Sie sind auf Seite 1von 335

This page intentionally left blank

Seeing the State

Poor people confront the state on an everyday basis all over the
world. But how do they see the state, and how are these engagements conducted? This book considers the Indian case where peoples
accounts, in particular in the countryside, are shaped by a series of
encounters that are staged at the local level, and which are also informed
by ideas that are circulated by the government and the broader development community. Drawing extensively on fieldwork conducted in eastern India and their broad range of expertise, the authors review a series of
key debates in development studies on participation, good governance,
and the structuring of political society. They do so with particular reference to the Employment Assurance Scheme and primary education
provision. Seeing the State engages with the work of James Scott, James
Ferguson and Partha Chatterjee, and offers a new interpretation of the
formation of citizenship in South Asia.
Stuart Corbridge is Professor of Geography at the London School
of Economics. He is the author or co-author of five books, including
Reinventing India (with John Harriss, 2000).
Glyn Williams is Senior Lecturer in Geography at Kings College,
London. He is the co-editor of a collection of essays on South Asia in a
Globalising World (2002).
Manoj Srivastava is a Research Associate in the Crisis State Programme, Development Studies Institute, London School of Economics.
He has worked for the Indian state for nearly twenty years.
Rene Veron is Assistant Professor in Geography at the University of
Guelph, Ontario. He is the author of a book on Real Markets and Environmental Change in Kerala (1999).

Contemporary South Asia 10

Editorial board
Jan Breman, G. P. Hawthorn, Ayesha Jalal, Patricia Jeffery, Atul Kohli

Contemporary South Asia has been established to publish books on the politics,
society and culture of South Asia since 1947. In accessible and compehensive
studies, authors who are already engaged in researching specific aspects of South
Asian society explore a wide variety of broad-ranging and topical themes. The
series will be of interest to anyone who is concerned with the study of South Asia
and with the legacy of its colonial past.
1 Ayesha Jalal, Democracy and Authoritarianism in South Asia: A Comparative
and Historical Perspective
2 Jan Breman, Footloose Labour: Working in Indias Informal Economy
3 Roger Jeffery and Particia Jeffery, Population, Gender and Politics: Demographic
Change in Rural North India
4 Oliver Mendelsohn and Marika Vicziany, The Untouchables: Subordination,
Poverty and the State in Modern India
5 Robert Jenkins, Democratic Politics and Economic Reform in India
6 Atul Kohli (ed.), The Success of Indias Democracy
7 Gyanendra Pandey, Remembering Partition: Violence and Nationalism in India
8 Barbara Harriss-White, India Working: Essays on Society and Economy
9 Baldev Raj Nayar and T. V. Paul, India in the World Order: Searching for
Major-Power Status

Seeing the State

Governance and Governmentality in India
Stuart Corbridge
London School of Economics

Glyn Williams
Kings College, London

Manoj Srivastava
London School of Economics

Rene Veron
University of Guelph

cambridge university press

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, So Paulo
Cambridge University Press
The Edinburgh Building, Cambridge cb2 2ru, UK
Published in the United States of America by Cambridge University Press, New York
Information on this title:
Stuart Corbridge, Glyn Williams, Ren Vron and Manoj Srivastava 2005
This publication is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format 2005

978-0-511-12916-2 eBook (EBL)

0-511-12916-5 eBook (EBL)


978-0-521-83479-7 hardback
0-521-83479-1 hardback


978-0-521-54255-5 paperback
0-521-54255-3 paperback

Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.


List of boxes, figures and tables

List of abbreviations

page vii

Part I The state and the poor

1 Seeing the state


2 Technologies of rule and the war on poverty


Part II The everyday state and society

3 Meeting the state


4 Participation


5 Governance


6 Political society


Part III The poor and the state

7 Protesting the state


8 Post-colonialism, development studies and spaces

of empowerment


9 Postscript: development ethics and the ethics

of critique




Appendix 1: Major national programmes and policies

related to poverty alleviation, 1999


Appendix 2: The 1999 general election in Hajipur




Boxes, figures and tables

Box 2.1
Box 3.1
Box 5.1
Figure 3.1
Figure 3.2
Figure 3.3
Figure 3.4
Figure 4.1
Figure 5.1
Figure 5.2
Figure 5.3
Figure 6.1
Table 1.1
Table 2.1

Table 2.2
Table 3.1
Table 3.2

Seeing and measuring the BPLs

page 75
Parental attitudes to education in Malda and
Midnapore Districts
The new public administration in India and poverty
alleviation in the countryside
The field sites
Schematic account of key sources of support for the
poor, by District
An official view of the developmental state: the EAS
in West Bengal
Chapati diagrams of government: Midnapore
field site
Sources of help in solving problems with education,
schools and teachers, West Bengal
EAS spending by panchayat and village, Sahar
Block, Bhojpur District, 19967 to 19989
EAS spending by panchayat and village, Murhu
Block, Ranchi District, 19934 to 19989
EAS spending by panchayat and village, Bidupur
Block, Vaishali District, 19967 to 19989
Local monitoring of the EAS in West Bengal: main
actors and responsibilities
Rent-seeking in the tree trade
Social Sector Plan outlays as a percentage of Total
Plan Outlays: Centre, States and Union Territories,
Literacy rates (age 7+) in 17 major states of India,
Census household by community and levels of
well-being and poverty
Household land ownership and income sources by
poverty ranking


List of boxes, figures and tables

Table 3.3
Table 3.4
Table 3.5
Table 3.6
Table 4.1

Table 4.2
Table 4.3

Table 4.4
Table 5.1
Table 5.2
Table 5.3
Table 5.4
Table 5.5

Poverty levels of female-headed households

Literacy rates (7+) by gender, class and caste
School attendance by gender, class and caste
Ability to support desired level of education
Proportion of male and female members of sample
households gaining one or more paid labour days
from the Employment Assurance Scheme
Number of meetings held of VECs of rural
P. S. schools, Bihar study Districts, 19979
Awareness of the existence and principal objectives
of the Employment Assurance Scheme (percentage
of sample households)
Evaluation of Malda village meeting: selected
questions and answers
Sector-wise breakdown of EAS schemes actually
implemented by the Blocks, Bihar
Schools with only one teacher, selected Districts of
Bihar, 19969
Schools with a toilet for girls, selected Districts of
Bihar, 19969
School infrastructure in Midnapore and Malda,
School infrastructure in selected primary schools,
Malda and Midnapore





This book draws on two linked projects supported by the United

Kingdoms Economic and Social Research Council (grant number
R000237761) and Department for International Development (grant
number CNTR 00 1553) in the years 19982001. We are grateful for
the financial support of these institutions, while noting that they are not
responsible for the findings we report here. As we also note in Chapter 1,
we are especially grateful to seven colleagues who worked with us in Bihar
(now Bihar and Jharkhand) and West Bengal on the ESRC project: Vishwