, 198.

Page 259
tion and antagonisms were heightened not only between but also within classes. The fragile facade of social order was cracked open and whole towns and villages
appeared to be on the edge of chaos. In virtually every town, the outbreak of plague paralysed trade and put its inhabitants to flight. From Karad, it was reported,
typically, in June 1897, 'The utter disorganization that prevailed in the town could hardly be imagined by one who had not seen it.'127 In Poona, it was said, 'The state
of the city was one of panic.'128 By February 1897, nearly half the inhabitants of Bombay city had fled, there was 'open bidding for labour at the street corners',129 and
the city's officials grew increasingly apprehensive that the social and political fabric of the city was about to disintegrate.130
At anarchy's edge, the panic created fresh opportunities for profit and power for those with the temerity and ruthlessness to seize them. Shortly after the first outbreak
of the disease in 1896, the Municipal Commissioner in Bombay reported: 'A gang of scoundrels took to blackmailing by personating the Police and Municipal servants,
and increased the general terror, extorting money as they did under threats of removal to hospital.'131 The evacuation of towns and villages, which it was believed had a
salutary effect on public health, was accompanied by an increased incidence, and fear, of crime. At the model segregation camp at Anand in Kaira district in Gujarat,
the banias having set up general provisions stores, were said to be 'in league with the badmashes in the town' to rob the detainees.132 During the epidemic, the police,
whose detectives 'gave us the fullest information',133 may also have found opportunities for gain.134 The implementation of the plague measures left much to the
discretion of the search parties and in the course of their activities, large informal powers accrued to their 'detectives'. Once a household or a neighbourhood became
the object of the search party's suspicion, its members were inextricably pulled into the vortex of the plague's terror. Their only means of escape was to negotiate and
bargain these suspicions to rest. Sometimes the official status of the search parties was difficult to ascertain, and some behaved no differently from 'scoundrels'. The
Government of Bombay became sufficiently anxious about their conduct to
127 Report from Satara District, in Condon, The Bombay Plague, p. 243.
129 S. M. Edwardes, The Rise of Bombay – A Retrospect (Bombay, 1902), p. 330.
130
Snow, Report on the Outbreak of Bubonic Plague, pp. 4–5, 7–9; Report by Executive Health Officer, ibid., pp. 70–8.
132 Report from Kaira District, in Condon, The Bombay Plague, pp. 160–3.
133 Report from Executive Health Officer, Bombay, in Snow, Report on the Outbreak of Bubonic Plague, p. 80.
134 For a general comment on such opportunities for the police during times of disturbances, see S. M. Edwardes, The Bombay City Police, p. 183. See also ch. 6 above.

Page 260
135
instruct its officers that 'The closest supervision is especially necessary over subordinates and they should be taught to treat people civilly.' If search parties fuelled
public anxieties, enterprising spirits exploited the uncertainty they created.
The uncertainties which shrouded medical opinion about the plague and the desperation with which people sought cures wherever they might be found cleared the way
for the proliferation of quacks. They were drawn from diverse social groups. As the Bombay Health Officer reported,
Many good men were spoiled by aspiring to the honours of being Plague Doctors. Mallees and Mahars, and even men employed in the service of the Tramway Company and in
the service of the city, persuaded themselves and encouraged others to believe that they had a cure for plague . . . Once a man became a Plague Doctor, he was seldom happy or
contented in any regular work. If he did not save his followers, he ruined himself for honest toil.136
Healing and curing, blackmail and extortion did not exhaust the commercial opportunities created by the panic. Evacuation and segregation camps opened up new
fields in trade. The collapse of the labour market offered marginal groups the chance to entrench themselves in particular occupations. Those who wielded enough
influence in a neighbourhood or a village could try to oust their rivals and establish themselves as jobbers and procurers and suppliers of labour. The growing
intervention of the state and its feverish search for collaborators enlivened the factions of the neighbourhood and as existing structures of power were subjected to
unprecedented stress, political rivalries were more freely pursued.
The style and method of the plague administration also enabled, indeed invited, people to prey on each other. It assumed that they would spy on their neighbours. It
expected that caste elders and local magnates would report suspicious cases and act as enforcers of the plague regulations. It hoped that, when detainees escaped
from segregation camps, the inhabitants of uninfected areas would 'refuse to harbour such fugitives'.137 These hopes were sometimes fulfilled and at other times
frustrated. Some people responded with an enthusiasm which could be readily mistaken for publicspiritedness. If suspicion was the defining characteristic of the
plague operations, it was also the hallmark of the panic.
The burden of this suspicion fell primarily upon those who could be defined as marginal and isolated as outsiders. With its vocabulary of
135 Circular from GOB, General (Plague), no. 2551 P43 Confidential, 6 April 1898, Home, Judicial, June 1898, 228–9 (B), NAI. I am grateful to Gordon Johnson for this reference.
136 Report from Executive Health Officer, Bombay, in Snow, Report on the Outbreak of Bubonic Plague, p. 80.

Page 261
'detection' and 'search', 'surveillance' and 'informants', 'suspects' and 'fugitives', the plague administration served to criminalize the disease and its victims. At one level,
officials identifying the disease with filth, directed their gaze naturally and primarily towards the poor. In October 1898, the 'Surveillance System' introduced on the
railways focused especially upon 'travellers who . . . were suspicious whether by reason of their appearance or symptoms, or the daily conditions of their clothes or
effects'.138 In what can only be assumed to be an extraordinary slip of the pen, the Bombay Health Officer boasted in 1897: 'From the beginning the greatest attention
was paid to the disinfection of houses and to the segregation of the poor.'139 The Anand detention camp in Kaira district was transformed in May 1898 to 'a
Disinfection Camp for dirty persons arriving from infected localities'.140 But, of course, the whole body of the poor could scarcely be stigmatized with the plague or
uniformly defined as criminal or marginal.
While plague officials may have worked with rather crude definitions of marginality, dominant groups and local majorities tried to deflect the impact of the plague
measures onto weaker and more peripheral groups. Certainly, those whose rights in the village or the neighbourhood appeared the most tenuous, or who could
otherwise be defined as out of the 'community', were the most likely to be reported for 'suspicion' of sickness. It was among untouchables, 'deviants' and outsiders that
the disease was first identified in previously unaffected localities. The illfated search party which visited the village of Ghori in Nasik district in September 1897 was
immediately directed by the village schoolmaster to the 'suspicious cases' in the Teli's quarter.141 The plague in Ankleshwar was said in August 1898 to have 'originated
among the Ghanchis, a very dirty class'.142 In Rajapur in Ahmednagar district, it was first registered among the Mahars and Chambars, but it was nonetheless
concealed by the village officers for over a fortnight. Its origin was attributed locally to an outsider: 'the brother of a Marwari from Sirur, who died of the plague,
presented the clothing of the deceased to a family of Mahars, of whom five caught the plague'.143 The 'infection', in Sirur taluka, Poona district, was said to have been
imported 'from prostitutes in the town'.144 The disease was introduced to Broach, it was claimed, not by rats but by
138 Ibid., p. 143.
139 Report by Executive Health Officer, Bombay, in Snow, Report on the Outbreak of Bubonic Plague, p. 87.
140 Report from Kaira District, in Condon, The Bombay Plague, p. 160.
141 J. K. Kabraji, Assistant Collector, to Collector, Nasik, 26 November 1897, GOB, General (Plague), vol. 389, Compilation no. 298, 1898, MSA.
142 Report from Broach District in Condon, The Bombay Plague, p. 155.
143 Report from Ahmednagar District, ibid., pp. 201–2.
144 Report from Poona District, ibid., p. 227.

Page 262
145
butchers and the quarter 'mostly inhabited by Mohammendans' was 'deeply infected with plague'. In Hubli, too, it was associated with 'the Mussalman community'.
They were regarded as 'turbulent',146 particularly hostile to the plague measures,147 and as Fakrudin Budansab, in charge of the local police and himself a Muslim,
advised, 'being arrogant and daring, it cannot be known what they would do in course of time'.148
In view of the havoc which could be wrought upon a locality or even a whole town, once it fell prey to official suspicion, it is not surprising that a common reflex was
simply to keep the plague administration at bay. Even local elites preferred to avert the official gaze altogether. This was in part why those who became suspects simply
denied that they were sick while the afflicted refused to believe that it was the plague that had struck. The proximity of the disease, let alone its arrival in the locality,
was also so terrifying to contemplate that people often refused to countenance the very idea of it. The progress of the search party in Ghori, for instance, was
intercepted by a large crowd armed with sticks whose spokesman was reported to have said that 'they did not want a doctor nor was there any sickness in the village'
and added that 'we shall not allow men of other creeds into our houses'.149 Buboes found on two patients in Jalgaon, as mortality rates rose, seemed to suggest that 'it
is plague the people are dying of'. But 'the opinion of the town is that it is not plague, and that the mortality is due to ordinary fever acting on constitutions undermined
by fever'.150 As 10 per cent of Karad's population was 'swept away' in three months, in 1897, the Hospital Assistant insisted that the disease was 'not plague but
remittent fever'.151
At the start of the epidemic in Bombay, an initial and spontaneous search for medical and official help was swiftly replaced by a general refusal of state intervention.
The municipal authorities and the provincial government, who had for decades done next to nothing to alleviate the social and sanitary conditions of the city, found it
impossible to persuade its inhabitants that the cause of the epidemic lay in 'the hopeless condition of their own dark, damp, filthy, overcrowded houses'. Instead, they
'raved about the sewers' and 'looked to everything except the buildings and the rooms in which they lived for the cause of the disease'.152 It was
145 Report from Broach District, ibid., p. 156.
146 Collector, Dharwar, to Commissioner, Southern Division, 13 March 1898, in GOB, General (Plague), vol. 389, Compilation no. 298, 1898, MSA.
147 District Magistrate, Dharwar to General Officer Commanding, Belgaum, 13 March 1898, ibid, MSA.
148 First Deputy Head Constable, in charge of the town of Hubli, to Assistant Superintendent of Police, Dharwar, 11 March 1898, ibid, MSA.
149 J. K. Kabraji, Assistant Collector, to Collector, Nasik, 26 November 1897, ibid, MSA.
150 Report from Khandesh District, in Condon, The Bombay Plague, p. 208.
151 Report from Satara District, ibid., pp. 243–4.

Page 263
preferable by far to flee the city or even simply pray that the plague would pass them by than to place their trust in a government which had shown little real or
sustained interest in their welfare, and now intervened in an arbitrary and brutal manner. The Municipal Commissioner found that 'the people refused all medical aid or
to listen to any advice'.153 To the official mind, the popular response sometimes appeared indiscriminately hostile. 'Nearly every hand was against the Municipal
Officers', complained the Health Officer.154 'Such a pass had we come to, that picking up a few sick pigeons . . . nearly led to a riot, and peaceful Bunnias, for 10 sick
pigeons threatened to raze the city.'155 It was almost as if to combat the disease and reorganize their lives around the epidemic people felt it imperative to resist the wise
and rational men who came bearing the officially prescribed remedies, like so many satanic gifts.
VII
It is perhaps easier to explain why panics begin that why they end and the plague epidemic in India was no exception. Bubonic plague was after all, the plague; it was
devastatingly and unremittingly deadly in its effects; it was a new and unfamiliar disease which no one fully comprehended and for which no remedies were known.
Official frenzy acted as the catalyst of the panic and the public response. This frenzy was informed by colonial perceptions of the threat which the epidemic posed to
Britain's empire and its international trade. It was conditioned and its expression inflected by colonial assumptions about Indian society and its cultural characteristics.
In the panic, such economy as may have guided the techniques of colonial rule was quickly spent and these assumptions were expressed with an unusually uninhibited
freedom. The excesses and desperation of official policies fed upon and fattened the terror which the epidemic unleashed upon those exposed to the disease. Many
responded to the desperation of official measures with an equally desperate resistance to and refusal of official, even medical, intervention. What colonial officials saw
as an irrational and obscurantist resistance to the dictates of science and reason only incited them to further, yet more ferocious and despairing executive action. In this
way, panic, terror and guilt engorged each other in a seemingly unending spiral.
The spiral was broken only when the frenzied temper of the plague measures relaxed. In part, this was a consequence of the identification of the plague as a disease of
the poor. At the same time, while the plague continued to flourish, it was impossible to escape the conclusion that the
153 Ibid., p. 5.
154 Report from the Executive Health Officer, Bombay, ibid., p. 71.
155 Ibid., p. 70.

Page 264
156
vigorously enforced policies of the state had achieved very little. They had failed to control the spread of the disease, protect the people or save the lives of the
afflicted. Rather, as the plague extended its sway, officials perceived more starkly the political risks inherent in these colonial policies. As early as July 1897, the
LieutenantGovernor of the NorthWestern Provinces had observed, 'If the plague regulations had been enforced in any city of these provinces in the way in which . . .
they were . . . enforced in Poona, there would certainly have been bloodshed here.'157 The formulation and implementation of the plague measures, indeed the official
construction of the panic, had been facilitated in the late 1890s, in two respects, by the prevailing political circumstances. First, in Bombay Presidency, British rule was
more firmly anchored and British government more developed than in the old Mughal heartland or the outlying frontier regions of the Punjab. In Bombay, the British
often fondly perceived the embodiment of the success of their technique of rule through collaboration. The Punjab, vital to the defence of the Indian empire and the
recruiting ground of the Indian Army, was, however, the home of the 'martial races' rather than the 'scribal classes'. At the height of the panic in 1897, the Bombay
Government had felt able to strike decisively at the apparently formidable political base which the 'extremist' Tilak had been building, through his famine campaigns,
Ganpati and Shivaji festivals and the recently captured Poona Sarvajanik Sabha.158 By contrast, to officials in the Punjab, it seemed extravagant to antagonize merely in
the name of science or in the interests of public health. It was here that the British rediscovered the fatalism with which they had long regarded questions of public
health and social conditions before the outbreak of plague. Second, colonial plague policies had also been facilitated by the decline of the Congress as an allIndia
focus of political opposition. By the mid1890s, the Indian National Congress had settled into torpor. It was only after 1903, while the plague continued unabated and
even reached its peak, that the Congress coincidentally began to revive and in this revival the contribution of Punjab and the NorthWestern Provinces was to be
substantial.159 As they emerged from the whirligig of the plague panic, the British could reassure themselves about
156 The first shifts of official thinking in this direction can be seen in Report of the Indian Plague Commission, 1898–99, vol. V, pp. 400–4, PP, 1902, vol. LXII.
157 Macdonnell to Elgin, 16 July 1897, KeepWith 5, Home Public A, May 1898, 329–44, NAI, cited by Johnson, Provincial Politics and Indian Nationalism, p. 97 fn. 1.
158
Cashman, The Myth of the Lokamanya; Johnson, Provincial Politics and Indian Nationalism, especially, chs. 2 and 3; Catanach, 'Poona Politicians and the Plague'.
159 N. G. Barrier, 'The Punjab Disturbances of 1907: The Response of the British Government in India to Agrarian Unrest', MAS, 1:4 (1967), 353–83; N. G. Barrier, 'The Arya Samaj and
Congress Politics in Punjab, 1894–1908', JAS, 26:3 (1967), 363–79; C. A. Bayly, The Local Roots of Indian Politics: Allahabad, 1870–1920 (Oxford, 1973).

Page 265
the resilience of their systems of collaboration in Bombay; but, more crucially, they had relearnt an old lesson: that the price of political intervention was liable to be the
destruction of these systems and, with it, the brittle and unsteady foundations of their rule.
The plague epidemic of the late 1890s in India carries many resonances of the AIDS epidemic of the 1980s in the United States. In the case of AIDS, state
intervention and public interest were limited by the belief that only marginal, deviant groups were likely to be affected. But this was not a function of calm. It was
accompanied by anxiety that this was Armageddon and rumours circulated about catching the virus from toilet seats, tea cups and toothbrushes. Vested interests, from
the homosexual bathhouseowners to the blood banks, at times impeded and deflected measures to control its spread. Rumours, rife at every stage of the epidemic,
explained its causes in terms of African initiation rites, the use of 'poppers' and a creation of germ warfare experiments gone wrong. A San Diego coroner, undoubtedly
trained at the best American medical schools, argued that the virus was in fact 'King Tut's curse', 'placed in the tomb to punish those who might later defile his grave'.
The legendary 'Orange County Connection' alone was widely believed to be responsible for the first hundred cases in the United States. He visited sex palaces
specially to spread the virus as an act of revenge but was so irresistible that the person who went to confront him about his behavior finished up having sex with him. In
May 1985, Burke's Peerage, compiled, of course, by literate and rational people, announced that to preserve 'the purity of the human race', it would omit all families
in which somebody was known to have AIDS. Their reason, they declared, was that 'AIDS may not be a simple infection, even if conveyed in an unusual way, but an
indication of a genetic defect'. American gays, often prosperous professionals, secular and modern in their outlook and highly educated, rather than prescientific,
superstitious and illiterate, showed at numerous points an antipathy to medical and policy initiatives as possible infringements on their civil liberties, when 'objectively'
they might have been seen as necessary and nonthreatening, and developed a whole vocabulary of euphemisms to maintain this position. As Randy Shilts, from whose
excellent book these instances are quoted, observed: 'Humans who have been subjected to a lifetime of irrational bigotry on the part of mainstream society can be
excused for harbouring unreasonable fears.'160 Perhaps this may provide a better clue to popular resistance to the plague measures in India than sacred traditions or
religious susceptibilities.
160 Randy Shilts, And the Band Played On, pp. 541–3 and passim.

Page 266
8—
Indian Nationalism, 1914–1947:
Gandhian Rhetoric, the Congress and the Working Classes
The period between 1914 and 1947 has frequently been characterized as the phase of mass nationalism. The Congress appeared in this period to be transformed from
the annual tamasha of a bombinating, mendicant elite to an irresistible mass movement inspired by a more radical leadership. Gandhi, it seemed, had marshalled the
Indian people behind the banner of the Congress and led them struggling rightfully and unitedly to be free. Yet on closer scrutiny the attitude of the Congress leadership
to the involvement of the working classes in this age of mass nationalism was suspicious and defensive. At one level, the working classes could no longer simply be
ignored: their own militancy, the impetus of political reform and the extension of the franchise, and the Congress claim to represent the nation as a whole pushed them
to the forefront of Indian politics. Moreover, while revolt in the countryside was what the British feared most, the great Congress agitations were largely urban affairs.
So for the Congress there was much to be gained by establishing their presence in the back streets and neighbourhoods of the towns.
On the other hand, no sooner was the Congress drawn into the sphere of workingclass politics than its leaders disclaimed or denounced their newly acquired
connections. Tilak, in whom Lenin's hopes for the future of the Indian proletariat were invested,1 exhorted the millworkers of Bombay to sink their differences with the
millowners and intoned sonor
1 After the strike which followed the arrest and sentencing of Tilak in Bombay in 1908, Lenin observed that the Tilakites had 'introduced to the Bombay working class certain
socialist ideas they had drawn primarily from the experience of the Russian working class' and were thus 'instrumental in awakening the class consciousness of the Indian
proletariat'. Now that the Indian 'proletariat had developed to conscious political mass struggle . . . the Russianstyle British regime was doomed'. V. I. Lenin, The National
Liberation Movement in the East, pp. 14–15, cited by Cashman, The Myth of the Lokamanya, p. 182 and A. I. Chicherov, 'Tilak's Trial and the Bombay Political Strike of 1908', in I.
M. Reisner and N. M. Goldberg (eds.), Tilak and the Struggle for Indian Freedom (New Delhi, 1966), p. 545.

Page 267
2
ously 'Bolshevism as it is preached in the West cannot succeed in India. Let us stick to our Vedanta and all our desires shall be fulfilled.' 'It does not require much
effort of the intellect', Gandhi declared in the early 1920s, 'to perceive that it is most dangerous to make political use of labour until labourers understand the political
condition of the country and are prepared to work for the common good'.3 During the Assam teacoolies' strike of 1921, at the height of the noncooperation
campaign, Gandhi defined his objectives thus:
In India, we want no political strikes . . . We do not need an atmosphere of unsettled unrest . . . We must gain control over all the unruly and disturbing elements or isolate them
even as we isolate government . . . We seek not to destroy capital or capitalists but to regulate the relations between capital and labour. We want to harness capital to our side.4
When he was elected the President of the AllIndia Trade Union Congress for 1929, Nehru confessed to the disappointed candidate, D. B. Kulkarni, a communist
railwayman, that he did not want the job, that he had been elected in a fit of absence of mind and that he would have certainly withdrawn in favour of his opponent, if
only he had been present at the session which elected him.5 In the year of his presidency, he had hoped to bring the 'National Congress' and the Trade Union Congress
'closer to each other', encouraging the former 'to become more socialistic' and 'organised labour to join the national struggle'. But in retrospect, he noted ruefully in his
Autobiography, this had been 'a vain hope, for nationalism can only go far in a socialistic or proletarian direction by ceasing to be nationalism'.6 In the mid1930s, the
Congress Socialist Party professed a more direct interest in the working class and played a more active role in trade union organization. But the Congress leadership
took the offensive against those whom Vallabhai Patel dismissed as 'young men with brain fever'7 and the Working Committee considered banning 'any Congressman
''who preaches class war" from membership of any executive committee'.8 In the late 1930s, the provincial Congress ministries, having taken office under the 1935
2 Quoted in Cashman, The Myth of the Lokamanya, p. 187.
3 Young India, 16 February 1921, CWMG, vol. XIX, p. 366.
4 Young India, 15 June 1921, CWMG, vol. XX, p. 228.
5 J. Nehru to D. B. Kulkarni, 10 September 1929, AICC Papers, File 16, 1929, NMML. See ch. 3 above.
6 J. Nehru, An Autobiography: With Musings on Recent Events in India (London, 1936), pp. 197–8.
7 Cited by B. R. Tomlinson, The Indian National Congress and the Raj: The Penultimate Phase, 1929–1942 (London, 1976), p. 52.
8 Ibid., p. 169, n. 103; Jamnalal Bajaj to Uma Nehru, 21 June 1934, AICC Papers, File G29, 1934, NMML, cited by Tomlinson, The Indian National Congress and the Raj, p. 52.

Page 268
9
Act, set about bashing trade unions and repressing strikes seemingly with more enthusiasm than the preceding colonial administration.
Popular support for the Congress has usually been measured in relation to its great agitational campaigns. However, the timing of the major Congress campaigns was
determined not by the groundswell of popular political action but rather by imperial and constitutional considerations. Noncooperation was the Gandhian alternative to
council entry; civil disobedience was the Congress response to the Round Table Conference, the Quit India movement its answer to Cripps's offer of a 'postdated
cheque on a crashing bank'. Popular participation in these campaigns was rooted in local grievances and social conflicts, which predated the national campaigns and
did not cease when the Congress agitations were called off.10
This discordance in the timing and objectives of the Congress agitations and the particular struggles which constituted them manifested itself in the uneven and variable
response of peasants and workers to the Congress. Not surprisingly, perhaps, workers who were drawn readily into noncooperation sometimes remained aloof from
civil disobedience; towns which appeared insurrectionary in the early 1930s remained relatively quiescent in the early 1940s. Noncooperation flourished among
various workingclass groups in Bengal11 but civil disobedience drew a more muted response, while labour, it appears, 'remained largely irrelevant to the Congress
strategy'.12 The strikes of 1920–22 in Jamshedpur fed conveniently into the noncooperation campaign, but civil disobedience failed as workers ignored the satyagrahis
and achieved record levels of production,13 but the Quit India
9 The policies of the Congress Ministry in Bombay towards labour are extensively documented in the papers of the Government of Bombay's Home (Special) Department. See
especially, GOB, Home (Special), Files 550 (25) IV A of 1938, 550 (25) IIIA of 1938 and 550 (25)B of 1938, MSA; see also AICC Papers, File PL2 of 1938, especially J. Nehru to B. G.
Kher, 5 August 1937, p. 583, NMML. For the case of Ahmedabad, see ATLA Papers, File 3, part 2, microfilm copy, reel 9, NMML; and also, Patel, The Making of Industrial
Relations, pp. 126–36. For the Madras Presidency, see Murphy, Unions in Conflict, pp. 148–219; Ramaswamy, The Worker and His Union, ch. 1; on Bihar, see D. Simeon, The
Politics of Labour under Late Colonialism: Workers, Unions and the State in Chota Nagpur, 1928–1939 (New Delhi, 1995), chs. 7–9.
10 This commonplace of the 'Cambridge school' in the 1970s has become a shibboleth of the 'subaltern school' in the 1990s.
11 Rajat K. Ray, 'Masses in Politics: The NonCooperation Movement in Bengal, 1920–22', IESHR, 11:4 (1974), 370 ff. See also, Basu, 'Workers' Politics in Bengal', ch. 5; S. Gourlay,
'Nationalists, Outsiders and the Labour Movement in Bengal During the NonCooperation Movement, 1919–1921', in Kapil Kumar (ed.), Congress and Classes: Nationalism, Workers
and Peasants (Delhi, 1988), pp. 34–57.
12 T. Sarkar, 'The First Phase of Civil Disobedience in Bengal, 1930–31', Indian Historical Review, 4:1 (1977), 94.
13 Vinay Bahl, 'TISCO Workers' Struggle, 1920–28', Social Scientist, 10:8 (1982), 44, fn. 34.

Page 269
14
campaign registered an impressive and dramatic response from the town's industrial workers.
The relationship between the Congress and its popular following was as problematic at the time as it has later proved elusive to historians. Despite the scholarly
attention paid to the subject, the past two decades have represented an interpretative Dark Age in the historiography of Indian nationalism. Although there have been
numerous interesting and illuminating studies, they have been characterized by a reluctance to address the relationship between the Congress and its fluctuating and
intermittent popular following, and, therefore, an inability to adequately explain its popular appeal. The protean character of the Congress and the plurality and diversity
of its followings, and the complex social formations which constituted them, has made it difficult to define and analyse this relationship. But in part its elusiveness has
also owed something to the ways in which it has been conceptualized.
The question first posed by Tilak in a newspaper column in 1895: 'Whose is the Congress? Of the Classes or of the Masses?'15 has, thus, led inexorably to a
historiographical stalemate. It is neither satisfactory to assume that nationalism was a reflexive response to, and natural outcome of, colonial exploitation nor to attribute
the growth of popular support for the Congress to a mass awakening from traditional passivity. Not only are the working classes thus perceived as a natural
constituency for the Congress, but this diffusionist approach also takes the political initiative out of the hands of the actors whose behaviour it seeks to explain. In a
Marxist variant, the nationalist movement was seen as serving a particular mix of class interests. While broadly appearing to work in coalition, their final shape was
invariably determined by the 'real' bourgeois character of the movement. The masses were preparing, and being prepared, for their own betrayal.
The 'mass awakening' thesis could now be extended into a 'betrayal' thesis. The ambivalence of the Congress towards its popular following has sometimes been
interpreted as an affirmation of its bourgeois character. The object of the Congress, in this view, was to control the masses and to divert them from their revolutionary
goals. But it is scarcely satisfactory to inveigh against the Congress for failing to lead a mass movement in a final assault against the colonial state. For if its leadership
proved inadequate to the task, and such an outcome was a realizable
14 N. Mansergh (ed.), The Transfer of Power, 1942–47, vol. II: Quit India, 30 April–21 September 1942 (London, 1971), especially Documents nos. 600–2, 612, 636, 650 and 672.
15 Mahratta, 27 October 1895, cited by Johnson, Provincial Politics and Indian Nationalism, p. 120.

Page 270
16
possibility, it is perhaps more pertinent to ask why the Congress was not replaced by a more effective popular agency of revolt. If 'workers, peasants and the urban
petty bourgeoisie' simply 'waited in vain' for the emergence of a 'revolutionary leadership',17 we might usefully ask why they did not manifest a greater impatience.
In reaction to the emphasis of the Cambridge school on factions competing for power and the squabbles for spoils, its critics sought to offer in its place the ideological
and cultural dimension of nationalism. But the old paradigm of 'mobilization', with all its limitations, continued to haunt their enquiries into the nature of popular
movements. Historians sought to reach the masses primarily through nationalist propaganda and the class interests of the peasantry. They focused on the divergence
between the latter and the bourgeois character of the Congress. The peasants were allowed to make their own history – if only when they rejected the importunities of
the Congress. But, for the most part, historians were concerned with how the Congress brought 'the people' into its orbit and the masses remained precisely what the
term described – a shadowy, undifferentiated lump which was faded into the background, and only intermittently brought to the fore by political activists.
For all their (sometimes acrimonious) differences, what all these approaches share in common is the attempt to comprehend the relationship between the Congress and
the masses as a process of mobilization. Indeed, this paradigm of mobilization has severely limited the interpretative scope of the historiography of Indian nationalism.
Arguments about political mobilization have perhaps been overly concerned with how parties spread their message and propagate their ideas and rather less with how
they were received or reinterpreted. Even those historians who are concerned with the latter have tended to measure the success of political mobilization by trying to
assess how directly it spoke to specific class interests or how literally it was lodged in an innate cultural idiom. But there is a certain circularity in such reasoning:
political mobilization is seen to have been successful when it attracted widespread support; the fact that it attracted widespread support is then taken as proof that its
rhetoric was appropriate whether in terms of class or culture.
16 For instance, Pandey, The Ascendancy of the Congress in Uttar Pradesh; D. Hardiman, Peasant Nationalists of Gujarat: Kheda District, 1917–1934 (Delhi, 1981). By the late
1930s, according to Pandey, 'there was, however, no longer any question of [the Congress] assigning a primacy . . . to political struggle for the advancement of consciousness and
the achievement of radical change in society, within which the nationalist leadership had been caught up for a short time'. G. Pandey, 'Congress and the Nation, 1917–1947', in R.
Sisson and S. Wolpert (eds.), Congress and Indian Nationalism: The PreIndependence Phase (Berkeley, 1988), pp. 130–1.
17 Ranajit Guha, 'On the Historiography of Colonial India', in Ranajit Guha (ed.), Subaltern Studies, vol. I, pp. 6–7.

Page 271
Faced with the 'elitist' bias of the historiography, whether intended or not, some historians set out in search of 'the relatively autonomous culture or mind' of the
subaltern.18 The notion of popular culture was now deployed to rescue the subaltern from the condescension of the historian. But, given the provenance of the notion
itself, it was more likely to mire her more deeply in it. In view of the diversity of the conditions and contexts in which the 'subordinated' operated, let alone the
indeterminacy of the boundary which separated them off from the 'dominant', this search for the collective 'mind' of the subaltern classes was bound to prove as elusive
as the Holy Grail. For where, after all, was the historian to find the common denominator by which these diverse elements could be identified collectively as a 'popular
culture', except in his own bag of tricks? Its definition was characterized by its arbitrariness, which in turn was liable to introduce an inherently circular analytic
procedure. Once the assumption was made that a 'popular culture', at one level, subsumed differences of class and occupation, age and gender, caste and religion,
language and region, it became possible to hold various and conflicting forms of social action within it. As soon as it was assumed that the 'subaltern mind' operated
primarily in a religious idiom, for instance, historians privileged metaphors and practices of religion in their accounts of the resistance of the subordinated.19
For an insight into the 'subaltern mind', moreover, historians have had to depend largely on the discourse of the dominant classes, in which, of course, the
characteristics of those whom they seek to control are delineated; and it need hardly be said that their perceptions of the 'subalterns' was an integral part of the
conditions of their domination.20 Studies of local responses to the Congress, rooted in their 'popular culture', have often found themselves resorting to rather traditional
18 S. Sarkar, 'The Conditions and Nature of Subaltern Militancy: Bengal from Swadeshi to NonCooperation, c. 1905–1922', in Guha (ed.), Subaltern Studies, vol. III, p. 277.
19 See, for instance, David Hardiman's account of how Kunvarji Mehta mobilized the adivasis in the Bardoli and Valod talukas of Surat district. After his initial efforts failed, Kunvarji
hit upon 'a new strategy'. He sought the name of the adivasis' most revered god, Simariyo Dev. In fact, 'he appears to have obtained the information in a rather confused form' – that is,
he got the wrong name and in any case thought Simariyo Dev was two gods. He began to tell the adivasis that Gandhi had been chosen by Simadiya Dev and Shiliya Dev as their
replacement, and that he was an avatar of Ram and Krishna. Although Kunvarji mistook the god's name, the religious adivasis believed him, and as a consequence began to shout
'MahatmaGandhinijai'. D. Hardiman, The Coming of the Devi (Delhi, 1987), pp. 168–9; see also S. Amin, 'Gandhi as Mahatma: Gorakhpur District, Eastern UP, 1921–2', in Guha (ed.),
Subaltern Studies, vol. III, pp. 1–61.
20 Thus, rumours and stories which indicate how the peasants perceived Gandhi and invested him with a magical status and divine powers are drawn from local newspapers, official
accounts, sometimes even the memoirs of Lord Ronaldshay. See Amin, 'Gandhi as Mahatma', pp. 1–57; Hardiman, The Coming of the Devi; Sarkar, 'The Conditions and Nature of
Subaltern Militancy'.

Page 272
counters of explanation – material discontent, class and colonial oppression and the diffusion of nationalist ideas by political agents – when they have had to explain the
political appeal of Gandhi and the Congress.21 What popular culture provides in these accounts is often the religious idiom in which these supposedly simple,
unsophisticated folk are said to interpret and express their political preferences and, thus, the appeal of their leaders.22 But in view of the prior assumption that 'the
consciousness of the subordinated' embraces, as Amin puts it, 'the complex relationship between popular culture, religiosity and inchoate political consciousness',23 this
is less remarkable than it is tautological. Moreover, to the extent that they insist upon the autonomy of subaltern politics, these accounts would tend to marginalize the
role of the Congress and its leadership. Yet the subject they define and the evidence they present on popular responses to the nationalist movement demonstrates the
apparently dominating influence of Gandhi and the Congress in local political discourse.
This emphasis on popular culture and the vigorous search for the 'subaltern mind' which it has unleashed has also served to stultify our understanding of popular
political responses to the nationalist movement. If in the 'animal politics' of the Cambridge school,24 there was little scope for grasping the terms on which the Congress
engaged with the working classes or the rural poor, the subaltern insistence on 'the autonomy of peasant insurgency'25 can scarcely allow for the systematic
investigation of the relationship between the Congress and its popular follow
21 Amin, 'Gandhi as Mahatma'; Hardiman, The Coming of the Devi. However, Amin's most recent and impressive work appears successfully to overcome the limitations imposed by
the notion of 'popular culture' upon 'Gandhi as Mahatma': see S. Amin, Event, Metaphor, Memory: Chauri Chaura, 1922–1992 (Berkeley, 1995).
22 Thus, according to Hardiman, the reason why the Patidars of Kheda, unlike their counterparts in Surat did not see visions of Gandhi at the bottom of their wells, was because they
were 'generally more sophisticated than those of South Gujarat'. Hardiman, The Coming of the Devi, pp. 50–1, n. 92. Significantly, it is at this point of his argument that Hardiman seeks
to set his findings beside those of Amin, 'Gandhi as Mahatma'.
23 S. Amin, 'Agrarian Bases of Nationalist Agitations in India: An Historiographical Survey', in D. A. Low (ed.), The Indian National Congress: Centenary Hindsights (Delhi, 1988), p.
105. The religiosity of the 'masses' was, of course, a recurrent theme in colonial discourse. It could serve conveniently to explain both the fanaticism of the mob as well as the appeal of
nationalist movements directed against the British. 'Those who wish to rouse the unpolitical masses', explained the Government of India in 1930, 'are able to play upon their feelings by
appealing to the interests which are intense and vivid in their lives. First among these interests must be placed the power of religion.' Government of India's Despatch on Proposals for
Constitutional Reform, dated 20th September 1930, PP, 1930–1, vol. XXIII, p. 695.
24 T. Raychaudhuri, 'Indian Nationalism as Animal Politics', Historical Journal, 22:3 (1979), 747–63.
25 R. Guha (ed.), Subaltern Studies, 6 vols. This notion of 'autonomy' has sometimes been combined with the contradictory thesis of betrayal.

Page 273
26
ing. In some versions, this relationship is not even problematized and its investigation is postponed or ruled out altogether. Thus, Amin exhorts historians to pursue the
'investigation of peasant political activity within nationalist activity', rather than 'concentrate on the relationship between the two'.27 However, since peasant politics
'within' nationalism entails a relationship, it is difficult to see why historians should, or, indeed, how they could, avoid the latter. The interpretative costs of the failure to
address this relationship is reflected in the historiographical and conceptual stagnation of the past two decades.
This chapter explores the relationship between the Congress and its popular following, in particular among the working classes. It will proceed by abjuring the notion
(or more accurately, the assumption) of 'popular culture'. No inherent solidarity will be attributed to the 'subordinate classes' for the conditions of their subordination
divided them more fundamentally than it united them. Rather, it will be argued that perceptions of mutuality among them, indeed the political vocabulary in which it was
expressed, was produced by the specificities of a particular intellectual and political context. Their solidarities, their coming together, should be observed and analysed;
they cannot be assumed. They merit investigation and demand explanation.
II
The success of political movements, their ability to attract widespread support, would depend to a large extent upon how far their rhetoric and programme could
penetrate the assumptions and ideologies of their putative followers, not simply how far their rhetoric spoke to specific and given social interests or political cultures.28
To achieve this fully, their political programme would have to do far more than express or reflect the specific interests of the working classes or find a resonance in the
daily struggles of workplace and neighbourhood. It would have to offer workers a means of comprehending and contextualizing their immediate situations, which were
necessarily diverse, while also indicating how they might be realistically transformed. Their programme would have to encapsulate and offer a social and political
alternative and enunciate a method by which it might be realized. In other words, their programmatic
26 Some recent work has, on the other hand, focused upon nationalist discourse while it has tended to neglect the wider social response to its programmes. See P. Chatterjee,
Nationalist Thought and the Colonial World: A Derivative Discourse (London, 1986); and P. Chatterjee, The Nation and Its Fragments: Colonial and PostColonial Histories
(Delhi, 1994).
27 Amin, 'Agrarian Bases of Indian Nationalism', p. 100.
28 See the seminal essay by G. Stedman Jones, 'Rethinking Chartism', in Stedman Jones, The Languages of Class, pp. 90–178.

Page 274
framework would have to be sufficiently flexible to be meaningful within particular and diverse contexts, while at the same time enabling workers to interpret and
develop this ideology in new directions and to forge fresh political strategies from within its parameters. Political ideologies rarely teach or convert, even if political
activists are fond of confessing to Pauline conversions. The effectiveness of political programmes is perhaps best judged, not in terms of how far they can squeeze
themselves into what their protagonists may perceive as an appropriate class or cultural idiom, but in terms of their dynamic and creative potential in the hands of their
adherents. In practice, the Congress was faced with the task of doing nothing less than defining the nation in terms which would find a continuing echo in the daily
struggles of the working classes.
These desiderata, never easy to achieve over any length of time, were to prove particularly difficult for the Congress in its relationship with the working classes. The
Congress neither became nor sought to become a party of the working class. This is not because, as was often implied in contemporary discourse and subsequently
too readily accepted by later historians, it was a bourgeois party. Conversely, it should not be taken for granted that a political programme for the working classes
could only be formulated within the language of socialism as if there was an organic – rather than a historical or a contingent – connection between them. It would be
misleading to assume that a political ideology – whether nationalism or socialism – has a given social embodiment or yet more metaphysically, that it expresses the
'consciousness' of a particular class.
In any case, the working classes represented less a single, coherent social category than a kaleidoscope of social groups. 'Factory' or 'industrial workers' composed a
small minority of the Indian workforce. But they were recruited largely from the smallholding peasantry. Frequently, they retained over several generations close
connections with their village base and drew upon its resources.29 Nor did they form a homogeneous social group. Industrialization often exacerbated the differences
between them.30 Labour for the coal mines and the tea gardens was recruited not only from the impoverished peasantry but also from groups of 'tribals',31
30 Ibid., passim; see ch. 3 above.
31 R. P. Behal and P. P. Mohapatra, 'Tea and Money Versus Human Life: The Rise and Fall of the Indenture System in the Assam Tea Plantations', in E. Valentine Daniel, H. Bernstein
and T. Brass (eds.), Plantations, Proletarians and Peasants in Colonial Asia (London, 1992), pp. 142–72; Simmons, 'Recruiting and Organising an Industrial Labour Force in Colonial
India'; R. Das Gupta, 'Plantation Labour in Colonial India', in Daniel et al. (eds.), Plantations, Proletarians and Peasants, pp. 172–98; P. P. Mohapatra, 'Coolies and Colliers: A Study of
the Agrarian Context of Labour Migration from Chotanagpur, 1880–1920', Studies in History, new series, 1:2 (1985), special issue, Essays in Agrarian History: India, 1850 to 1940,
edited by S. Bhattacharya, 13–42; C. Bates and

Page 275
32
and they were sometimes indentured to serve in the British plantation colonies overseas. By the early twentieth century, a growing number of rural households
required offfarm earnings in order to maintain themselves, but most sought employment on the land, some travelled for the harvest season to the neighbouring district
or sought wage employment in nearby towns. It was a relatively small proportion who became longdistance migrants to industrial labour in the large cities.33 The
population of most small towns, local market or administrative centres, was drawn from the neighbouring villages. Here, the working classes were employed in manual
labour, in service occupations, small workshops or petty trade.34 Such people were identified in the large cities as the 'casual poor' or, more recently, lumped together
in the 'informal sector'. In fact, casual and permanent factory labour described overlapping social spheres. They did not constitute separate labour markets
distinguished by special social characteristics peculiar to each.35 Similar, certainly overlapping, social processes by which 'tribals' or 'artisans' were turned into
'peasants', or each of them into 'workers', had been operating in Indian society for over two centuries. Significantly, the processes by which these working classes
were constituted intermeshed. As they did so, they traversed a wide expanse of Indian society and diverse social forms. Their diversity greatly complicated the task of
developing a rhetoric or a political programme with a significance for the working class.
It is interesting to speculate that the Congress might most readily have
M. Carter, 'Tribal Migration in India and Beyond', in G. Prakash (ed.), The World of the Rural Labourer in Colonial India (Delhi, 1992), pp. 205–47; R. Das Gupta, 'Migrants in Coal
Mines: Peasants or Proletarians, 1850s–1947', Social Scientist, 151 (December 1985), 18–43; R. Ghosh, 'A Study of the Labour Movement in the Jharia Coalfield, 1900–1977',
unpublished Ph.D. thesis, Calcutta University, 1992.
32 H. Tinker, A New System of Slavery: The Export of Indian Labour Overseas 1830–1920 (London, 1974); J. Breman, Taming the Coolie Beast: Plantation Society and the Colonial
Order in Southeast Asia (Delhi, 1989); Marina Carter, Servants, Sirdars and Settlers: Indians in Mauritius, 1834–1874 (Delhi, 1995).
33 G. Omvedt, 'Migration in Colonial India: The Articulation of Feudalism and Capitalism by the Colonial State', Journal of Peasant Studies, 7:2 (1980), 185–212; Mohapatra, 'Coolies
and Colliers'; Brahma Nand, 'Agricultural Labourers in Western India: A Study of the Central Division Districts of the Bombay Presidency During the Late Nineteenth and Early
Twentieth Century' Studies in History, new series, 1:2 (1985), 221–46; Pradipta Chaudhury, 'Labour Migration from the United Provinces, 1881–1911' Studies in History, 8:1 (1992), 13–
42; J. Pouchepadass, 'The Market for Agricultural Labour in Colonial North Bihar, 1860–1920', in M. Holmstrom (ed.), Work for Wages in South Asia (Delhi, 1990), pp. 11–27;
Chandavarkar, The Origins of Industrial Capitalism, ch. 4; Das Gupta, 'Migrants in Coal Mines'; Sen, 'Women Workers in the Bengal Jute Industry', especially chs. 1–3.
34 N. Gooptu, 'The Political Culture of the Urban Poor in North India, 1920–47', unpublished Ph.D. dissertation, University of Cambridge, 1991; Vijay Prashad, 'Chuhras and Colonialism',
unpublished paper.
35 Chandavarkar, The Origins of Industrial Capitalism, especially ch. 3.

Page 276
been able to move a massive broadly based and inclusive political alliance against British rule in the 1870s and 1880s, before or immediately after its foundation. In this
period, Indians were most completely excluded and distanced from the political structure of colonial rule.36 At the same time, the bankrupt colonial state was driving
deeper into Indian society in its feverish search for revenues, seeking new sources of taxation, bearing down upon social groups which had previously been allowed a
relatively free hand. To hypothesize the existence of conditions conducive to such a broadbased nationalist movement in the 1870s and 1880s is not to suggest that it
was imminent or even practicable. Certainly, the early nationalists were 'indifferent or hostile to the efforts being made to ameliorate the conditions of work of the
factory workers' and even the radicals among them, including Tilak, are found to have been 'insensitive and even opposed to the cause of labour'.37 Subsequently, the
changing policies of the state made this impossible. From the 1880s onwards, the policies of the state served to fragment what earlier might have provided the basis for
a broad front organized around a political programme, both radical and inclusive. Political and bureaucratic reform not only began to remedy the narrow and exclusive
character of the political system but it also worked in favour of some and to the disadvantage of others. It helped to bring caste and communal differences emphatically
into the political arena. At the same time, the social and economic policies of the state, while claiming to encourage the social transformation of the economy, served to
protect the agrarian base from the forces of capital, and contributed increasingly to the disintegration of the broad alliance that might hypothetically have been put
together and moved laboriously against the structure of colonial rule. In the twentieth century, the difficulties inherent in seeking to reconcile class conflict within the
framework of a representative nationalist movement were increasingly intensified.
From the late nineteenth century onwards, a whole range of frustrated 'bourgeois' interests were being driven into the Congress. But they were a disparate lot, from
large merchant shroffs and millowners to liquor distillers in South India and stultified rural capital of diverse sorts. Most of these interests were divided in their response
to Congress campaigns. Nonetheless, between the two world wars, the Congress came to establish closer links with Indian capitalists and derived both power and
resources from these connections.38 At the same time, after 1918, the working
36 D. Washbrook, The Emergence of Provincial Politics: Madras Presidency, 1870–1920 (Cambridge, 1976), ch. 2.
37 Bipan Chandra, The Rise and Growth of Economic Nationalism in India: Economic Policies of the Indian National Leadership, 1880–1905 (New Delhi, 1966), p. 330.
38 C. Markovits, Indian Business and Nationalist Politics, 1931–39: The Indigenous Capitalist Class and the Rise of the Congress Party (Cambridge, 1985); A. D. D. Gordon,
Businessmen and Politics: Rising Nationalism and a Modernising Economy, 1918–1933 (New Delhi,

Page 277
classes established an increasingly prominent presence in the political arena. Strikes became more frequent, involved more workers and occurred on a larger scale than
ever before. In any case, at the level of local and secondary leadership, there were a growing number of Congressmen who were drawn, sometimes by conviction,
sometimes by expediency, into acting as spokesmen for labour in industrial disputes. At times, national leaders stepped forward as champions of labour in specific
cases, notably those of Gandhi in Ahmedabad in 1918–2039 and Bose in Jamshedpur in the late 1920s.40 Indeed, the extension of the franchise made it increasingly
important for politicians to pay closer attention to this constituency. Above all, it was imperative for the Congress to demonstrate its representative character and this
depended in part upon its ability to bring the working classes into the nationalist fold. As labour registered an increasingly prominent presence in the political arena, the
Congress developed closer links with a variety of commercial interests and especially with industrial capital. Fresh life was breathed into dilemmas and contradictions
which had been with the Congress since 1885: how was it to attract workingclass support without dividing its own ranks? How could it champion labour's cause
without becoming the prisoner of the working classes?
At every stage of its development, the effectiveness of the Congress as a bargaining agent with the colonial state and, therefore, its ability to attach significant local
interests, rested on its claim to represent the nation as a whole. To fortify this claim, the Congress leadership recognized that the party would have to speak with one
voice. To demonstrate its representative character, it was essential for the Congress to expand and diversify its social base. But to speak with one voice, it had to ride
the range on its socially diverse constituents and marshal them behind an inevitably thin facade of unity. This was always liable to narrow the range of options before the
Congress as a movement of opposition to colonial rule. Since it orchestrated a complex and fragile network of alliances between disparate
1979); A. Mukherjee, 'The Indian Capitalist Class and Foreign Capital, 1927–47', Studies in History, 1:1 (1979), 105–48; Bipan Chandra, 'The Indian Capitalist Class and Imperialism
before 1947', Journal of Contemporary Asia, 5:3 (1974), 309–26; Bipan Chandra, 'Jawaharlal Nehru and the Indian Capitalist Class in 1936', Economic and Political Weekly, 10:33–5
(1975), 1307–24; S. Sarkar, 'The Logic of Gandhian Nationalism: Civil Disobedience and the Gandhi–Irwin Pact (1930–31)', Indian Historical Review, 3:1 (1976), 114–46; B. Chatterji,
Trade, Tariffs and Empire: Lancashire and British Policy in India, 1919–1939 (Delhi, 1992).
39 M. Desai, A Righteous Struggle: A Chronicle of the Ahmedabad Textile Labourers' Fight for Justice (Ahmedabad, 1951); E. Erikson, Gandhi's Truth: On the Origins of Militant
NonViolence (London, 1969); S. Patel, The Making of Industrial Relations: The Ahmedabad Textile Industry, 1918–1939 (Delhi. 1987), ch. 3.
40 Simeon, The Politics of Labour under Late Colonialism, chs. 2–3; V. Bahl, 'TISCO Workers' Struggle, 1920–28'; V. Bahl, The Making of the Indian Working Class: The Case of the
Tata Iron and Steel Company (Delhi, 1995).

Page 278
social groups and conflicting political interests negotiated on terms of limited advantage to each, it was committed to a strategy of glossing over what might divide and
concentrating upon what was sure to unite its followers.41 Programmes, propaganda and policies designed to 'harness capital to our side' were unlikely to attract labour
and, further, they could lead the Congress to act antagonistically to the interests of the working classes. The constraint of speaking with one voice to London reduced
the Congress to a babel of tongues when it spoke to the working classes. The price of accommodation at the top was often proliferating confusion at the base.
Moreover, there were several and persistent attempts to challenge the Congress's attempts to define or indeed represent the nation. Most obviously, the British
repeatedly questioned the Congress claim to speak for the nation as a whole. Indeed imperial ideologues frequently went one step further and denied that there was, or
ever had been, or indeed ever could be an Indian nation. In the 1880s, John Strachey had enunciated the doctrine that 'the first and most essential thing to learn about
India – [is] that there is not, and never was an India . . . possessing . . . any sort of unity'.42 If the subcontinent was teeming with countless distinct nationalities, it
followed that the Congress could not possibly represent them all, let alone weld them into a single harmonious whole. Whereas, in the 1880s, colonial ideologues had
dismissed the Congress as merely a clerisy of the Westerneducated, who represented no one other than themselves, by the 1920s, they portrayed it as the instrument
of the political agitator, seeking to manipulate the illiterate and the innocent, if contented, masses for their own selfish purposes. The Congress was unrepresentative,
not only because 'the politicallyminded class' was divorced from the 'real people', but also because it had failed to win and retain the confidence of the Muslims.43 In
this apparent failure to represent the Muslims, colonial discourse fixed on one of its perennial, indeed defining, themes. For colonial rulers feared above all that where
the forces of darkness and superstition reigned, the masses, 'when captured by those who wilfully or
41 A. Seal, The Emergence of Indian Nationalism: Competition and Collaboration in the Later Nineteenth Century (Cambridge, 1968); Johnson, Provincial Politics and Indian
Nationalism, ch. 1.
42 John Strachey, India (London, 1888), p. 5. See also Sir Auckland Colvin, Audi Alterem Partem: Being Two Letters on Certain Aspects of the Indian National Congress Movement
(Simla, 1888); Richard Temple, India in 1880 (London, 1880); Valentine Chirol, Indian Unrest (London, 1910); Sir Verney Lovett, A History of the Indian Nationalist Movement
(London, 1920); P. C. Bamford, Histories of the NonCooperation and Khilafat Movements (Delhi, 1925; reprinted Delhi, 1974).
43 For an excellent, recent account of colonial views of the Congress, see W. H. MorrisJones, 'If It Be Real, What Does It Mean?: British Perceptions of the Indian National Congress',
in Sisson and Wolpert (eds.), Congress and Indian Nationalism, pp. 90–118.

Page 279
44
recklessly pour jars of paraffin upon their ignorance and credulity, can break into fanatical fury'. Not only did colonial ideologues take it for granted that this
fanaticism sprang from sources which lay outside the universe of reason and, therefore, beyond their grasp, but they were also apprehensive that the fury which it
unleashed would exceed their power to control. This innate unreasonableness, this inherent fanaticism made it imperative for the British to rule India. But it could also
swiftly render India ungovernable. That India might be granted independence at all could be nothing other than a tribute to the British achievement in forging a nation
from such unpromising and fissiparous material; for some, it would constitute the 'fulfilment' of British rule.
The oldest apology for British rule, which, indeed, has outlasted it, was that in a land with a thousand definitions of the word, only the Collector could rule impartially
and was therefore indispensable. Colonial suspicions were then enshrined in the principle of special and communal representation. Similarly when Ambedkar declaimed
against what Gandhi and the Congress had done to the untouchables, he was interrogating the credibility and content of swaraj.45 The 'twonations theory', one Hindu,
the other Muslim, proved to be the most devastating blow to the definition of a single nation, although there was no innate reason why this should have been the case.46
Within the fold of the Congress, too, there were ambiguities about the character of the nation which could only be resolved by refusing to define. Congress rhetoric in
the UP acquired a Hindu idiom and a revivalist tone which alienated Muslims.47 Brahmins and 'nonBrahmins' in the Deccan were agreed that the untouchables must be
kept firmly in their place.48 The association of the Congress with locally dominant peasants or urban capitalists often suggested that class divisions might obtrude upon
equal rights in the nation. Thus, during the noncooperation and civil disobedience campaigns, numerous businessmen grew anxious that the Congress agitations would
'create a feeling of disregard for authority' far beyond
44 Sir Verney Lovett, A History of the Indian Nationalist Movement, pp. 247–8, cited by MorrisJones, 'British Perceptions', p. 104.
45 B. R. Ambedkar, What Congress and Gandhi Have Done to the Untouchables (Bombay, 1945).
46 F. C. R. Robinson, Separatism among Indian Muslims: The Politics of the United Provinces' Muslims, 1860–1923 (Cambridge, 1974); D. Page, Prelude to Partition: The Indian
Muslims and the Imperial System of Control (Delhi, 1982); A. Jalal, The Sole Spokesman: Jinnah, the Muslim League and the Demand for Pakistan (Cambridge, 1985); G. Pandey, The
Construction of Communalism in Colonial North India (Delhi, 1990); J. Chatterji, Bengal Divided: Hindu Communalism and Partition, 1932–1947 (Cambridge, 1994).
47 C. A. Bayly, 'Patrons and Politics in Northern India', in J. Gallagher, G. Johnson and A. Seal (eds.), Locality, Province and Nation: Essays on Indian Politics (Cambridge, 1973), pp.
29–68; Pandey, The Ascendancy of the Congress, ch. 5.
48 GOB, Home (Special), File 363 (5) of 1928, MSA.

Page 280
49
the capacity of even a swaraj government to remedy and, in 1930, even the normally astute Ambalal Sarabhai predicted in panic that 'Bolshevik propaganda will find
fertile soil in India', if things continued as they were.50
It was in this context of ambiguity, contradiction and conflict that the rhetoric and programme of the Congress began to take shape. Crucially, the very object or slogan
of swaraj, selfrule, was redefined and used by Gandhi to mean all things to all men: from dominion status to independent parliamentary government, from an
expression of the grievances of Khilafat Muslims to the promise of a golden age, from devotion to the charkha to selfdiscipline and inner selfrealization. The meanings
of swaraj could range from the personal to the political, from the spiritual to the constitutional, from millenarian promise to immediate and concrete objectives. It
offered flexibility but remained far too opaque to establish realizable connections between the sphere of daily social conflict and the sense of a programmatic politics.
Its strength was that it offered a receptacle into which the aggrieved might pour whatever they chose. Its persistent weakness, however, was that it seemed to offer a
great deal in general and nothing very much in particular. It offered a programme which was not so much flexible as formless, and highly vulnerable to competing
programmes and ideologies.
Although Congress was not, as Nehru once pointed out, 'a labour organization and to expect it to act as a pure labour organization is a mistake',51 the terms which
Congress offered in its attempt to pull the working classes into a nationalist alliance were not fashioned to consistently encourage them. Indeed, for the Congress, the
imperative of placing its representative credentials before the colonial state was always likely to narrow its freedom of manoeuvre in developing a programme and
rhetoric which might embed itself within the political assumptions and vocabulary of the working classes. Since the Congress was always less a political party than a
field of political action and an arena for the brokerage of particular interests, there was a bewildering variety of publicists who spoke in its name. The audiences which
they addressed were similarly diverse and the interests which they represented, and which the Congress, whether at the local, provincial or the allIndia level, sought to
coalesce, were heterogeneous and divided. As a result, it is impossible to capture the rhetoric of nationalism in a single voice or a consistent and coherent social
message. Indeed, its rhetoric was often cacophonous. But it is not merely by generalization of sweeping and magisterial simplicity
49 Lalji Naranji to Thakurdas, 28 March 1930, Thakurdas Papers, File 91, NMML.
50 Ambalal Sarabhai to Thakurdas, 28 March 1930, ibid, NMML.
51 J. Nehru to D. B. Kulkarni, 10 September 1929, AICC Papers, File 16, 1929, NMML.

Page 281
that we might identify its dominant forms. It is possible to discern the broad parameters within which the Congress sought to address and lead the working classes.
Gandhi's rhetoric came increasingly to inflect nationalist discourse and to provide it with its distinct vocabulary. Gandhi's political rhetoric will, thus, be used here as a
point of entry to the examination of the relationship between the Congress and the working classes.
The rhetoric of nationalism rarely sought, least of all before its heterogeneous audiences, to define a particular workingclass interest, or to speak, even within a given
context, exclusively to its immediate objectives. Of course, the Congress did not readily admit the distinction between the interests and aspirations of the working
classes and those of the nation, and when it did so, it was usually not in order to represent the former but to control and constrain them in the name of the latter. From
a perspective which assumed the coherence of a workingclass interest or insisted on distinguishing it from national aspirations, the Congress appeared as the
purveyors of the doctrine of class conciliation.
Moreover, the relationship of these politicians and publicists to their popular base, as well as their influence within the Congress as a whole, was often determined by
local circumstances and, especially, their own place within local structures of power. Yet, while many publicists addressed the working classes in the name of the
Congress, the latter engaged or distanced itself from these local movements and publicists, according to its perception and evaluation of their social and political
character. In the 1930s, as the Congress became increasingly centralized, and the unitary structure of its formal organization was strengthened, only those could
champion the particular interests of the working classes who had nothing to lose by alienating local party bosses and propertied and commercial magnates.
There were, in particular, two recurrent themes in nationalist rhetoric in relation to the working classes. First, considerable emphasis was laid upon the partnership
between labour and especially Indian capital. This theme had a long history. Tilak, who had once opposed factory legislation, on the grounds that the millworkers of
Bombay preferred to work long hours, advised the latter in 1919, in the midst of a wave of strikes, and on a rare occasion when he addressed a workers' meeting, that
they should sink their differences with their employers and combine against the British.52 Lala Lajpat Rai, in his presidential address to the AllIndia Trade Union
Congress, added a plea for equity to this doctrine of class harmony. 'Labour and capital', he declared 'must meet on equal ground
52 Cashman, The Myth of the Lokamanya, p. 185.

Page 282
53
and join hands to develop Indian industries . . . at present neither the government nor the capitalist is disposed to treat the worker fairly and equally.' Similarly, Joseph
Baptista, once a Fabian in Cambridge and later a Tilakite in India, in his welcoming address, looked forward to 'the higher ideal of partnership', without which 'the
wellbeing of workers will never be secured'. For workers and capitalists, he observed, were 'partners and coworkers, and not buyers and sellers of labour. They are
all engaged in promoting the wellbeing of society.' But this ideal of partnership, he argued, would only be realized 'by the power of unions, strikes and boycotts'.54
To Gandhi, of course, partnership was scarcely a novel idea and it was a theme to which he was repeatedly to return. 'When millhands learn to identify themselves
with the rest of the millowners', Gandhi told the Ahmedabad millworkers in 1920, 'they will rise and with them will rise the industries of our country.'55 While
suggesting thereby that the interests of labour and capital were intimately linked, Gandhi also eschewed the notion that strikes may be used as a threat, or as an
instrument of coercion. But it would be misleading to conclude that Gandhi was merely offering a mystical rendition of the truisms and dogma of the capitalist classes.
In this instance, as in many others, Gandhi intervened in political discourse to transform it and, indeed, he developed and deployed the existing discussion in a strikingly
fresh direction.
During the 1918 strike, Gandhi developed a code of ethics, a philosophy of practice, to govern industrial relations and he elaborated it over the next two decades. It
was founded on a supposition of the interdependence of labour and capital. In 1918, Gandhi sometimes lamented the strike, rather whimsically, as a family dispute. In
so doing, he referred at the same time to the interdependence of the interests of labour and capital, rich and poor, majur and seth, as well as to the fact that Ambalal
Sarabhai, the President of the Millowners' Association, was the brother of the workers' leader, Anasuyabehn. Earlier, when, in December 1917 Seth Ambalal had first
sought his advice about how to deal with the weavers' demand for higher wages, Gandhi had told him to 'remove their discontent' by 'binding them with the silken
thread of love'.56 In February 1918, in the midst of the strike, even the Collector of Ahmedabad was led to report to his less sanguine and less credulous superiors in
Bombay that 'a
53
AITUC – Fifty Years On: Documents, vol. I (AITUC, New Delhi, 1973), p. 29.
54 Ibid., p. 12.
55 Young India, 6 October 1920.
56 Gandhi to Ambalal Sarabhai, 21 December 1917, CWMG, vol. XIV, p. 115. Indeed, Gandhi went so far as to read him the lesson that he 'should satisfy the weavers for the sake of
Shrimati Anasuyabehn' who 'has a soul which is absolutely pure', thereby placing Ambalal, as he pointed out, 'under a double obligation: to please the workers and earn a sister's
blessings'.

Page 283
piquant feature of the whole affair' was that the dispute 'partakes of the nature of a family one and unlike many family disputes is conducted with a complete lack of
acerbity'.57 Many years later, in 1925, Gandhi restated his 'ideal' that 'capital and labour . . . should be a great family living in unity and harmony' with the capitalists
serving as 'the trustees' for the moral and material 'welfare of the labouring classes'.58
In seeking to reconcile the interests of labour and capital within the figurative framework of the family, Gandhi argued that workers had a right to their needs but these
could only be properly assessed in terms of the economic condition of the industry or the employers. But he did not dwell too long or define too carefully the criteria by
which the workers' needs or the financial condition of industry might be assessed. Gandhi did insist, however, that the workers' demands, like their means of securing
them, should be 'just'. Thus, workers should 'ask only for what is our right'59 and not seek 'merely to take advantage of the capitalist's position'.60 They were to refrain,
therefore, from making demands 'at will', 'irrespective of the employer's financial condition'.61 In this light, it may be supposed that it was the selfcontrol of the workers
which would set the limit on their demands. But he was 'aware', he wrote in 1921, 'that the labouring classes in India has not yet become enlightened enough to have
the ability to regulate the relations between labour and capital on a just basis'.62 Until that time came, labour was in need of 'disinterested friendship'.63 To such potential
friends of labour, his advice was that they should aim
to elevate the workmen by creating between the two parties a family relationship . . . And to secure this end there is no path like Truth. Mere increase in wages should not satisfy
you; you must also watch by what means they get it and how they spend it.64
It was also, then, the duty of the workers' adviser, in the case of Ahmedabad in 1918 happily himself, to curb his followers, if this became necessary, and to ensure
that their demands were just. The workers' adviser, in Gandhi's view, should also seek to effect a consensus, and persuade the employers of the justice of the workers'
demands. If the employers rejected a 'just' demand, then the claim would have to be submitted to a neutral 'umpire' or even a 'panch' for arbitration, whose
57 Collector of Ahmedabad to Secretary, GOB, Judicial, 25 February 1918, Bombay Confidential Proceedings, Judicial Department, for March 1918, vol. 36, p. 33, OIOC.
58 Young India, 20 August 1925.
59 Ahmedabad Millhands' Strike, Leaflet no. 7, 4 March 1918, CWMG, vol. XIV, p. 236.
60
Young India, 6 October 1920.
61 Desai, A Righteous Struggle, p. 49; CWMG, vol. XIV, p. 233.
62 Young India, 8 June 1921.
63 Ibid., 1 June 1921.
64 Ibid.

Page 284
65
decision would be binding on both sides. Employers who refused to consider the just needs of the workers or, in the case of disagreement, countenance arbitration,
had succumbed to 'the Western or modern devilish kind of justice' in which 'each thus thinks only of himself and is bound not to think of the other'.66 If the workers had
no option but to embark upon a strike, they should act as satyagrahis. 'The workers' struggle', he wrote in 1918, 'depends solely on the justice of their demand and the
rightness of their action'.67 By using just and noncoercive means to struggle for just demands, the workers would ensure that the employers would 'rectify their
mistake'.68 In fact, the satisfaction of a 'just' demand 'will not harm the employers' interests but will do them good'.69 Thus, during the 1918 strike, Gandhi had insisted:
'We can never wish or do ill to the employer, and in every action of ours the idea of their good is also always present. We want to secure the good of the workers
while safeguarding the good of the employers.'70
Gandhi's ruminations on industrial relations may be interpreted as simply an arcane expression of capitalist interests. It is possible to argue – indeed, it has frequently
been suggested – that once Gandhi's rhetoric is stripped of reference to the principles of satyagraha and ahimsa, truth and nonviolence, little remains to distinguish it
from the case which employers throughout the world had argued for over a hundred years. Certainly, employers in various industries in India in the 1920s and 1930s
had sought to defuse workers' grievances – whether wage demands or protests against job losses – by resolutely placing them in the context of the trade conditions, or
what the market could bear. They would warmly welcome the argument that wage demands ought to be governed strictly by the financial condition of the employer
and of the industry as a whole, which in any case was difficult to ascertain without their fullest cooperation. Employers were by no means enthusiastic, not even in
Ahmedabad, about subjecting their business practices and their labour relations to the scrutiny and the judgement of outsiders. At the same time, arbitration
proceedings, which appeared equitable in theory, were quite the opposite in practice, for they tended to favour the stronger side in any dispute – which for the most
part meant the employers. Otherwise, arbitration tribunals, established to effect an agreed and viable settlement, were liable to find that their recommendations were set
aside or simply ignored and their objectives subverted. It has even been suggested that, as the colonial state elaborated arbitration and conciliation procedures in trade
disputes legislation after 1929, they proceeded along lines which were not fundamentally different from those which Gandhi
65 Ibid.
66
Desai, A Righteous Struggle, p. 49.
67 Ibid., p. 61.
68 Ibid. p. 42.
69 Ibid., p. 46.
70 Ibid., p. 45.

Page 285
71
had elaborated in Ahmedabad a whole decade earlier. In fact, this body of legislation, including the Trade Disputes Act which was introduced in Bombay by the
Congress provincial ministry in 1938, had its origins perhaps rather more clearly in the Whitley Councils and the legislation being formulated in Britain than in Gandhi's
thought. Each of these arguments employed by capitalists, however, can be said to have found a resonance in Gandhi's precepts only by ignoring other parts of his
strictures and obligations represented in his ideal state of capital–labour relations. It is significant that while some traders and industrialists, notably G. D. Birla, became
devoted followers or at least grasped that Gandhi's influence served as a force for conservatism, there were many capitalists in the 1920s and 1930s who viewed
satyagraha with suspicion and feared that civil disobedience would lead to chaos and anarchy.
If it is implausible to portray Gandhi as a capitalist ideologue, it is, conversely, misleading to suppose that his rhetoric was devised to appeal to the class interests of
workers while at the same time somehow compromising or undermining them. To suggest that Gandhi simply wrapped a message of class conciliation in a vocabulary
of resistance and opposition is to take an overly simplistic and instrumental view of his rhetoric. Nor can it begin to explain why his rhetoric exercised any attraction for
the working classes. The discourse of class was alien to Gandhi's intellectual and political idiom. When he addressed the issue, it was within terms whose scope and
reach were wider and more expansive. It would be highly reductivist to measure his precepts and practice, and their appeal to workers, strictly by the ideology of
class. On the other hand, it is important to recognize that the 'class interests' of the working classes, far from being uniform and homogeneous, were themselves
fractured by relations of power, emanating from village and neighbourhood, caste and religion, gender and, indeed, the very process of production.
To grasp why (say) the working classes were sometimes able to find a resonance in his rhetoric, it is essential to appreciate Gandhi's ability to develop the blandest
metaphor and the most platitudinous axiom in a distinctly subversive direction. This, for instance, is how Gandhi extrapolated his notion of the industrial 'family' and of
the role of capitalists as 'trustees' for the workers. 'Speaking as a labourer, like you', he told the workers of the Maharaja Mills in Bangalore in August 1927,
I do not think there need be any clash between capital and labour. Each is dependent on the other. What is essential today is that the capitalist should not lord it over the labourer.
In my opinion, the millhands are as much the proprietors of their mills as the shareholders, and when the millowners realise that the
71 Patel, The Making of Industrial Relations.

Page 286
millhands are as much millowners as they, there will be no quarrel between them . . . When you know that the mill is as much yours as of the millowners, you will never damage
your property, you will never angrily destroy cloth or machinery with a view to squaring your quarrel with the millowners.72
The notion of the interdependence of capital and labour could now be read as something akin to a syndicalist argument for workers' control. Several months later, he
returned to a similar theme while elaborating his fond metaphor of the family. 'The relation between millagents and millhands', he declared in May 1928 (while the
communist general strike in Bombay was in full swing), 'ought to be one of father and children, or as between bloodbrothers.' He deplored the tendency of the
Ahmedabad millowners to describe themselves sometimes as 'masters' and their employees as their 'servants' as 'a negation of ahimsa'. Thus, he told the millowners:
What I expect of you, therefore, is that you should hold all your riches as a trust to be used solely in the interests of those who sweat for you, and to whose industry and labour
you owe all your position and prosperity. I want you to make your labourers copartners of your wealth.
Indeed, he advised them to respect 'the mutual obligations of love' as they exist between father and son, for thereby they would be able to bring about 'an end to all
labour disputes' and render trade unions redundant. But this ideal could not be realized, he warned, so long as 'there is a single millhand who does not regard the mill
in which he works as his own'.73
For the millhand to 'regard' the mill as his own was not, of course, for him to own it. If the millowners, presumably all capitalists, owed 'all [their] position and
prosperity' to the 'industry and labour' 'of those who sweat' for them, it did not necessarily follow that the workers could ask for control and ownership as theirs by
right. For workers to seek control, let alone ownership, would in 1918, for instance, have amounted to a negation of ahimsa, to the extent that it sought to usurp and
undermine the capitalist. But there is no reason to suppose that the working classes were, for Gandhi, precluded from justly making such demands under changed
circumstances as relations between capital and labour changed, and especially as his social and economic ideal of a world of selfsufficient producers came to be
realized. Nor did the notion that the workers should be encouraged to regard the mill as their own imply that Gandhi was seeking to destroy the distinction between
labour and capital. On the contrary, by seeking to ensure 'harmony' and mutual trust between them, he appeared, if anything, to affirm it. It is true that he sometimes
appeared to espouse a labour theory of value, for which he owed something to
72 Young India, 4 August 1927.
73 Ibid., 10 May 1928.

Page 287
74
Ruskin. Labour, he insisted, represented a form of capital. Without labour, capital would not 'fructify', just as the very existence of labour depended upon capital.
What capital offered to complement the skill and strength of labour was not wealth alone, but also, as Gandhi was sometimes led to observe, 'intelligence and tact'; and
'labour will never attain to that intelligence', for if it did, 'labour will cease to be labour and become itself the master'.75 If it was incumbent upon the capitalist to act as
the 'trustee' for the moral and material welfare of the labouring classes, to act as the father of his workers, it was obviously not for the son to damage, hurt or replace,
let alone destroy the father. Yet, this relationship did impose upon both sides a 'mutual obligation of love', which would enable workers to come to see the capitalist's
property as their own.
So Gandhi was no advocate of workers' control or theorist of the dissolution of capitalism. But there is room within his account to imagine a time when 'labour will
cease to be labour', when the son, to develop Gandhi's metaphor, might inherit his father's wealth, 'intelligence and tact', and when it could be wholly appropriate and
just, rather than coercive, for workers to demand to own the capital they had helped to create as no more than 'what is our right'. In the meanwhile, what Gandhi
offered the working classes, and insisted upon in their name, was their greater empowerment. For as he was to write in the late 1930s, when labour and capital
recognize the true nature of their interdependence, they will learn 'to respect and appreciate each other as equal partners in a common enterprise' and realize that 'they
need not regard each other as inherently irreconcilable antagonists'. The working man, he lamented, was insufficiently organized, which
prevented him from realizing the power and full dignity of his status. He has been taught to believe that his wages have to be dictated by capitalists instead of demanding his own
terms . . . It is the grossest of superstitions for the working man to believe that he is helpless before the employers.76
There is so little correspondence between the language of class and Gandhi's own inimitable vocabulary that it will always be misleading to seek to measure the specific
gravity of 'classness' in his rhetoric and extremely narrow to ask if he served in this context to encourage class conciliation or class solidarity. At the same time, once it
is recognized that for all the ambiguities and contradictions of his pronouncements, he sometimes developed his ideas, and they could similarly be interpreted and
extended by his audience, in more radical and subversive directions, it becomes far less plausible to cast him in the role of a capitalist ideologue or damn him as the
agent of workingclass betrayal.
74 Harijan, 23 March 1934.
76 Harijan, 3 July 1937.

Page 288
The second theme in Congress rhetoric relating to labour and industrial relations was the notion that workers should sacrifice current needs for future bliss. Present
grievances which could not be resolved under British rule would be redressed by swaraj, the ultimate goal for which the masses should rally behind the Congress
banner. Strikes in Indianowned industries only served, it was argued, to exacerbate foreign domination. It was considered incumbent upon labour to refrain from
industrial action, but it was also imperative that Indian capitalists fulfil their obligation to satisfy the basic needs of the workforce. Thus when Subhas Chandra Bose
intervened in the TISCO strike in 1929, Sir Ibrahim Rahimtoola was able to reassure Birla that the workers' champion would seek an early settlement because he took
the view that, 'if as a result of the strike, the industry is forced into other than Indian hands, it would be highly detrimental to national progress'.77 This doctrine could
serve not only to alienate workers, especially those in the midst of a strike, but it also placed a weapon in the hands of the employers to wield against trade unions and
their leaders, not infrequently Congressmen, who attempted to intervene in industrial disputes. It was an argument which might have seemed plausible when
predominantly Indian employers in Bombay, Ahmedabad or Jamshedpur sought to retrench their labour force or increase workloads or cut wages in the interests of
national progress. But the situation could become infinitely more complex when, for instance, the expatriate British owners of the Calcutta jute mills justified their own
rationalization schemes during the 1929 general strike by arguing that any alternative to reducing jobs and wages, and increasing workloads, would simply surrender a
share of their market to Dundee and foreign competitors in general.78 Certainly, a doctrine which invited workers to reconcile themselves to job losses, lower wages
and higher workloads in the national interest was likely to have limited sectional appeal. Furthermore, it was a doctrine which employers could turn against trade unions
in general, and those Congressmen in particular, who attempted to intervene in industrial disputes. The recurrent themes of Congress rhetoric on the labour question, it
might seem, were so unlikely to attract workingclass support that it may reasonably be asked whether this was ever their intention.
In the rhetoric of the Congress, social conflicts and class struggles were frequently subordinated to its political objectives. Sometimes, the con
77 Sir Ibrahim Rahimtoola to G. D. Birla, 9 July 1929, Thakurdas Papers, File 42 (II), NMML.
78 Bengal Legislative Proceedings, 5–9 August 1929 (Calcutta, 1930), p. 302, quoted by B. Chakrabarty, Subhas Chandra Bose and Middle Class Radicalism: A Study in Indian
Nationalism, 1928–1940 (London, 1990), p. 72; P. Saha, History of the Working Class Movement in Bengal (New Delhi, 1978), pp. 114–31; Goswami, 'Multiple Images: Jute Mill Strikes
of 1929 and 1937', 560–3.

Page 289
sideration of the former was postponed in favour of the realization of the latter. But the interpretation of this rhetoric as conclusive proof of the social character of the
Congress as a 'bourgeois party' would be misleading. The Congress did not represent a conspiracy to dupe or betray the working classes and the peasantry or to
deflect their natural propensity to revolutionary action. It was not, indeed, the agency of a 'passive revolution'. Nationalism was primarily a discourse of political
exclusion, not of social conflict or economic grievance. The social evils of British India were perceived and portrayed as the outcome of the political injustice of foreign,
imperial rule. To end this political injustice would not, as Gandhi frequently warned, necessarily lead to the correction or the resolution of the society's wrongs. But
without independence, Indians could not emancipate themselves from its thraldom. On the contrary, independence from British rule would merely constitute a 'formal'
freedom, if Indians did not effect 'reform from within', achieve 'inward freedom' or attain individual 'selfrealization'. But even this would not suffice. 'I am afraid I must
repeat the gospel to you and remind you', Gandhi told a meeting of village workers at Nagpur in March 1935, 'that when you demand swaraj, you do not want swaraj
for yourself alone, but for your neighbour too'.79 When the Congress advised the postponement of social struggle until the achievement of independence, this was not
simply an expression of its determination to subordinate the class interests of the poor to its own longterm political goals – indeed, this would be a rather mechanistic
and narrow interpretation. Rather, it reflected its diagnosis of the underlying cause of existing social evils in terms of the larger system of colonial oppression, political
coercion and tyrannical power.
Of course, from the earliest days of their rule, the British had drawn Indians into the structure of governance at its various levels. As the British began to associate more
Indians with their rule, the differential access to power, which Indians thus gained, began to divide them. Increasingly, in the nineteenth century, the British were drawn
into reconciling two contradictory aims in their government of India. On the one hand, they sought to unify and centralize their government in order to enable them to
appropriate and deploy Indian resources more effectively for imperial purposes. On the other hand, they had to offset the consequences of an increasingly meddlesome
government and to secure the acquiescence of their subjects by allowing them larger areas of influence and control within the colonial system, preferably at points
which would not limit their freedom to work their dominion in their own larger, global imperial interests. It was a persistent, perhaps defining, theme of colonial dis
79 Harijan, 1 March 1935, in The Moral and Political Writings of Mahatma Gandhi, vol. III: NonViolent Resistance and Social Transformation, ed. R. Iyer (Oxford, 1987), p. 262.

Page 290
course that India was a museum of tribes, castes, races, religions and even languages, each of which needed to be classified and represented. It is not surprising,
therefore, that as the British searched for a principle of differentiation and classification, they should have found these categories of ethnicity, language, caste and
religion, however fictive the unities they presumed, to be the most compelling. Indeed, it became essential for Indians to define themselves along these communal lines
to seek representation within, and to widen their access to, the structure, and the resources, of British government. Ethnicity thus became a basic principle of social and
political competition. In this way, the British imparted to the protean languages of Indian politics a fresh vocabulary of sectarianism, a political language of social and
communal interests within which competition of diverse sorts and a wide range of conflicts were likely to be expressed.
To the extent that its protagonists may have appeared to be in quest of a larger share of political power, nationalism might be seen as the political programme of Indian
elites with limited social interest or appeal. But its significance was widened by two factors. First, the terms on which Indians might extend their place within the
representative system and bureaucratic structure of British rule was of considerable interest to those groups who were unlikely to benefit personally from it. As some
social groups obtained and consolidated their position within the structure of governance, or the educational system, it made their protection and patronage accessible
to their friends and villagers, castefellows and coreligionists, and created among them at least the promise and opportunity for social mobility. Substantial interests,
however defined, which were seemingly backward in gaining access to English education, government employment or political office, might receive official
encouragement to advance or be offered special representation. The demand for a greater share of power, especially when it was couched in terms of the
representation of caste, religious and ethnic groupings, could discover an appeal among those who gained nothing directly from it.
Second, Gandhian ideology, which increasingly inflected the rhetoric of the Congress, could be stretched to include those whom the electoral system did not reach and
who were unlikely to wield any influence within the bureaucracy. Significantly, Gandhians abjured power and their influence derived in part from their dissociation from
the structure of power. Some noncooperators became Swarajists; many satyagrahis wished to accept office. But Gandhi and his followers were not tainted by the
unseemly rush for a limited share of power to wield against those who had none. Their rhetoric of selfdenial and selfsacrifice, and their ennobling of 'garibi' identified
them with the 'deprived' and 'downtrodden'. 'Swaraj

Page 291
for me', Gandhi had written in 1924, 'means the freedom for the meanest of our countrymen . . . I am not interested in freeing India from merely the English yoke. I am
bent upon freeing India from any yoke whatsoever.'80 To be realized, swaraj would have to extend beyond independence. It would have to encompass universal adult
suffrage and require the transcendence of the coercion, manipulation and exploitation of the powerless. It could bring about a society 'in which every man and woman
knows . . . that no one should want anything that others cannot have with equal labour'.81 In a speech to the Ahmedabad millworkers, who were on the point of taking
industrial action in response to a threatened wage cut of 20 per cent in 1936, Gandhi told them that the 'key to Swaraj' lay in the villages which were 'in a worse plight'
than even those who lived in the city: 'When I succeed in ridding the villages of their poverty, I have won Swaraj for you and the whole of India . . . Their deliverance is
also yours.'82
III
Political programmes are not wafted on rhetoric alone. How the working classes interpreted the rhetoric of the Congress, as well as the nature of their response to its
programme, was also influenced by the role which the party played in workingclass life and workingclass politics. Indeed, popular perceptions of the political
interventions of the Congress could also shape how its rhetoric was interpreted. As government policies, at any level from district and municipality to the centre,
increasingly impinged upon the working classes, the scope for politicians to act as mediators between them and the state increased. How they performed this role
would determine how they were perceived by the working classes. From the 1890s, in the aftermath of the plague epidemic, urban improvement and town planning,
housing and public provision, slum clearance and sanitation, acquired a growing significance on local political agendas and in the public domain. They were matters of
immediate concern to the working classes. The response of local politicians to such issues, in addition to their role in labour disputes, became important criteria by
which they were assessed.
As Indian politicians exercised greater power in municipal and provincial politics, so workers came to assess them and their political associates by the initiatives they
took to ameliorate the social conditions of their neighbourhoods and to improve their access to housing and employment.
81 Harijan, 28 July 1946, in Iyer (ed.), The Moral and Political Writings, vol. III, p. 232.
82 Harijan, 7 November 1936.

Page 292
Thus, the Ahmedabad Textile Labour Association's role in municipal politics played an important part in enabling it to rebuild its organization, regain its credibility and
revive its fortunes after the mill strike of 1923. In addition, following Gandhi into 'constructive work', the ATLA's activities now focused upon education, social reform,
housing and flood relief. By the mid1920s, its membership exceeded a quarter of the workforce and between 1928 and 1930, it doubled yet again.83 Similarly, in the
mid1920s, the working classes in Madras 'became disillusioned' with, and turned against, the Swarajists because they 'had shown little real interest in the industrial
divisions of the city in such pressing needs as drainage, lighting and adequate housing'.84 Similar issues dominated the political responses of the North Indian urban poor
and shaped their political identity. Indeed, the success of the Congress Socialists in the mid1930s in attracting the support of the urban poor hinged on their ability and
willingness to write these needs into their political agendas.85
Workingclass responses to the Congress were also informed by the role which the latter played in the sphere of industrial relations. What the Congress as a national
institution could offer the working classes was a wider focus for specific disputes. Its influence with the state and the access of its politicians to the councils of
government sometimes made the Congress a powerful ally. This influence enabled it, for instance, to secure the position of the Ahmedabad Textile Labour Association
as the sole bargaining agent in the city's textile industry. With the exception of the Ahmedabad Textile Labour Association, the Congress eschewed any direct
connection with the trade unions. Nevertheless, numerous Congressmen intervened in industrial disputes and offered themselves as publicists of labour's cause. They
often acted in this context in their individual capacities and not as agents of the Congress. Some, like Gandhi and Subhas Chandra Bose, boasted a national
prominence; most were drawn from the secondary leadership or locallevel publicists of the Congress. The growing rewards for representing labour in the formal
institutions of the state attracted such men to this field of political activity. Their calculations were not entirely misplaced. Many of these publicists rose to prominence
within the Congress itself or came to dominate local or provincial politics. Gulzarilal Nanda, the main force of the Ahmedabad Textile Labour Association, and V. V.
Giri, the railway trade unionist of South India, became Labour Ministers in the provincial Congress governments of 1937–39 and occupied important central cabinet
offices in the 1950s and 1960s. Nanda was twice a caretaker Prime
84 Murphy, Unions in Conflict, p. 103.
85 Gooptu, 'The Political Culture of the Urban Poor in North India', ch. 5.

Page 293
86
Minister in the 1960s while Giri was to become the President of India in the 1970s.
The increasing interest of a growing number of publicists in the conditions of the working classes, which first became fully evident in labour politics in the immediate
aftermath of the first world war, is perhaps most clearly grasped in relation to two developments. First, between 1900 and 1920, famines and epidemics, the debate
about the causes of Indian poverty and the destruction of Indian industry, had directed attention to the social conditions of the urban poor. In the context of the
swadeshi campaign of 1905–8 and anxieties about the social consequences of industrialization, it is not surprising that humanitarian sentiment should have drawn
middleclass youth, many of them lawyers and doctors, into 'social service', the amelioration of the conditions of the poor and necessarily into the representation of
their interests to the state and the mediation of their disputes with their employers. N. M. Joshi's involvement with labour's cause developed from his activities on behalf
of the Social Service League, which was inextricably linked with the Bombay Textile Labour Union.87 S. K. Bole and H. A. Talcherkar were social workers who
organized the Kamgar Hitwardhak Sabha, which played a prominent role in the general strike of 1919 in Bombay, and they were subsequently to carry their
connections into the Bombay Textile Labour Union in 1925.88 Anasuyabehn Sarabhai had been engaged in setting up schools for the children of millworkers when they
approached her to represent them in their disputes with their employers.89 Chelvapathy Chetty and Ramanujulu Naidu, who led the Binny's strike in 1918, were
primarily associated with a religious association to moralize the working classes.90 B. P. Wadia, the founder of the Madras Labour Union, N. S. Ramaswamy Iyengar,
protagonist of the Coimbatore Labour Union, and B. Shiva Rao, labour leader, public servant and the outstanding writer on Indian industrial workers, came to the
labour movement through the Theosophical Society, while the charismatic Tiru Vi Kalyansundaram Mudaliar discovered his interest in workingclass politics through
Saiva Siddhanta philosophy.91 Social work and a humanitarian, even philanthropic, concern for the plight of the working classes did not necessarily precede – it could
sometimes follow – the interest and engagement of these publicists with trade unions. The selfproclaimed mother of the
86 V. V. Giri, My Life and Times, vol. I (Delhi, 1976).
87 V. B. Karnik, N. M. Joshi – Servant of India (Bombay, 1972).
88 'The Kamgar Hitwardhak Sabha: A Brief Sketch', Indian Textile Journal, 29 (July 1919), 177–9 and ibid., 29 (August 1919), 209–10.
89 Desai, A Righteous Struggle, p. 4; Erikson, Gandhi's Truth, pp. 300–2; Patel, The Making of Industrial Relations, p. 38.
90 Murphy, Unions in Conflict, pp. 64–5.
91 Ibid., p. 90.

Page 294
Bengal jute workers, Santosh Kumari Devi, recalled how she 'started the labour movement' in the early 1920s:
I went to almost all the jute mills at Naihati, Garifa, Halisahar, Bhatpara, Sodepur to organize the labourers. Over and above making the labourers members of the jute workers'
union, we set up night schools and even health centres for working women and children in some centres . . . Unfortunately the national leaders of that time gave little or no thought
for the toiling masses. Of course, I worked there in the name of the Congress and hence some of the more conscious [sic] workers were attracted towards the Congress. My
concept was that the Congress should come forward to organize the working class, so that the workers in their turn realised the importance of the Congress.92
Second, in the first two decades of the twentieth century, political reform, the extension of the franchise and the proliferation of new municipal councils created fresh
opportunities in public life. Not only did it allow politicians a certain autonomy from their patrons in the mofussil, but it granted those with weaker access to power the
opportunity to shake the grip of established politicians and their networks on the existing political institutions, by widening their base of support.93 Some publicists
attempted to develop this following by representing the new 'backward caste' or 'communal' constituencies which were being brought into existence. Others presented
themselves as protectors of the poor and guardians of the deprived, and stepped forward readily as spokesmen for the workers during industrial disputes.
Congress publicists who thus styled themselves as advocates of the workers' cause often had an ambiguous relationship with the party. In the 1910s and 1920s, the
character of the Congress changed quickly, its ideological and political identity appeared constantly in flux and its fortunes flowed and ebbed dramatically. Its
organizational machinery, first fashioned during the early 1920s – possibly the most enduring legacy of noncooperation – fell into disuse later in the decade and was
revived as an effective force only in the mid1930s.94 It was only then – and then, too, not very consistently – that the freedom with which publicists moved into or out
of the Congress, and sometimes just as readily into or out of other political groupings, was curtailed. As the Congress geared itself to
92 Santosh Kumari Devi, 'How I Started The Labour Movement', unpublished memoir, cited by M. Chattopadhyay, 'Santosh Kumari Devi: A Pioneering Labour Leader', Social
Scientist, 128 (January 1984), 65; Sen, 'Women Workers in the Bengal Jute Industry', esp. ch. 6.
93 Washbrook, The Emergence of Provincial Politics; Bayly, The Local Roots of Indian Politics; Rajat Ray, Social Conflict and Political Unrest in Bengal, 1875–1927 (Delhi, 1986);
Robinson, Separatism Among Indian Muslims; Cashman, The Myth of the Lokamanya.
94 Gopal Krishna, 'The Development of the Congress as a Mass Organization, 1918–1923', JAS, 25:3 (1966), 413–30; Tomlinson, The Indian National Congress and the Raj, pp. 33–5.

Page 295
contest the 1937 elections, the incentive to move into other groupings, less likely to triumph at the polls, declined. Membership of the Congress, or the adoption of its
nomenclature, does not provide a ready or reliable guide to the political and ideological identity of the individual publicist.
Between 1914 and 1922, a wide range of politicians identified themselves with workingclass causes. For instance, we are told, in Madras Presidency, 'most of the
labour leaders in the immediate postwar period were either supporters or members of . . . the Indian National Congress'.95 'Noncooperators', who are said to have
formed 15 per cent of all trade union office holders in Bengal between 1918 and 1921, made up the largest single group of 'outsiders' among labour leaders, while
more than a further 10 per cent styled themselves simply as 'nationalists'.96 Some of these 'outsiders' in Bengal, notably S. N. Haldar and Byomkesh Chakrabarti, were
invited by the steelworkers of Jamshedpur to represent them during the strikes of 1920.97 Local publicists of the Home Rule League in Bombay intervened in labour
disputes and sought to establish their presence in the mill districts after 1917 and played a prominent role in the early stages of the general strike of 1919.98
It should not be supposed, however, that by virtue of their intervention, these publicists determined the shape of workers' politics. On the contrary, their role was more
usually governed by the momentum of workers' industrial and political action. Although a majority of trade union posts were held by 'outsiders' in Bengal, they were
involved in less than onefifth of the strikes which occurred between January 1918 and December 1921.99 Similarly, 'almost all the strikes between 1918 and 1922
and later on' in Madras Presidency were initiated by the workers 'very often without knowledge of the outside leaders'. Since these leaders were generally 'extremely
nervous and cautious about initiating strikes', they were frequently 'embarrassed' by workers' actions.100 Similarly, in Ahmedabad, the aftermath of the 1918 dispute
witnessed numerous strikes in individual mills in which the local Gandhians played only the most minimal role. The noncooperation movement provided a further spur
to industrial action. Strikes now continued to occur without the direction of
96 Gourlay, 'Nationalists, Outsiders and the Labour Movement in Bengal', p. 39.
97 Bahl, 'TISCO Workers' Struggle', 37–40.
98 'A Report from the Commissioner of Police, Bombay, to Government of Bombay, Concerning Political Developments before and during 1919', Curry Papers, Box IV, nos. 54 and 55,
Centre of South Asian Studies, Cambridge.
99 Gourlay, 'Nationalists, Outsiders and the Labour Movement', pp. 36–7.
100 Murphy, Union in Conflict, pp. 86–7. For an account of a strike in Coimbatore which thus embarrassed Shiva Rao so that 'he returned to Madras in a huff', see ibid., pp. 120–4.

Page 296
the ATLA leadership, sometimes in flat contradiction to their wishes. The union, officially formed in February 1920, could scarcely claim to control the workers, let
alone exert any influence over its increasingly militant members, especially when it was seeking to restrain them.101 It was a characteristic feature of the development of
workingclass politics in Bombay, as elsewhere, that trade unions usually formed only after a strike began and collapsed soon after it had ended. Frequently, as the
Home Rule League as well as Bole's Kamgar Hitwardhak Sabha discovered during the 1919 general strike, they were swept aside by the workers' refusal to accept a
settlement on the terms they obtained, or even simply by the groundswell of popular action.102 Until at least the late 1920s, when it became easier for officials,
millowners and political observers to point an accusing finger at communist troublemakers, they were frequently bemused by and hard pressed to explain the ability of
workers to sustain lengthy strikes with at best only the most rudimentary formal organization.103
Since the Gandhians represented a dissident faction within the Congress between 1915 and 1920, and the noncooperation movement was explicitly oppositional to
the state, it is scarcely surprising that they provided a refuge for irreconcilables and brought into the nationalist fold varied and often conflicting political aspirations.
While Congress in its agitational mode became a party of allcomers, and a wide range of publicists sought to represent workers' grievances, it should not be supposed
that all labour leaders lent their weight to the noncooperation movement. Prominent trade union leaders, like N. M. Joshi, S. K. Bole and B. Shiva Rao, opposed the
movement. Others, like C. R. Das, doyen of the Bengal Congress, 'broker between irreconcilables' and the political guide of Santosh Kumari Devi, were drawn into
labour disputes as part of the noncooperation movement but rejected the path of the 'nochangers' after 1922, successfully contested the council elections of 1923
and swiftly lost interest in the proletariat.104
The noncooperation movement did not rupture the old linkages which had constituted the political system before 1920. Nor did it radically transform the character of
the Congress. As the movement stut
101 Patel, The Making of Industrial Relations, pp. 51–63, 126–36.
102 For an account of the 1919 strike in Bombay, see Newman, Workers and Unions, pp. 120–9; Kumar, 'The Bombay Textile Strike, 1919'.
103 Secretary, GOB, Home to Secretary GOI, Home, 7/15 February 1919 in Bombay Confidential Proceedings, vol. 46, 1919, OIOC; Report of the Industrial Disputes Committee, 1922, in
Labour Gazette, 1:8 (April 1922), 24; Bombay Chronicle, 21 February 1924.
104 J. A. Gallagher, 'Congress in Decline: Bengal, 1930–39', in Gallagher, Johnson and Seal (eds.), Locality, Province and Nation, pp. 269–325; Ray, Social Conflict and Political
Unrest, ch. 5; Basu, 'Workers Politics', ch. 6.

Page 297
tered violently to a close, provincial politicians and local bosses began to desert it. The 'nochangers' failed to prevent the Swarajists from contesting the 1923 elections
or the Cocanada Congress from ratifying their decision. In most provinces, there were enough politicians willing to work the Montford constitution to make dyarchy
effective. By allowing Indian elected representatives a real, if modest, share of power in the provinces, the Montford reforms rendered the linkages of allIndia politics
redundant and allIndia parties gradually became moribund. By contrast, provincial politics were in ferment, animated by deepening factional rivalries and by unstable
ministerial coalitions, whose meddling served inevitably to disappoint more clients than their patronage could satisfy. During the 1920s, the Congress divided between
several parties and disintegrated into numerous factions. Political interests did not necessarily retain the same identity at local and provincial levels. Thus, the
Ahmedabad Textile Labour Association, seemingly the creature of the Congress, joined forces in the municipal corporation with the Independent Party of the
millowners and acted increasingly in opposition to the Congress.105 Trade union leaders who may have been noncooperators in the early 1920s now found that their
connections with the Congress were becoming more attenuated. As more Indian elected politicians acquired provincial power under dyarchy, and exercised it under
various, often changing, party labels, so these publicists of labour's cause had less need to operate under the banner of the Congress. The alliances which they forged
with the ministerial coalitions which governed the province could be deployed to widen their base among the workers or to strengthen their negotiating status with the
employers.
Furthermore, in the 1920s, labour began to be absorbed into the representative system of colonial rule. Spokesmen for labour, notably, in 1920, K. C. Roy
Chowdhury from Bengal and N. M. Joshi from Bombay, were nominated to the central legislative assembly, and both of them initiated important pieces of legislation
during the decade. More crucially, perhaps, the extension of the franchise meant that workers and their unions exerted a growing, sometimes a decisive influence, on
elections, especially to the municipal councils, but also to the provincial legislatures. In the early 1920s, in Bombay, over 11 per cent of the residents of the
predominantly workingclass municipal Ward F were entitled to vote, although only about oneeighth of them actually did. In Wards E and G, which were also
predominantly working class, the turnout was much higher – roughly onethird of the electorate – but the proportion of the residents who were eligible to vote was
about 3 per cent and 5 per

Page 298
106
cent respectively. Although workingclass voters generally formed as yet a small proportion of the electorate, they sometimes exercised a disproportionate influence
on the result. In the Madras municipal elections of 1924, it was said, 'they largely determined the outcome of the elections in the industrial wards as the union was able
to mobilise them as a bloc'.107 When the Madras Labour Union contested the 1927 election to the provincial legislature, they were able, together with their allies, to
defeat the Swarajists in each constituency ''where the vote of the industrial workers was most effective'.108 By 1936, when the Madras Labour Union supported the
Congress, the workingclass vote was shown to be influential not only in the constituencies which they won but also in those which they lost because of the local
influence of a rival trade union clique.109
During the 1930s, however, this fluid and flexible relationship between labour leaders, on the one hand, and the Congress, on the other, underwent fundamental
change. The Congress imparted to its organization a greater formality and tightened and centralized its control through its 'High Command'. Increasingly, it defined the
criteria of party membership and it became less tolerant of divided loyalties within its ranks. In the unseemly rush for Congress 'tickets' which gathered force after
1934, it acquired the opportunity and the power to discriminate between its followers. By the mid1930s, it was becoming increasingly clear that there were important
electoral gains to be made by organizing labour and forming trade unions, while, at the same time, labour leaders, no less than other publicists, recognized that the route
to electoral success and ministerial power lay through the Congress. In the 1920s, local circumstances had largely determined the attitude of labour leaders to the
Congress. Where hostile elements held sway over the local Congress (or the Swarajists), it was sometimes preferable to seek leverage upon the political system
through a rival political grouping. On the other hand, if their access to the Congress remained open, and where, for instance, antagonistic local bosses remained
suspicious of the party, they might prefer to operate beneath the nationalist banner. The possible permutations seemed numerous and complex. In the 1930s the
options narrowed. Publicists who stepped forward as champions of the urban poor had already, in the 1910s and 1920s, been most commonly drawn from among
politicians with relatively little access to power and few firm and powerful alliances to put at risk. But in the 1930s, as landlords, merchants and even some industrial
capitalists, acquired greater influence
106 K. T. Shah and G. J. Bahadurji, Constitutions, Functions and Finance of Indian Municipalities (Bombay, 1925), pp. 31–2.
108 Ibid., p. 105.
109 Ibid., pp. 130–1.

Page 299
and control within the Congress, so labour publicists in the party, especially the more radical among them, came increasingly to be marginalized. The Congress was no
longer a party of allcomers, even less a refuge for irreconcilables. If, in the early 1920s, publicists on the margins of the political system were able to forge wider
political alliances and achieve a greater political prominence by organizing and representing labour, by the mid1930s it was becoming far more difficult to build such
careers within the Congress. Increasingly, the Congress representatives of labour and workingclass interests, with wider political ambitions, had to act in ways which
did not offend their other political constituents. Whereas, in the 1920s, labour was only one of a range of political concerns for the publicist, and often a rather marginal
concern, more labour leaders emerged in the following decade who were concerned exclusively with labour matters. At the same time, these labour leaders, if they
chose to operate within the fold of the Congress, were more effectively subordinated to the High Command and they were often precluded from rising to greater
prominence within the movement. The outcome, in Bengal and Madras, in Kanpur and Ahmedabad, in Sholapur and Coimbatore, was the growth of socialist and
communist organizations within the labour movement.110 The 'rise' of the left was not limited to workingclass movements alone in the 1930s.111 In part, it reflected the
sharpening and proliferation of class conflicts in Indian society during the depression. It also reflected the changing character of the Congress and its relationship with its
social base, which in itself was affected by changes in the nature of the relationship between the Congress and the colonial state and, in particular, its structure of
governance. Finally, the growth of socialist organization was stimulated by the hopes raised and expectations generated among the working classes, and the followers
of the Congress more generally, by its electoral victories in
110 On Bengal, see Chakrabarty, Rethinking Working Class History; Basu, 'Workers' Politics in Bengal'; on Madras, see C. J. Baker, The Politics of South India, 1920–37,
(Cambridge, 1976), pp. 190–2; Baker, An Indian Rural Economy, ch. 5; Murphy, Unions in Conflict; Ramaswamy, The Worker and His Union, ch. 1; on Kanpur, see Joshi, 'Bonds
of Caste, Ties of Religion'; Pandey, 'Ideological Conflict in the Kanpur Trade Union Movement'; Gooptu, 'Political Culture of the Urban Poor in North India'; on Ahmedabad, see
Patel, The Making of Industrial Relations; GOB, Home (Special), File 550 (25) IIIA of 1938, MSA; on Sholapur, see GOB, Home (Special), File 550 (14) of 1933, File 543 (53) A of
1934 and File 550 (25) IVA of 1938, MSA.
111 Rajnarayan Chandavarkar, 'From Communism to Social Democracy: The Rise and Resilience of Communist Parties in India, 1920–1995', Science and Society, 61:1 (1997), 99–106; D.
N. Dhanagare, Peasant Movements in India, 1920–1950 (Delhi, 1983); A. Cooper, Sharecropping and Sharecroppers' Struggles in Bengal, 1930–1950 (Calcutta, 1988); K.
Damodaran, 'Memoir of an Indian Communist', New Left Review, 93 (1975), 35–59; D. Menon, Caste, Nationalism and Communism in South India: Malabar 1900–1948 (Cambridge,
1994).

Page 300
1937. To some, it appeared, in their enthusiasm, that the elections had brought the irreconcilables to office.
IV
It should not be supposed, however, that these interventions in labour disputes and attempts at labour organization always worked in favour of the Congress inevitably
securing its place within the diverse political cultures of the working class. As labour organizers, these publicists had to act within the constraints imposed upon them by
the structure of industrial relations. They were always liable to be ground between the irreconcilable grievances of labour and the implacable resistance of capital.
Historians have perhaps too readily assumed that the formation of trade unions reflected the level of industrialization in the economy or a particular stage of working
class consciousness. But this is to minimize the extent to which the growth of trade unions depends upon the tolerance of the state and the acquiescence of
employers.112 In India, the employers as well as the state disclosed a low threshold of tolerance for workers' combinations. Workers' grievances, when they could not
be resolved within the apparatus of labour discipline at the workplace, were rarely given serious consideration, unless they went on strike. But if a strike did not result
in a complete closure, workers ordinarily risked dismissal. By dismissing the irreconcilable, locking out the implacable, cajoling the sullen, bullying the waverers,
discriminating against those whom they could not dismiss, and choosing whom they would negotiate with, the employers could decisively influence the complex, if
fragile, alliances both within and outside the workplace, upon which trade unions rested. Most commonly, they could undermine and destroy workers' combinations at
the point of production.
Under these adverse circumstances, politicians seeking to organize labour could follow two courses of action. First, they could try to persuade the employers and the
state to accept their credentials and negotiate with them as representative, responsible and legitimate spokesmen for the workers. Certainly, the political influence of
some Congressmen gave them a certain leverage upon employers when seeking recognition as bargaining agents on behalf of the workforce. They could deploy their
wider connections or their positions in the councils of state to the advantage of their followers or the embarrassment of the employers and the government. Their
influence at these levels might encourage employers to invite them to settle disputes while the confidence they commanded

Page 301
with employers or the state could allow them to build up a significant following at the workplace. In a sense, the effectiveness of trade union leaders in these
circumstances depended on their skill not in putting the workers' grievances to the employers but in representing the employers to the workforce. Their success and
their survival depended upon their ability to please both masters and frequently they were to find themselves ground between the upper and the nether millstone.
On the other hand, trade unions whose existence was not safeguarded by the relative tolerance of the employers and the state often had to impose themselves forcibly
upon industrial politics, by adopting a strategy of direct action and confrontation: meddling in the daily disputes of the workplace, taking up minor grievances,
demonstrating their ability to secure favourable settlements, attempting to generate sufficient momentum for largescale action. By intervening energetically in minor
disputes and assiduously airing workers' grievances, they attempted to generate sufficient momentum for largescale action. Necessarily, they established themselves
most successfully during general strikes. Excluded from the workplace, these unions often paid closer attention to building up connections in the neighbourhood. They
were sometimes able to deploy these connections, conversely, to forge closer linkages with the workplace as well as to generate and sustain industrial action, whether
by combating blacklegs or cajoling landlords and grain merchants to extend credit to their members.113
Of course, this distinction between unions committed to strategies of confrontation and conciliation is too starkly drawn. There were many who moved adroitly
between both styles of trade unionism and some who wholly embraced neither. Trade unions which flourished on their accreditation by the employers and the state
could not for long ignore those who cut a figure on the street corner. Strategies of confrontation were futile if they never yielded concessions. At some point or other,
'company unions' had to lead strikes and yet protect their favoured status, while the most radical unions had to face the dilemma of how to reflect the militancy of their
followers without permitting it to go too far. But the confrontational style is most readily associated with the communist trade unions which operated in Bombay in the
late 1920s and 1930s, in Sholapur in the 1930s and in Kanpur in the late 1930s and early 1940s. This style of trade unionism was aptly captured by the jute workers
of Calcutta who during the strikes of 1929 began to refer to the communist trade union leaders as 'the strike babus'.114 Congress politicians could not always follow the
path of confrontation. They often brought to trade
114 Mitra, 'Growth of Trade Union Consciousness among Jute Mill Workers', p. 1841.

Page 302
union organization wider and more secure political connections which were jeopardized by the pursuit of any highly adventurist strategy of confrontation. By allowing
labour militancy its head, they could alienate capital when they were seeking to harness it to their side and put at risk their own local political alliances. This meant that
they frequently found themselves trying to move against the tide of workingclass opinion. It often required them to restrain their followers from industrial action or to
advise them to return when they wanted to prolong the strike or indeed to negotiate an early settlement even at the expense of accepting terms less favourable than
their original demands. In such circumstances, these spokesmen for labour could rapidly discover that they had nobody to speak for. Even when leaders of national
prominence, like Gandhi or Bose, intervened in labour disputes, able to deploy their reputations and their extensive political connections to the advantage of the
workers and to command, however reluctantly it was given, the confidence of the employers, they could emerge with little to show for their efforts.
Despite all his bluster, Bose, involved in the closing stages of the Jamshedpur strike of 1928, obtained a deal from the employers which, in the interests of a quick
settlement, studiously ignored the most vital demands behind the strike.115 Thus, when Bose returned to Jamshedpur in the following year, as the champion of the
Golmuri tinplate workers, Birla, a Calcutta industrialist, assured his Bombay counterpart, Thakurdas, that the tiger of Bengal might roar but he certainly would not
spring. 'Mr Bose', he wrote, 'can be relied upon to help Tata Iron and Steel Works whenever necessary.' For 'Mr Bose', he went on to explain,
is a very sincere and scrupulous man and appreciates the necessity of cooperation with reasonable and advanced type of capitalists. He himself belongs to the aristocratic class,
although he voluntarily renounced many luxuries. His main object in labour matters no doubt is service to the labour but not necessarily inimical to the capitalist.116
This style of intervention in labour disputes often served to alienate workingclass support. It is scarcely surprising that the Jamshedpur workers simply ignored the
Congress satyagrahis during the civil disobedience campaign in 1930.117
In Ahmedabad, the relationship between the Congress and the working classes followed a different pattern, even if the outcome was in fact all too familiar. The
Congress, as we have seen, developed an uniquely intimate relationship with the Ahmedabad Textile Labour Association, and while it did not seek to forge similar links
with other trade unions, it guarded its
115 Simeon, The Politics of Labour under Late Colonialism, chs. 2–3; Bahl, 'TISCO Workers' Struggle, 1920–28'; Bahl, The Making of the Indian Working Class.
116 Birla to Thakurdas, 16 July 1929, Thakurdas Papers, File 42 (I), NMML.
117 Bahl, 'TISCO Workers' Struggle', p. 44, n. 34.

Page 303
118
connections with the ATLA closely. Gandhi's involvement with the city's textile workers began with the satyagraha of 1918, before he was firmly established as a
dominant figure within the Congress. Although Gandhi withdrew from active involvement with the union after 1920, he continued to keep a watchful eye on the conduct
of its affairs, which were left in the charge of his loyal and talented followers, Anasuyabehn Sarabhai and, increasingly, Gulzarilal Nanda. Perhaps in part as a
consequence, the textile workers developed after 1918 a closer affinity and loyalty to Gandhi than the ATLA was able to evince. The national standing of the Congress
and the wider political influence of Gandhi encouraged the millowners and the colonial state to raise the threshold of their tolerance of workers' combinations in this
instance. The leverage which Gandhi and the Congress exercised on the colonial state and their potential for mobilizing mass support enabled them to lodge the ATLA
as the principal bargaining agent on behalf of the textile workers. The commercial possibilities which the swadeshi campaign opened up for the millowners in the
domestic market must have also entered their calculations. Moreover, the millowners, like local officials, recognized as early as 1918 that Gandhi's charisma might
serve to restrain the working classes and pose a lesser threat to the public order than any imagined, or available, alternative. While Gandhi's advice to the workers in
1918 'was all in the direction of sobriety and conciliation', the fact remained that 'a few of the wilder spirits' among the workers 'question what he can do'.119 Clearly, if
for the millowners it was preferable to deal with Gandhi than with 'the wilder spirits', it was essential for them to strengthen his hand. Of course, neither the millowners
nor the colonial state were unwavering in this belief. Indeed, the Ahmedabad Millowners' Association was often engulfed by bitter and acrimonious disputes over the
formulation of a labour policy, which not only threatened at times to destroy their own organization but led them periodically to severely undermine the ATLA to the
point of virtually wrecking it.120
For nearly a decade after the 1923 strike the millowners entertained the existence of something remotely resembling a system of collective bargaining, and roughly
along the lines envisaged by Gandhi in 1918 – where industrial relations were conducted through the ATLA and, if necessary, a board of arbitration. It was facilitated
by the relative prosperity of the Ahmedabad mills, by the ATLA's own stance of 'collaboration' in its 'indigenous experiment in trade unionism' and the millowners'
recognition that it was best to deal with the devil they knew. If they were to destroy the ATLA, as they very nearly did in 1923, who knew what
119 BPP, SAI, 1918, p. 164, para 235.
120 G. L. Nanda to S. G. Banker, 23 March 1933, in File 40, pp. 77–83, ATLA Papers, microfilm copy, reel no. 1, NMML.

Page 304
manner of beast might be raised up from the deep? Thus, the ATLA dedicated itself 'purely to promote the moral and social welfare of the workers and to raise their
value as citizens'.121 Seth Maganlal Girdhardas, no friend of the trade union, had warned the Millowners' Association, over which he presided, that 'we should not
commit the error of taking all labour organizations as inimical to our interests' but rather support and strengthen the hand of those unions which sought 'to evolve order
out of chaos and deal with rough illiterate masses'.122
Once the depression which had afflicted other centres of the cotton textile industry several years earlier now began to affect the Ahmedabad mills, however, and their
markets slumped in the early 1930s, the millowners showed that they preferred the freedom to make 'private settlements' with their workers to the rigours of collective
bargaining and arbitration, which lumped them all together even as local rivalries grew sharper and more antagonistic. The millowners began once more to deal with the
ATLA as often as they ignored or sidestepped it, followed the rules of arbitration more often in the breach than in the observance and submitted, if they were forced,
to agreements over wages and conditions, only to steadfastly evade their implementation. It was only the growing threat of the communist Mill Mazdoor Sangh in the
mid1930s, drawing upon the support of the mainly Muslim weavers,123 which led the millowners to close ranks and to seek to blunt the red menace by consolidating
the position of the ATLA, while the latter, apprehensive of the pressure of their communist rivals on their flank, and the Congress, now in office, and anxious about the
possibility of a socialist challenge to the control of the local party, were only too willing to collaborate.124 It was the pressure of this rivalry within the labour movement
in Ahmedabad, rather than, for instance, the power of capitalist interests within the Congress, which drove the highly repressive policies towards labour – including the
formulation and passage of the Bombay Trade Disputes Act of 1938, and directed primarily against the communist unions – adopted by the provincial Congress
ministry in Bombay between 1937 and 1939.125 Significantly, it was the threat of labour militancy which persuaded the millowners to accept in 1939 the
121 'Memorandum on the Reorganization of the Administration of the Textile Labour Association, 1933–34', File 47, ATLA Papers, ibid.
122 Annual Report of the Ahmedabad Millowners' Association, 1923, pp. 32–3, cited by Salim Lakha, Capitalism and Class in Colonial India: The Case of Ahmedabad (New Delhi,
1988), pp. 111–12.
123 GOB, Home (Special), File 550 (25) IIIA of 1938, MSA.
124 ATLA Papers, File 3, Part 2, microfilm copy, reel no. 9, NMML.
125 GOB, Home (Special), Files 550(24) of 1938, 550(25) III A, MSA; AITUC Papers, File 59, NMML; Markovits, Indian Business and Nationalist Politics, especially pp. 150–78; see
also ch. 3 below.

Page 305
principles over which the 1918 and 1920 strikes had, in part, been fought: that, under the terms of the Industrial Disputes Act, the ATLA would be the exclusive
representative of the workers and that arbitration would be mandatory.126
Large claims have been made for the Ahmedabad Textile Labour Association and many of them should be treated with scepticism. It is true that the ATLA, quite
unlike most other unions in India, achieved a certain permanence and boasted an impressive longevity. But it benefited from the local and national standing of Gandhi
and from its own early entrenchment, through its alliance with the Congress, within the wider structure of politics. It was also blessed at times by the favour and
tolerance of officials and employers. The readiness of the employers to deal with workers' grievances only when they were routed through the machinery of the union
was a privilege enjoyed by few other effective trade unions. In the quotidian context of the workplace, it was difficult for most workers, except, perhaps, the most
skilled, to operate consistently outside its framework, let alone in opposition to it. This fact helped to lodge the ATLA as an agent of collective bargaining in the
industry. It is not intended to suggest, however, that the ATLA was merely the stooge of the millowners. Despite the continued impression that industrial conflict in
Ahmedabad 'partakes the nature of a family [dispute] and unlike many family disputes is conducted with a complete lack of acerbity',127 the ATLA was not simply a
company union. The tensions which its presence generated, after all, served to divide the Ahmedabad Millowners' Association for twenty years.
In the long view, however, its record is far less impressive. Its appeal appears to have been rather sectional, manifesting itself most strongly among Harijan spinners.
Weavers, and especially the Muslims, initially remained sceptical of Gandhi in 1918, joined the union subsequently and then left it in large numbers in 1923, were
increasingly alienated from it in the 1930s, when less than onetenth of their numbers counted among the members, and formed what was effectively their own union in
the Mill Mazdoor Sangh under communist leadership.128 Despite its acceptance as the workers' representative, and for all the influence it appeared to command in the
wider political sphere, employers were able to dismiss strikers with greater regularity and in larger numbers in Ahmedabad than was ever possible in Bombay.
Similarly, the ATLA was
127 Collector of Ahmedabad to Secretary, Government of Bombay, Judicial, 25 February 1918, Bombay Confidential Proceedings, Judicial Department, for March 1918, vol. 36, p. 33,
OIOC.
128 GOB, Home (Special), File 550 (25) IIIA of 1938, MSA.

Page 306
unable to defend the wage levels of its members. In 1918, the general strike was settled with some reference to the existing wage levels in Bombay. For the next two
decades, the Bombay millowners watched with envy, and with regret, as their Ahmedabad rivals were able to cut wages without always paying the price of the bitter
and protracted general strikes on the scale which they had endured. Fred Stones, managing director of the E. D. Sassoon group of mills in Bombay, observed
somewhat ruefully in 1939, 'Had we Gandhiji in Bombay I say our firm would have saved lakhs and lakhs of rupees.'129 It is doubtful that Gandhi would have denied
the charge. 'You have given me the credit', he told the Ahmedabad millowners, a whole decade earlier, in 1928, 'for keeping the city of Ahmedabad free from a labour
upheaval such as Bombay is at present passing through. Well, I cannot quite disclaim that credit.'130 Indeed, the ATLA often found itself following the lead of the
workers with some embarrassment or alternatively distancing itself from the strikes of its members. Although the arbitration machinery established the ATLA as the
main and most effective channel through which the workers could express their grievances, the millowners frequently ignored this machinery to make their own private
settlements outside the formal institutions so elaborately designed for the industry as a whole. While labour policy became the cause of open schism in the Millowners'
Association, Gulzarilal Nanda had begun to complain vociferously by 1933 that the arbitration machinery was on the point of collapse and threatened to take much of
the union's organization away with it. So for all the longevity and institutionalization of the ATLA, this is what the 'indigenous experiment in trade unionism' had come to.
'The difficulty', as he put it, was rather 'fundamental':
It is the mental reservations on the side of the employer in accepting the Union and the procedure of negotiations and arbitration. As long as it is a question of tolerating the
organization as a necessary nuisance when it is too strong for them and constantly trying to undermine its strength and influence, the conditions for successful cooperation do
not exist.131
The fact that political linkages with the Congress could bring various workingclass groups tangible benefits whether in the workplace or the neighbourhood may help
to explain why the Congress could attract a significant workingclass following. But the limitations of Congress interventions in the problems and politics of the working
classes, combined
129 Proceedings of the TLIC, Main Inquiry, Oral Evidence, Mr. F. Stones, Director, Messrs. E. D. Sassoon and Co., File 72, p. 3624, MSA.
130 Young India, 10 May 1928.
131 Gulzarilal Nanda to Shankarlal Banker, 23 March 1933, ATLA Papers, File 40, 'Papers Relating to the Depression, 1933', pp. 77–83, microfilm copy, reel no. 1, NMML.

Page 307
with the ambivalence towards labour which was inherent in its rhetoric and programme, may also help to explain why workingclass support for the Congress was so
often sporadic and intermittent. At certain times and some places, the Congress was able to win the adherence of specific groups of workers.
The hybridity and sectionalism of the working class and its fragmentary and uneven response to the Congress has complicated the task of identifying the content and
character of workingclass nationalism. Clearly, this nationalism did not arise out of a universal commitment to a territorial principle. To some extent, workingclass
participation in the nationalist campaigns invited the Congress to redefine its own varied, sometimes unformulated concept of the nation. Workingclass support for the
Congress did not stem from a general concern that the interests of the Indian people were homogeneous and indeed it occurred despite the persistent and often painful
recognition that they were not. Moreover, the work experiences of different groups of workers in the same industry and the same town, let alone in the same industry
in different centres or different occupations in different towns, varied considerably. But these experiences and conditions of work clearly inflected the response of
particular groups of workers to the rhetoric, programme and ideology of the Congress.
Nationalist ideologies were likely to disclose a more accessible and immediate social meaning on the railways. Racist and discriminatory practices pervaded the
operations and organization of work in the railways. Railway colonies, inhabited by 'European' and 'Eurasian' officers and supervisors, fostered conditions of apartheid.
'Europeans' enjoyed a virtual monopoly of the higher paid jobs and the vital operations. Indian workers, or 'coolies', were employed in the lowestpaid 'native grades'.
Structures of pay, promotion and apprenticeship were highly discriminatory. The jobs and promotion prospects of most Indian employees, even in the more skilled
occupations or in the intermediate wage categories, often depended on the whim or the good will of their 'European' or 'AngloIndian' supervisors, who were
sometimes less skilled, if better paid. Considerable violence and brutality, we are told, were used in the discipline and 'management' of labour132 – although whether
there was
132 Proceedings of the MCC, Defence Statement, K. N. Joglekar, pp. 1768–76; Jagga, 'Colonial Railwaymen and British Rule', pp. 106–14; BPP SAI, 1916, no. 29, 22 July, para 958;
Arnold, 'Industrial Violence', pp. 249–54; I. J. Kerr, Building the Railways of the Raj, 1850–1900 (Delhi, 1995).

Page 308
133
anything specifically 'colonial' about this is a more problematic and vastly underrated question.
Work on the railways was characterized by a complex division of labour and intricately devised structure of grades and functions, finely drawn hierarchies and fiercely
maintained demarcations of status, all of which was overlaid by patterns of racial exclusion and segregation. Railwaymen, even in the native grades, were notoriously
fond of their uniforms and, like university bureaucrats, revelled in the pomp of office, however lowly. This 'petty bourgeois sentimentalism', as K. N. Joglekar, the
Bombay communist, once described it, was often deemed to be an obstacle to the unity of workers across the various grades, ranks and departments. The service and
clerical occupations looked down upon the workshop hands and both groups held those who worked 'on the line' in the deepest contempt. Nonetheless, their
resentment of the privileges of European and AngloIndian workers could override these prejudices and sometimes created the possibility of alliances across
entrenched, customary divisions among Indian railwaymen. The nationalism of railway workers was nourished by racial conflicts between European and Eurasian
foremen and Indian workers.134
It is not intended to suggest that the frequency of strikes on the railways between 1917 and 1922 and, indeed, between 1928 and 1931, demonstrated the submission
of workers to the nationalist programme. Of course, several hartals and stoppages did occur on various railways in 1921–22 explicitly in response to the Khilafat and
noncooperation movements. Most strikes, however, focused on issues directly related to wages and the conditions of work, and not infrequently to the abuse and
violence meted out by supervisors and managers. Nonetheless, their occurrence was not unrelated to the rhetoric and objectives of the anticolonial agitations.
Occasionally, as in the case of the North Western Railwaymen's Association, the conduct of a strike threatened to effect an alliance across the racial divide. The North
Western Railwaymen's strikes in 1919 and 1920 were inextricably bound up with the Rowlatt satyagraha, the Khilafat movement and the noncooperation agitation. In
May 1920, however, a strike which began in response to the dismissal of seven fitters of the Moghulpura workshop in Lahore turned into a battle for the recognition of
the union. The support which this action secured from the European engine drivers, briefly appearing to knit together the whole workforce, forced the employers to
accede to the union's demands. The inspiration behind the union's campaign was the charismatic John 'the Mahatma' Miller, formerly an engine driver himself and a
'European',
134 Proceedings of the MCC, Defence Statement, K. N. Joglekar, pp. 1770 ff.

Page 309
135
who had since 1907 been dismissed by at least three railway companies for his role in organizing strikes. But this solidarity was both too difficult to sustain and too
good to last. In the early 1920s, 'Mahatma' Miller crumbled under the pressures of his own charisma. Following its triumphs of 1919–20, 'corruption set in' and the
union was said to have became 'a grazing ground' for its leadership. In 1925, a campaign of passive resistance in the Rawalpindi workshop against retrenchment on the
North West Railway led to a lockout and the strike spread along the line from Karachi to Delhi. The union, led by Miller, now opposed the strike; the workers
sustained it for four months. By the time the strike collapsed, the union had become moribund and 'Mahatma' Miller was 'rewarded . . . with a kushi job on the North
West Railway'.136
Where 'Europeans' were found in profusion and dominated the structure of supervision, discipline and, indeed, ownership, it was for workers far easier to identify
nationalist sentiments with their everyday experiences or, conversely, to develop a social interpretation of the nationalist question. But it would be simplistic and
misleading to conclude that workers' nationalism arose only from, or directly in relation to, their personal experience of 'European' capitalists, managers and overseers.
The Congress found its most devoted, loyal and strongest workingclass following in Ahmedabad, where owners, managers and jobbers were not only Indian but
predominantly Gujarati.137 In Bombay, the millworkers manifested a greater enthusiasm for the nationalist cause even as the ranks of supervisors and mill managers
became increasingly Indianized – although the situation was complicated by the fact that they frequently identified Parsi supervisors, managers and millowners with the
colonial regime.138 In fact, it is one of the least observed peculiarities of Indian nationalism that its rhetoric was largely free of racism – or at least that it saved its racism
primarily for other Indians. However, it was when the lines of antagonism, which marked their social relations both within and beyond the workplace, could be readily
traced to the wider context of the colonial state that the protean languages of nationalism most obviously provided the working classes with an explanation, and a
remedy, for the social and political conditions in which they found themselves.
135 Jagga, 'Colonial Railwaymen and British Rule', pp. 120–7, esp. p. 120, n. 51.
136 Proceedings of the MCC, Defence Statement, K. N. Joglekar, pp. 1781–4.
137 Mehta, The Ahmedabad Cotton Textile Industry; Gillion, Ahmedabad; Patel, The Making of Industrial Relations.
138 Chandavarkar, The Origins of Industrial Capitalism, pp. 411–20; S. Bhattacharya, 'Swaraj and the Kamgar: The Indian National Congress and the Bombay Working Class, 1919–
1931', in Sisson and Wolpert (eds.), Congress and Indian Nationalism, pp. 223–49; R. Kumar, 'From Swaraj to Purna Swaraj: Nationalist Politics in the City of Bombay 1920–32', in D. A.
Low (ed.), Congress and the Raj: Facets of the Indian Struggle (London, 1977), pp. 77–107.

Page 310
It might be supposed that in Bengal, where capital was overwhelmingly dominated by 'Europeans', where the jute mills were both owned and managed by them, where
expatriates observed and maintained the finest caste distinctions among themselves as well as in relation to Indian merchants, the working classes would more readily,
indeed naturally, identify themselves with the Indian nation. In fact, workingclass nationalism owed as much to the social relations of the workplace as to the political
conflicts of the milltowns. Here, subordinated bhadralok elites sought to wrest control of the town municipalities from the Scottish mill managers and their Indian allies
by drawing on workingclass support. They developed this workingclass following partly by their willingness to intervene in the disputes of the workplace but, more
especially, by the promise that placed in power they might act to improve conditions of housing and sanitation and restrain the police.139
The formation of a labour force in the plantations and mines was as deeply marked by colonial precept and practice as it was on the railways. It was a function of the
generic weakness of expatriate capital in South Asia that it had sought its fortune on the margins of agrarian settlement since the early nineteenth century. The rough
andready capitalism which expatriate entrepreneurs created so cheaply was particularly exploitative and oppressive of labour. The discourse of labour deployment
and discipline, primarily, perhaps even quintessentially, the colonial discourse, and one which served, and continues to serve, the interests of Indian capital so
handsomely, was fashioned in the coal mines, plantations and in systems of labour indenture. In the Jharia coalfields, the support for noncooperation was driven in part
by an alliance between the Indian owners of the small, surface mines and motley interests drawn from the largely 'European'owned sector of large, deep mines –
Bengali clerks and literate service groups, raising contractors and migrant labour from UP (the 'paschimas' to be distinguished from the local 'dehati' and 'jungli' labour
of Santhals, Bauris and other tribal and lowcaste groups employed under particularly harsh conditions in the Indian sector). It did not take long for this alliance to
come unravelled in the 1930s. But it was motivated as much by the search of Indian capital for a larger stake in the industry as by the often economic grievances of the
Indian employees of the larger European mineowners.140
In Madurai, by contrast, similar motives on the part of Indian merchants who dominated the local Congress led them first to seek an alliance with local labour and then
repudiate it. In 1931, the Madurai Congress Committee decided to take on the redoubtable Harvey industrial empire
139 Basu, 'Workers' Politics in Bengal', ch. 3.
140 Ghosh, 'The Labour Movement in the Jharia Coalfield'.

Page 311
during the civil disobedience movement. They named the Harvey mills as importers of foreign yarn, boycotted their products and, with the help of 'the tough,
unscrupulous union boss', Varadarajulu Naidu, they effectively disrupted production. They demanded that Harveys should appoint more Indians to their board of
directors, employ more Indians as mill executives and allow Indians to buy a larger stake in their mills. In fact, the local Congress committee was dominated by the
town's leading merchants, some of whom, including its leading figure, N. M. R. Subbaraman, had already invested in the mills. The prospect that the boycott, and the
financial pressures which it could generate during the depression, could create an opportunity for them to acquire a larger share and extend their control of the mill
would not have been lost on these entrepreneurs, and leading protagonists of the satyagraha. The struggle for truth might yield handsome dividends. On the other hand,
liberating Indian capital from the dominance of British businessmen did not necessarily mean that they should embrace the rights of labour. Certainly, Varadarajulu
Naidu for his part had entertained the hope that by supporting civil disobedience he might extract the employers' recognition of his union and thus secure its position as
the principal bargaining agent in the local industry. However, the leading lights of the Congress campaign approached Varadarajulu Naidu with the utmost caution.
They were reluctant to allow the millworkers to participate in the civil disobedience campaigns for fear that such mass participation might threaten their own control
over the movement. Moreover, as cloth merchants and industrialists themselves, they had little wish to aggravate the problems of labour control in the town. Having
now acquired a larger stake in the mill, they could scarcely countenance Varadarajulu Naidu's suggestion that the authority of his union to represent the workers be
recognized. The larger their financial stake in the industry, the less they could afford to risk the entrenchment of trade unions within its structure of management and
control.141
In the labour unrest witnessed in Assam between 1920 and 1922, the strikes, violence and desertions of the tea garden labourers arose out of disputes about wages
and working conditions. Although the planters insisted on explaining these conflicts in terms of the machinations of the Congress agitators, local officials found, as the
Deputy Commissioner of Lakhimpur told the Assam Labour Enquiry Committee, 'little or no evidence that the strikes were due to the influence of the Congress
agitators', while his counterpart at Darrang was willing only to go so far as to say that the noncooperators 'had created an atmosphere which was favourable to the
occurrence of strikes and outbreaks among the ignorant
141 Murphy, Unions in Conflict, pp. 139–41.

Page 312
142
coolies'. On the other hand, the socalled sympathetic strike on the Assam–Bengal Railway, which followed the Chargola exodus and the Chandpur incident, does
not appear on detailed scrutiny to have been particularly sympathetic. The railwaymen who went on strike wished to register their protest, not primarily against the
violence which the troops visited upon the tea garden labourers at Chandpur railway station, but especially against the Gurkhas who, while attacking the fleeing tea
garden labourers, had been sufficiently indiscriminate to assault them as well. Once the strike began, both railwaymen and steamer crews focused upon their wages and
working conditions – including racially discriminatory wage rates – and the union, led by the then noncooperator, J. M. Sen Gupta, quickly distanced itself from the
nationalist agitation.143 Ironically, at the same time, C. R. Das, the Bengal Congress leader, was busy declaring that 'if it had been a labour strike, a mere question
between the employer and the employed, I should certainly have discouraged it from the Congress point of view'.144
The response of workers to specific forms of political action adopted in the various Congress agitations also varied considerably. Thus, in Bombay, the picketing of
liquor shops found the most enthusiastic response during the noncooperation campaign while the boycott of foreign cloth was received most readily in the early 1930s.
The reasons are not far to seek. First, in the initial stages of noncooperation, picketing concentrated primarily on foreign liquor shops. Foreign liquor symbolized
wealth and was readily identified with the customs and habits of the colonial rulers. Second, in the context of the Khilafat wrongs, the theme of colonial injustice was
also infused with religious prohibitions on the consumption of liquor, and, indeed, although both Hindus and Muslims participated in the picketing, the excise
administration claimed that 'where picketing is done by Hindus, it is done in a less objectionable way than where it is done by Mahomedans'.145 Crosscommunal
picketing did not diminish the use of communalist abuse to customers, who when they emerged from the shops were greeted with 'such choice language as . . . if the
customer is a Hindoo whether he had been there to drink cow's blood or in case of a Mahomedan, pig's blood'.146 Theorists of popular culture
142 Report of the Assam Labour Enquiry Committee, 1921–22, p. 19; cited by R. P. Behal, 'Forms of Labour Protest in Assam Valley Tea Plantations, 1900–1930', Economic and
Political Weekly, 20:4 (26 January 1985), Review of Political Economy, pp. PE21 to PE22.
143 Gourlay, 'Nationalists, Outsiders and the Labour Movement in Bengal', pp. 49–53.
144 Mussalman, 17 June 1921, cited by Sarkar, 'The Conditions of Subaltern Militancy', p. 295, fn 97.
145 Chief Excise Inspector, Bombay, to Superintendent of Salt and Excise, Bombay, 7 November 1921, in GOB, Home (Special), File 355 (21) F of 1921, p. 15, MSA.
146 Ibid.

Page 313
might have discovered a syncretism on the picket lines which overrode the effects of a language, often taken at other times and places to explain the occurrence of
communal riots. Third, the bulk of liquor shops were owned by Parsis, from whose ranks were also drawn the most prominent and the most loyal millowners, a
significant proportion of mill managers and officials and a large number of skilled workers and supervisors in the cotton mills and the railway workshops. These were
the 'Biryaniwallahs of Willingdon Club' who supported the Willingdon Memorial in 1918 and seemingly refused to observe the hartal which greeted the Prince of
Wales on his visit to Bombay in November 1921. In fact, despite the opposition of K. F. Nariman, Burjorji Bharucha and the nationalists of the Bombay Chronicle
office, the Parsi Panchayat had insisted on presenting an address to the Prince and in the midst of the general observance of the hartal, extending to him 'an enthusiastic
welcome'.147 The fact that communal riots in Bombay in the nineteenth century had primarily occurred between Parsis and Muslims – arising out of commercial rivalries
in the shipping and export trades – added a further dimension to these tensions. What were commonly described at the time as the Prince of Wales riots in November
1921 at times acquired the character of a communal riot in which Parsis became the target of the violence of mainly, but not exclusively, Muslim crowds enthusiastically
assisted by Hindu 'satyagrahis'. In the case of the boycott campaigns in the early 1930s, the connections between the imports of foreign cloth and the economic
troubles of the industry were easily made. Indeed, the millowners in their excise duty, rupee ratio and tariff campaigns as well as the Currency League had repeatedly
made them. Foreign competition, it was said, had resulted in wage cuts in 1923 and 1925 and it was also supposed by millworkers to lie at the root of the
retrenchment and rationalization schemes which were central to the strikes of 1928–29.148
Attempts by the Congress to attract a workingclass following could serve to accentuate the sectionalism of the labour force, not solely to bind its solidarities. In the
1920s, as labour became increasingly peripheral to the concerns of the Congress, labour leaders who styled themselves as Congressmen often found themselves in
conflict with the local Congress party, with Swarajist ministries or local mercantile and propertied interests which espoused and constituted the Congress. Conversely,
as its own organizational machinery disintegrated, the Congress exercised less direct control over its local publicists, including those who acted as labour
147
Note by Deputy Commissioner of Police, CID, Bombay, 1 December 1921, enclosed in Commissioner of Police, Bombay to Secretary, GOB, Home, 1 December 1921, in Bombay
Confidential Proceedings, Home Department, December 1921, vol. 62, p. 777, OIOC.
148 Chandavarkar, The Origins of Industrial Capitalism, chs. 8 and 9.

Page 314
leaders. This situation was to change in the 1930s, when the High Command tightened its control over the rank and file and when leading labour became a rather more
specialized profession within the fold of the Congress. Although one consequence of this changed context in the 1930s was the growth of socialist and communist
agitations, Bombay city had witnessed the left rising on the crest of a wave of workingclass militancy much earlier in the late 1920s. The decay and decline of the city's
Congress was manifested in the growing prominence of communist activists in the ward and district Congress committees.149 During the civil disobedience movement,
and in its immediate aftermath, the Congress, in seeking to revive its local organization, tried also to extend its linkages with workingclass politics. But since the general
strike of 1928, the mill districts of Bombay had been dominated by the communists of the Girni Kamgar Union. As the conduct and the collapse of the strike in 1929
laid bare the divisions within the working classes, and the Meerut arrests by removing the more prominent leadership created new opportunities for, and sharpened old
rivalries among, those who had been left behind, the Congress sought to develop a more extensive workingclass following. At first, they cultivated those who had long
harboured an antagonism against the communists, notably Mayekar and Keshav dada Borkar. But it soon became apparent that it was essential to break into the
domain of the Girni Kamgar Union. Thus, S. K. Patil, the doyen of the city's Congress, with the assistance of Vallabhai Patel, courted the Kandalkar faction of the
Girni Kamgar Union.150 During the 1930s, as unemployment and repression combined to undermine the solidarities forged by the Girni Kamgar Union, the Congress
became an alternative source of support for those labour leaders who were disappointed and disgruntled with the dominant factions within the labour movement. Of
course, it was not unusual for labour leaders driven by conflicting ideological commitments or diverse political connections to compete with each other to develop a
following. Nor is it surprising that this competition could extend into workplace and neighbourhood and attach itself to deeper and more fiercely contested rivalries in
the daily social relations of the working classes. In Bombay, however, its significance lay in the fact that it was the Congress which by trying to acquire a stake in this
seemingly volatile political constituency fed the rivalries, deepened the antagonisms and accentuated the sectionalism of the working classes. Its effect was to further
weaken the solidarities which had been forged in the strikes of 1928–29.
The sectionalism of the working class could sometimes take caste and
149 GOB, Home (Special), File 143 (K) Part IV (a) of 1928; File 143 (K) VII of 1928; and File 543 (18) C of 1928, MSA.
150 GOB Home (Special), File 750 (39) – II of 1930, MSA.

Page 315
'communal' forms. Since competition for jobs (as well as housing and credit) could follow lines of caste and communal difference, trade union rivalries, sometimes
promoted by the Congress, could acquire a communal edge. Employers frequently sought to diversify the caste and communal composition of their labour force to
extend their control over it.151 When, during strikes, they tried to recruit workers of a different religious, caste or regional identity from those who had struck, their
attempts to manipulate the social composition of the workforce could deepen rivalries, provoke violent conflict and lead – as it did in Bombay in February and May
1929 and again in May 1932 – to communal riots.152
Congress agitations, especially the picketing of liquor and cloth shops could provoke communal antagonisms and sometimes led to largescale violence. Young men,
collected together in large groups, intoxicated by the carnivalesque spirit which prevailed on these occasions, sometimes puffed up by their righteousness, were unlikely
to proceed with sensitivity and care or attend the disciplines of nonviolence, and they readily and frequently gave offence. It has already been shown how the
picketing of liquor shops and the hartal organized for the Prince of Wales's visit to Bombay in November 1921 led to a communal riot between Parsis and Muslims.153
Similarly, picketing generated resentment and led to communal violence in Banaras in 1931.154 In 1932, the picketing of piecegoods shops in Bombay which sold
foreign cloth exacerbated rivalries in the trade and, because Muslim traders were largely, though not exclusively, involved in the import trade, it intensified communal
antagonisms and fed into the riots which engulfed the city in May.155
151 Examples may be found in Bengal in the 1890s, Madras in the early 1920s, and in Kanpur in the late 1930s. For Bengal, see Das Gupta, 'Factory Labour in Eastern India', pp. 289–
303; Chakrabarty, 'Communal Riots and Labour', pp. 150–4; Basu, 'Workers' Politics in Bengal', ch. 4; on Kanpur, see Joshi, 'Bonds of Caste, Ties of Religion', and Joshi, 'Kanpur
Textile Labour', p. 1827; on Madras, see E. D. Murphy, 'Class and Community in India: The Madras Labour Union, 1918–21', IESHR, 14:3 (1977), 292–321.
152 For the 1929 strike, see G. D. Birla to P. Thakurdas, 4 May 1929, Thakurdas Papers, File 81 (II) of 1929, pp. 287–85; G. D. Birla to P. Thakurdas, 4 May 1929, Thakurdas Papers, File 42
(II), of 1923–34, NMML. See also, GOB, Home (Special), File 348 Part II of 1929 and GOB, Home (Poll), File 344 of 1929, MSA. On the 1932 riots, see Commissioner of Police, Bombay to
Secretary, GOB, Home (Special), 4 April 1932 in GOB, Home (Special), File 792 of 1932, MSA. Of course, explanations for communal antagonisms and communal riots, to be plausible
and to be adequate, will have to extend beyond the social and economic conflicts which informed them. Crucially, the hiring of Muslims and Pathans to break strikes was not the only
cause of these three Bombay riots.
153 Bombay Confidential Proceedings, Home, December 1921, vol. 62, pp. 771–81, OIOC.
154 Pandey, The Ascendancy of the Congress in UP, pp. 129–30 ff.; Gooptu, 'Political Culture of the Urban Poor in North India', ch. 5.
155 GOB, Home (Special), Files 792 of 1932, 793 (1) of 1932 and 793 (1), Parts I –III of 1932, MSA.

Page 316
The attempt by Congress 'volunteers' to enforce a hartal in Kanpur on the day of Bhagat Singh's execution led to a ferocious communal riot which left 400 people
dead and 1,200 injured in three days of violence.156 Such riots are detonated on long fuses and many of these had been working their way through the region for over
a decade. The reluctance of Muslim traders, shopkeepers, artisans and the urban poor to participate in the hartal grew in part out of their resentment about Gandhi's
withdrawal of noncooperation, which many perceived as a betrayal of the Khilafat cause. The intensification of communal tensions in the 1920s found expression in
the revivalist shuddhi and tanzeem movements. The decline of the trades which Muslims had once dominated, the rapid rise of some Hindu traders on the back of local
economic expansion during the first world war and in its immediate aftermath, and the increasingly fierce competition for jobs, housing and credit as the flow of
migrants to the city gathered pace, each contributed to the entrenchment of communal antagonisms. More generally, political discourse in the early twentieth century
was fashioned increasingly by a sectarian vocabulary, as political groupings defined themselves in caste or communal terms to gain access to the structure of colonial
power and its system of representation. Thus Harijans and backward castes sought to define their own 'communities' by inventing past glories, which, it was inevitably
said, had been flattened by Muslim invaders, thus explaining their current lowly status in communal terms.157 Factional rivalries within the provincial Congress often led
their protagonists to cultivate a wider constituency by espousing an increasingly explicit Hindu rhetoric.158 In a sense, the policies of government and the politics of the
provincial Congress sanctioned the increasingly free expression of communal antagonisms in public discourse.
It has been a central contention of this book that relations of power which constituted the working classes cannot be understood in terms of class alone. The formation
of the working classes also encompassed relations of power described by caste and kinship, village and neighbourhood, skill and occupation, age and gender, religion
and nation. None of these signified a permanent and unchanging, or a compelling and necessary social and cultural identity or a state of 'consciousness'. How and why
workers identified with one or other of these affinities was determined by
156 Report of the Congress Cawnpore Riots Enquiry Committee in N. G. Barrier (ed.), Roots of Communal Violence (Delhi, 1976); Report of the Commission of Inquiry into the
Cawnpore Riots and Resolution of the Government of the United Provinces, PP, 1930–1, vol. XII, Cmd. 3891; Pandey, The Ascendancy of the Congress in UP, pp. 129–42; Joshi,
'Bonds of Community, Ties of Religion', pp. 269–73 and passim; Gooptu, 'Political Culture of the Urban Poor in North India', ch. 5.
157 Gooptu, 'Political Culture of the Urban Poor in North India', ch. 4.
158 Pandey, The Ascendancy of the Congress, pp. 115–27.

Page 317
the specificities of a particular political conjuncture. Furthermore, it should be apparent that the efforts of the Congress to 'mobilize the masses' could serve to divide
them; the latter possessed no inherent or natural propensity to unite, let alone behind the Congress. The rhetoric of nationalism yielded competing images of the nation.
This competition rested on the claim made by some groups that they represented the nation more fully and effectively than their rivals. Thus, the rhetoric of a Hindu
nationalism might claim to distil the essence of the Indian nation, to represent the 'real people' better than their noncommunal variants; and it is clear that to some
sections of the working classes this rhetoric appeared plausibly to explain their circumstances and hold out the promise of redress. While the Congress claim to
represent the nation as a whole precluded, or at least set firm limits to, its espousal of a particular class or religious identity, it also declared an intent to transcend and
subsume communal difference, even if the rhetoric of some Congressmen was suffused by an increasingly strident Hindu idiom. In the 1920s and 1930s, the Congress,
in the mainstream, sought to place their nation on a firmly popular footing, not only through mass mobilization, but by representing the nation in terms which made it
interchangeable with 'the real people', 'the common man' and 'the meanest of our countrymen'. But here lay the rub. For, in nationalist, as in colonial discourse,
'communalism' was also explained as the product of ignorance and superstition and its symptoms were thought to be most commonly displayed by simple and ordinary
people. It followed, therefore, that they needed to be protected from themselves, not simply from the machinations of bigots and reactionaries or from colonial pro
consuls seeking to divide and rule. In this way, one of the many contradictions of Congress rhetoric was elaborated. For nationalism was called upon to transcend the
people and, at the same time, to represent them, to protect them from themselves and yet become of them. 'The Congress', as Jawaharlal Nehru had told the Lucknow
Congress in his famous presidential address, 'must be not only for the masses, as it claims but of the masses'.159
VI
However, there was one important theme of Congress rhetoric which had significant implications for the development of workingclass politics. But its consequences
were unintended and unforeseen. In the 1920s and 1930s, the colonial state was subject to increasingly stringent political criticism. The rhetoric of the Congress
became more explicitly anti
159 The Selected Works of Jawaharlal Nehru, vol. VII, p. 178.

Page 318
colonial. It no longer spoke of the unBritishness of British rule in India, but questioned the morality of the colonial state. It challenged the justice of its laws; it withheld
its cooperation from the Government; it elevated civil disobedience into an act of heroism. Although some Congressmen entered the councils in 1923, council entry
could only be justified by the objective of wrecking the legislatures from within. Of course, this had a greater symbolic or rhetorical value than any consistent practical
effect.
In the domain of workingclass politics, its benefits accrued not to conciliatory Congress trade union organizers but to political groups who, seeking to represent
labour, were marginalized by employers as well as the state. These publicists, notably communists, were forced by their own marginality in industrial politics, by their
exclusion from the workplace, not only to adopt a stance of continuous opposition to the employers and the state but also, through their repeated interventions in the
disputes of workplace and neighbourhood, to realize this rhetoric in political action. In the 1920s and 1930s, one of the most significant factors in workingclass
politics was the growing presence of the state. Attempts to recruit workers for a distant and irrelevant war caused considerable disquiet in 1917. The economic
dislocations of the war and its immediate aftermath were widely perceived as a failure of government. The grain riots in Madras in 1918 were the result, we are told, of
'popular resentment . . . against the government [which] seemed to be exercising no control to ensure supplies and keep prices down'.160 The tariff, excise and
monetary policies of the state impinged directly upon working conditions, employment and wages, and sharpened the lines of antagonism between workers and the
state. The Bombay millowners frequently justified their wage cuts or rationalization schemes as the necessary, though undesirable, consequences of the government's
tariff and exchange policies. Lower wages, higher workloads and greater unemployment lay at the heart of the communistled general strikes of 1928–29 in
Bombay.161 Largescale, sometimes industrywide, strikes, frequently prolonged and bitter disputes, stoked official anxieties about the public order. Strikers were
more closely policed. The intervention of the state to settle disputes worked more often to defeat their objectives and to negate their demands than to secure them. In
the 1930s, the effect of trade union legislation was largely to narrow the already limited freedom of manoeuvre which workers and their representatives had previously
possessed.
The political experience of the working classes was constituted in relation to the state. In the late 1920s and 1930s, in Bombay and
160 D. Arnold, 'Looting, Grain Riots and Government Policy in South India, 1918', Past and Present, 84 (1979), 145.

Page 319
Sholapur, and later in the decade in Coimbatore, Kanpur and Calcutta, communist trade unions gained considerably from their stance of consistent opposition to the
state. The anticolonial rhetoric of the Congress, by challenging the legitimacy of the state, paid handsome dividends to trade unions and political groups which pursued
an active strategy of confrontation. Opposition to the state provided a focus around which a fragmented and sectionalized working class could at times coalesce,
though not necessarily behind the banner of the Congress – indeed often under the hammer and sickle of various socialist and communist groupings.
VII
In the 1940s, however, this situation was reversed. It became much harder for the working classes to perceive the left, and especially the communists, in this
oppositional role. After 1941, the communists abandoned their oppositional stance altogether and adopted an increasingly conciliatory method in industrial disputes.
They sought actively to prevent strikes, to encourage greater productivity, even to dissuade workers from participation in nationalist agitations and tried to convince
their once enthusiastic followers that they were fighting the people's war.162 On the other hand, the Congress now appeared to be the only party which was willing to
move against the state. Having accepted office in 1937, they had spurned it after 1939. The Quit India movement was a far more impressive campaign than civil
disobedience had been and it attracted widespread support in Bombay and Gujarat, UP and Bihar, Bengal and Orissa.163 According to the Viceroy, it was 'by far the
most serious rebellion since that of 1857'. The immediate future, he feared, would bring 'a formidable attempt to renew this widespread sabotage of our war effort',
which could if successful 'damage India irretrievably as a base for future allied operations'.164 In 1942, however, the British commanded
162 The fate of the communists in Bombay during the second world war can be followed in GOB Home (Special), File 543 (13) – B (4) of 1941–3 and File 543 (13) – B (5) of 1943–5,
MSA.
163 A. C. Bhuyan, The Quit India Movement: The Second World War and Indian Nationalism (Delhi, 1975); F. G. Hutchins, Spontaneous Revolution: The Quit India Movement (New
Delhi, 1971); F. G. Hutchins, India's Revolution: Gandhi and the Quit India Movement (Cambridge, Mass., 1973); Y. B. Mathur, Quit India Movement (Delhi, 1979); S. Henningham,
'Quit India in Bihar and the Eastern United Provinces: The Dual Revolt', in R. Guha (ed.), Subaltern Studies, vol. II, pp. 130–79; P. Greenhough, 'Political Mobilization and the
Underground Literature of the Quit India Movement, 1942–44', MAS, 17:3 (1983), 353–86; P. N. Chopra (ed.), Quit India Movement: British Secret Report (Faridabad, 1976); G. Pandey
(ed.), The Indian Nation in 1942 (Calcutta, 1988).
164 Linlithgow to Churchill, 31 August 1942 in P. N. Mansergh (ed.) The Transfer of Power, 1942–47, vol. II: 'Quit India', 30 April–21 September 1942 (HMSO, London, 1942),
Document no. 662, pp. 853–4.

Page 320
165
greater military force than at any time in the history of their rule. This was no time for 'roughhousing Indian rioters' with 'bandobast and bluff' – the habitual role of
the Indian Army in the 1920s and 1930s. The British moved their artillery and their aeroplanes swiftly into action and the Quit India movement was suppressed, rather
ruthlessly, within a month. In parts of Bihar and east UP it rumbled on for months and even years.166
Between 1942 and 1945, Congressmen were imprisoned in droves, while the communists, standing shoulder to shoulder with the British to fight the people's war,
were ironically able to operate openly and legally. During this period, communist labour leaders entrenched themselves effectively in political institutions, and especially
in the trade union movement. Where the communists had been relatively marginal to the labour movement in the 1930s, for instance in Bengal, they took their
opportunity to capture the provincial Trade Union Congress gleefully. But in the 1946 elections in Bengal, the Congress swept the polls in the labour constituencies,
with only two exceptions, one of them being Jyoti Basu, the future leader of the CPI (M) and the chief minister of West Bengal.167 In Bombay, however, where the
communists had dominated the labour movement, they knew only too well that dominating the committees of evanescent trade unions did not ensure influence and
control at the level of the workplace and the neighbourhood. It was in the 1940s that their fortunes began to wane. But the loyalty which the communists could inspire
among the working classes in Bombay, especially in the cotton textile industry, was not wholly eroded until the 1960s and 1970s. The People's War was to cast a
lengthening shadow over its strategists as well as its foot soldiers.
It has already been argued that Indian nationalism, as it was represented by the Congress, was primarily a discourse of political exclusion. As the second world war
ended, and swaraj seemed realizable, political liberation brought with it for many in India the promise of social renewal and transformation as well. Now that the
political aims of swaraj were about to be achieved, the social questions which had been subordinated to it would be more explicitly addressed. Between 1945 and
1947, the
165 J. Gallagher, The Decline, Revival and Fall of the British Empire, ed. Anil Seal (Cambridge, 1982), p. 137.
166 M. Harcourt, 'Kisan Populism and Revolution in Rural India: The 1942 Disturbances in Bihar and East United Provinces', in Low (ed.), Congress and the Raj; G. Omvedt, 'The Satara
Prati Sarkar' in Pandey (ed.), The Indian Nation, pp. 223–62; G. Pandey, 'The Revolt of August 1942 in Eastern UP and Bihar', in ibid., pp. 123–64; and C. Mitra, 'Popular Uprising in
1942: The Case of Ballia', in ibid., pp. 165–84.
167 Singh, Life is a Struggle, p. 63; N. Basu, Political Parties and Labour Politics, 1937–1947 (Calcutta, 1992) especially ch. 5.

Page 321
working classes could more closely identify the Congress with a political programme which appeared most consistently and effectively to encompass their interests.
Ironically, at the same time, the objectives of the Congress now focused less on mobilizing these aspirations than, increasingly, on curbing them.168 In 1946–47,
however, as independence took shape before their eyes, it seemed to many, including large sections of the working classes, that there was room for all in the nation of
the future. Nothing suggested this more strongly than the fact that despite the comprehensive failure of Ambedkar's Scheduled Caste Federation in the 1946 elections,
this inveterate champion of separate electorates for the untouchables was first brought into the Constituent Assembly, and then appointed to chair the committee which
drafted the constitution, by the Congress, whom he had in the previous year made the subject of his coruscating invective. Article 17 of the constitution briskly
abolished untouchability. Even some of the temples which had most stubbornly resisted the entry of untouchables opened their doors to them on the eve of
independence.169 Nobody, it seemed, was to be excluded from India's 'tryst with destiny' – nobody that is, except the Muslims of Bengal and Punjab and of the
remote Muslimmajority provinces of the northwest.
Those who remained in India could now be embraced by an inclusive concept of the nation. Significantly, the constitution adopted the principle of universal adult
suffrage and it abolished special and communal electorates. As the Congress drew closer to inheriting the Raj, its social base was dominated by propertied, mercantile
and scribal elites whose interests would ostensibly have been better served by limited electorates than by universal franchise. The adoption of universal adult suffrage,
therefore, requires explanation. It cannot, of course, be explained as the culminating triumph of a 'mass' nationalism. Nor, in view of the many Gandhian precepts which
the Congress flouted, can the decision be understood as deferential to Gandhi's enunciation of universal suffrage as one of the essential preconditions of swaraj in its
most comprehensive sense. Nor can it simply be explained as the logical expression of the inherently democratic impulse of Indian nationalism. There is little to justify
quite such a romantic view. On the contrary, shrewd observers of Indian politics in the 1940s noted with disquiet and apprehension the undemocratic tendency of the
Congress to seek to substitute itself for the apparatus of the colonial state. In the last decades of British rule, the
168 S. Sarkar, 'Popular Movements and National Leadership, 1945–47', Economic and Political Weekly, 17:14, 15 and 16 (annual number, April 1982), 677–89.
169 E. Zelliot, 'Congress and the Untouchables, 1917–1950', in Sisson and Wolpert (eds.) Congress and Indian Nationalism, pp. 193–4.

Page 322
170
Congress had presented itself not merely as a party of nationalist opposition to the colonial state, but as a parallel government, even a stateinwaiting.
The commitment to universal suffrage in 1950 owed something to the lessons which the Congress had learnt in 1937. When the British extended the franchise in 1935,
their aim was to secure their rule on a firmer foundation by creating a political system which would reward their collaborators, like the zamindars of Bengal and Bihar
and the taluqdars of the UP, the big capitalists and moribund Liberals like Tej Bahadur Sapru. They calculated that by extending the vote beyond the thirty million who
were enfranchised by the Government of India Act of 1935, they would favour the Congress and that by excluding the masses, they would favour the Liberals and
their other allies. In fact, they made the mistake of wishing to attach their rule to men whose future lay behind them and whose collaboration would not secure their
position; and they compounded it by assuming that the Congress was the party of the masses. Plainly, it was not. If the British persuaded themselves that they knew
their India, the Congress sometimes took its own rhetoric too literally. As much to the surprise of the British as to the Congress, the latter was elected to power in
seven out of eleven provinces. By limiting the electorate to thirty million, the British had, in fact, enfranchised precisely those intermediate castes and substantial
peasants who in the 1930s had entrusted their hopes and their interests to the safekeeping of the Congress and excluded those who might from deference or
dependence, or even antagonism against dominant peasants, have voted for the taluqdars and zamindars.171 By the late 1940s, it must have been clear to any
discerning Congressman that it was imperative to avoid the mistakes which the British had made in the previous decade.
This imperative was made more, rather than less, urgent by the extent to which the Congress benefited from the settlement of 1947. As the constitutional negotiations
were deadlocked in 1946, and the Congress had pressed for the partition of India, it emerged with a centralized political structure and strong unitary powers. In the
light of its experience of the 1920s, which had demonstrated beyond doubt that it would not
170 R. Coupland, India – A Restatement (London, 1945), esp. pp. 173–6; R. Coupland, A Report on the Constitutional Problem in India, 3 vols. (London, 1942–3); George E.
Schuster and Guy Wint, India and Democracy (London, 1941).
171 Government of India Despatch on Proposals for Constitutional Reform, dated 20th September 1930, PP, 1930–31, vol. XXIII, pp. 71–14, Cmd. 3700; Report of the Indian
Franchise Committee, 1932, vol. I, PP, 1931–32, vol. VIII, Cmd. 4086; Coupland, India, pp. 127–54; Tomlinson, The Indian National Congress and the Raj, pp. 65–85; Baker, The
Politics of South India, chs. 3 and 4; R. J. Moore, The Crisis of Indian Unity, 1917–1940 (Oxford, 1974).

Page 323
flourish in a federation of strong provinces, these were major assets. The sheer size of the electoral system also worked in favour of the Congress. None of its rivals –
now that the Muslim League was removed from the scene – had the resources, experience and political networks to compete on an equal footing for the support of
such a massive, dispersed and heterogeneous electorate. The Congress leadership learned from the experience of 1937 that no political party or social interest with a
firm grip on a strong centre could hope, especially in the context of the 1950s, to long survive a narrowly restricted franchise. To thus leave a significant section of the
population out of the political nation was for the Congress in power to offer its rivals an opportunity to unleash a popular movement against them.
If nationalism had been primarily a discourse of political exclusion before 1947, its main protagonist, having inherited the state, now continued to accommodate the
citizens of independent India within its political domain. The rhetoric of planning and economic development, national integration and nationbuilding offered a stake in
the nation for each of its citizens; and as it came to be embellished more explicitly with the idioms of socialism and the slogan of garibi hatao, it held out the promise that
each might eventually have an equal share. Having been nurtured on the claim that it represented the nation as a whole, the Congress was reluctant radically to alter its
character in the 1950s and 1960s. In 1957, the general secretary of the Congress, stressing the need for ideological clarity, and 'for bringing about the socialist order
through the democratic process', described it as the party of the
landlords as well as the tenant class; there are businessmen as well as representatives of the working class. When natural differences arise as a result of existing conflict of class
interests we try to strike a balance in order to satisfy both sections. We must recognize that there do exist class conflicts and it is necessary to resolve them not through hatred
and violence but through persuasion and democratic legislation. The Congress stands for the ultimate welfare of all sections of the population and desires to hate none.172
It was only in the late 1960s, as foreign exchange and balance of payments crises, and then industrial stagnation began to take effect, that this rhetoric lost its force and
it became increasingly difficult for the Congress to sustain its allembracing definition of nation.173 As deeper and more
172 Shriman Narayan, 'Need for Ideological Clarity', AICC Economic Review, 15 June 1957, cited by F. Frankel, India's Political Economy, 1947–1977: The Gradual Revolution
(Princeton, 1978), p. 160.
173 On the crisis of the 1960s, see P. Patnaik, 'Imperialism and the Growth of Indian Capitalism', in R. Owen and R. Sutcliffe (eds.), Studies in the Theory of Imperialism (London, 1972),
pp. 210–29.

Page 324
ferocious social conflicts revealed how difficult it would be to manage the Indian state on the old terms, the definition of the Indian nation was to be debated within a
new, narrower and more sectarian idiom.
VII
Despite the proliferation of local studies of the Congress, and of nationalist agitations, explanatory frameworks for understanding the relationship between the Congress
and its fluctuating and intermittent 'mass following' have been not only scarce but also rudimentary. Circumstances have made cases. And as circumstances have been
infinitely varied, cases have proliferated. It has been the intention of this chapter to propose that a way out of this historiographical stagnation is to address the
relationship between the Congress and its following specifically as a relationship, or as a 'problematic' in its own right. Studies of 'mass mobilization', by their diffusionist
assumption that ideas and a consciousness of resistance trickled down from the leaders to their followers, have taken the initiative out of the hands of those whose
behaviour needs to be explained, and have thus cast more darkness than light upon the nature of popular politics. Studies of 'popular culture', by seeking polemically to
dissociate the politics of nationalism from 'peasant' or 'subaltern' action, have undermined the very notion of a relationship between them.
In examining this relationship between the Congress and its popular following, this chapter has proceeded on the assumption that popular nationalism was more than
the sum of their diverse class interests or their particular and varied social relations. Thus, it will not be enough to conclude that popular discontent found 'articulation' in
nationalist campaigns – although at specific moments this may indeed have happened.174 Moreover, popular responses to the Congress were not always contained
within the ideological parameters or discursive framework of an Indian nationalism. It could develop a more specific affinity to caste or religious community, or it might
be extended in a more explicitly anticolonial direction, seeking to register its hostility to the colonial state. To suggest that popular support for the Congress could
amount to more than the sum of its social and sectional parts should not invite the assumption that there was a latent nationalism in Indian society waiting to be picked
up and developed by an appropriate leadership, rhetoric or political programme.
The argument that nationalism proved attractive to the class interests which it served is not our concern, not least because class interests were
174 Or, for that matter, they might have found expression in communal conflicts.

Page 325
not naturally solidaristic but inherently fragmented and sectionalized by the diverse relations of power through which they were constituted. The investigation of the
relationship between the Congress and its popular following should begin with a scepticism of the axiom that social groups can be ascribed a single common interest
derived from their relationship to the means of production. Not only the supposedly objective interests of given social groups but even their defining identities were
subject to the flux of changing historical circumstances. Their recognition of the mutuality of their interests, no less than the language in which they were described, were
contingent upon particular and shifting intellectual and political contexts. Theoretical assumptions about the nature and effects of production relations, which have
yielded such abstractions of common interest, have been repeatedly and often rudely brought to ground by the subversive play of politics.
To challenge the assumption of a solidaristic working class is not, of course, to foreclose the possibility that workers might perceive the mutuality of their interests
within a particular historical conjuncture, but rather to reject the notion that its permanence was inscribed in the relations of production and that its permanence
transcended its formation in, and reshaping by, changing historical circumstances. There is no reason to suppose that industrial workers had a natural propensity to
grasp their social situation consistently, or exclusively, in the language of class. If the working classes sometimes perceived their common interests in the language of
class, they could individually and severally under different circumstances grasp their identity in the language of caste, religion or nation. The working classes did not
constitute a natural, if underutilized, resource for nationalism. On the contrary, the ideology of nationalism, like the organization of the Congress, had to be socially
constituted and reproduced. The nationalism of the working classes, perhaps any nationalism, could take diverse forms and find a variety of contradictory expressions.
Its development did not necessarily work to the benefit of the Congress but in some cases and at certain times profited the communists. Similarly, the definitions of
nation produced by these developments could serve to divide a fragmented working class yet further, accentuating differences between Hindus and Muslims,
untouchables and caste Hindus, sometimes between occupations, though probably mediating and temporizing those between the sexes.
The appeal of the Congress depended largely on how it represented its followers to themselves both in its rhetoric and its political action. Whether workers identified
themselves with the Congress, or the nation, was contingent upon the plausibility of its political language, upon whether the working classes could explain their social
and political cir

Page 326
cumstances in terms of its rhetoric and find within its programmatic framework both a convincing diagnosis of their ills and a realizable charter for the redressal of their
grievances, the resolution of their problems and the transformation of their conditions. No conception of nationalist discourse, however, would be sufficient if it
excluded what its protagonists did. The political stratagems and campaigns of the Congress influenced the ways in which workers interpreted their rhetoric. After the
first world war, the rhetoric of the Congress, never univocal even at its most harmonious, came increasingly to be inflected by Gandhi. For all the inspirational quality of
his rhetorical message, Gandhi's popularity among the Ahmedabad millworkers during the 1918 strike was rapidly slipping until he turned the tables on the millowners
and the workers with his fast. In addition, during the strike, his rhetoric was grasped and interpreted in the light of his own imaginative conduct of the dispute: his
accessibility (and that of his lieutenants) to the workers, his attention to detail, his willingness to associate with the needs of the workers and to intervene on behalf of
individual workers on any matter however insignificant, his daily leaflets, his evening meetings on the banks of the Sabarmati.175 In the early 1920s, Gandhian rhetoric
represented the masses more persuasively than, for instance, the nationalism of the preceding decade. Its significance lay in its ability to offer materials from which its
followers could not only grasp their situation with greater clarity but also forge strategies to ameliorate, remedy and alter its conditions. This in itself was not a
permanent, or even necessarily an enduring condition. On the other hand, the ability of the Congress to generalize disputes beyond the locality enabled it to attach itself
to specific and localized struggles and at the same time made it worthwhile for workingclass groups to recruit the Congress at particular moments. The limitations of
the rhetoric of the Congress, and its political context, were sometimes to be cruelly exposed in the following decades and sometimes, as conditions changed, to make
fresh political strategies and opportunities available to its constituents. Political mobilization should be considered, therefore, not as a process of the diffusion of ideas
and beliefs from leaders to followers, from the literate to the ignorant, from the sophisticated to the simple, which stimulates the masses into action, but as a process of
empowerment in which the latter are able to interpret a political discourse not only for an explanation of their present situation but also for a realizable method of
changing it.
175 Desai, A Righteous Struggle.

Page 327
9—
South Asia and World Capitalism:
Towards a Social History of Labour
In the past three decades, the work of dependency theorists and 'world systems' analysts has ensured that it is no longer tenable to study the history of any region,
whether in the West or the Third World, in isolation from the world economy.1 But it is also becoming something of a commonplace that a perspective centred upon
the 'metropole' or the 'core' has tended to peripheralize the history of large parts of the world. The view from the periphery has sometimes suggested that the leading
question is misdirected:2 while it is impossible to deny that the world economy has impinged upon social relations and economic development in India, it is also
important to ask how Indian society shaped and channelled the impact of the West. This chapter will seek to situate the development of
An earlier version of this chapter was presented to a conference at Tufts University in December 1986 on the theme of South Asia and World Capitalism. Some of the papers
presented at the conference were published in S. Bose (ed.), South Asia and World Capitalism (Delhi, 1990). The conference focused upon the work of Immanuel Wallerstein,
which is reflected in the particular emphasis placed in this chapter upon his treatment of world capitalism.
1 I. Wallerstein, 'The Rise and Future Demise of the World System: Concepts for Comparative Analysis', Comparative Studies in Society and History, 16 (1974), 387–415; I.
Wallerstein, The Capitalist World Economy (Cambridge, 1979). For Wallerstein's attempt to place the history of India in relation to the world system, see I. Wallerstein, 'Incorporation
of the Indian Subcontinent into the Capitalist World Economy', Economic and Political Weekly, 21:4 (25 January 1986), Review of Political Economy, PE28 to PE39. For useful
surveys of dependency theories, see A. Brewer, Marxist Theories of Imperialism: A Critical Survey (London, 1980), especially, Part III; K. Griffin and J. Gurley, 'Radical Analyses of
Imperialism, the Third World and the Transition to Socialism', Journal of Economic Literature, 23 (1985), 1089–143. For an interesting and important critique, see R. Brenner, 'The
Origins of Capitalist Development: A Critique of NeoSmithian Marxism', New Left Review, 104 (1977), 25–94.
2 Wallerstein, 'Incorporation of the Indian Subcontinent into the Capitalist World Economy'. For a detailed critique of Wallerstein, see D. Washbrook, 'South Asia, the World System
and World Capitalism', JAS, 49:3 (1990), 479–508. For a different perspective, see A. K. Bagchi, 'Colonialism and the Nature of ''Capitalist" Enterprise in India', Economic and Political
Weekly, 23:31 (30 July 1988), Review of Political Economy, PE38 to PE58.

Page 328
capital–labour relations in South Asian industry within the context of world capitalism as well as the subcontinent's economic history. In seeking to examine how the
formation of an industrial labour force shaped the nature of capitalist development in India, it will also attempt to pull together some of the leading themes which this
book has investigated.
II
How to classify social change or grasp the meaning of capitalism in South Asia, however, is a vexed question. Marx defined and elaborated the concept of the
capitalist mode of production specifically in the context of the industrial revolution. Its defining feature was its inherent tendency to reproduce itself on an increasing
scale. This was most fully realized with the commoditization of labour power, facilitating its progressive subsumption by capital. Production for the market, and the
increasing commoditization of labour power, propelled the development of capitalism when it was accompanied by a complementary process of industrialization.
To a large extent, Marx inferred the characteristics of precapitalist modes of production from his more systematic exposition of capitalism. Few subsequent theorists
have matched his range and historical insight and precapitalist modes of production have remained weakly formulated. Most studies of the rise of capitalism and the
dissolution of earlier modes of production in the Third World identify elements of 'commercialization' as markers towards capitalist development, but usually these
processes are observed without a corresponding growth of industry and its effects are unlikely, although they are often assumed, to be the same. As Wallerstein
illustrates effectively, historians of capitalism in the West have also found it especially difficult to conceptualize social relations in transitional periods characterized by
the absence of significant industrialization.3 Another response to the problem of transition and the task of identifying the emergence of the capitalist mode of production
has been to acknowledge the existence of diverse forms of production relations in precapitalist societies and to focus upon their articulation within a given social
formation. But these modes of production have often appeared so diverse and varied and their articulation so remote, complex and abstract that this has proved a
difficult method to apply in concrete empirical or historical analysis.4
3 Wallerstein, The Capitalist World Economy, pp. 15–17.
4 G. Arrighi, 'Labour Supplies in Historical Perspective: A Study of the Proletarianization of the African Peasantry in Rhodesia', Journal of Development Studies, 6:3 (1970), 197–234;
Brewer, Marxist Theories of Imperialism, ch. 8. For the mode of production debate conducted in relation to India, see Utsa Patnaik (ed.), Agrarian Relations and Accumulation: The
'Mode of Production' Debate in India (Bombay, 1990).

Page 329
The roots of this problem may well lie in a rather absolutist application of definitions of capitalism in terms of 'free' wage labour. Such an application may be facilitated
by cases where all the processes of commercialization in Marx's model were working simultaneously and together. But these were special cases. Where these
processes occurred partially, historians may be better served by a looser and more flexible understanding of capitalism. The readiness of the world systems approach
to accept that capitalist exploitation could occur through several forms of labour use should facilitate a broader definition of both capital and labour in South Asian
history. With the emergence of world capitalism, a wide range of production relations manifested in large parts of the world could no longer simply be taken to reflect
precapitalist modes for they were being constituted in relation to, and were indeed passing under the hegemony of capitalism. Paradoxically, by relaxing the rigidities
of the concept, the world systems approach may offer South Asianists most usefully, not so much a method of investigating its incorporation into the world economy,
as a means of recovering the subcontinental context and history of capitalist development.
While the world systems approach may have directed our attention to 'the problems incurred by using the pervasiveness of wage labour as a defining characteristic of
capitalism',5 it has generated problems of its own. The capitalist mode of production was conceived as an abstract model, which nowhere manifested itself in real
circumstances, but provided a key to understanding a wide variety of cases. The world systems approach while inveighing against the reification of its parts has tended
to reify capitalism as a whole. Its explanatory value seems to lie in its claim that the world system constituted a real observable totality, for which it provided the most
suitable generalizing description. The world system is offered as the only real 'totality' since no other categories are selfsustaining units of analysis or able to provide an
adequate basis for comparison. But the notion of totality is, of course, beset with epistemological problems, for its integrity lies most securely in the eyes of its
beholder.
Furthermore, the world system, it would appear, is created primarily through the expansion of networks of exchange. But the mechanism of capitalist expansion for
Marx was not simply the driving force of production for sale and profit but the appropriation of surplus value as it was shaped by class conflict between labour and
capital at the point of production. In its steady diffusion of market relations, world capitalism often appears to be an inexorable force hammering ceaselessly upon
passive, traditional societies, moulding them according to its needs. On the other hand, class struggle and contradiction were integral and indis
5 Wallerstein, The Capitalist World Economy, p. 17.

Page 330
pensable to Marx's conception of capitalist development. To recognize that their outcome would determine the nature and characteristics of capitalism in particular
cases appears also to promise a more specifically historical approach to the investigation of social change.
III
Industrialization within the periphery raises a special problem for the analysis of the modern world system. If the expansion of the European world economy in its phase
of industrial capitalism, driven forward by its thirst for markets and raw materials, created a single division of labour within the capitalist world economy, and
progressively appropriated and transferred the surplus to the core, it is easier to conceptualize the incorporation within the world economy of an agrarian periphery
than an industrializing one. Already, by the early twentieth century, India had a substantial industrial sector and one of the ten largest labour forces in the world.6
The conventional explanation of the origins of industrialization in India has been that it formed part of a steady process of technological diffusion working gradually
outwards from eighteenthcentury Western Europe.7 Seen as a process of technological diffusion, industrialization in India can be understood as a function of the
expansion of networks of exchange throughout the world and the incorporation of new regions into a world economy centred upon Western Europe. Some strength is
lent to this argument by the fact that the capital for industry was raised predominantly from the profits of trade, and in the case of the Bombay textile industry in the
China trade, where peddling cotton and pushing opium were integral to the mechanisms by which Western European capitalism penetrated the periphery and
undermined preexisting 'world empires'.8 Similarly, expatriate British capital which sponsored industrialization in eastern India had been fattened on the gains of the
private trade and nurtured by the Company's mercantilism. Moreover, expatriate capital engaged primarily in the extraction, processing and export of raw materials,
enabled the incorporation of the subcontinent into the capitalist world economy, and helped to foster the division of labour
6 B. R. Tomlinson, The Political Economy of the Raj: The Economics of Decolonization in India (London, 1979), p. 31. See ch. 2 above.
7 See ch. 2 above.
8 Chandavarkar, The Origins of Industrial Capitalism, pp. 44–67; Siddiqi, 'The Business World of Jamsetjee Jejeebhoy'; A. M. Vicziany, 'The Cotton Trade and the Commercial
Development of Bombay, 1853–1875', unpublished Ph.D. thesis, University of London, 1975; A. K. Bagchi, 'Reflections on Patterns of Regional Growth in India During the Period of
British Rule', Bengal Past and Present, 95:1 (1976), 247–89; on the China trade, see M. Greenberg, British Trade and the Opening of China, 1800–1842 (Cambridge, 1951).

Page 331
9
based upon it. That the domination of European capital was nurtured by the mercantilist Company state, acting as an agency for the incorporation of the Indian
economy, only serves to strengthen the argument that the rise of industry was integral to the incorporation of the subcontinent to the world economy.
But the logic of the notion that industrialization was part of a Eurocentric diffusion of technology is tied to social theories of modernization, tends to remove the growth
of factory industry from larger patterns of social and economic change and understands attenuated and incomplete processes of industrialization in terms of a teleology
derived from what we think we know about its final outcome, which in turn has usually emerged from special or even unique cases.10 Historical evidence, both in South
Asia and Western Europe, suggests a different view: it points to an understanding of industrialization, not in terms of the forces of production, of technological advance
sustained by its own inner logic, but in terms of the social relations of production, as an outcome of class conflict.11 In this perspective, industrialization was not simply
the consequence of the widening flows of exchange and their effects upon the development of the forces of production. On the contrary, its emphasis upon production
relations and class conflict suggested that the history of industrialization would most fruitfully be returned to its South Asian context, albeit one which was influenced by
its relationship to the capitalist world economy.
It is not intended to suggest either that industrial development in India was the natural or inevitable outcome of the relationships between capital and labour or that the
South Asian context of class formation and class conflict was somehow autonomous of world capitalism. But it is doubtful that the expansion of the capitalist world
economy may be adequately explained in terms of 'sources . . . internal to the European worldeconomy', irrespective of its interaction with the regional economies and
polities which it sought to pull into its orbit.12 We might, indeed, accept,
9 Bagchi, 'Reflections on Patterns of Regional Growth in India'; S. Bhattacharya, 'Regional Economy (1757–1857): Eastern India, Part I', and B. B. Chowdhury, 'Regional Economy
(1757–1857): Eastern India, Part II', in Kumar (ed.), CEHI, vol. II, pp. 270–95, 295–332; Wallerstein, 'Incorporation of the Indian Subcontinent'; C. Fisher, 'Planters and Peasants: The
Ecological Context of Agrarian Unrest on Indigo Plantations of North Bihar, 1820–1920', in C. J. Dewey and A. G. Hopkins (eds.), The Imperial Impact: Studies in the Economic
History of Africa and India (London, 1978), pp. 114–31; Sir P. Griffiths, The History of the Indian Tea Industry (London, 1967); J. Richards, 'The Indian Empire and Peasant
Production of Opium in the Nineteenth Century', MAS, 15:1 (1981), 59–82.
10 See ch. 2 above.
11 See, especially, the highly suggestive essay by Stephen Marglin, 'What Do Bosses Do?' Review of Radical Political Economy, 6 (1974), 60–112.
12 Wallerstein, 'The Incorporation of the Indian Subcontinent', PE30.

Page 332
as Wallerstein has advised, that 'we should not look too far for reasons peculiar to the Indian subcontinent' to explain its 'incorporation' into the world economy. Yet it
was precisely these reasons of specific, historical context which determined the manner and pattern of its 'incorporation'. Indeed, they led Wallerstein to admit, in
contradiction, that in the manner of its incorporation India was 'the exception and exceptions need to be accounted for'.13 Recent research has increasingly directed
attention to the domestic origins of Indian capitalism.14 In the late eighteenth and early nineteenth centuries, diverse social groups, with some command of capital, were
able to exploit the mechanisms of local state systems and growing commercial opportunities to consolidate their position and accumulate capital while chipping away at
communal reciprocities and social obligations upon it, and eroding labour's customary rights to its share of the product.15 The expansion of British power in India
picked up these tendencies, was even facilitated by their workings, and indeed developed them further. Its effects, which only became apparent in the 1830s and
1840s, were manifested in the subordination of Indian capital to the interests of the colonial state and metropolitan capitalism.16 The economic function of mercantile
capital was progressively reduced to petty trade and usury. The relative acquiescence of local elites, or their lack of resistance to the expansion of British power, has
been attributed to the increasing control which capital gained over labour.17 The development of Indian capital thus came to be predicated upon the more intensive
exploitation of labour.
The nature and contradictions of indigenous capitalist subordination may not have become fully apparent until the long depression of the 1820s–50s18 began to lift and
rising prices, a buoyant and expanding
13 Ibid., PE32.
14 See the seminal article by Irfan Habib, 'The Potentialities of Capitalistic Development in the Economy of Mughal India', Journal of Economic History, 29:1 (1969), 32–78.
15 These arguments are most clearly stated in D. Washbrook, 'Progress and Problems: South Asian Social and Economic History, c. 1720–1860', MAS, 22:1 (1988), 57–96. See also C. A.
Bayly, Rulers, Townsmen and Bazaars: North Indian Society in the Age of British Expansion, 1770–1870 (Cambridge, 1983), chs. 5 and 6; E. T. Stokes, The Peasant and the Raj:
Studies in Agrarian Society and Peasant Rebellion in Colonial India (Cambridge, 1978), chs. 1–3; Perlin, 'ProtoIndustrialization in PreColonial South Asia'.
16 Bayly, Rulers, Townsmen and Bazaars, ch. 7; Washbrook, 'Progress and Problems'; Bagchi, 'Reflections on Patterns of Regional Growth'; Siddiqi, 'Jamsetjee Jejeebhoy'; D. A.
Washbrook, 'Land and Labour in Late Eighteenth Century South India: The Golden Age of the Pariah', in P. Robb (ed.), Dalit Movements and the Meanings of Labour (Delhi, 1993), pp.
68–86. See also Wallerstein, 'The Incorporation of the Indian Subcontinent', which takes a longer view of this process of subordination.
17 Washbrook, 'Progress and Problems'.
18 Ibid.; Bayly, Rulers, Townsmen and Bazaars, chs. 6 and 7; A. Siddiqi, Agrarian Change in a Northern Indian State: Uttar Pradesh, 1819–1833 (Oxford, 1973); S. Guha, The
Agrarian Economy of the Bombay Deccan, 1818–1941 (Delhi, 1986), ch. 2.

Page 333
export demand and favourable exchange rates held out fresh opportunities for capitalist development at the very moment when an unfavourable legal system and
political initiatives to prevent land alienation, protect the debtor and define tenants' rights served to restrict the possibilities of its development and limit its scope.19 The
rapidity with which the agrarian buoyancy of the later nineteenth century collapsed in the late 1890s exposed most clearly the contradictions inherent in the
development of Indian capital.20 The growth of nationalism, particularly among a wide range of 'bourgeois' groups in the 1880s, may perhaps be seen as a political
manifestation of this contradiction.
The consequence of limited capitalist development in the countryside was the stagnation of the agrarian structure. The limitations of capitalist development signified its
weakness as an agent of fundamental or farreaching changes in the structure of the agrarian economy. On the other hand, economic buoyancy served to shore up the
fragile smallholding agrarian base. Agrarian stagnation increased the risks of investment and left the rural economy vulnerable to demographic pressures, the vagaries of
the weather and international price fluctuations.21
Of course, some of the impulses which served to subordinate indigenous capital and subsume labour to it originated in the capitalist world economy or stemmed from
the colonial state. Bullion inflows had been crucial to the consolidation of indigenous capital while their contraction in the 1830s conversely weakened its position. In
any case, bullion began to be replaced by land revenue as the driving force of commercialization.22 Deindustrialization undermined the interests of mercantile capitalists
but it also forced artisans onto the land and thus swelled the supply
19 Stokes, The Peasant and the Raj, chs. 9–12; Washbrook, 'Law, State and Agrarian Society in Colonial India'; Baker, An Indian Rural Economy; on the Punjab, see N. G. Barrier,
The Punjab Land Alienation Bill of 1900 (Duke University, 1966); N. Bhattacharya, 'The Logic of Tenancy Cultivation: Central and SouthEast Punjab, 1870–1935', IESHR, 20:2
(1983), 121–70; Bhattacharya, 'Lenders and Debtors'; on the Deccan, see N. Charlesworth, Peasants and Imperial Rule: Agriculture and Agrarian Society in the Bombay
Presidency, 1850–1935 (Cambridge, 1985), pp. 95–203; J. Banaji, 'Capitalist Domination and the Small Peasantry: Deccan Districts in the Late Nineteenth Century', Economic and
Political Weekly, 12:33–4 (1977), special number, 1375–404; Guha, The Agrarian Economy of the Bombay Deccan, chs. 3–6.
20 Stokes, The Peasant and the Raj, ch. 12; Washbrook, 'Law, State and Agrarian Society', 681–94; Banaji, 'Capitalist Domination and the Small Peasantry'; Guha, The Agrarian
Economy of the Bombay Deccan, esp. ch. 5; S. C. Mishra, 'Agricultural Trends in the Bombay Presidency, 1900–1920: The Illusion of Growth', MAS, 19: 4 (1985), 733–59.
21 Stokes, The Peasant and the Raj, ch. 12; Washbrook, 'Law, State and Agrarian Society'; Banaji, 'Capitalist Domination and the Small Peasantry'; Guha, The Agrarian Economy of the
Bombay Deccan, ch. 5; S. C. Mishra, 'Commercialization, Peasant Differentiation and Merchant Capital in Late Nineteenth Century Bombay and Punjab', Journal of Peasant Studies,
10:1 (1982), 3–51.
22 Bayly, Rulers, Townsmen and Bazaars, chs. 6–7.

Page 334
23 24
of labour. Land revenue became the main engine for the appropriation and remittance of the surplus to England. While the colonial bureaucracy destroyed the
revenue farming function of mercantile capital and reduced its function within the state and agrarian society, the weight of the revenue demand increased the demand for
credit from cultivators especially when prices began to fall in the 1830s, and thus facilitated their subordination to merchant capital. While productive investment
declined, rising population, the settlement of mobile peasants, pastoralists, 'tribals' and forest dwellers under the twin imperatives of policing and taxation, and the
effects of deindustrialization meant that competition of labour for the land increased and while the economy became less diversified and more dependent upon
agriculture, so poorer and marginal lands were brought under the plough.25
These were scarcely promising beginnings for capital to embark upon industrialization. Yet on a comparative timescale, modern factory industry appeared in India
before it came to be established in Japan or Russia. In the face of the steady transfer of surplus to the core, industrial development in the periphery might be explained
as a consequence of capitalist interests 'seeking to distort [a single world] market for their benefit by organizing to exert influence on states'.26 Yet this argument would
be difficult to sustain in the case of the colonial state whose aim was to act as the instrument of incorporation into the capitalist world economy on particular terms of
advantage to Britain. Of course, the imperatives of political stability within India made the colonial state vulnerable to some Indian pressures. But in fact industrialization
in India was driven forward by responses to the general subordination of indigenous capital rather than its development.27 Those fractions of capital which invested in
industry were usually marginal groups, often distanced from the institutions of the colonial state or even facing decline in valuable areas of the
23 On the 'deindustrialization' debate, see the literature cited in ch. 2, fn 56, above. See also T. Roy (ed.), Cloth and Commerce: Textiles in Colonial India (New Delhi, 1996); for a
recent general survey, see B. R. Tomlinson, The New Cambridge History of India, vol. III, part 3: The Economy of Modern India, 1860–1970, (Cambridge, 1993), pp. 101–9.
24 A. Siddiqi, 'Money and Prices in the Earlier Stages of Empire', Indian Economic and Social History Review, 18:3 and 4 (1981), 231–62; Bagchi, 'Reflections on Patterns of Regional
Growth'.
25 C. A. Bayly, The New Cambridge History of India, vol. II, part 1: Indian Society and the Making of the British Empire (Cambridge, 1988), pp. 138–50; Washbrook, 'Problems and
Progress'; M. Gadgil and R. Guha, 'State Forestry and Social Conflict in British India', Past and Present, 123 (1989), 141–77; M. Rangarajan, 'Imperial Agendas and India's Forests: The
Early History of Indian Forestry, 1800–1878', IESHR, 31:2 (1994), 147–67.
27 See ch. 2 above; and Chandavarkar, The Origins of Industrial Capitalism, pp. 44–71, 239–77, 335–96.

Page 335
export trade. For the risks of industrial investment were high and the returns uncertain. Parsi and Bhatia entrepreneurs invested in the first cotton mills in Bombay when
they had been undermined in the export trade in raw cotton.28 Marwaris provided the major Indian competitors within the jute industry.29 Expatriate capital appears to
have been rather timid and, in particular, slow to introduce mill production of jute textiles, which in any case were eventually able to compete effectively with the
handloom sector largely because of their control over the export outlets and carrying trade and their access to the Presidency banks, the capital of the agency houses
and the ear of the government.30 Moreover, expatriate capital was invested in plantation industries, like indigo and tea, in the hope of generating raw material supplies
and primary products within the British empire, seeking, in other words, to define an autarkic union within the capitalist world economy.31
The subordination of Indian capital and the structural constraints within which it developed conditioned the circumstances in which it might be drawn forward for
industrial investment. Undercapitalization was a general characteristic and indeed constraint upon the development of Indian industry. Capital was characteristically
mobilized in small pools, specifically for enterprises whose risks were known, whose viability was proven and whose directors were trusted. As a result investment
tended to be imitative. Successful lines of production could attract a flood of investment until they became oversubscribed and unprofitable. At or preferably before
this point, entrepreneurs would seek to diversify their investment, thus perpetuating the instability inherent in conditions where investment was so susceptible to fashion
and believed to be so exposed to risk. The low ratio of fixed capital, the intensity of competition and the general tendency to undercapitalization called for a rapid
turnover to
28 Marika Vicziany, 'Bombay Merchants and Structural Changes in the Export Community, 1850 to 1880'; see ch. 2 above; and Chandavarkar, The Origins of Industrial Capitalism,
ch. 6. Rutnagur (ed.), Bombay Industries: The Cotton Mills.
29 Bagchi, Private Investment in India, chs. 6 and 8; T. A. Timberg, The Marwaris: From Traders to Industrialists (New Delhi, 1978); Goswami, 'Then Came the Marwaris'; Goswami,
'Collaboration and Conflict'; Goswami, 'Sahibs, Babus and Banias'; Chakrabarty, Rethinking Working Class History, ch. 1.
30 Bagchi, Private Investment in India, ch. 6; A. K. Sen, 'The Commodity Pattern of British Enterprise in Early Indian Industrialization, 1854–1914', in The Second International
Conference of Economic History, AixenProvence, 1962, vol. II: Middle Ages and Modern Times (Paris, 1962), pp. 781–808; Tomlinson, 'Colonial Firms and the Decline of Colonialism
in Eastern India'; B. R. Tomlinson, 'British Business in India, 1860–1970', in R. P. T. DavenportHines and Geoffrey Jones (eds.), British Business in Asia since 1860 (Cambridge, 1989),
pp. 92–116; Goswami, 'Sahibs, Babus and Banias'.
31 Indigo, raw silk, opium and cotton accounted for nearly twothirds of the total value of exports between 1814 and 1850. K. N. Chaudhuri, 'Foreign Trade and Balance of Payments
(1757–1947)', in Kumar (ed.), CEHI, vol. II, p. 844; Wallerstein, 'The Incorporation of the Indian Subcontinent', PE30 to PE31.

Page 336
finance debts, and production strategies sufficiently flexible to adapt to unpredictable markets. This pattern of investment also committed entrepreneurs to operate on
narrow margins, tailoring production to shortterm fluctuations of demand, minimizing their overheads and varying their labour force according to need.32
How far capital operated in this way varied according to the nature of the industry, its labour supply and labour relations, and its markets. But even in the case of the
iron and steel or the coal industry, there were periodic crises of undercapitalization and the manipulation of working capital often held the key to the organization and
deployment of labour.33 The cotton textile industry operated on a larger scale in Bombay than anywhere else in the subcontinent; yet here too the millowners forsook
the advantage of realizing the full value of their machinery costs and other fixed investments through the maintenance of steady levels of production. Despite an
apparently oligopolistic structure of ownership in Bombay, the industry was rarely able to impose its wishes on individual entrepreneurs and attempts to fix price or
output levels, or even to reach agreement on production conditions, seemed doomed to failure.34 More surprisingly, in view of the relative weakness of foreign
competition, attempts to introduce shorttime working and curtail production in Bengal proved similarly problematic in the late 1920s and early 1930s.35
The causes of these production strategies were varied and complex ranging from the nature and institutions of industrial finance to the effects of foreign and to a lesser
extent domestic competition in the home market. They reflected the interplay between structural constraints within the internal economy and the effects of the world
economy upon the accumulation and reproduction of capital in India. Their consequence was a fluctuating demand for labour and the extensive use of casual hiring.
Cumulatively these conditions, characterized by flexible patterns of labour use and the extensive deployment of casual labour, tended to produce a labour supply in
excess of the usual needs of local industry in
32 See ch. 2 above; Chandavarkar, The Origins of Industrial Capitalism, pp. 67–71, chs. 6–8. The evidence for the jute industry may be read in a similar fashion, even if it has been
interpreted rather differently, for instance, in Chakrabarty, Rethinking Working Class History, ch. 1.
33 Simmons, 'Indigenous Enterprise in the Indian Coal Mining Industry', 189–218; C. P. Simmons, 'Vertical Integration and the Indian Steel Industry', MAS, 11:1 (1977), 127–48; V. Bahl,
'The Emergence of LargeScale Steel Industry in India under British Colonial Rule, 1880–1947', IESHR, 31:4 (1994), 413–60; S. B. Datta, Capital Accumulation and Workers' Struggle in
Indian Industrialization: The Case of the Tata Iron and Steel Company, 1910–1970 (Stockholm, 1986); Simeon, The Politics of Labour Under Late Colonialism; Bahl, The Making of
the Indian Working Class.
34 Chandavarkar, The Origins of Industrial Capitalism, chs. 6–8.
35 Goswami, Industry, Trade and Peasant Society; Goswami, 'Collaboration and Conflict'; Chakrabarty, Rethinking Working Class History, ch. 1.

Page 337
order to meet the peak periods of production. In turn, this situation of excess supply perpetuated the labourintensive strategies of employers, for if labour appeared to
be abundant they had little incentive to maintain steady levels of employment in the face of volatile markets.36 The extensive use of casual labour is sometimes attributed
to the lack of dynamism in the industrial sector, a case of inefficient entrepreneurs following anachronistic preindustrial methods seeking and obtaining an equally
inefficient and traditionbound labour supply.37 But a more plausible view would be that these stratagems were adaptations to a given economic context and, judged
within these terms, they may have yielded a reasonably efficient pattern of labour deployment.
It is perhaps more significant, in understanding the nature of class formation, that the overwhelming majority of the industrial labour force experienced uncertain
conditions of employment. Employers and experts alike preferred to distinguish between permanent and casual labour in Indian industry. But these were categories of
employer policies; they did not reflect accurately either attitudes to work or the nature of the labour supply. For the demand for labour itself was scarcely permanent: it
fluctuated substantially as employers took workers on or laid them off as the market for their products required. Permanent workers could easily and often arbitrarily
lose their jobs and thus be thrown on to the casual labour market, whether for reasons of trade or because they were attempting to defend their position at work
through industrial action. Neither considerations of 'mentalities' and consciousness nor those of wage differentials and skill provide the basis for a tenable distinction
between industrial workers and the casual poor.38
If the effects of the structural stagnation of the Indian economy by the late nineteenth century shaped the options before industrial capital, they also influenced the
formation of an industrial labour force. Demographic and commercial pressures, playing upon the fragile smallholding agrarian base, served to swell the supply of
labour to the towns. Rural migration constituted the most important source of labour supply for urban industry. The purpose of migration was to earn cash to pay off
debts or buy seed or simply to relieve, however temporarily, the subsistence pressures on shrinking family holdings in the village base.39 This was sometimes achieved
by moving to neighbouring districts or even further afield for
37 Most recently, for instance, in Chakrabarty, Rethinking Workingclass History, ch. 1.
38 Chandavarkar, The Origins of Industrial Capitalism, ch. 3. This finding informs the arguments of chs. 2, 4, 5 and 8 above.
39 Ibid., ch. 4; for an account of how this pattern of migration and the rural connections of workers were shaped by, and in turn influenced, gender relations, see Sen, 'Women Workers
in the Bengal Jute Industry', esp. chs. 1–3.

Page 338
40
agricultural work. Longdistance migration accounted for a small proportion of the movement of people within the subcontinent.
Among migrants in search of wage labour, the majority were male. This male bias was greater in the case of longdistance migration, urban and industrial employment
and indentured labour especially in its early phase. Consequently, the struggle of smallholding peasant households to retain their land depended increasingly upon the
more intensive exploitation of female and child labour. Whole families migrated in search of work only in periods of distress or in a fit of prosperity. Family migration
may have been more common in the case of seasonal agricultural labour seeking employment in neighbouring districts. A higher proportion of women migrated to the
towns among dalits and low castes, who had a lower rate of migration overall until the 1920s, and among higherstatus artisan castes. The effect of factory legislation,
which regulated the conditions of women's work after 1881, was to deprive them of any staple, largescale employers. In the 1870s, for instance, women, including
migrants, had formed a growing proportion of the labour force in the Bombay cotton mills.41 Female labour was limited to the casual, manual, unskilled and poorly paid
occupations. If women's work was confined to the domestic sphere in the peasant household and thus devalued, their alternative employment opportunities carried low
status, considerable uncertainty and meagre rewards. However central women's work may have been to the reproduction of the household economy of the working
classes, it was increasingly perceived and treated as marginal.
As the status of women's work diminished, more substantial peasant households sought to withdraw their labour when they could. On the other hand, women workers
in the cotton and jute mills counted a substantial proportion of single, 'deserted' and widowed among them. They entered public discourse with the stigma and the
threat of existing beyond the control of the family or, implicitly, of men. Confined to the lowstatus and lowwage sectors of the urban economy, women often
withdrew from the labour market precisely when conditions were buoyant and the demand for labour was growing, and sought to enter it when conditions were
depressed and jobs were scarce. The fact that the supply of female labour was determined in this sense by the demand for male labour ensured that women
commanded little bargaining power and
40 Omvedt, 'Migration in Colonial India', p. 188; Chandavarkar, The Origins of Industrial Capitalism, ch. 4; Sen, 'Women Workers in the Bengal Jute Industry', esp. chs. 1–3; Das
Gupta, 'Migrants in Coal Mines'; Mohapatra, 'Coolies and Colliers'; Brahma Nand, 'Agricultural Labourers in Western India'; Chaudhury, 'Labour Migration from the United
Provinces'; Pouchepadass, 'The Market for Agricultural Labour in Colonial North Bihar'; Bhattacharya, 'Agricultural Labour and Production'.

Page 339
found it difficult to entrench themselves in more favourable positions within the labour market. The identification of women's work with low status had significant social
consequences. Respectability became the exclusive attribute of households which were able to withdraw the labour of women. If women's work was confined to their
home, their public presence degraded them. Women workers in Bombay as well as Bengal were labelled as sexually promiscuous and they were not infrequently
associated with prostitution. The domestication of female labour and its withdrawal from the public domain influenced the shift from bride price to dowry. Similarly, it
facilitated the moralizing discourse of philanthropists and social reformers and set up the spectacle of notables lecturing unselfconsciously to the poor about the secrets
of household management and healthy diets, ideals of motherhood and childrearing.42
Throughout the colonial period, it remained predominantly the case that the industrial labour force was composed largely of male migrants, most of whom had spent
their childhood in their villages and moved to the cities in their adolescence. They left their families behind and when their wives migrated with them they sometimes
returned to the village to have their children. In retirement or old age they returned to the land. While they worked in the factories they maintained their rural
connections: returning to the village to help with the harvest or to recover their health or to participate in family and village celebrations or religious observances; and
remitting money to their relatives to supplement their incomes and liquidate their debts.43
The relationship between their rural base and their industrial employment varied according to region and industry. Some important industrial centres recruited workers
from adjacent villages;44 but workers migrated over long distances to Bombay, Calcutta and Jamshedpur, where the labour force poorly represented their contiguous
districts.45 The railways were the largest single employers of 'industrial' labour in India; yet line
42 Ibid., pp. 94–9; Sen, 'Women Workers in the Bengal Jute Industry'.
43 Chandavarkar, The Origins of Industrial Capitalism, ch. 4; Sen, 'Women Workers in the Bengal Jute Industry', esp. chs. 1–3.
44 For instance, twothirds of the mill labour force in Ahmedabad were recruited within the district; onefifth were exclusively dependent upon urban, industrial employment, had no
rural connections and were permanently resident in the city. See RCLI, Evidence, Bombay Presidency, Memorandum of the Government of Bombay, vol. I, part i, p. 4.
45 On Bengal, see Das Gupta, 'Factory Labour in Eastern India'; and Sen, 'Women Workers in the Bengal Jute Industry', esp. chs. 1–3; on Jamshedpur, where unskilled labour was
recruited locally, see M. D. Morris, 'The Labor Market in India', in W. E. Moore and A. S. Feldman (eds.), Labor Commitment and Social Change in Developing Areas (New York,
1960), pp. 173–200; Bahl, The Making of the Indian Working Class; Simeon, The Politics of Labour, ch. 1; on Bombay, see RCLI, Evidence, Bombay Presidency, vol. I, part. i,
Memorandum of the Government of Bombay, p. 4; Morris, Industrial Labour Force, pp. 62–5; Chandavarkar, The Origins of Industrial Capitalism, ch. 4.

Page 340
46
workers were often recruited from and remained in the villages along the tracks. Migration to the tea plantations involved movement to a rural setting and involved for
many what was perceived as primarily 'agricultural' work.47 In the coal industry, attempts were made to induce tribals from the neighbouring areas to work
underground by offering them allotments of cultivable land near the mines.48 The Raniganj and Jharia coalfields experienced their worst labour shortages in the harvest
season when workers were drawn away by higher agricultural wages.49 There was an even more intimate connection between agricultural and industrial work in the
seasonal industries which were concerned primarily with the processing of raw agricultural products and offered employment as soon as the harvest ended.50
While the major motivation behind migration was to supplement the rural resources of the family, their ability to fulfil these intentions registered differences between
workers. Labour in the tea plantations was recruited from the tribal areas of Chota Nagpur and Bihar frequently as whole families, and once in the gardens they were
governed by draconian laws of contract and the general system of indenture.51 Labour contractors also secured workers for the coal mines by buying out their debts
and simply renewing their obligations within a changed context.52 Casual work and general labouring in the towns did not yield either a regular or sufficient income for
workers to supply the rural base, and in periods of sickness and unemployment they would be forced back upon their rural connections.53 Large numbers of industrial
workers found themselves in a similar situation. In the early 1920s, 64 per cent of the adult male labour force in the Bombay mills earned less than the daily wage rate
for skilled rural labour in the Konkan, the major supplier of
46 Proceedings of the MCC, Defence Statement, K. N. Joglekar, pp. 1766–965; Jagga, 'Colonial Railwaymen and British Rule'; Kerr, Building the Railways of the Raj.
47 Behal, 'Forms of Labour Protest in Assam Valley Tea Plantations'; Griffiths, The Indian Tea Industry, pp. 267–420.
48 See Simmons, 'Recruiting and Organising an Industrial Labour Force in Colonial India'; Das Gupta, 'Migrants in Coal Mines'.
49 Report of the RCLI, pp. 115–17.
50
S. Amin, Sugarcane and Sugar in Gorakhpur: An Inquiry into Peasant Production for Capitalist Enterprise in Colonial India (Delhi, 1984); J. Breman, Of Peasants, Migrants and
Paupers: Rural Labour Circulation and Capitalist Production in West India (Delhi, 1985).
51 Behal and Mohapatra, '''Tea and Money Versus Human Life"'; Das Gupta, 'Plantation Labour in Colonial India'; Mohapatra, 'Coolies and Colliers'; Bates and Carter, 'Tribal Migration
in India and Beyond'; Carter, Servants, Sirdars and Settlers.
52 For evidence of similar practices followed in the coal mines, see Report of the RCLI, p. 15; Simmons, 'Recruiting and Organising an Industrial Labour Force', p. 473–5; Das Gupta,
'Migrants in Coal Mines'; Omvedt, 'Migration in Colonial India', 193–4.

Page 341
54 55
labour to the city. Low wages, irregular employment and intolerable housing conditions characterized the urban experience of most migrant workers. Millworkers in
the large industrial centres may have been in closer touch with, and better placed to supply, their rural base than plantation labour, but the conditions of urban life
required them to repeatedly seek the support of their village base, both because their social connections, based upon village and kinship, might help them to find work,
credit and housing in the towns and relief and support in periods of unemployment, sickness and old age, and because their ability to maintain and renew their rural ties
improved their access to rural resources. Town and country should be seen as mutually supportive spheres in the family economy of the working class. Remittances
from industrial earnings may have enabled some workingclass families to extend their holdings or invest in the education of their (male) children; but, for most,
remittances simply enabled them to stay where they were, to roll their debts and to hold on to their stake in the village. This symbiosis between the urban resources and
rural base of the family economy of the working class was sometimes perpetuated and reproduced over several generations. Underlying this continuity was the
instability of the urban economy, characterized by the fluctuations in the flows of capital and the demand for labour.56
The difficulties of maintaining an adequate supply in the face of the fluctuating demand for labour posed special problems of recruitment and discipline.57 Employers had
to devise methods of recruitment which would enable them to employ additional workers at short notice or lay them off as the market required and, in discharging
them, still hold them within the urban economy or the industry's pool of labour supply. To this end, employers in a large number of industries delegated responsibility
for the hiring and firing of workers to intermediaries known variously as jobbers and sirdars, muccadams, mistries and serangs and by sundry other names. This
practice, apparently pervasive throughout the subcontinent, has often been understood as a culturally specific phenomenon. Yet the term 'jobber' came from
Lancashire, where it was associated with
54 Labour Office, Bombay, Report on an Enquiry into the Wages and Hours of Labour in the Cotton Mill Industry, May 1921 (Bombay, 1923), pp. 6–7. For a different view, see
Mazumdar, 'Labour Supply in Early Industrialisation'.
55 Chandavarkar, The Origins of Industrial Capitalism, ch. 4; Basu, 'Workers' Politics in Bengal'; Sen, 'Women Workers in the Bengal Jute Industry'.
56 See ch. 2 above. These issues are examined more extensively in Chandavarkar, The Origins of Industrial Capitalism, ch. 4.
57 See ch. 2 above; and Chandavarkar, The Origins of Industrial Capitalism, pp. 99–110, 195–200, 295–307.

Page 342
a technical function, and involved no stated recruiting or supervisory duties. In fact, foremen in British industry were frequently involved in recruiting labour and similar
agents of recruitment and discipline were to be found in the world beyond South Asia.58 Indeed, Sikh foremen were employed in the Shanghai cotton mills because
they were believed to be particularly good at disciplining the workforce.59
It is frequently assumed that the jobber system was a function of an early stage of industrialization reflecting the immaturity of both entrepreneurs and workers, their
insufficient adaptation to the industrial setting and their general and mutually deserved inefficiency. This is a harsh view, somewhat distanced from the daily politics of
the workplace. The prevalence and pervasiveness of the jobber is also attributed to the need felt by urbane managers and supervisory staff for an intermediate cadre to
mediate the linguistic and cultural chasm that divided them from the labour force. Doubts about one of these views undermine the plausibility of the other.
If cultural factors shaped methods of recruitment, it is probable that the 'colonial' perceptions of employers were more important than the rural connections of the
workforce. Their social distance and isolation from the workingclass life and their apprehension of labour's social threat may have led them to suppose that the
maintenance of discipline and control hinged upon factors inward and specific to South Asian culture, calling for mediation through selected individuals deemed to be
the leaders or 'headmen' of the working classes.60 Employers may have supposed that the diversity of South Asian culture implied that workers from the same village
or kinship group would be more compatible and perhaps more efficient. One historian, who took it for granted that these systems of recruitment were adaptations to a
migrant labour force, nonetheless adduced evidence to show to the contrary that plantation owners held romantic notions of labour contracting as a patriarchal system
particularly suited to the culturally specific needs of Indian workers.61
On the other hand, employers profited more extensively from the fact that these systems of recruitment accentuated the sectionalism of the workforce than that the
jobber mediated its cultural heterogeneity. In fact, entrepreneurs in most industries attempted to diversify the ethnic composition of their workforce and sought to
maintain its heterogeneity.
58 P. Joyce, Work, Society and Politics: The Culture of the Factory in Late Victorian England (London, 1980); J. Melling, 'NonCommissioned Officers: British Employers and
their Supervisory Workers, 1880–1920', Social History, 5:2 (1980), 183–221; R. Gray, The Aristocracy of Labour in Nineteenth Century Britain, c. 1850–1900 (London, 1981).
59 J. Chesneaux, The Chinese Labour Movement, 1919–1927 (Stanford, 1974).
60 On the character of this colonial discourse, see ch. 1 above.
61 Omvedt, 'Migration in Colonial India', 185–212.

Page 343
The jobber was valued more for his managerial than his cultural functions: to maintain an adequate labour supply at low wage levels in the face of a fluctuating demand
for labour. He also fulfilled certain pacemaking and disciplinary tasks. These roles were intimately linked. Thus, absenteeism was not simply a problem of labour
supply, but it could also be a challenge to the discipline of a jobber's team, and in either case was related to the maintenance of production levels. Most crucially, these
functions of discipline and control, recruitment and diversification enabled employers to see the jobber as an invaluable bulwark against industrial action. Even when
they complained of being locked out of the management of their own workforce or considered the possibility that the abolition of this supposedly anachronistic,
dysfunctional institution might permit the development of a perfectly responsive labour market, they remained respectful of the political functions of the jobber system.
Significantly, it was the momentum of workingclass politics in Bombay which most clearly revealed the weaknesses of the jobber system and led millowners to
contemplate its reform and the modification of their methods of labour control.62
Historians, following contemporary observers, have almost universally exaggerated the strengths and overlooked the weaknesses of the jobber system. The jobber has
been repeatedly cast as a figure of awesome power but his dominance has more often been assumed than demonstrated.63 Indeed, assessments of whether the jobber
system was functional to industry simply took it for granted that the jobber could deliver what was expected of him or, alternatively, that he could consistently impede
the realization of the common good. But the jobber system was by no means homogeneous. Its form varied according to industry and region. Its diversities were
shaped by the nature of the industry, the character of the labour process and the wider political context.
The power of the jobber within a single industry varied according to occupation and the skill and bargaining position of the workers he supervised. Labour for the tea
plantations and the coal mines was recruited either through sardars, who were employees, or contractors.64 Most factory industries in the cities were able to recruit
their labour force from
62 Chandavarkar, The Origins of Industrial Capitalism, pp. 99–110, 295–307.
63 Most recently, for instance, in Chakrabarty, Rethinking Working Class History.
64 High rates of mortality in transit as well as on the plantations led to the passing of legislation which required that these contractors were licensed. See Griffiths, The Indian Tea
Industry, pp. 267–96; for coal mining, see Simmons, 'Recruiting and Organising an Industrial Labour Force', 473–4; R. Shlomowitz and L. Brennan, 'Mortality and Migrant Labour in
Assam, 1865–1921', IESHR, 27:1 (1990), 85–110; R. Shlomowitz and L. Brennan, 'Mortality and Migrant Labour en route to Assam, 1863–1924', ibid., 27:3 (1990), 313–30.

Page 344
65
those who had already migrated to the town and did not need to scour the countryside for workers. Jobbers were, thus, often selected for their influence in the
workplace rather than their patronage in the village, and, for migrant workers, jobbers formed only the most decorative thread in the warp and woof of the social
connections through which they found work.66
How wages were paid was also crucial in determining the power of the jobber. The sardars who recruited for the tea plantations were often responsible for distributing
wages; at times, they 'procured' labour by relieving men of their debt bonds to moneylenders or landlords and, by appropriating a proportion of their wages, held them
firmly in relationships of subordination.67 In addition, the position of the sardar was strengthened by the attenuated nature of workers' rural connections, their
dependence upon plantation housing and draconian legislation governing breach of contract widely construed.68 Coal mines at Raniganj and Jharia hired raising
contractors, who in return for an agreed rate per tonnage, recruited, disciplined and paid labour.69 Jobbers in the cotton mills and sardars in the jute industry may have
often accepted bribes and commissions for hiring workers and retained some power to influence wages by choosing how to allocate tasks, machines and raw
materials, but in not having direct responsibility for calculating and distributing wages, lacked a vital instrument of control. Jobbers often acted as moneylenders or petty
landlords or their agents. But for most workers in the industrial centres there were competing sources of credit and patronage, so that migrants did not simply move
from one form of debt bondage to another form of personal dependence. In these contexts, too, the jobber's authority and influence varied with his position in the
division of labour. He was subject to the competition of rivals and was often squeezed between the workers' demands and managerial expectations. Moreover, the
role assigned to jobbers was not fulfilled exclusively by them. Some workers were
65 By 1918 the Buckingham and Carnatic Mill in Madras relied largely on their school where the children of their employees were being educated for a lifetime in the mills. Murphy,
'Class and Community in India'.
66 RCLI, Evidence, Bombay Presidency, Seth Ambalal Sarabhai, Ahmedabad Manufacturing and Calico Printing Co. Ltd, vol. I, part. i, p. 277.
67 Report of the RCLI, p. 15; Omvedt, 'Migration in Colonial India', 193–4; Behal and Mohapatra, '''Tea and Money Versus Human Life"'; Mohapatra, 'Coolies and Colliers'; Bates and
Carter, 'Tribal Migration'.
68 It was not until 1926, following the recommendations of the Assam Labour Enquiry Committee (1921–2) and N. M. Joshi's persistent efforts in the Legislative Assembly that this Act
was repealed. But the specificity of the Assam case needs at least slightly to be modified. As V. B. Karnik observed, in the view of officials and employers, 'a strike was regarded as an
act of insubordination and indiscipline, an illegal breach of contract'; see Karnik, Strikes in India, pp. 24–5.
69 Simmons, 'Recruiting and Organising an Industrial Labour Force', 473–5.

Page 345
recruited by friends and castefellows in the industry; managers and lesser supervisory personnel were also often involved in recruiting labour and sometimes
developed similar relationships with workers both in the workplace and the neighbourhood. To the extent that the jobber was required to build up a following outside
the workplace, he was forced to remain responsive to the needs and satisfy the expectation of the workingclass neighbourhood. The wider his range of intervention in
the material structure of the neighbourhood, the greater would be the competition to which he was subjected.70
The nature of the industrial labour force, and its relationship with capital, formed around the fitful and spasmodic patterns of investment and an inherently volatile urban
and industrial sector. By the early twentieth century, price fluctuations in the world economy were registering their impact on an internal economy which had already
been frozen into stagnation.71 The depression of the 1930s was symptomatic of the incorporation of the Indian subcontinent into the capitalist world economy, though
this process was mediated to a significant degree by the class structure, the social and economic institutions of the internal economy and the character of local
production relations. In the depression, the tightening of money supply, dramatically falling prices and the difficulties of recovering dues in the countryside combined
with the effects of tariff protection, the disruption of international trade and the flow of cheap labour to attract capital increasingly to the towns. At the same time,
scarce credit and rising interest rates only made it more important for smallholders, trapped within a stagnant agrarian economy subject to unprecedented pressures
and finding their rural base shrinking, to find wageemployment in the towns.72
It was not so much that the demand for labour in largescale industry expanded rapidly to absorb this migration, but that the large cities offered the widest range of
economic opportunities. For migrant workers, contraction in one area of the urban economy could still leave them the possibility of finding work in another.73 In the
1930s, the population of the largest towns increased by threequarters while the total urban population rose by less than a third. In the 1940s, grain rationing, public
distribution systems in the largest towns, most seriously affected by the disruption of food supplies, inflation and wartime scarcities, made these
71 Baker, An Indian Rural Economy, esp. ch. 2; Tomlinson, The Economy of Modern India, 1860–1970, ch. 2.
72 Baker, An Indian Rural Economy; Tomlinson, The Political Economy of the Raj.
73 Baker, An Indian Rural Economy, ch. 5; Chandavarkar, The Origins of Industrial Capitalism, ch. 3.

Page 346
74
particularly attractive to migrants from the countryside. From the 1930s onwards, the largest cities recorded the fastest growth rates, some of the smaller or middling
towns entered decline or stagnated. Until the 1930s, the fluctuations of the urban economy had often led to the movement of labour, especially those in casual
employment, between town and country. Subsequently, expanding urban opportunities have drawn in the slack of the countryside while their contraction has often left
labour stranded in the towns. This does not mean that migrant workers became progressively urbanized or proleterianized. On the contrary, most migrants still
attempted to maintain their village connections, but those who were most securely employed were now most likely to succeed whereas the loss of rural ties was
generally the result of distress. The rapid growth of India's largest cities over the last halfcentury is symptomatic of agrarian decline and rural deprivation rather than a
function of economic growth.
At one level, we might suppose that there is nothing very surprising about this pattern of labour force formation. It could be understood simply as a labour force in an
early stage of industrial development. But then we would also have to acknowledge that in this case its development remained at an early stage for considerably over a
century. This line of argument could be carried a stage further: perhaps the formation of the labour force was a function of inefficient entrepreneurs engaging an
inefficient, unskilled, unstable supply of migrant labour, of capitalists trying to scamper up backwardsloping curves. But it has often been misleading to measure
entrepreneurial behaviour against some ideal of neoclassical or Marxist theory and highly problematic to arrive at a meaningful comparative standard of skill or labour
efficiency. Such an argument, moreover, can be so completely circular as to be happily selfsustaining. Thus, for instance, when metropolitan and expatriate capitalists
entering Kenya or Rhodesia in the 1920s continued the systems of migrant labour utilized by inefficient settlers, they are said to have brought their methods of labour
management intact with them from South Africa or India.75 Within a given economic and political context, it is possible that capitalists as well as migrant workers were
making optimal choices in India.
74 Christopher Baker, 'Colonial Rule and the Internal Economy in Twentieth Century Madras' in Baker, Johnson and Seal, Power, Profit and Politics, pp. 575–602.
75 For instance, J. Forbes Munro observed that 'Metropolitan and expatriate firms coming in during the 1920s, to mine copper in Northern Rhodesia or plant tea in Kenya . . . had needs
for large amounts of unskilled labour and brought with them, from South Africa and India, methods which were compatible with migrant labour'. See J. Forbes Munro, Britain in
Tropical Africa, 1880–1960: Economic Relationships and Impact (London, 1984), p. 56.

Page 347
Finally, the pattern of labour force formation in India may be explained in terms of the inadequate penetration of the world economy. The incorporation of the Indian
subcontinent into the world economy may be variously dated in its origins but it had progressed substantially by 1900. The logic of further incorporation and the
progressive transfer of surplus to the core would suggest the prevention rather than the expansion of capitalist development. On the other hand, Wallerstein has also
argued that the 'further development of the worldeconomy' advances class formation in the periphery, so that class consciousness becomes 'a more relevant political
tool', although 'a class conscious proletariat cannot emerge before in fact it represents a larger sector of the population'.76
Such arguments build upon evolutionary assumptions which can only be justified by the prognosis that the Indian economy was always becoming more like the
capitalist West. Frequently, the ensuing question is counterfactual: why did it not increasingly approximate to the capitalist West? In India, however, a migrant working
class, retaining its rural connections, operating within an overstocked labour market, apparently divided by caste and religion, lacking the support of trade union
organization, showed a remarkable propensity for collective industrial action on a considerable scale.77 The nature of their political action cannot readily be deduced
from the characteristics of labour force formation. Nor is it usefully illuminated by the common theories, or expectations, of workingclass political behaviour. Migrant
workers were, for instance, often the most militant strikers. They migrated to the town with the aim of preserving their stake in rural society and, consequently, they
were deeply committed to the defence of their wages, their jobs and their position at work. The importance of their urban, industrial interests combined with their
access to an alternative base, however fragile, made migrant workers often the most resilient in the course of industrial action. Their political consciousness can only be
flattened and distorted by notions of their 'rural mentalities'.
Similarly, there was no clear divide between the unskilled and skilled labour, the casual poor and industrial workers, but rather numerous gradations between them,
complicating the attribution of a specific mentality to any given economic or social status within the working class. Caste, kinship and communal ties could facilitate
association and solidarity: they did not work simply, permanently or irretrievably to divide the working class along predictable lines. On the other hand, the rise of a
manufacturing sector by aggregating workers did not simply forge solidarities among them; on the contrary, its development often acted to
77 See chs. 3, 4, 5 and 8 above.

Page 348
divide workers and intensify the competition between them. There is no reason to suppose that an increase in their numbers, or more generally the progress of
manufacturing industry, would lead inevitably to class consciousness.
The absence of trade unions did not reflect a lack of consciousness among workers but suggested rather that they could not easily be sustained in the face of hostility
from employers and the state, which in turn was facilitated or circumscribed by the wider political context.78 The fact that most workingclass struggles arose out of or
focused upon the issues of the workplace should not lead us to conclude, as Wallerstein suggests, that they were economistic. For the obstacles to industrial action
sometimes appeared so awesome, from the manipulation of the wage relationship to the legal framework in which it operated and the repression by the agencies of the
state, that to resist the initiatives of employers often entailed stark political and moral choices, and the conduct of and participation in industrial action was a powerful
driving force in the development of class consciousness. Indeed, the scale of some strikes in India, most notably the 1928–29 strikes in Bombay organized in the teeth
of such opposition and sustained more or less continuously for eighteen months,79 was unprecedented and scarcely repeated in the history of the Western working
class, in cases where the development of capitalism had ostensibly met Wallerstein's desiderata – a proletariat of significant proportions and conditions which facilitated
its awareness of its class identity.
Workingclass resistance in India cannot be contained within the limitations imposed upon it by the categories of Wallerstein's analysis. The strikes of Indian workers
may have amounted to nothing more than 'the political struggles of . . . segments of classes within national boundaries', but they cannot be defined reductively as 'the
daily bread and butter of local politics'.80 Nor did they represent the crystallization of class consciousness as the discovery of 'the clearest route to the acquisition of
power within a given state structure'.81 Indeed, the Indian working classes could scarcely be said to represent a 'syndical interest group' at the national level, bound
together by their 'collective relationship to the world economy'.82 Of course, the momentum of workingclass action exerted an influence upon the policies of the
colonial state. But workingclass politics in early twentiethcentury India was characteristically local
78 See ch. 3 above.
79 See ch. 4 above; for an account of these strikes, see Newman, Workers and Unions, pp. 168–250; see also S. Bhattacharya, 'Capital and Labour in Bombay City, 1928–29', Economic
and Political Weekly, 16:42 and 43 (17–24 October 1981), Review of Political Economy, PE36 to PE44.
81 Ibid., p. 226.
82 Ibid., p. 24.

Page 349
ized and the formal or indeed informal connections across industrial centres were invariably weak. Industrial action centred upon the workplace and extended
outwards into the urban neighbourhood. The conduct of industrial action frequently generalized the scope of workingclass politics: when the state appeared to
intervene on the side of the employer in an industrial dispute it tended to nurture the growth of political consciousness. If class consciousness was not national in scope,
the working classes scarcely organized to act at the level of the 'national' state and bend its workings to safeguard workingclass interests within the world economy.
In the context of these localized struggles, the world economy did not weigh very heavily in the formation of workingclass consciousness. Of course, there were
moments when it was prominent. The Bombay millworkers appreciated that the excise duty impinged upon their wage cuts of 1925 and that the tariff and rupee
questions had an important bearing upon the redefinition of work, increased workloads and wage reductions that passed under the sobriquet of rationalization in 1928
and 1929.83 But their battle was waged in each case against the initiatives of the employers. There was some suspicion that the employers and the state, while passing
the responsibility onto each other, were in equal measure undermining the social and economic position of the workforce. However, their attitude to the state was
conditioned largely by its role in the development or settlement of their conflicts with their employers.84
Since class consciousness did not emerge in terms of a clear awareness of how the working class was situated within the world economy and since their struggles were
informed principally by the tensions and antagonisms of the local, rather than the national, context, it is difficult to credit Wallerstein's assertion that in 'peripheral areas
of the worldeconomy', 'the primary contradiction is between the interests organized and located in the core countries and their local allies on the one hand, and the
majority of the population on the other'.85 The periphery was scarcely a nest of singing birds. And such an argument subsumes beneath 'the primary contradiction' the
fundamental class differences and conflicts which divided the majority of the population, let alone the fact that the local allies of core interests were often
interchangeable with their local opponents, unless, of course, we understand class as an ontological thing, so that this primary contradiction was always immanent in
social relations, even if it was never fully realized.
Historians have often approached the question of class consciousness rather negatively. The expectation that class consciousness flowed nat
84 See ch. 4 above.

Page 350
urally from the development of largescale production and advanced industrial capitalism has frequently obliged historians to explain its absence. Conversely, in India, it
has been something of a commonplace to dismiss class and class consciousness altogether as a possible outcome of social conflict. Thus, in an argument which sought
to situate capital–labour relations at the heart of an analysis of the development of 'world capitalism', one historian has recently observed, 'Of the many forces which
have threatened South Asian society's political stability over the last one hundred and fifty years, that represented by the fur sich class power of the deeply
impoverished masses would not rank as the most significant.'86 If historians have approached the question of class so defensively, it is not because of the economic
deficiencies in class formation in India or even its political weakness, but rather because of the conceptual difficulties inherent in the term. Class and class
consciousness, categories grounded in a historical theory, have given rise to ahistorical approaches to the subject. For it has frequently led historians either to conceive
of political behavior, a priori, in terms of assumed social collectivities, or to measure a given set of conditions against an ideal construct of class and then to seek to
explain the gap between them as well as the shortcomings of the latter. If class formation could not occur simply within the nationstate, historical enquiry has focused
upon determining the functions of class and class conflict within the capitalist world economy. A certain functionalism underlies Wallerstein's definition of 'ethnonational
consciousness' as 'the constant resort of all those for whom class organization offers the risk of a loss of relative advantage through the normal workings of the market
and class dominated political bargaining'.87 Class and ethnicity are perhaps more fruitfully conceived as discursive categories. The principal historical questions which
their analysis raises are then why at particular moments diverse groups with often conflicting interests and varied cultures came together in often fragile political
coalition, why these apparently common interests are perceived in class terms, and why in particular contexts social and political conflicts appear to be illuminated by
ethnic, religious or national identities. It would follow, therefore, that nationality itself entailed a discursive reality, so that its definition could exclude groups among 'the
majority of the population' from the boundaries of national or ethnic identity or include them on strictly unequal terms.
86 Washbrook, 'South Asia, the World System and World Capitalism', 486–87.

Page 351
Bibliography
Manuscript Sources
Proceedings of the Secretary of State for India, in the Public and Judicial, and the Economic and Overseas Departments; the Private Office Papers; the Proceedings of
the Government of India; and Proceedings of the Government of Bombay. Oriental and India Office Collections, British Library, London.
Records of the Government of India, Home, Commerce and Industries, Industries and Labour Departments. National Archives of India, New Delhi.
Records of the Government of Bombay, Judicial, Home (Political), Home (Special), Finance, Revenue Development, General, General (Plague), Public Works
Departments. Maharashtra State Archives, Bombay.
Records of the Commissioner of Police, Bombay. Office of the Commissioner of Police, Bombay.
Records of the Deputy InspectorGeneral of Police, Criminal Investigation Department, Maharashtra, Secret Abstracts of Intelligence, Bombay Presidency Police,
Office of the Deputy Inspector General of Police, Criminal Investigations Department, Maharashtra, Bombay.
Unpublished Proceedings of Committees of Enquiry
The Extension of the City of Bombay, 1887, GOB, Public Works Department (General), Vol. 1162, Compilation no. 4133 W, 1868–89. Maharashtra State
Archives, Bombay.
Bombay Strike Enquiry Committee, 1928. Maharashtra State Archives, Bombay.
Bombay Riots Inquiry Committee, 1929. Maharashtra State Archives, Bombay.
Court of Enquiry into a Trade Dispute between Several Textile Mills and their Workmen, 1929. Maharashtra State Archives, Bombay.
Bombay Provincial Banking Enquiry Committee, 1929. Maharashtra State Archives, Bombay.
Meerut Conspiracy Case, Statements and Exhibits, 1929–33. National Archives of India, New Delhi.
Bombay Disturbances Enquiry Committee, 1938. Government of Bombay, Home (Special) Files. Maharashtra State Archives, Bombay.
Textile Labour Inquiry Committee, Interim and Main Inquiry, 1937–40. Maharashtra State Archives, Bombay.

Page 352
Private Papers
Papers of J. C. Curry, Centre of South Asian Studies, Cambridge.
Papers of N. M. Joshi, Nehru Memorial Museum and Library, New Delhi.
Papers of Sir P. Thakurdas, Nehru Memorial Museum and Library, New Delhi.
Papers of Organizations
Ahmedabad Textile Labour Association, Microfilm copy. Nehru Memorial Museum and Library, New Delhi.
Minute Books of the Committee of the Bombay Millowners' Association. Office of the Bombay Millowners' Association.
All India Congress Committee. Nehru Memorial Museum and Library, New Delhi.
AllIndia Trade Union Congress. Nehru Memorial Museum and Library, New Delhi.
Official Publications
Parliamentary Papers
Annual Report on Sanitary Measures in India.
Royal Commission on Labour, Foreign Reports, vol. II, The Colonies and the Indian Empire with an Appendix on the Migration of Labour, 1892, vol.
XXXVI, Cmd. 6795–XI.
'Papers Relating to the Outbreak of Bubonic Plague in India with Statement Showing the Quarantine and Other Restrictions Recently Placed Upon Indian Trade, up to
March 1897', 1897, vol. LXIII, Cmd. C8386.
Report of the Indian Plague Commission, 1898–99, with Appendices and Summary, vol. V, 1902, vol. LXXII, Cmd. 810.
Report of the Select Committee on the Employment of the Military in Cases of Disturbances, Parliamentary Papers, 1908, vol. VII, Cmd. 236.
Report of the Committee Appointed by the Government of India to Investigate the Disturbances in the Punjab etc. vol. XIV, 1920, Cmd. 681.
Reports on the Punjab Disturbances, April 1919. vol. XIV, 1920, Cmd. 534.
Disturbances in the Punjab, Statement by BrigadierGeneral R. E. H. Dyer. vol. XXXIV, 1920, Cmd. 771.
Report of the Commission of Inquiry into the Cawnpore Riots and Resolution of the Government of the United Provinces. vol. XII, 1930–1, Cmd. 3891.
Government of India's Despatch on Proposals for Constitutional Reform, dated 20th September 1930, vol. XXIII, 1930–31, Cmd. 3700.
Report of the Indian Franchise Committee, 1932, vol. 1, vol. VIII, 1931–32, Cmd. 4086.
His Majesty's Stationery Office (HMSO)
Report of the Indian Statutory Commission, vol. I, Survey. London, 1930.
Report of the Royal Commission on Labour in India, and Evidence (11 vols.). London, 1931.

Page 353
Government of India
Report of the Indian Factory Commission, Appointed in September 1890, under the Orders of His Excellency, the GovernorGeneralinCouncil, with
Proceedings and Appendices. Calcutta, 1890.
Census of India, 1891–1941.
Report of the Indian Factory Labour Commission 1908, 2 vols. Simla, 1908.
Indian Industrial Commission, Vol. I – Report, vols. II–V – Evidence. Calcutta, 1918.
The Indian Tariff Board (Cotton Textile Industry Enquiry), 1927. Vol. I – Report, vols. II–IV – Evidence. Bombay, 1927.
Indian Central Banking Enquiry Committee. Vol. I – Reports, vols. II–III – Evidence. Calcutta, 1930–32.
Report of the Indian Tariff Board Regarding the Grant of Protection to the Cotton Textile Industry. Calcutta, 1932.
Report of the Indian Tariff Board Regarding the Grant of Additional Protection to the Cotton Textile Industry. Calcutta, 1932.
Report of the Special Tariff Board on the Enquiry Regarding the Level of Duties Necessary to Afford Adequate Protection to the Indian Cotton Textile
Industry Against Imports from the United Kingdom of Cotton Piecegoods and Yarn, Artificial Silk Fabrics and Mixture Fabrics of Cotton and Artificial
Silk. Delhi, 1936.
Government of India, Terrorism in India, 1917–1936. Simla, 1937.
Main Report of the Labour Investigation Committee. Delhi, 1946.
Government of Bombay
Coneybeare, H. Report on the Sanitary State and Requirements of Bombay, Selections from the Records of the Bombay Government, new series, vol. XI.
Bombay, 1855.
Report of the Commissioner Appointed by the Governor of Bombay in Council to Inquire into the Conditions of the Operatives in the Bombay Presidency,
1875. Bombay, 1875.
Report by W. O. Meade King on the Working of the Indian Factories Act in Bombay together with Certain Suggestions and Proposals. Bombay, 1882.
Report of the Commission Appointed to Consider the Working of Factories in the Bombay Presidency, 1888. Bombay, 1888.
Fawcett, C. G. H. A Monograph on Dyes and Dyeing in the Bombay Presidency. Bombay, 1896.
Enthoven, R. E. The Cotton Fabrics of the Bombay Presidency. Bombay, 1897.
Snow, P. C. H. Report on the Outbreak of Bubonic Plague in Bombay, 1896–97. Bombay, 1897.
Haffkine, Mons. W. M. Experiment on the Effect of Protective Inoculation in the Epidemic of Plague at Undhera Taluka, Baroda, February and March,
1898. Bombay, 1898.
A Monograph on Evacuation as a Protective and Combative Plague Measure, Compiled Under the Orders of Sir A. Wingate, KCIE, Acting Chief Secretary
to Government by Lt. J. K. Condon, UnderSecretary to Government (Plague

Page 354
Department) for the use of the Indian Plague Commission, 30 March 1899. Bombay, 1899.
Report of the Municipal Commissioner on the Plague in Bombay in the Year Ending 31st May, 1899. Bombay, 1899.
Condon, Capt. J. K. The Bombay Plague, Being A History of the Progress of Plague in the Bombay Presidency from September 1896 to June 1899. Bombay,
1900.
Gazetteer of Bombay City and Island, 3 vols. Bombay, 1909.
Report of the Bombay Development Committee. Bombay, 1914.
Report of the Industrial Disputes Committee. Bombay, 1922.
Report of the Bombay Strike Enquiry Committee, vol. I. Bombay, 1929.
Report of the Bombay Riot Inquiry Committee. Bombay, 1929.
Report of the Court of Inquiry into a Trade Dispute Between Several Textile Mills and their Workmen. Bombay, 1929.
Report of the Bombay Provincial Banking Enquiry Committee. Vol. I – Report, vols. II–IV – Evidence. Bombay, 1930.
Report of the Rent Enquiry Committee, with Evidence. Bombay, 1939.
Report of the Textile Labour Inquiry Committee 1937–38. Vol. I – Interim Report. Bombay, 1938.
Report of the Textile Labour Inquiry Committee. Vol. II – Final Report. Bombay, 1953.
Labour Office, Bombay
Labour Gazette (monthly).
Report on an Enquiry into the Wages and Hours of Labour in the Cotton Mill Industry, May, 1921 by G. Findlay Shirras. Bombay, 1923.
Report on an Enquiry into Working Class Family Budgets in Bombay by G. Findlay Shirras. Bombay, 1923.
Report on an Enquiry into the Deductions from Wages or Payments in Respects of Fines. Bombay, 1928.
Wages and Unemployment in the Bombay Cotton Textile Industry. Bombay, 1934.
Report on Wages, Hours of Work and Conditions of Employment in the Textile Industries (Cotton, Silk, Wool and Hosiery) in the Bombay Presidency
(excluding Sind), May 1934. General Wage Census, Part I – Perennial Factories. Third Report. Bombay, 1937.
Annual Reports
Administration Report of the City of Bombay Improvement Trust.
Administration Report of the Government of Bombay.
Administration Report of the Municipal Commissioner for the City of Bombay.
Bombay Presidency Police, Secret Abstracts of Intelligence.
Factory Report for the Bombay Presidency.
Report of the Department of Industries in the Bombay Presidency.
Report on the Native Newspapers Published in the Bombay Presidency.
Report on the Police in the Town and Island of Bombay for 1884–1912.
Report on the Police in the City of Bombay, 1913–47.

Page 355
Prices and Wages in India. Calcutta.
Report of the Sanitary Commissioner for Bombay.
Report of the Sanitary Commissioner with the Government of India.
Report of the Sanitary Commissioner for the NorthWestern Provinces.
Report on the Working of the Development Department in Bombay.
Report on the Working of the Indian Factories Act.
NonOfficial Reports
Annual Reports of the Bombay Millowners' Association.
Manchester Chamber of Commerce. Bombay and Lancashire Cotton Spinning Inquiry: Minutes and Evidence. Manchester, 1888.
Great Indian Peninsular Railway, Report for the Royal Commission on Labour in India. Bombay, 1929.
Newspapers
Bombay Chronicle
Bombay Sentinel
Congress Socialist
Harijan
Indian National Herald
Kranti (Marathi)
Nava Kal (Marathi)
Times of India
Young India
Published Sources
AITUC – Fifty Years On: Documents, Introduction by S. A. Dange, vol. I. AITUC, New Delhi, 1973.
Chopra, P. N. (ed.). Quit India Movement: British Secret Report. Faridabad, 1976.
The Collected Works of M. K. Gandhi, 90 vols. Delhi, 1958–84.
Desai, M. A Righteous Struggle: A Chronicle of the Ahmedabad Textile Labourers' Fight For Justice. Ahmedabad, 1951.
Mansergh, P. N. S. (ed.). The Transfer of Power, 1942–47, vol. II: Quit India, 30 April–21 September 1942. HMSO, London, 1971.
The Moral and Political Writings of Mahatma Gandhi, 3 vols., ed. R. Iyer. Oxford, 1987.
Punjab Disturbances, 1919–20, 2 vols., Report of the Commissioners Appointed by the Punjab SubCommittee of the Indian National Congress to Look
into the Jallianwalla Bagh Massacre. Delhi, 1976.
Report of the Congress Cawnpore Riots Enquiry Committee in N. G. Barrier (ed.), Roots of Communal Violence Delhi, 1976.
The Selected Works of Jawaharlal Nehru, 15 vols., ed. S. Gopal. New Delhi, 1972–82.
Source Material for a History of the Freedom Movement in India, vol II, 1885–1920. Bombay, 1958.

Page 356
Source Material for a History of the Freedom Movement in India, vol VI, NonCooperation Movement, Bombay City, 1920–1925, ed. B. G. Kunte. Bombay,
1978.
Books, Pamphlets and Articles
Abrams, P. Historical Sociology. London, 1983.
Ambedkar, B. R. What Congress and Gandhi Have Done to the Untouchables. Bombay, 1945.
Amin, S. Sugarcane and Sugar in Gorakhpur: An Inquiry into Peasant Production for Capitalist Enterprise in Colonial India. Delhi, 1984.
'Gandhi as Mahatma, Gorakhpur District, Eastern UP, 1921–22', in R. Guha (ed.), Subaltern Studies, vol. III, pp. 1–57.
'Agrarian Bases of Nationalist Agitations in India: An Historiographical Survey', in D. A. Low (ed.), The Indian National Congress: Centenary Hindsights, pp.
98–128. Delhi, 1988.
Event, Metaphor, Memory: Chauri Chaura, 1922–1992. Berkeley, 1995.
Amin, S. and D. Chakrabarty (eds.). Subaltern Studies, vol. IX. Delhi, 1996.
Arnold, D. 'The Armed Police and Colonial Rule in South India, 1914–1947', Modern Asian Studies, 11:1 (1977), 101–25.
'Labour Relations in a South India Sugar Factory 1937–39' Social Scientist, 65 (December 1977), 16–33.
'Dacoity and Rural Crime in Madras, 1860–1940', Journal of Peasant Studies, 6:2 (1979), 140–67.
'Industrial Violence in Colonial India', Comparative Studies in Society and History, 22:2 (1980), 234–55.
'Cholera and Colonialism in British India', Past and Present, 113 (1986), 118–51.
Police Power and Colonial Rule: Madras, 1859–1947. Delhi, 1986.
'Touching the Body: Perspectives on the Indian Plague, 1896–1900', in R. Guha (ed.), Subaltern Studies, vol. V, pp. 55–90.
'Smallpox and Colonial Medicine in Nineteenth Century India', in D. Arnold (ed.), Imperial Medicine and Indigeneous Societies, pp. 45–65. Manchester, 1988.
'Cholera Mortality in British India, 1817–1947', in T. Dyson (ed.), India's Historical Demography: Studies in Famine, Disease and Society, pp. 261–84.
London, 1989.
Colonizing the Body: State Medicine and Epidemic Disease in Nineteenth Century India. Berkeley and Los Angeles, 1993.
Arnold, D. and D. Hardiman (eds.), Subaltern Studies, vol. VIII. Delhi, 1994.
Arrighi, G. 'Labour Supplies in Historical Perspective: A Study of the Proletarianization of the African Peasantry in Rhodesia', Journal of Development Studies, 6:3
(1970), 197–234.
BadenPowell, B. H. The Indian Village Community. London, 1896.
Bagchi, A. K. Private Investment in India 1900–1939. Cambridge, 1972.
'Foreign Capital and Economic Development in India: A Schematic View', in K. Gough and H. P. Sharma (eds.), Imperialism and Revolution in South Asia, pp.
43–76. London, 1973.

Page 357
'DeIndustrialisation in Gangetic Bihar, 1809–1901', in B. De (ed.), Essays in Honour of Professor S. C. Sarkar, pp. 499–522. Delhi, 1976.
'DeIndustrialisation in India in the Nineteenth Century: Some Theoretical Implications', Journal of Development Studies, 12:2 (1976), 135–64.
'Reflections on Patterns of Regional Growth in India During the Period of British Rule', Bengal Past and Present, 95:1 (1976), pp. 247–89.
'A Reply', Indian Economic and Social History Review, 16:2 (1979), 147–61.
'Colonialism and the Nature of ''Capitalist" Enterprise in India', Economic and Political Weekly, 23:31 (30 July 1988), Review of Political Economy, PE38 to
PE58.
Bahl, V. 'Attitude of the Indian National Congress Towards the Working Class Struggle in India, 1918–1947', in Kapil Kumar (ed.), Congress and Classes:
Nationalism, Workers and Peasants, pp. 1–33.
'TISCO Workers' Struggle, 1920–28', Social Scientist, 10:8 (1982), 32–44.
'The Emergence of LargeScale Steel Industry in India under British Colonial Rule, 1880–1947', Indian Economic and Social History Review, 31:4 (1994),
413–60.
The Making of the Indian Working Class: The Case of the Tata Iron and Steel Company. Delhi, 1995.
Bairoch, P. 'International Industrialization Levels from 1750 to 1980', Journal of European Economic History, 11:2 (1982), 269–333.
Baker, C. J. The Politics of South India 1920–37. Cambridge, 1976.
'Debt and Depression in Madras, 1929–1936', in C. J. Dewey and A. G. Hopkins (eds.), The Imperial Impact: Studies in the Economic History of Africa
and India, pp. 233–42.
'Colonial Rule and the Internal Economy in Twentieth Century Madras', Modern Asian Studies, 15:3 (1981), special issue, Power, Profit and Politics: Essays
on Imperialism, Nationalism and Change in Twentieth Century India, 1870–1940, ed. C. Baker, G Johnson and A. Seal, 575–602.
'Economic Reorganization and the Slump in South and SouthEast Asia', Comparative Studies in Society and History, 23:3 (July 1981), 325–49.
An Indian Rural Economy: The Tamil Nad Countryside, 1880–1950. Oxford, 1984.
Balibar, E. 'Basic Concepts of Historical Materialism', in L. Althusser and E. Balibar (eds.), Reading Capital, pp. 199–308. London, 1970.
Bamford, P. C. Histories of the NonCooperation and Khilafat Movements. Delhi, 1925; reprinted Delhi, 1974.
Banaji, J. 'Capitalist Domination and the Small Peasantry: Deccan Districts in the Late Nineteenth Century', Economic and Political Weekly, 12:33–4 (1977), special
number, 1375–404.
Baran, P. The Political Economy of Growth. New York, 1957.
Barrier, N. G. The Punjab Land Alienation Bill of 1900. Duke University, Program in Comparative Studies in Southern Asia, Monograph and Occasional Papers
Series, no. 2, 1966.
'The Arya Samaj and Congress Politics in Punjab, 1894–1908', Journal of Asian Sudies, 26:3 (1967), 363–79.
'The Punjab Disturbances of 1907: The Response of the British Government in India to Agrarian Unrest', Modern Asian Studies, 1:4 (1967), 353–83.

Page 358
Bates, C. N. 'The Nature of Social Change: The Kheda District, 1818–1918', Modern Asian Studies, 15:4 (1981), 771–821.
Bates, C. and M. Carter. 'Tribal Migration in India and Beyond', in G. Prakash (ed.), The World of the Rural Labourer in Colonial India, pp. 205–47. Delhi,
1992.
Bayly, C. A. 'Patrons and Politics in Northern India' in J. Gallagher, G. Johnson and A. Seal (eds.), Locality, Province and Nation: Essays on Indian Politics, pp.
29–68.
The Local Roots of Indian Politics: Allahabad, 1870–1920. Oxford, 1975.
'Indian Merchants in a ''Traditional" Setting: Benares, 1780–1830', in C. J. Dewey and A. G. Hopkins (eds.), The Imperial Impact: Studies in the Economic
History of Africa and India, pp. 171–93.
Rulers, Townsmen and Bazaars: North Indian Society in the Age of British Expansion, 1770–1870. Cambridge, 1983.
'The PreHistory of "Communalism"? Religious Conflict in India, 1700–1860', Modern Asian Studies, 19:2 (1985), 177–203.
The New Cambridge History of India, vol. II:1 Indian Society and the Making of the British Empire. Cambridge, 1988.
Beals, A. R. 'Strategies of Resort to Curers in South India', in Charles Leslie (ed.), Asian Medical Systems: A Comparative Study, pp. 184–200. Berkeley and Los
Angeles, 1976.
Beattie, J. M. Crime and the Courts in England, 1660–1800. Oxford, 1986.
Behal, R. P. 'Forms of Labour Protest in Assam Valley Tea Plantations, 1900–1930', Economic and Political Weekly, 20:4 (26 January 1985), Review of Political
Economy PE19 to PE26.
Behal, R. P. and P. P. Mohapatra. '"Tea and Money Versus Human Life": The Rise and Fall of the Indenture System in the Assam Tea Plantations, 1840–1908', in E.
V. Daniel, H. Bernstein and T. Brass (eds.), Plantations, Proletarians and Peasants in Colonial Asia, pp. 142–72.
Bhattacharya, N. 'The Logic of Tenancy Cultivation: Central and SouthEast Punjab, 1870–1935', Indian Economic and Social History Review, 20:2 (1983), 121–
70.
'Agricultural Labour and Production: Central and SouthEast Punjab, 1870–1940', in K. N. Raj, N. Bhattacharya, S. Guha and S. Padhi (eds.), Essays on the
Commercialization of Indian Agriculture, pp. 105–62. Delhi, 1985.
'Lenders and Debtors: Punjab Countryside, 1880–1940', Studies in History, 1:2 (1985), 305–42.
Bhattacharya, S. 'Cotton Mills and Spinning Wheels: Swadeshi and the Indian Capitalist Class, 1920–22', Economic and Political Weekly, 11:47 (1976), 1828–34.
'Capital and Labour in Bombay City, 1928–29', Economic and Political Weekly, 16:42 and 43 (17–24 October 1981), Review of Political Economy, PE36
to PE44.
'Regional Economy (1757–1857): Eastern India, Part I', in D. Kumar (ed.), The Cambridge Economic History of India, vol. II: C. 1757–c. 1970, pp. 270–
331.

Page 359
'Swaraj and the Kamgar: The Indian National Congress and the Bombay Working Class, 1919–1931', in R. Sisson and S. Wolpert (eds.), Congress and Indian
Nationalism: The PreIndependence Phase, pp. 223–49. Berkeley and Los Angeles, 1988.
Bhuyan, A. C. The Quit India Movement: The Second World War and Indian Nationalism. Delhi, 1975.
Bose, Sanat. '''Industrial Unrest" and the Growth of Labour Unions in Bengal, 1920–1924', Economic and Political Weekly, 16:44–6 (November 1981), special
number, 1849–60.
Bose, S. (ed.) South Asia and World Capitalism. Delhi, 1990.
Brahma Nand. 'Agricultural Labourers in Western India: A Study of the Central Division Districts of the Bombay Presidency During the Late Nineteenth and Early
Twentieth Century', Studies in History, new series, 1:2 (1985), special issue, Essays in Agrarian History: India, 1850 to 1940, ed. S. Bhattacharya, pp. 221–46.
Breman, J. Of Peasants, Migrants and Paupers: Rural Labour Circulation and Capitalist Production in West India. Delhi, 1985.
Taming the Coolie Beast: Plantation Society and the Colonial Order in Southeast Asia. Delhi, 1989.
Beyond Patronage and Exploitation: Changing Agrarian Relations in South Gujarat. Delhi, 1993.
Breman, J. and E. Valentine Daniel. 'Conclusion: The Making of a Coolie', in E. V. Daniel, H. Bernstein and T. Brass (eds.), Plantations, Proletarians and
Peasants in Colonial Asia, pp. 268–95.
Brenner, R. 'The Origins of Capitalist Development: A Critique of NeoSmithian Marxism', New Left Review, 104 (July–August 1977), 25–94.
Brewer, A. Marxist Theories of Imperialism: A Critical Survey. London, 1980.
Brogden, M. The Emergence of the Police: The Colonial Dimension', British Journal of Criminology, 27:1 (1987), 4–14.
BurnettHurst, A. R. Labour and Housing in Bombay: A Study in the Economic Condition of the WageEarning Classes of Bombay. London, 1925.
Cannadine, D. 'The Present and the Past in the English Industrial Revolution, 1880–1980', Past and Present, 103 (May 1984), pp. 131–71.
Carter, M. 'Strategies of Labour Mobilisation in Colonial India: The Recruitment of Indentured Workers for Mauritius', in E. V. Daniel, H. Bernstein and T. Brass
(eds.), Plantations, Proletarians and Peasants in Colonial Asia, pp. 228–45.
Servants, Sirdars and Settlers: Indians in Mauritius, 1834–1874. Delhi, 1995.
Cashman, R. I. The Myth of the Lokamanya: Tilak and Mass Politics in Maharashtra. Berkeley and Los Angeles, 1975.
Catanach, I. J. 'Plague and the Indian Village, 1896–1914', in P. Robb (ed.), Rural India: Land, Power and Society under British Rule, pp. 216–43. London,
1983.
'Poona Politicians and the Plague', in J. Masselos (ed.), Struggling and Ruling: The Indian National Congress, 1885–1985, pp. 198–215. New Delhi, 1987.
'Plague and the Tensions of Empire, 1896–1918', in D. Arnold (ed.), Imperial Medicine and Indigeneous Societies, pp. 149–71. Manchester, 1988.

Page 360
Cell, J. W. 'AngloIndian Medical Theory and the Origins of Segregation in West Africa', American Historical Review, 91:2 (1986), 307–35.
Chakrabarty, B. Subhas Chandra Bose and Middle Class Radicalism: A Study in Indian Nationalism, 1928–1940. London, 1990.
Chakrabarty, D. 'Sasipada Bannerjee: A Study in the Nature of the First Contact of the Bengal Bhadralok with the Working Class of Bengal', Indian Historical
Review, 2:2 (1976), 339–64.
'Communal Riots and Labour: Bengal's Jute Mill Hands in the 1890s', Past and Present, 91 (May 1981), 140–69.
'Conditions for Knowledge of WorkingClass Conditions: Employers, Government and the Jute Workers of Calcutta', in R. Guha (ed.), Subaltern Studies, vol.
II, pp. 259–310.
'On Deifying and Defying Authority: Managers and Workers in the Jute Mills of Bengal, circa 1890–1940', Past and Present, 100 (August 1983), 124–46.
Rethinking WorkingClass History: Bengal, 1890–1940. Delhi, 1989.
Chandavarkar, R. 'Workers' Politics in the Mill Districts of Bombay Between the Wars', Modern Asian Studies, 15:3 (1981), special issue, Power, Profit and
Politics: Essays on Imperialism, Nationalism and Change in Twentieth Century India, 1870–1940, ed. C. Baker, G. Johnson and A. Seal, 603–47.
'Workers' Resistance and the Rationalization of Work in Bombay between the Wars', in D. Haynes and G. Prakash (eds.), Contesting Power: Resistance and
Everyday Social Relations in South Asia, pp. 109–44. Berkeley and Los Angeles, 1991.
The Origins of Industrial Capitalism in India: Business Strategies and the Working Classes in Bombay, 1900–1940. Cambridge, 1994.
'From Communism to ''Social Democracy": The Rise and Resilience of Communist Parties in India, 1920–1995', Science and Society, 61:1 (1997), 99–106.
Chandra, B. The Rise and Growth of Economic Nationalism in India: Economic Policies of Indian National Leadership 1880–1905. New Delhi, 1966.
Charlesworth, N. 'Rich Peasants and Poor Peasants in Late Nineteenth Century Maharashtra', in C. J. Dewey and A. G. Hopkins (ed.), The Imperial Impact:
Studies in the Economic History of Africa and India, pp. 97–114.
British Rule and the Indian Economy, 1800–1914. London, 1983.
Peasants and Imperial Rule: Agriculture and Agrarian Society in the Bombay Presidency, 1850–1935. Cambridge, 1985.
Chatterjee, P. Nationalist Thought and the Colonial World: A Derivative Discourse. London, 1986.
The Nation and Its Fragments: Colonial and PostColonial Histories. Delhi, 1994.
Chatterjee, P. and G. Pandey (eds.), Subaltern Studies, vol. VII. Delhi, 1992.
Chatterji, B. 'Business and Politics in the 1930s: Lancashire and the Making of the IndoBritish Trade Agreement', Modern Asian Studies, 15:3 (1981), special issue,
Power, Profit and Politics: Essays on Imperialism, Nationalism and Change in Twentieth Century India, 1870–1940, ed. C. Baker, G Johnson and A. Seal,
527–73.
'The Political Economy of "Discriminating Protection": The Case of Textiles in the 1920s', Indian Economic and Social History, 20:3 (1983), 239–75.

Page 361
Trade, Tariffs and Empire: Lancashire and British Policy in India, 1919–1939. Delhi, 1992.
Chatterji, J. Bengal Divided: Hindu Communalism and Partition, 1932–1947. Cambridge, 1994.
Chattopadhyay, M. 'Santosh Kumari Devi: A Pioneering Labour Leader', Social Scientist, 128 (January 1984), 62–73.
Chaudhuri, K. N. 'The Structure of the Indian Textile Industries in the Seventeenth and Eighteenth Centuries', Indian Economic and Social History Review, 11:2–3
(1974), 127–82.
The Trading World of Asia and the English East India Company, 1660–1760. London, 1978.
'Foreign Trade and Balance of Payments (1757–1947)', in D. Kumar (ed.), The Cambridge Economic History of India, vol. II: C. 1757–c. 1970, pp. 804–77.
Chaudhuri, K. N. and C. J. Dewey (eds.). Economy and Society: Essays in Indian Economic and Social History. Delhi, 1979.
Chaudhury, P. 'Labour Migration from the United Provinces, 1881–1911', Studies in History, new series, 8:1 (1992), pp. 13–42.
Chesneaux, J. The Chinese Labour Movement, 1919–1927. Stanford, 1974.
Chicherov, A. I. 'Tilak's Trial and the Bombay Political Strike of 1908', in I. M. Reisner and N. M. Goldberg (eds.), Tilak and the Struggle for Indian Freedom,
pp. 545–626.
Chirol, V. Indian Unrest. London, 1910.
Cholia, R. P. Dock Labourers in Bombay. Bombay, 1941.
Chowdhury, B. B. 'Regional Economy (1757–1857): Eastern India, Part II', in D. Kumar (ed.), The Cambridge Economic History of India, vol. II: C. 1757–c.
1970, pp. 270–331.
Cobb, R. The Police and the People: French Popular Protest, 1789–1820. Oxford, 1970.
Cockburn, J. S. 'Patterns of Violence in English Society: Homicide in Kent, 1560–1985', Past and Present, 130 (February 1991), 70–106.
Cohen, S. P. The Indian Army: Its Contribution to the Development of the Indian Nation. Berkeley, 1971.
Colvin, Sir Auckland. Audi Alterem Partem: Being Two Letters on Certain Aspects of the Indian National Congress Movement. Simla, 1888.
Cooper, A. Sharecropping and Sharecroppers' Struggles in Bengal, 1930–1950. Calcutta, 1988.
Coupland, R. A Report on the Constitutional Problem in India, 3 vols. London, 1942–3.
India – A Restatement. London, 1945.
Cox, E. C. Police and Crime in India. London, 1910.
Crouzet, F. Capital Formation in the Industrial Revolution. London, 1972.
Curry, J. C. The Indian Police. London, 1932.
Dale, S. F. Islamic Society on the South Asian Frontier: The Mappilas of Malabar, 1498–1922. Oxford, 1980.
Damodaran, K. 'Memoir of an Indian Communist', New Left Review, 93 (1975), 35–59.

Page 362
Daniel, E. V., H. Bernstein and T. Brass (eds.). Plantations, Proletarians and Peasants in Colonial Asia. London, 1992.
Das, S. Communal Riots in Bengal in the Twentieth Century. Delhi, 1992.
'The ''Goondas": Towards a Reconstruction of the Calcutta Underworlds through Police Records', Economic and Political Weekly, 29:44 (29 October 1994),
2877–83.
Das, V. 'Introduction: Communities, Riots and Survivors: The South Asian Experience', in Das (ed.), Mirrors of Violence: Communities, Riots and Survivors in
South Asia.
Das V. (ed.), Mirrors of Violence: Communities, Riots and Survivors in South Asia. Delhi, 1990.
Das Gupta, R. 'Factory Labour in Eastern India: Sources of Supply, 1855–1946: Some Preliminary Findings', Indian Economic and Social History Review, 13:3
(1976), 277–328.
'Migrants in Coal Mines: Peasants or Proletarians, 1850s–1947', Social Scientist, 151 (December 1985), 18–43.
'Plantation Labour in Colonial India', in E. V. Daniel, H. Bernstein and T. Brass (eds.), Plantations, Proletarians and Peasants in Colonial Asia, pp. 173–98.
Datta, P. 'Strikes in the Greater Calcutta Region, 1918–1924' Indian Economic and Social History Review, 30:1 (1993), 57–84.
Datta, S. B. Capital Accumulation and Workers' Struggle in Indian Industrialization: The Case of the Tata Iron and Steel Company, 1910–1970. Stockholm,
1986.
Datta, V. N. Jallianwalla Bagh. Ludhiana, 1969.
Davis, J. 'The London Garotting Panic of 1862: A Moral Panic and the Creation of a Criminal Class in midVictorian England', in V. A. C. Gatrell, B. Lenman and G.
Parker (eds.), Crime and the Law: The Social History of Crime in Western Europe since 1500, pp. 190–213. London, 1980.
'From "Rookeries" to "Communities": Race, Poverty and Policing in London, 1850–1985', History Workshop Journal, 27 (Spring 1989), 66–85.
'Jennings' Buildings and the Royal Borough: The Construction of an Underclass in midVictorian England', in D. Feldman and G. Stedman Jones (eds.),
Metropolis – London: Histories and Representations since 1800, pp. 11–39. London, 1989.
Davis, K. The Population of India and Pakistan. Princeton, N.J., 1951.
Desai, A. V. 'The Origins of Parsi Enterprise', Indian Economic and Social History Review, 5:4 (1968), 307–17.
Dewey, C. 'Images of the Village Community: A Study in AngloIndian Ideology', Modern Asian Studies, 6:3 (1972), 291–328.
'Some Consequences of Military Expenditure in British India: The Case of the Upper Sind Sagar Doab, 1849–1947', in C. Dewey (ed.), Arrested Development
in India: The Historical Dimension, pp. 93–169. Delhi, 1988.
'The End of the Imperialism of Free Trade: The Eclipse of the Lancashire Lobby and the Concession of Fiscal Autonomy to India', in C. J. Dewey and A. G.
Hopkins (eds.), The Imperial Impact: Studies in the Economic History of Africa and India, pp. 35–67.
Dewey, C. J. and A. G. Hopkins (eds.). The Imperial Impact: Studies in the Economic History of Africa and Asia. London, 1978.

Page 363
Dhanagare, D. N. Peasant Movements in India, 1920–1950. Delhi, 1983.
Dirks, N. The Hollow Crown: Ethnohistory of an Indian Little Kingdom. Cambridge, 1987.
Ditton, J. Controlology: Beyond the New Criminology. London, 1979.
Drummond, I. British Economic Policy and the Empire, 1919–1939. London, 1972.
Dumont, Louis 'The ''Village Community" from Munro to Maine', Contributions to Indian Sociology, 9 (December 1965), 67–89.
Edwardes, S. M. The Rise of Bombay – A Retrospect. Bombay, 1902.
The ByWays of Bombay. Bombay, 1912.
Memoir of Rao Bahadur Ranchodlal Chotalal. Exeter, 1920.
The Bombay City Police – 1672–1916. A Historical Sketch. London, 1923.
Emsley, C. Policing and Its Context, 1750–1870. London, 1983.
Erikson, E. Gandhi's Truth: On the Origins of Militant NonViolence. London, 1969.
Fabian, J. Time and the Other: How Anthropology Makes Its Object. New York, 1983.
Farnie, D. A. The English Cotton Industry and the World Market, 1815–96. Oxford, 1979.
Fisher, C. 'Planters and Peasants: The Ecological Context of Agrarian Unrest on Indigo Plantations of North Bihar, 1820–1920', in C. J. Dewey and A. G. Hopkins
(eds.), The Imperial Impact: Studies in the Economic History of Africa and India, pp. 114–31.
Frank, A. G. Capitalism and Underdevelopment in Latin America: Historical Studies of Chile and Brazil. New York, 1967.
Frankel, F. India's Political Economy, 1947–1977: The Gradual Revolution. Princeton, 1978.
Freitag, S. 'Collective Crime and Authority in North India', in A. Yang (ed.), Crime and Criminality in British India, pp. 140–63.
'Crime in the Social Order of Colonial North India', Modern Asian Studies, 25:2 (1991), 227–61.
Gadgil, M. and R. Guha. 'State Forestry and Social Conflict in British India', Past and Present, 123 (1989), 141–77.
Gallagher, J. 'Congress in Decline: Bengal, 1930–39', in J. Gallagher, G. Johnson and A. Seal (eds.), Locality, Province and Nation: Essays on Indian Politics, pp.
269–325.
The Decline, Revival and Fall of the British Empire, ed. A. Seal. Cambridge, 1982.
Gallagher, J. and A. Seal. 'Britain and India Between the Wars', Modern Asian Studies, 15:3 (1981), special issue, Power, Profit and Politics: Essays on
Imperialism, Nationalism and Change in Twentieth Century India, 1870–1940, ed. C. Baker, G. Johnson and A. Seal, 387–414.
Gallagher, J., G. Johnson and A. Seal (eds.). Locality, Province and Nation: Essays on Indian Politics. Cambridge, 1973.
Gatrell, V. A. C. 'Crime, Authority and the Policeman State, 1750–1950', in F. M. L. Thompson (ed.), The Cambridge Social History of Britain, 1750–1950, vol.
III, pp. 243–310. Cambridge, 1990.
Geary, R. European Labour Protest, 1848–1918. London, 1981.

Page 364
Gillion, K. Ahmedabad: A Study in Indian Urban History. Berkeley and Los Angeles, 1968.
Giri, V. V. My Life and Times, vol. I. Delhi, 1976.
Gopal Krishna. 'The Development of the Congress as a Mass Organization, 1918–1923', Journal of Asian Studies, 25:3 (1966), 413–30.
Gordon, A. D. 'Businessmen and Politics in a Developing Colonial Economy: Bombay City, 1918–1933', in C. J. Dewey and A. G. Hopkins (eds.), The Imperial
Impact: Studies in the Economic History of Africa and India, pp. 194–215.
Businessmen and Politics: Rising Nationalism and a Modernising Economy in Bombay, 1918–1933. New Delhi, 1978.
Gordon, S. 'Scarf and Sword: Thugs, Marauders and State Formation in 18th Century Malwa', Indian Economic and Social History Review, 6:4 (1969), 403–29.
Goswami, O. 'Collaboration and Conflict: European and Indian Capitalists and the Jute Economy of Bengal, 1919–1939', Indian Economic and Social History
Review, 19:2 (1982), 141–79.
'Then Came the Marwaris: Some Aspects of the Changes in the Pattern of Industrial Control in Eastern India', Indian Economic and Social History Review,
22:3 (1985), 225–49.
'Multiple Images: Jute Mill Strikes of 1929 and 1937 Seen Through Other's Eyes', Modern Asian Studies, 21:3 (1987), 547–83.
'Sahibs, Babus and Banias: Changes in Industrial Control in Eastern India, 1918–1950', Journal of Asian Studies, 48:2 (1989), 289–309.
Industry, Trade and Peasant Society: The Jute Economy of Eastern India, 1900–1947. Delhi, 1991.
Gourlay, S. 'Nationalists, Outsiders and the Labour Movement in Bengal During the NonCooperation Movement, 1919–1921', in Kapil Kumar (ed.), Congress and
Classes: Nationalism, Workers and Peasants, pp. 34–57.
Gray, R. The Aristocracy of Labour in Nineteenth Century Britain, c. 1850–1900. London, 1981.
Greenberg, M. British Trade and the Opening of China, 1800–1842. Cambridge, 1951.
Greenhough, P. 'Political Mobilization and the Underground Literature of the Quit India Movement, 1942–44', Modern Asian Studies, 17:3 (1983), 353–86.
Griffin, K. and J. Gurley. 'Radical Analyses of Imperialism, the Third World and the Transition to Socialism', Journal of Economic Literature, 23 (1985), 1089–
143.
Griffiths, P. To Guard My People: The History of the Indian Police. London, 1971.
Griffiths, Sir Percival. The History of the Indian Tea Industry. London, 1967.
Guha, R. 'On Some Aspects of the Historiography of Colonial India', in R. Guha (ed.), Subaltern Studies, vol. I, pp. 1–8.
Guha, R. (ed.), Subaltern Studies, 6 vols. Delhi, 1982–89.
Guha, S. The Agrarian Economy of the Bombay Deccan, 1818–1941. Delhi, 1986.
Guha Ray, S. 'Tramworkers of Calcutta: Some Reflections on their Unionisation and Political Experience', Social Scientist, 156 (May 1986), 15–32.
Gupta, A. The Police in British India, 1861–1947. Delhi, 1979.

Page 365
Gupta, P. S. 'Notes on the Origin and Structuring of the Industrial Labour Force in India – 1880 to 1920', in R. S. Sharma (ed. with V. Jha), Indian Society:
Historical Probings: In Memory of D. D. Kosambi, pp. 414–34. New Delhi, 1979.
Gurr, T. R. 'Historical Trends in Violent Crime: A Critical Review of the Evidence', Crime and Justice: An Annual Review of Research, 3 (1981), 295–353.
Habakkuk, H. J. 'The Historical Experience of Economic Development', in E. A. G. Robinson (ed.), Problems in Economic Development, pp. 112–38. London,
1965.
Habib, I. 'The Potentialities of Capitalistic Development in Mughal India', Journal of Economic History, 39:1 (1969), 32–78.
'The Technology and Economy of Mughal India', Indian Economic and Social History Review, 17:1 (1980), 1–34.
'Monetary Systems and Prices', in T. Raychaudhuri and I. Habib (eds.), Cambridge Economic History of India, vol. I, pp. 360–81. Cambridge 1982.
Hall, S., C. Critcher, T. Jefferson, J. Clarke and B. Roberts. Policing the Crisis: Mugging, the State and Law and Order. London, 1978.
Harcourt, M. 'Kisan Populism and Revolution in Rural India: The 1942 Disturbances in Bihar and East United Provinces', in D. A. Low (ed.), Congress and the Raj:
Facets of the Indian Struggle, 1917–1947, pp. 315–48.
Hardiman, D. Peasant Nationalists of Gujarat: Kheda District, 1917–1934. Delhi, 1981.
The Coming of the Devi. Delhi, 1987.
Hasan, A. 'The Silver Currency Output of the Mughal Empire and Prices in India in Sixteenth and Seventeenth Centuries', Indian Economic and Social History
Review, 6:1 (1969), 85–116.
Haynes, D. 'The Dynamics of Continuity in Indian Domestic Industry: Jari Manufacture in Surat, 1900–1947', Indian Economic and Social History Review, 23:2
(1986), 128–49.
Henningham, S. 'The Contribution of ''Limited Violence" to the Bihar Civil Disobedience Movement', South Asia, new series, 2:1 and 2 (1979), 60–77.
'Quit India in Bihar and the Eastern United Provinces: The Dual Revolt' in R. Guha (ed.), Subaltern Studies, vol. II, pp. 130–79.
Heston, A. and R. Summers. 'Comparative Indian Economic Growth: 1870 to 1970', American Economic Review, 70:2 (1980), 96–101.
Hirst, L. Fabian. The Conquest of Plague. A Study of the Evolution of Epidemiology. Oxford, 1953.
Hobsbawm, E. J. Primitive Rebels: Studies in Archaic Forms of Social Movement in the 19th and 20th Centuries. Manchester, 1959.
Labouring Men: Studies in the History of Labour. London, 1964.
Industry and Empire. London, 1968.
Hoselitz, B. E. and W. E. Moore (eds.). Industrialisation and Society. Paris, 1963.
Hunter, Sir W. W. Bombay, 1885 to 1890: A Study in Indian Administration. London, [1892?].
Hutchins, F. G. Spontaneous Revolution: The Quit India Movement. New Delhi, 1971.

Page 366
India's Revolution: Gandhi and the Quit India Movement. Cambridge, Mass., 1973.
Inden, R. Imagining India. Oxford, 1991.
Jagga, L. 'Colonial Railwaymen and British Rule: A Probe into Railway Labour Agitation in India, 1919–1922', in B. Chandra (ed.), The Indian Left: Critical
Appraisals, pp. 103–45. New Delhi, 1983.
Jalal, A. The Sole Spokesman: Jinnah, the Muslim League and the Demand for Pakistan. Cambridge, 1985.
James, R. C. 'Trade Union Democracy: Indian Textiles', The Western Political Quarterly, 11:3 (1958), 566–72.
'The Casual Labour Problem in Indian Manufacturing', The Quarterly Journal of Economics, 74:1 (February 1960), 100–16.
Jeffrey, K. and P. Hennessey. States of Emergency: British Governments and Strikebreaking since 1919. London, 1983.
Jeffrey, R. The Decline of Nayar Dominance. Brighton, 1976.
Johnson, G. Provincial Politics and Indian Nationalism: Bombay and the Indian National Congress, 1880–1915. Cambridge, 1973.
Jones, E. L. (ed.). Agricultural and Economic Growth in England, 1650–1815. London, 1967.
Joshi, C. 'Kanpur Textile Labour: Some Structural Features of Formative Years', Economic and Political Weekly, 16:44–6 (November 1981), special issue, 1823–
38.
'Bonds of Community, Ties of Religion: Kanpur Textile Workers in the Early Twentieth Century', Indian Economic and Social History Review, 22:3 (1985), pp.
251–80.
'Worker Protest, Managerial Authority and Labour Organization: Kanpur Textile Industry', Nehru Memorial Musem and Library, Occasional Papers on History
and Society, no. 27. Delhi, 1985.
Joshi, H. and V. Joshi, Surplus Labour and the City: A Study of Bombay. Delhi, 1976.
Joshi, N. M. The Trade Union Movement in India. Poona, 1927.
Urban Handicrafts of the Bombay Deccan. Poona, 1936.
Joshi, P. C. (ed.). Rebellion – 1857: A Symposium. Delhi, 1957.
Joyce, P. Work, Societry and Politics: The Culture of the Factory in Late Victorian England. London, 1980.
Kadam, Manohar. Narayan Meghaji Lokhande: Bharatiya Kamgar Calvalliche Janak. Bombay, 1995.
Karnik, V. B. Strikes in India. Bombay, 1967.
N. M. Joshi: Servant of India. Bombay, 1972.
Kaye, J. W. A History of the Sepoy War in India, 3 vols. London, 1864, 1870 and 1876.
Keer, D. Dr Ambedkar: Life and Mission. Bombay, 1954.
Kelly, J. D. '''Coolie" as a Labour Commodity: Race, Sex and European Dignity in Colonial Fiji', in E. V. Daniel, H. Bernstein and T. Brass (eds.), Plantations,
Proletarians and Peasants in Colonial Asia, pp. 246–67.
Kennedy, M. Notes on the Criminal Classes in the Bombay Presidency. Bombay, 1909.

Page 367
Kerr, C. 'Changing Social Structures', in W. E. Moore, and A. S. Feldman (eds.), Labour Commitment and Social Change in Developing Areas, pp. 348–59.
Kerr, C. and A. Siegel. 'The Structuring of the Labour Force in Industrial Society', Industrial and Labour Relations Review, 8:2 (1955), 151–68.
Kerr, C., F. H. Harbison, J. T. Dunlop and C. A. Myers. 'Industrialism and Industrial Man', International Labour Review, 82:3 (1960), 236–50.
Industrialism and Industrial Man: The Problems of Labour and Management in Economic Growth. London, 1962.
Kerr, I. J. Building the Railways of the Raj, 1850–1900. Delhi, 1995.
Kirk, R. and C. P. Simmons. 'Lancashire and the Equipping of Indian Cotton Mills: A Study of Textile Machinery and Supply, 1854–1939', in K. Ballhatchet and D.
Taylor (eds.), Changing South Asia: Economy and Society, pp. 169–81. London, 1984.
Klein, I. 'Malaria and Mortality in Bengal', Indian Economic and Social History Review, 9:2 (1972), 132–60.
'Death in India', Journal of Asian Studies, 32:4 (1973), 639–59.
'Urban Development and Death: Bombay City, 1870–1914', Modern Asian Studies, 20:4 (1986), 725–54.
'Plague, Policy and Popular Unrest in British India', Modern Asian Studies, 22:4 (1988), 723–55.
Kling, B. B. Partner in Empire: Dwarkanath Tagore and the Age of Enterprise in Eastern India. Calcutta, 1981.
Kooiman, D., 'Jobbers and the Emergence of Trade Unions in Bombay City', International Review of Social History, 22:3 (1977), 313–28.
Krishnamurty, J. 'Deindustrialization in Gangetic Bihar during the Nineteenth Century: Another Look at the Evidence', Indian Economic and Social History Review,
22:4 (1985), 399–416.
Kumar, D. (ed.). The Cambridge Economic History of India, vol. II: C. 1757–c. 1970. Cambridge, 1982.
Kumar, D. and J. Krishnamurthy. 'Regional and International Economic Disparities since the Industrial Revolution: The Indian Evidence', in P. Bairoch and M. Levy
Leboyer (eds.), Disparities in Economic Development since the Industrial Revolution, pp. 361–72. London, 1981.
Kumar, Kapil (ed.). Congress and Classes: Nationalism, Workers and Peasants New Delhi, 1988.
Kumar, Radha. 'Family and Factory: Women in the Bombay Cotton Textile Industry, 1919–1939', Indian Economic and Social History Review, 20:1 (1983), 81–
110.
Kumar, R. 'The Bombay Textile Strike, 1919', Indian Economic and Social History Review, 8:1 (1971), 1–29.
'From Swaraj to Purna Swaraj: Nationalist Politics in the City of Bombay 1920–32', in D. A. Low (ed.), Congress and the Raj: Facets of the Indian Struggle,
pp. 77–107.
Lakha, Salim. Capitalism and Class in Colonial India: The Case of Ahmedabad. New Delhi, 1988.
Lakshman, P. P. Congress and the Labour Movement in India. Congress Economic

Page 368
and Political Studies, no. 3, Economic and Political Research Department, AllIndia Congress Committee, Allahabad, 1947.
Lal, B. 'Kunti's Cry: Indentured Women on Fiji's Plantations', Indian Economic and Social History Review, 22:2 (1985), 55–71.
Landes, D. S. The Unbound Prometheus: Technological Change and Industrial Development in Western Europe from 1750 to the Present. Cambridge,
1969.
Lewis, W. A. 'Economic Development with Unlimited Supplies of Labour', The Manchester School of Economic and Social Studies, 22:2 (1954), 139–91.
Lovett, V. A History of the Indian Nationalist Movement. London, 1920.
Low, D. A. (ed.). Congress and the Raj: Facets of the Indian Struggle. London, 1977.
Macfarlane, A. The Justice and the Mare's Ale: Law and Disorder in Seventeenth Century England. Oxford, 1981.
Maine, H. S. Village Communities in the East and West. London, 1871.
Marglin, Stephen. 'What Do Bosses Do?' Review of Radical Political Economy, 6 (1974), 60–112.
Markovits, C. Indian Business and Nationalist Politics, 1931–39: The Indigenous Capitalist Class and the Rise of the Congress Party. Cambridge, 1985.
Masani, R. P. The Evolution of Local SelfGovernment in Bombay. Oxford, 1929.
Masselos, J. C. 'Power in the Bombay ''Moholla" 1904–15: An Initial Exploration into the World of the Indian Urban Muslim', South Asia, 6 (1976), 75–95.
'Change and Custom in the Format of the Bombay Mohurram During the Nineteenth and Twentieth Centuries', South Asia, new series, 5:2 (1982), 47–67.
Mathur, Y. B. Quit India Movement. Delhi, 1979.
Mazumdar, D. 'Labour Supply in Early Industrialization: The Case of the Bombay Textile Industry', Economic History Review, second series, 26:3 (1973), 477–96.
Mehta, M. J. The Ahmedabad Cotton Textile Industry: Genesis and Growth. Ahmedabad, 1982.
Mehta, M. M. Structure of Indian Industries. Bombay, 1955.
Mehta, S. D. The Cotton Mills of India, 1854–1954. Bombay, 1954.
Melling, J. 'NonCommissioned Officers: British Employers and their Supervisory Workers, 1880–1920', Social History, 5:2 (1980), 183–221.
Menon, D. Caste, Nationalism and Communism in South India: Malabar 1900–1948. Cambridge, 1994.
Metcalf, T. R. Land, Landlords and the British Raj: Northern India in the Nineteenth Century. Berkeley and Los Angeles, 1979.
Mills, I. D. 'Influenza in India During 1918–19', Indian Economic and Social History Review, 23:1 (1986), 1–40.
Mishra, S. C. 'Commercialization, Peasant Differentiation and Merchant Capital in Late Nineteenth Century Bombay and Punjab', Journal of Peasant Studies, 10:1
(1982), 3–51.
'Agricultural Trends in the Bombay Presidency, 1900–1920: The Illusion of Growth', Modern Asian Studies, 19:4 (1985), 733–59.
Misra, B. 'Factory Labour During the Early Years of Industrialisation: An Appraisal in the Light of the Indian Factory Commission 1890', Indian Economic and
Social History Review, 12:3 (1975), 203–28

Page 369
Mitra, C. 'Popular Uprising in 1942: The Case of Ballia' in G. Pandey (ed.), The Indian Nation in 1942, pp. 165–84.
Mitra, Ira. 'Growth of Trade Union Consciousness among Jute Mill Workers, 1920–40', Economic and Political Weekly, 16:44–6 (November 1981), special
number, 1839–48.
Mohapatra, P. P. 'Coolies and Colliers: A Study of the Agrarian Context of Labour Migration from Chotanagpur, 1880–1920', Studies in History, new series, 1:2
(1985), special issue, Essays in Agrarian History: India, 1850 to 1940, ed. S. Bhattacharya, 13–42.
Moore, R. J. The Crisis of Indian Unity, 1917–1940. Oxford, 1974.
Moore, W. E. and A. S. Feldman (eds.). Labour Commitment and Social Change in Developing Areas. New York, 1960.
Morgan, Jane. Conflict and Order: The Police and Labour Disputes in England and Wales, 1900–1939. Oxford, 1987.
Morris, M. D. 'The Labour Market in India', in W. E. Moore and A. S. Feldman (eds.), Labour Commitment and Social Change in Developing Areas, pp. 173–
200.
The Emergence of an Industrial Labour Force in India: A Study of the Bombay Cotton Mills, 1854–1947. Berkeley and Los Angeles, 1965.
'The Growth of Large Scale Industry to 1947', in D. Kumar (ed.), The Cambridge Economic History of India, vol. II: C. 1757–c. 1970, pp. 553–676.
MorrisJones, W. H. '''IfIt Be Real, What Does It Mean?": British Perceptions of the Indian National Congress', in R. Sisson and S. Wolpert (eds.), Congress and
Indian Nationalism: The PreIndependence Phase, pp. 90–118. Berkeley, 1988.
Mukherjee, A. 'Crime and Criminals in Nineteenth Century Bengal (1861–1904)', Indian Economic and Social History Review, 21:2 (1984), 153–83.
Munro, J. Forbes. Britain in Tropical Africa, 1880–1960: Economic Relationships and Impact. London, 1984.
Murphy, E. D. 'Class and Community in India: The Madras Labour Union, 1918–21', Indian Economic and Social History Review, 14:3 (1977), 292–321.
Unions in Conflict: A Comparative Study of Four South Indian Textile Centres, 1918–1939. New Delhi, 1981.
Myers, C. A. Labour Problems in the Industrialization of India. Cambridge, Mass., 1958.
Nehru, J. An Autobiography: With Musings on Recent Events in India. London, 1936.
Newman, R. 'Social Factors in the Recruitment of the Bombay Millhands', in K. N. Chaudhuri and C. J. Dewey (eds.), Economy and Society: Essays in Indian
Economic and Social History, pp. 277–95.
Workers and Unions in Bombay, 1918–29: A Study of Organisation in the Cotton Mills. Canberra, 1981.
Nigam, S. 'Disciplining and Policing the "Criminals by Birth", Part I: The Making of a Colonial Stereotype – the Criminal Tribes and Castes of North India', Indian
Economic and Social History Review, 27:2 (1990), 131–64.

Page 370
Nigam, S. 'Disciplining and Policing the ''Criminals by Birth", Part II: The Development of a Disciplinary System, 1871–1900', Indian Economic and Social History
Review, 27:3 (1990), 257–87.
Omvedt, G. 'Migration in Colonial India: The Articulation of Feudalism and Capitalism by the Colonial State', Journal of Peasant Studies, 7:2 (1980), 185–212.
'The Satara Prati Sarkar', in G. Pandey (ed.), The Indian Nation in 1942, pp. 223–62.
Omissi, D. The Sepoy and the Raj: The Indian Army, 1860–1940. London, 1994.
Page, D. Prelude to Partition: The Indian Muslims and the Imperial System of Control. Delhi, 1982.
Pandey, G. The Ascendancy of the Congress in Uttar Pradesh, 1926–1934: A Study in Imperfect Mobilization. Delhi, 1978.
'Economic Dislocation in NineteenthCentury Eastern Uttar Pradesh: Some Implications of the Decline of Artisanal Industry in Colonial India', in P. Robb (ed.),
Rural South Asia: Linkages, Change and Development, pp. 89–129. London, 1983.
'Congress and the Nation, 1917–1947', in R. Sisson and S. Wolpert (eds.), Congress and Indian Nationalism: The PreIndependence Phase, pp. 121–33.
Berkeley, 1988.
'The Revolt of August 1942 in Eastern UP and Bihar', in G. Pandey (ed.), The Indian Nation in 1942, pp. 123–64.
The Construction of Communalism in Colonial North India. Delhi, 1992.
Pandey, G. (ed.). The Indian Nation in 1942. Calcutta, 1988.
Pandey, S. M. 'Ideological Conflict in the Kanpur Trade Union Movement: 1934–1945', Indian Journal of Industrial Relations, 3:2 (1967), 243–68.
As Labour Organises: A Study of Unionism in the Kanpur Cotton Textile Industry. New Delhi, 1970.
Panikkar, K. N. Against Lord and State: Religion and Peasant Uprisings in Malabar, 1836–1921. Delhi, 1989.
Parvatibai Bhor. Eka Rannaraginichi Hakikat, as told to Padmakar Chitale. Bombay, 1977.
Patel, K. Rural Labour in Industrial Bombay. Bombay, 1963.
Patel, S. The Making of Industrial Relations: The Ahmedabad Textile Industry, 1918–1939. Delhi. 1987.
Patnaik, P. 'Imperialism and the Growth of Indian Capitalism', in R. Owen and R. Sutcliffe (eds.), Studies in the Theory of Imperialism, pp. 210–29. London, 1972.
Patnaik, Utsa (ed.). Agrarian Relations and Accumulation: The 'Mode of Production' Debate in India. Bombay, 1990.
Pearse, A. The Cotton Industry of India: Being the Report of a Journey to India. Manchester, 1930.
Perlin, E. 'Eyes Without Sight: Education and Millworkers in South India, 1939–1976', Indian Economic and Social History, 18:3 and 4 (1981), 263–86.
Perlin, F. 'Protoindustrialization and Precolonial South Asia' Past and Present, 97 (February 1983), 30–95.

Page 371
Pollard, S. Peaceful Conquest: The Industrialization of Europe, 1760–1970. Oxford, 1981.
Pollitzer, R. Plague. Geneva, 1954.
Pouchepadass, J. 'The Market for Agricultural Labour in Colonial North Bihar, 1860–1920', in M. Holmstrom (ed.), Work for Wages in South Asia, pp. 11–27.
Delhi, 1990.
Prakash, G. Bonded Histories: Genealogies of Labor Servitude in Colonial India. Cambridge, 1990.
Pryde, A. W. 'The Work of the Labour Officer', in C. Manshardt (ed.), Some Social Services of the Government of Bombay. Bombay, 1937.
Radhakrishna, M. 'The Criminal Tribes Act in the Madras Presidency: Implications for the Itinerant Trading Communities', Indian Economic and Social History
Review, 26:3 (1989), 269–95.
'Surveillance and Settlements under the Criminal Tribes Act in Madras', Indian Economic and Social History Review, 29:2 (1992), 171–98.
Raghunathji, K. The Hindu Temples of Bombay. Bombay, 1900.
Ramasamy, P. Plantation Labour, Unions, Capital and the State in Peninsular Malaysia. Kuala Lumpur, 1994.
Ramaswamy, E. A. The Worker and His Union: A Study in South India. New Delhi, 1977.
Rangarajan, M. 'Imperial Agendas and India's Forests: The Early History of Indian Forestry, 1800–1878', Indian Economic and Social History Review, 31:2
(1994), 147–67.
Ray, R. K. 'Masses in Politics: The NonCooperation Movement in Bengal, 1920–22', Indian Economic and Social History Review, 11:4 (1974), 343–410.
Industrialization in India: Growth and Conflict in the Private Corporate Sector, 1914–47. Delhi, 1979.
Urban Roots of Indian Nationalism: Pressure Groups and Conflict of Interests in Calcutta City Politics, 1875–1939. New Delhi, 1979.
'Pedhis and Mills: The Historical Integration of the Formal and Informal Sectors in the Economy of Ahmedabad, Indian Economic and Social History Review,
19:3 and 4 (1982), 387–96.
Social Conflict and Political Unrest in Bengal, 1875–1927. Delhi, 1986.
Reiner, R. The Politics of the Police. Brighton, 1985.
Reisner, I. M. and N. M. Goldberg (eds.), Tilak and the Struggle for Indian Freedom. New Delhi, 1966.
Revri, C. The Indian Trade Union Movement: An Outline History, 1880–1947. New Delhi, 1972.
Richards, J. F. 'The Indian Empire and Peasant Production of Opium in the Nineteenth Century', Modern Asian Studies, 15:1 (1981), 59–82.
Richards, J. F. 'Mughal State Finance and the Premodern World Economy', Comparative Studies in Society and History, 23:2 (1981), 285–308.
Robb, P. 'The Ordering of Rural India: The Policing of Nineteenth Century Bengal and Bihar', in D. M. Anderson and D. Killingray (eds.), Policing the Empire:
Government, Authority and Control, 1830–1940, pp. 126–50. Manchester, 1991.

Page 372
Robinson, F. C. R. Separatism Among Indian Muslims: The Politics of the United Provinces' Muslims, 1860–1923. Cambridge, 1974.
Robinson, R. 'Oxford in Imperial Historiography', in F. Madden and D. K. Fieldhouse (eds.), Oxford and the Idea of the Commonwealth: Essays Presented to
Sir Edgar Williams, pp. 30–48. London, 1982.
Rostow, W. W. The Stages of Economic Growth: A NonCommunist Manifesto. Cambridge, 1960.
Rowe, W. D. 'Caste, Kinship and Association in Urban India', in A. Southall (ed.), Urban Anthropology: Cross Cultural Studies of Urbanization, pp. 211–49.
New York, 1973.
Roy, T. Artisans and Industrialization: Indian Weaving in the Twentieth Century. Delhi, 1993.
Roy, T. (ed.). Cloth and Commerce: Textiles in Colonial India. New Delhi, 1996.
Russell, W. H. My Diary in India in the Year 1858–59, 2 vols. London, 1860.
Rutnagur, M. C. The Indian Textile Journal Directory of Indian Manufactories, 1894. Bombay, 1894.
Rutnagur, S. M. (ed.). Bombay Industries: The Cotton Mills – A Review of the Progress of the Textile Industry in Bombay from 1850 to 1926 and the Present
Constitution, Management and Financial Position of Spinning and Weaving Factories. Bombay, 1927.
Saha, P. History of the Working Class Movement in Bengal. New Delhi, 1978.
Said, E. Orientalism. New York, 1978.
Sandilands, J. 'The Health of the Bombay Worker', Labour Gazette, 1:2 (October 1921), 14–16.
Sarkar, S. 'Popular Movements and National Leadership, 1945–47', Economic and Political Weekly, 17:14, 15 and 16 (April 1982), annual number, 677–89.
'The Conditions and Nature of Subaltern Militancy: Bengal from Swadeshi to NonCooperation, c. 1905–1922', in R. Guha (ed.), Subaltern Studies, vol. III, pp.
271–320.
Sarkar, T. 'The First Phase of Civil Disobedience in Bengal, 1930–31', Indian Historical Review, 4:1 (1977), 75–95.
Bengal, 1928–34: The Politics of Protest. Delhi, 1987.
Saul, S. B. Studies in British Overseas Trade. Liverpool, 1960.
Sayer, D. 'British Reactions to the Amritsar Massacre, 1919–1920', Past and Present, 131 (May 1991), 130–64.
Schuster, George E. and Guy Wint. India and Democracy. London, 1941.
Scott, H. H. A History of Tropical Medicine, 2 vols. London, 1939.
Scott, J. C. Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven, 1985.
'Everyday Forms of Peasant Resistance' in J. C. Scott and B. J. T. Kerkvliet (eds.), Everyday Forms of Peasant Resistance in Southeast Asia, pp. 5–35.
Scott, J. C. and B. J. T. Kerkvliet (eds.). Everyday Forms of Peasant Resistance in SouthEast Asia. London, 1986.
Seal, A. The Emergence of Indian Nationalism: Competition and Collaboration in the Later Nineteenth Century. Cambridge, 1968.
Seers, D. 'The Birth, Life and Death of Development Economics', Development and Change, 10:4 (1979), 707–19.

Page 373
Sen, A. K. 'The Commodity Pattern of British Enterprise in Early Indian Industrialization, 1854–1914', in The Second International Conference of Economic
History, AixenProvence, 1962, vol. II: Middle Ages and Modern Times, pp. 781–808. Paris, 1962.
Sen, S. N. Eighteen Fifty Seven. New Delhi, 1957.
Seth, B. R. Labour in the Indian Coal Industry. Bombay, 1940.
Shah, K. T. and G. J. Bahadurji. Constitutions, Functions and Finance of Indian Municipalities. Bombay, 1925.
Sharpe, J. A. 'The History of Violence in England: Some Observations', Past and Present, 108 (1985), 206–15.
Shilts, Randy. And the Band Played On. Politics, People and the Aids Epidemic. London, 1988.
Shiva Rao, B. The Industrial Worker in India. London, 1939.
Shlomowitz, R. and L. Brennan. 'Mortality and Migrant Labour in Assam, 1865–1921', Indian Economic and Social History Review, 27:1 (1990), 85–110.
'Mortality and Migrant Labour en route to Assam, 1863–1924', Indian Economic and Social History Review, 27:3 (1990), 313–30.
Siddiqi, A. 'Money and Prices in the Earlier Stages of Empire', Indian Economic and Social History Review, 18:3 and 4 (1981), 231–62.
'The Business World of Jamsetjee Jejeebhoy', Indian Economic and Social History Review, 19:3 and 4 (1982), 301–24.
Simeon, D. The Politics of Labour under Late Colonialism: Workers, Unions and the State in Chota Nagpur, 1928–1939. New Delhi, 1995.
Simmons, C. P. 'Indigenous Enterprise in the Indian Coal Mining Industry, c. 1835–1939', Indian Economic and Social History Review, 13:2 (1976), 189–218.
'Recruiting and Organizing an Industrial Labour Force in Colonial India: The Case of the Coal Mining Industry c. 1880–1939', Indian Economic and Social
History Review, 13:4 (1976), 455–85.
'Vertical Integration and the Indian Steel Industry', Modern Asian Studies, 11:1 (1977), 127–48.
Singh, Moni. Life is a Struggle, trans. Mrs Karuna Banerjee. New Delhi, 1988.
Singha, Radhika. '''Providential" Circumstances: The Thuggee Campaign of the 1830s and Legal Innovation', Modern Asian Studies, 27:1 (1993), 83–146.
Slack, P. 'The Response to Plague in Early Modern England: Public Policies and Their Consequences', in J. Walter and R. Schofield (eds.), Famine, Disease and the
Social Order in Early Modern Society, pp. 167–87. Cambridge, 1989.
Smelser, N. J. Social Change and the Industrial Revolution. London, 1959.
Stedman Jones, G. Languages of Class: Studies in English Working Class History, 1832–1982. Cambridge, 1983.
Stokes, E. The English Utilitarians and India. Cambridge, 1959.
The Peasant and the Raj: Studies in Agrarian Society and Peasant Rebellion in Colonial India. Cambridge, 1978.
Stone, L. 'Interpersonal Violence in English Society, 1300–1980', Past and Present, 101 (1983), 22–33.
'A Rejoinder', Past and Present, 108 (1985), 206–15.

Page 374
Strachey, J. India. London, 1888.
Sutcliffe, R. 'Imperialism and Industrialisation in the Third World', in R. Owen and R. Sutcliffe (eds.), Studies in the Theory of Imperialism, pp. 171–92. London,
1972.
Temple, Richard. India in 1880. London, 1880.
Thomas, P. J. The Growth of Federal Finance in India. London, 1939.
Thorner, D. and A. Thorner. Land and Labour in India. Bombay, 1962.
Tilly, C. 'The Changing Place of Collective Violence', in M. Richter (ed.), Essays in Theory and History: Approaches to the Social Sciences, pp. 139–64.
Cambridge, Mass., 1970.
Tilly, C., Louise Tilly and Richard Tilly. The Rebellious Century, 1830–1930. London, 1975.
Timberg, T. A. The Marwaris: From Traders to Industrialists. New Delhi, 1978.
Tinker, H. A New System of Slavery: The Export of Indian Labour Overseas, 1830–1920. Oxford, 1974.
Tomlinson, B. R. 'India and the British Empire, 1880–1935', Indian Economic and Social History Review, 12:4 (1975), 339–80.
'India and the British Empire, 1935–1947', Indian Economic and Social History Review, 13:3 (1976), 331–52.
The Indian National Congress and the Raj: The Penultimate Phase, 1929–1942. London, 1977.
The Political Economy of the Raj: The Economics of Decolonization in India. London, 1979.
'Colonial Firms and the Decline of Colonialism in Eastern India, 1914–1947', Modern Asian Studies, 15:3 (1981) special issue, Power, Profit and Politics:
Essays on Imperialism, Nationalism and Change in Twentieth Century India, 1870–1940, ed. C. Baker, G. Johnson and A Seal, 455–86.
'Writing History Sideways: Lessons for Indian Economic Historians From Meiji Japan', Modern Asian Studies, 19:3 (1985), 669–98.
'British Business in India, 1860–1970', in R. P. T. DavenportHines and Geoffrey Jones (eds.), British Business in Asia since 1860. Cambridge, 1989.
The New Cambridge History of India, vol III:3: The Economy of Modern India, 1860–1970. Cambridge, 1993.
Townshend, P. The British Campaign in Ireland, 1919–1921: The Development of Political and Military Policies. Oxford, 1975.
Trebilcock, R. C. The Industrialization of the Continental Powers, 1750–1914. New York, 1981.
Tripathi, D. The Dynamics of a Tradition: Kasturbhai Lalbhai and His Entrepreneurship. New Delhi, 1981.
Tripathi, D. and M. J. Mehta. 'The Nagarsheth of Ahmedabad: The History of an Urban Institution in a Gujarati City', Proceedings of the Indian History Congress
(1978?), 481–96.
Twomey, M. 'Employment in Nineteenth Century Indian Textiles', Explorations in Economic History, 20:1 (1983), 37–57.
Vicziany, A. M. 'Bombay Merchants and Structural Changes in the Export Community, 1850 to 1880', in K. N. Chaudhuri and C. J. Dewey (eds.), Economy and
Society: Essays in Indian Economic and Social History, pp. 163–96.

Page 375
'The DeIndustrialization of India in the Nineteenth Century: A Methological Critique of Amiya Kumar Bagchi', Indian Economic and Social History Review,
16:2 (1979), 105–46.
Visaria L. and P. Visaria, 'Population, 1757–1947' in D. Kumar (ed.), The Cambridge Economic History of India, vol. II: 1757–1970, pp. 463–532.
Wagle, D. M. 'Imperial Preference and the Indian Steel Industry', Economic History Review, 34:1 (1981), 120–31.
Wallace, D. R. The Romance of Jute. Calcutta, 1909.
Wallerstein, I. 'The Rise and Future Demise of the World System: Concepts for Comparative Analysis', Comparative Studies in Society and History, 16:4 (1974),
387–415.
The Capitalist World Economy. Cambridge, 1979.
'Incorporation of the Indian Subcontinent into the Capitalist World Economy', Economic and Political Weekly, 21:4 (25 January 1986), Review of Political
Economy, PE28 to PE39.
Warboys, M. 'The Emergence of Tropical Medicine: A Study in the Establishment of a Scientific Speciality', in G. Lemaine, R. Macleod, M. Mulkay and P. Weingart
(eds.), Perspectives on the Emergence of Scientific Disciplines, pp. 75–98. The Hague, 1976.
Warren, Bill, Imperialism – The Pioneer of Capitalism. London, 1980.
Washbrook, D. A. The Emergence of Provincial Politics: Madras Presidency, 1870–1920. Cambridge, 1976.
'Law, State and Agrarian Society in Colonial India', Modern Asian Studies, 15:3 (1981), special issue, Power, Profit and Politics: Essays on Imperialism,
Nationalism and Change in Twentieth Century India, 1870–1940, ed. C. Baker, G Johnson and A. Seal, 649–721.
'Progress and Problems: South Asian Economic and Social History, c. 1720–1860', Modern Asian Studies, 22:1 (1988), 57–96.
'South Asia, the World System and World Capitalism', Journal of Asian Studies, 49:3 (1990), 479–508.
'Land and Labour in Late Eighteenth Century South India: The Golden Age of the Pariah', in P. Robb (ed.), Dalit Movements and the Meanings of Labour, pp.
68–86. Delhi, 1993.
Weber, Max. The Religion of India, trans. H. H. Gerth and D. Martindale. Glencoe, Ill., 1958.
Williams, R. Keywords: A Vocabulary of Culture and Society. London, 1976.
Wood, C. 'Peasant Revolt: An Interpretation of Moplah Violence in the Nineteenth and Twentieth Centuries', in C. J. Dewey and A. G. Hopkins (eds.), The Imperial
Impact: Studies in the Economic History of Africa and India, pp. 132–51.
Yang, A. 'The Agrarian Origins of Crime: A Study of Riots in Saràn District, India, 1886–1920', Journal of Social History, 13:2 (1979), 289–306.
Yang, A. (ed.). Crime and Criminality in British India. Arizona, 1985.
Yolland, Z. Boxwallahs: The British in Cawnpore, 1857–1901. Norwich, 1994.
Zelliot, E. 'Congress and the Untouchables', in R. Sisson and S. Wolpert (eds.), Congress and Indian Nationalism. The PreIndependence Phase, pp. 182–97.
Berkeley and Los Angeles, 1988.

Page 376
Unpublished Papers and Dissertations
Basu, S. 'Workers' Politics in Bengal, 1890–1929: MillTowns, Strikes and Nationalist Agitations'. Unpublished Ph.D. thesis, University of Cambridge, 1994.
Dewey, C. J. 'The Agricultural Output of an Indian Province: The Punjab, 1870–1940'. Paper read to the Economic History Seminar, Institute of Commonwealth
Studies, 30 April 1973. Centre of South Asian Studies Library, Cambridge.
Ghosh, R. K. 'A Study of the Labour Movement in the Jharia Coal Field, 1900–1977'. Unpublished Ph.D. thesis, University of Calcutta, 1992.
Gooptu, N. 'The Political Culture of the Urban Poor in North India, 1920–47'. Unpublished Ph.D. thesis, University of Cambridge, 1991.
Haynes, D. 'From Merchant Capital to Weavers' Capital: The Slow Transformation of Artisanal Production in the Bombay Presidency, 1900–1950'.
Narain, N. 'Cooption and Control: The Role of the Colonial Army in India, 1918–1947'. Unpublished Ph.D. thesis, University of Cambridge, 1993.
Pradhan, G. R. 'The Untouchable Workers of Bombay City'. Unpublished M.A. thesis, University of Bombay, 1936.
Sen, Samita. 'Women Workers in the Bengal Jute Industry, 1890–1940: Migration, Motherhood and Militancy'. Unpublished Ph.D. thesis, University of Cambridge,
1992.
Vicziany, A. M. 'The Cotton Trade and the Commercial Development of Bombay, 1853–1875'. Unpublished Ph.D. thesis, University of London, 1975.

Page 377
Index
absenteeism 343
Acland, George 48
agrarian economy 333
in colonial discourse 23, 24
Ahmedabad
capitallabour relations 277, 282, 283, 284, 286
cotton textile industry 53–5, 56–7, 58, 59, 64, 302–6
strikes 295, 326
trade unions 97
workforce 76–7
workingclass nationalism 309
Ahmedabad Millowners' Association 87, 303, 304, 306
Ahmedabad Textile Labour Association (ATLA) 77, 81, 84, 86–7, 96, 292
AIDS (Acquired Immune Deficiency Syndrome) 18, 265
AllIndia Trade Union Congress (AITUC) 85–6, 95, 267, 281
Alwe, Arjun 134
Amin, S. 272, 273
Amritsar massacre (1919) 186, 218–19, 221, 230
army
and the plague epidemic 241–2, 249–50, 253
and the police 219–25, 229, 231–3
Arnold, David 146, 251, 252
Asavale, R. S. 160
Assam, workingclass nationalism 311–12
ATLA see Ahmedabad Textile Labour Association
'bad characters'/hooligans
and the police 204–7, 209
and workingclass violence 160–3, 174–5
Bajirao, Jaysingrao 124
Baker, Christopher 227
study of Tamil Nadu handloomweaving 11, 46–7
Baker, Surgeon Major 257
Bakhale, R. R. 92
Balubhai Desai 115
Baptista, Joseph 282
Barry, Surgeon Major 257
Bengal
military initiatives against terrorism 232
Bengal Iron and Steel Company 49
Bengal Iron Works Company 49
Bharucha, Burjorji 313
Bhumgara, M. S. 120
Bingley, A. H. 251
Birla, G. D. 285, 302
blackleg labour 92, 93, 98, 119, 120, 121–6, 135, 136
clashes between strikers and 149
and neighbourhood leaders 193, 195
Blainey, Dr Thomas 254
boarding houses, and jobbers 106
Bole, S. K. 116–17, 293, 296
Bombay
cotton textile industry 51, 53, 54, 56, 58, 59, 60–2
and the labour force 68, 76
mill committees 131–3
neighbourhoods 102–4, 105
and the Northrop (automatic) loom 64–6
and workers' politics 100–42
and world capitalism 330, 335, 336
see also Sassoon cotton mills
gymnasiums (akhadas) 112–13, 114–16
Home Rule League 295, 296
and industrialization 40
Kamgar Hitwardhak Sabha 77
Municipal Corporation 84

Page 378
(cont.)
Bombay
plague epidemic 234–43, 247, 248, 256, 259, 262–3, 264, 265
police and public order 180–233
riots
Borkar riot (1928) 137–40
communal riots 141, 150, 160, 162, 172–5
plague 248, 251
Tilak riots (1908) 153–5
strikes 75, 76, 81, 90, 110–11, 119–27, 144, 192, 198–9, 208, 211
trade unions 82, 84, 86, 96, 117–42
urban policing 15
workers' politics 8, 13, 100–42
workingclass violence 156–79
workingclass voters 297–8
see also Girni Kamgar Union
Bombay Labour Group 130
Bombay Millowners' Association 119, 124
and the Girni Kamgar Union 127
and workingclass violence 168
Bombay Riots Inquiry Committee (1929) 163, 173, 197
Bombay Textile Labour Union (BTLU) 78, 86, 95, 116–17, 129, 130, 134, 293
Borkar, Keshav dada 78, 112, 123, 137, 138, 193, 314
Bose, S. C. 277, 288, 292, 302
Brenner, R. 36
Britain
cotton textile industry 56, 57
Industrial Revolution 32–4, 39, 40
plague 235
police force 225
political culture 230–1
strikes 231
brothelkeepers 108
and the police 200–4
bubonic plague see plague epidemic
Budansab, Fakrudin 262
business failures, and industrialization 48–51
Calcutta
strikes 75, 90, 145
trade unions 82, 83, 84, 88
workforce characteristics 75–6
see also jute industry
capital
Congress links with 277, 278
and the cotton textile industry 54–5, 56–6, 60, 62–3
expatriate 330, 335
and industrialization 38, 49–50
capital accumulation, and economic development 33
capital–labour relationships
and the Congress 281–7
and world capitalism 328–50
capitalism 3, 20–1
and the Marxist theory of industrialization 35–7
violence in transition to 177
see also world capitalism
caste
and class 141–2
in colonial discourse 23–4, 26, 27, 28
and criminalization 226, 228, 229
and independence 321
and jobbers 106
and the plague epidemic 252, 256
and policing 16
and politics 276
and skilled labour 26
and trade unions 77, 78
and workingclass nationalism 315, 316
and workingclass neighbourhoods 107, 191, 195
casual poor 7, 337
in Britain 231
and the police 206
and violence 14, 15, 177
casual workers 26, 337, 340
censorship, and the police 207–8
Chakrabarti, Byomkesh 295
Chakrabarty, D. 145
Chakraborty, Gopen 88
chawls (tenements)
in Bombay 103
committees 106, 116, 133, 191
and jobbers 105, 106
and strikes 125
Chetty, Chelvapathy 293
China, and the cotton textile industry 56, 57, 60–1
cholera, and the plague epidemic 244, 245
Chotalal, Ranchodlal 57
Chowdhury, K. C. Roy 297
civil disobedience movement 268, 279, 285, 318
and the police 204, 212–13, 214, 217

Page 379
and trade unions 94
and workingclass nationalism 311, 314
class
and caste 141–2
and the Congress 270, 271, 276–7, 279, 289, 293
and ethnicity 350
and Gandhi 285
and the industrial labour force 337
and nationalism 324–5
and the neighbourhood 142
power relations and political conflict 2
see also working classes
class consciousness
and the formation of the working classes 8–9
and the history of the working classes 6
and industrial disputes 127
and industrialization 12, 68, 348
and power relations 13–14
and trade unions 75, 98, 99
and the world economy 348–50
coal mining
and the capital market 44, 47–8
entrepreneurs 48, 50
jobbers 344
labour force 340
Cobb, Richard 153
Coimbatore
cotton textile industry 60
strikes 75
colonial discourse 21, 22
and crime 229
on differentiation in India 290
on labour 23–9
and the mob 216
and the police 222–5, 229–30
on workingclass violence 14–15
colonial rule
and labour representatives 297–8
colonial state
and the Congress 276, 277–8, 280, 299, 303, 317–19
and industrial disputes 92–4
and military force 231–3
and neighbourhood leaders 192
and the police 16, 180–1, 183, 217, 219, 228
and political surveillance 211–12
and workingclass violence 150, 166–7, 172, 175–6, 178–9
see also Government of India; state intervention
communists
political surveillance of 210–12
trade unions 77–8, 86, 96, 127–31, 133–4, 301, 319
see also Girni Kamgar Union
and workingclass violence 157–8, 164–9, 172, 173
conflict
social conflict and power relations 2–3
see also riots; violence
Congress see Indian National Congress
contracts (labour), and colonial discourse 26, 27–8
cotton textile industry 52–66
Ahmedabad 53–5, 56–7, 58, 59, 302–6
Bombay see under Bombay
capital 54–5, 56–6, 60, 62–3
Coimbatore 60
early attempts to establish 49, 50
entrepreneurial strategies 52–5
growth in the 1930s 58–61
and the Northrop (automatic) loom 64–6
rationalization schemes 61–3, 64, 100
strikes 61, 63–4, 68, 100–1
and trade unions 90
see also handloomweaving industry; millowners
Cox, Edmund 185, 187, 205
crime 2, 157
and 'bad characters'/hooligans 204–7
and colonial discourse 229
and 'dacoity' 226
and the neighbourhood 191–5, 227–8
and the plague epidemic 259–60
and the police 182, 183, 226
reported and recorded 189–91, 228
and 'thuggee' 226, 229
see also police
'criminal' violence 149
Cripps, Sir Stafford 268
crowd control, and the police 212–16, 228–30
'crowd' violence 149
culture
and imperial expansion 20
and industrialization 38–9
'traditional Indian', in colonial discourse 28
see also popular culture
Curry, J. C. 217–18

Page 380
dadas (neighbourhood leaders)
in Bombay 108, 111–13, 115, 116, 117, 121, 195
and dispute settlement 191
and the police 204
Dange, S. A. 121, 131, 132, 136, 158
Das, C. R. 84, 312
Das, Veena 145–6
dependency theory, and capitalist development 36
Desai Balubhai 112
Devi, Santosh Kumari 294, 296
Dhaku Janu Lad 113
Dharavi Tannery and Leather Workers' Union 199
Dinshaw, F. E. 44
discourse
and power 14
and the plague epidemic 17
see also colonial discourse; public discourse
diseases, and mortality 234
Dyer, General 218–19, 230
East India Company 48
economic depression (1930s), and world capitalism 345
economic development
and dependency theory 36
and industrialization 32–3, 39–40, 43, 44, 69–72
preconditions of 39–40, 43, 69
and trade unions 74
economy see agrarian economy; capitalist world economy
Eden Commission (1879) 219, 220
Edwardes, S. M. 154–5, 204
elections
workingclass voters 297–8
electoral system, and independence 321–3
employers
and characteristics of the workforce 75–6
and the Congress 300
Gandhi on workers and 283–5
and jobbers 342
and trade unions 80–4, 85, 88, 89–90, 98, 99, 129
see also millowners
England
violence in early modern 148–9
see also Britain
entrepreneurs
and industrialization 40–1, 43, 48–51
ethnicity
and class 350
and nationalism 290
ethnonational consciousness 350
Eurocentricity 21
factory system
and entrepreneurs 52
and the formation of the working classes 8
and industrialization 34, 37, 48
and largescale industry 44
Favel, Inspector 200–4
Fawcett Committee 169
festivals, Bombay workingclass neighbourhoods 109–10
Finckelstein, Maurice 201–2, 204
First World War 135
Fooks, Mary 203
Foucault, M. 21, 28
Fritza (brothelkeeper) 202–3
functionalism 6
and industrialization 11, 30, 31–5, 37
Gandhi, M. K. 18–19, 86, 97, 210, 216, 266, 267, 277
and capital–labour partnerships 282–7
on independence 289
and labour disputes 292, 302, 303, 305, 306, 326
and swaraj 290–1
gangs 206–7
Gell, H. G. 153–4
Ghori village, and the plague epidemic 261, 262
Ginwalla, F. J. 85
Girdhardas, Seth Maganlal 304
Giri, V. V. 95, 292, 293
Girni Kamgar Mahamandal (GKM) 130, 134
and the Borkar riot (1928) 137
Girni Kamgar Union (GKU) 15, 75, 81, 87–8, 90, 91
and the Borkar riot (1928) 137, 138, 140
and dadas 121
and government legislation 93, 94
and the neighbourhood 104, 126, 131, 132, 133
and the police 211
and strikes 121, 127
and workingclass nationalism 314

Page 381
and workingclass violence 150, 157, 158, 164, 168, 169, 170
Gordon, D. D. 144
Goswami, Dharani 88
Government of Bombay
and the plague epidemic 247, 251, 259–60
and the police 186, 188–90, 205, 211, 215, 223
and political surveillance 209
and workingclass violence 157, 162–3, 165, 166–9, 170–1
Government of India
and the police 221, 224
and the South India Railway strike (1928) 152
and trade unions 84, 94, 135–6
and workingclass violence 164–9
see also colonial state; state intervention
Government of India Act (1935) 322
grain dealers
Bombay workingclass neighbourhoods 108, 109, 111
as neighbourhood leaders 191, 193
and the workforce 67
Guha, Ranajit 147
Gurkhas, and protection rackets 195
gymnasiums (akhadas), in Bombay 112–13, 114–16, 193
Habib, Irfan 70
Haffkine, M. 255
Haldar, S. N. 295
handicraft industries 45–7
handloomweaving industry 11–12, 45–7
Ahmedabad 56
Harvey mills, and the Madurai Congress Committee 310–11
Hashim Dada of Nagpada 247
Heston, A. 70
Hindus
and Muslims 192, 195, 213–14
'Pardeshi'
in the Bombay police force 186–7
and protection rackets 195
and picketing of liquor shops 312
and workingclass nationalism 316, 317, 325
and workingclass violence 159, 173, 174
history, and the working classes 2–7
Hobsbawm, Eric 39
Home Rule League 83
housing
Bombay policemen 188
Bombay workingclass neighbourhoods 107
overcrowding 103
see also chawls
Hunter Committee 218
Hurst, A. R. Burnett 105, 111
indentured servitude 25
independence, and the Congress 320–4
Indian National Congress
and capital–labour partnerships 281–7
and class 270, 271, 276–7, 279, 289, 293
and the colonial state 276, 277–8, 280, 299, 303, 317–19
and communists 164
and the electoral system 321–3
and independence 320–4
and industrial relations 300–7
and labour leaders 298–9, 313–14
and political mobilization 270, 326
and popular culture 271–3
as representing the nation 278–81, 323–4
and social policies 291–2, 293–4
and swaraj 288–91
and trade unions 19, 77–8, 91, 96–7, 267–8, 292, 297, 301–2, 306, 320
and the working classes 19, 266–73, 274–81, 291–326
Indian Plague Commission 244–5
Indian Statutory Commission 223
Indian Trades Union Federation 95
Industrial Disputes Act 305
Industrial Disputes Committee 75
industrial revolution in Britain 32–4, 39, 40
industrialization 2, 10–12, 30–73
and caste 141
and colonial rule 69, 71–3
diffusion of 40–1
and entrepreneurs 40–1, 43, 48–51
failures 48–51
and functionalism 11, 30, 31–5, 37
and the history of the working classes 6, 8

Page 382
(cont.)
industrialization
and labour 38, 66–8, 69, 71, 336–41
and largescale production 43–5, 48
and Marxism 11, 30, 35–8, 41
and preindustrial societies 34–5
and trade unions 19, 35, 74–5, 79–80, 97–8
and violence by workers 146
Western model of 30, 31, 32–4, 38, 40, 42, 79
Whig historiography of 42
and the workforce 274–5
iron industry, early attempts to launch 48–9, 51
Iyengar, N. S. Ramaswamy 293
Jadhav, Parashuram 104
Jamshedpur strike (1928) 277, 302
Jatadhari Babu (K. C. Mitra) 88
Jeeva, Govind 255
'Jenny's gang' 206
Jhabvala, S. H. 85, 104
jobbers 341–5
in Bombay workingclass neighbourhoods 105–7, 108–9
and colonial discourse 25–6
and dadas 113
and gymnasiums 115–16
and the labour force 67–8, 76, 105–6
and mill committees 132–3
and millowners 105, 106–7
as neighbourhood leaders 195
and strikebreaking 107, 118, 121, 124, 141
and strikes 81, 125
and trade unions 79, 80, 82, 83, 126, 128, 129
and workers' politics 101
Joglekar, K. N. 308
Joshi, Chitra 145
Joshi, N. M. 78, 85–6, 95, 125, 130, 293, 296, 297
Joshi, P. C. 78
Joshi, V. H. 112
jute industry
business failures 48, 50
entrepreneurs 50–1, 335
handicraft sector 47
labour force 76, 77
rationalization schemes 288
strikes 75, 90, 145
trade unions 82, 83, 301
and workingclass nationalism 311–12
Kalyansundaram, Tiru Vi 293
Kamgar Hitwardhak Sabha 129, 293, 296
Kandalkar, G. L. 133
Kanpur
communist trade unions 301
violence by workers 145
Kanpur Mazdoor Sabha (KMS) 78
Karad, plague epidemic 259, 262
Karnataka, police in 184
Karnik, V. B. 113, 116, 121
Kasale, Govind 138
Kennedy, Hartley 195–6
kinship
and jobbers 106
and the police 227
and policing 16
and workingclass neighbourhoods 107, 191
Kitchener, Lord 226
Koilpatti, trade unions 82
Konkan, police in 184
Kubes, Milton 80–1, 118, 193–4
Kulkarni, D. B. 95, 267
Kulkarni, Shripat Gopal 249–50
Kumar, Kanchan 209–10
labour
capital–labour partnerships, and the Congress 281–7
colonial discourse about 23–9
Indian labour as premodern 26, 27–8
and industrialization 38, 66–8, 69, 71, 336–41
see also workforce
labour movement, history 4–5, 6
land revenue 333–4
Landes, David 33–4, 37
landlords
jobbers as 106
largescale industry 43–5, 48
and trade unions 79
law enforcement methods 14
Laxman, Tukaram 111
legislation, and trade unions 93–4
Lenin, V. I. 266
Lewis, W. A. 33

Page 383
liquor shops, picketing of 312–13, 315
living conditions
local government, and the police 188–9
Madras
municipal elections 298
trade unions 85
violent demonstrations 145
Madras Labour Union 298
Madurai, trade unions 82
workingclass nationalism 310–11
Mahars, in the Bombay workforce 141
Majid, Abdul 134
Mallik, Surendra Nath 84
Maloney, T. 63
Marathas
in the Bombay police force 186, 187
in the Bombay workforce 141
Markovitch, Mrs 202, 203
Marxism 6, 148
and industrialization 11, 30, 35–8, 41
Mapillas, suppression of the 231–2
material world 21, 22–3, 29
Meerut Conspiracy Case 15, 82, 85, 157, 169, 172, 194, 211, 314
Meyer, D. 203
migrant workers 337–8, 339–41, 345–6, 347
military see army
Miller, John 'the Mahatma' 308–9
millowners
and capitallabour relations 286
and informal neighbourhood leaders 193
and jobbers 105, 106–7
rationalization schemes 61–3, 64, 100, 131, 288, 318
and the state 135, 136
and strikes 124
and trade unions 86–7, 117–19
and workingclass violence 159, 164–5, 167, 172, 175
Millowners' Association 63, 80, 87, 91, 128, 133, 136, 282
Ahmedabad 87, 303, 304, 306
and workingclass violence 169, 170
see also Bombay Millowners' Association
Mir Abdul Ali 196
modernization theory
and Marxist theories of industrialization 37
and world capitalism 331
Mohammedan Association 77
Mohurram festival 109–10, 114
toli bands 160
Mohurram riots (1908) 192, 193
moneylenders
Bombay workingclass neighbourhoods 108, 110–11
jobbers as 106, 344
and the police 200
Moni Singh 88–9
Morris, M. D. 41, 45, 50–1, 144
Moses, M. N. 199–200
Munawar, Syed 115, 128, 161
municipal representation, trade unions 84–5
Muslim League 323
Muslims
in the Ahmedabad mills 304, 305
as blackleg labour 213–14
in the Bombay police force 187
in the Bombay workforce 141
and the Congress 278, 279, 280
dadas 115
Mohurram riots 192
neighbourhood leaders 192
and picketing of liquor shops 312
religious festivals 109–10, 115
and workingclass nationalism 313, 315, 316, 325
and workingclass violence 159, 173, 174
Mutiny (1857) 231
Nagpur
strikes 75
workforce 77
Naidu, Ramanujulu 293
Naidu, Varadarajulu 311
Nanavaty, H. D. 160
Nanda, Gulzarilal 292–3, 303, 306
Napier, Sir Charles 184
Naranji, Lalji 161
Nariman, K. F. 125, 141, 313
nationalism 2, 18–19, 266–326
and capitalism 333
and class 324–5
and imperialism 4

Page 384
(cont.)
nationalism
and the police 228
rhetoric of, and the working classes 281–91
workingclass 14, 18–19, 307–17
see also Indian National Congress
Nehru, Jawaharlal 95, 267, 280, 317
neighbourhoods
Bombay 102–44
and the police 181, 196–8, 227–8
and class consciousness 8–9
informal leaders 191–5
policing 16
social connections 67, 68
and trade unions 78–9
workingclass in Bombay 13
and workplace connections 8, 77, 102
Newman, Richard 144
Nimbkar, R. S. 122, 138
noncooperation movement 267, 268, 279
and industrial action 295–6
NorthWestern Provinces, and the plague epidemic 264
Orientalism 1, 4, 6, 22
Origins of Industrial Capitalism (Chandavarkar) 7
Pandey, Gyanendra 145
Parsis
entrepreneurs 335
and liquor shops 313
and Muslims 313, 315
Parulekar, S. V. 199
Patel, Vallabhai 210, 267, 314
Pathans
moneylenders 110–11, 194
and protection rackets 194, 204
and workingclass violence 168–9, 173
Patil, S. K. 123, 314
patronage, and trade unions 88, 91, 140
Pendse, Lalji 122
piecegoods dealers, and the Ahmedabad cotton textile industry 55
Pitambar, Mancharam (Barny) 201–2
plague epidemic 14, 17–18, 234–65
and the army 241–2, 249–50, 253
Bombay 234–43, 247, 248, 256, 259, 262–3, 264, 265
and crime 259–60
deaths 234–5, 237, 238, 240, 256
domestic care of the sick 257–8
and Europeans 239, 242
evacuation of towns and villages 259
flight of the population 247–8
and hospitalization 250, 252, 254, 256–7, 258
and living conditions 262–3
and medical science 243–6
and plague doctors 260
and the poor 239, 261
riots 235, 239, 246–7, 248, 251, 263
rumours 251, 252–3
sanitary measures 238, 241, 245
and segregation 238, 242, 250, 252, 258, 260
surveillance at ports and railways 238, 248, 261
and Western medicine 235, 254–6, 258
pleaders' offices, Bombay workingclass neighbourhoods 104
police 15–17
and the Amritsar massacre (1919) 186, 218–19, 221
arms 186
and the army 219–25, 229, 231
and 'bad characters'/hooligans 204–7, 209
and the Borkar riots (1928) 137–8, 139–40
in Britain 225
costs and financing 184, 188–9
and crowd control 212–16, 228–30
detectives 195–6, 208
efficiency 188
'Europeanization' 221–2
and industrial disputes 93, 119, 134–5, 136–7, 198–200
methods of policing 14
and the neighbourhood 181, 196–8, 227–8
overworking of 185
physical condition of recruits 185
and power relations 16
and prostitutes 200–3
and public order in Bombay 180–233
recruitment by geographical area 186–7
and reported stolen property 188
rural areas 184–5, 226–7
turnover 187–8
wages 185, 188
and workingclass violence 154, 161–2, 163, 165, 168, 169, 170, 171, 172, 175

Page 385
see also crime
political conflict, and social relations 2
political culture, Britain 230–1
political mobilization, and the Congress 270, 326
political programmes, effectiveness of 273–4
political surveillance, and the police 207–12
Poona, plague epidemic 248, 249, 259, 264
Poona Plague Committee 250, 252
popular culture 2, 19–20
and class consciousness 9
and nationalism 324
poverty, and the plague epidemic 239, 261
power
and discourse 14
power relations
and social conflict 2–3
and violence 178
workplace and neighbourhood 9
preindustrial societies 34–5, 79
prostitutes, and the police 200–3
public discourse, and social groups 18
Punjab
Rowlatt satyagraha 231
'putting out' system
combined with factory system 44
handloomweaving 46
Quit India movement 319–20
and military force 232, 233
and the workers 268–9
Rahimtoola, Sir Ibrahim 288
Rai, Lala Lajpat 281
railways
labour force 339–40
South India Railway strike (1928) 151–2
strikes 75
workers and racial conflict 97
and workingclass nationalism 307–9, 312
Ramaswamy, N. G. 81
Ramji, Sir Manmohandas 132
Rand, W. C. 237, 243, 247, 248
Rane, Babaji 80–1
Rao, B. Shiva 95, 293, 296
Rao, Narayan 116
Rashtriya Mill Mazdoor Sangh 94
Ray, R. K. 45
Reade, W. L. 243
Red Flag Union 118, 123
Religion of India (Weber) 27
religious festivals
Bombay workingclass neighbourhoods 109–10, 114
and 'roughs' 160
repression
and colonial rule 233
legitimizing measures of 28
riots
communal 141, 150, 160, 162, 172–5
and the police 208, 212, 213–14
and workingclass nationalism 313, 315–16
Mohurram riots (1908) 192, 193
plague 235, 239, 246–7, 248, 251, 263
and the police 208, 226–7
Tilak riots (1908) 153–5, 197
Roebuck, J. H. 158–9
rural migrants 7, 337–8, 339–41, 345–6
and industrialization 66–7, 68
in the labour force 76, 143
and the police 206
and violence 145–7, 160, 161, 163, 177
Ruskin, John 287
Sabbaji Rao 196
Said, Edward 21, 22, 28
Saklatwalla, S. D. 124, 125
Samad Khan 197–8
Sangh, Mill Mazdoor 304
Sapru, Tej Bahadur 322
Sarabhai, Ambalal 106, 280, 282
Sarabhai, Anasuyabehn 293, 303
sardars 76, 344
Sarma, P. K. K. 85
Sassoon cotton mills 60, 63, 306
and the Borkar riot (1928) 138
and trade unions 80–1, 118, 123
satyagraha 285, 290, 302
Savarkar, Dr N. D. 116
Schlamp, Sophie 203
Sen Gupta, J. M. 312
Shepherd, W. D. 257
Shilts, Randy 265

Page 386
Shiva Rao, B. 82
Sholapur, industrial action 75, 77, 78, 301
shopkeepers
Bombay workingclass neighbourhoods, 109–11
shroffs, and the Ahmedabad cotton textile industry 54, 55
Simon Commission 152, 164
Simond, P. L. 244–5
Singh, Bhagat 316
Sircar, N. C. 44
skilled labour 26
Snow, P. C. 252
social conflict, and power relations 2–3
social groups
defining 14
and public discourse 18
Social Service League 117–18, 129, 293
socialism
South Gujerat, police in 184
state intervention
in Bombay workingclass neighbourhoods 134–6
and communist trade unions 127
and the jobber system 67–8
and workingclass violence 163–72, 175–6
steel industry, early attempts to launch 48–9, 51
StevensonMoore, C. J. 197
Stone, Lawrence 148
Stones, Fred 63, 306
Strachey, John 278
street life, Bombay workingclass neighbourhoods 103–4
strikebreaking
blacklegging 92, 93, 98, 119, 120, 121–6, 135, 136
and violence 149
and dadas 111, 116, 121
and gymnasiums 115–16
and jobbers 107, 118, 121, 124, 141
strikes
in Bombay 75, 76, 81, 90, 119–27, 144
and informal neighbourhood leaders 192
and the police 198–9, 208, 211
and shopkeepers 110–11
in Britain 231
and capitalism 348
communistled 129, 172
and the Congress 267–8, 277, 292, 295–6, 301, 302
and the cotton textile industry 61, 63–4, 68, 100
Gandhi and 282–3, 284, 286
general 15, 61, 63–4, 86, 87, 90, 91, 110, 122, 157
and the labour market 81
lightning 91, 92, 143, 144, 149, 157, 161
and migrant workers 347
and the noncooperation movement 268
and the police 225
on the railways 308, 309, 312
and trade unions 83–4, 87–8, 97–8
wages and working conditions 92, 100
and workingclass violence 149, 150, 157, 158, 163–4, 165–6, 167, 171, 172
see also trade unions
Subbaraman, N. M. R. 311
Summers, R. 70
Surat, and the plague epidemic 248, 258
Swarajists 288–91, 292, 297
Talcherkar, H. A. 293
Tamhanekar, P. T. 138
Tamil Nadu, handloomweaving industry 11, 46–7
Tariff Board 61, 64–5
tariff protection, and the cotton textile industry 61
Tata, J. N. 49
Tata Iron and Steel Company 44
technology, and industrialization 38, 44
Textile Labour Association 64
Thakurdas, Sir Purshottamadas 115
Theosophical Society 293
Thevar, Ayyaswami 200
Tilak, Bal Gangadhar 153, 266–7, 269, 276, 281
Tilak riots (1908) 153–5, 197
Toster, M. S. 201
trade, and the cotton textile industry 52–3
Trade Disputes Act 93, 285, 304

Page 387
trade unions 12–13, 74–99
and Bombay workingclass neighbourhoods 117–42, 192
and the colonial state 92–4
and the Congress 19, 77–8, 91, 96–7, 267–8, 292, 297, 301–2, 306, 320
and employers 80–4, 85, 88, 89–90, 98, 99
and industrialization 35, 74–5, 79–80, 97–8
and jobbers 79, 80, 82, 83, 126, 128, 129
leadership 84–9
legislation 93, 285, 304, 318
neighbourhood recruitment 104
organization 74, 97–8, 101–2
and the police 211
studies of 5
and violence 148
weakness of 143–4
see also Bombay Girni Kamgar Union; communist trade unions; strikes
tribes, and criminalization 226, 228, 229
Tuticorin, trade unions 82
unemployment
cotton textile industry 63, 133
and neighbourhood connections 107
and workingclass violence 160–1
United States, AIDS epidemic of the 1980s 18, 265
urban poor, and colonial discourse on violence 15
Varerkar, Mama, Dhavta Dhota 159
Vicziany, M. 53
Viegas, Dr A. C. 237, 239, 254
village community, in colonial discourse 23, 24, 25
village connections
and Bombay workingclass neighbourhoods 107, 108
and jobbers 106
and policing 16
and the workforce 77, 274
villages
rural migrants from 7
and swaraj 291
Vincent, Fatty 198–9, 202
violence 2, 143–79
and 'bad characters'/hooligans 160–3, 174–5
and collective action 155–6
and colonial discourse 14–15, 28
crowd control and the police 212–16
discourse of 14, 150, 176–7
historians' conceptualization of 147–50, 152, 176
industrial 155, 177
Marxist theories of 148
mass, and colonial discourse 216
reporting of 150, 152–3
and 'the other' 174
threat of 151
and trade unions 98, 144–5
in workers' politics 9
see also riots
Wadia, B. P. 293
wages
adult male labour force 340–1
entrepreneurs and industrialization 40
and the jobber system 344
policemen 185, 188
and strikes 92, 100
Wallerstein, I. 328, 332, 347, 348, 349, 350
Washbrook, David 145
Watts, T. 158
weavers, in the Bombay workforce 141
Weber, Max, The Religion of India 27
Weir, T. S. 242, 256
Western Europe, and industrialization 30, 31, 32–4, 38, 40, 42, 79
Whig historiography of industrialization 42, 44
women
and police violence 214
sexual harassment by soldiers 249, 253
in the workforce 338–9
workforce
absenteeism 343
and jobbers 67–8, 76, 105–6, 341–5
labour market collapse, and the plague epidemic 260
labour market segmentation, and neighbourhood 9
labour patterns, and largescale industries 44
migrant workers 337–8, 339–41, 345–6, 347
preindustrial character 79, 101

Page 388
(cont.)
workforce
structural characteristics 75–7
women 338–9
see also labour
working classes
casual poor 7
and the Congress 19, 266–73, 274–81, 291–300
formation 7–10
history 2–7, 8
and industrialization 10–12, 37
and nationalism 14, 18–19, 307–17
and the 'Orientalist' tradition 1
and the police 16
and popular culture 19–20
and the rhetoric of nationalism 281–91
rural migrants 7
sectionalism 9–10, 12, 19, 37, 76, 77, 79, 314–15, 342
and studies of Indian capitalism 3
violence, colonial discourse on 14–15
see also class consciousness; labour force; trade unions
working conditions
in colonial discourse 27
in the cotton textile industry 63
and strikes 92
workplace
and neighbourhood 8, 9
in Bombay 104–44
world capitalism 2, 20–1, 327–50
and Marxism 328–30

Rajnarayan Chandavarkar - Imperial Power and Popular Politics - Class, Resistance and The State in India, 1850-1950-Cambridge University Press (1998)

Uploaded by

Copyright:

Available Formats

Rajnarayan Chandavarkar - Imperial Power and Popular Politics - Class, Resistance and The State in India, 1850-1950-Cambridge University Press (1998)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rajnarayan Chandavarkar - Imperial Power and Popular Politics - Class, Resistance and The State in India, 1850-1950-Cambridge University Press (1998)

Uploaded by

Copyright:

Available Formats

You might also like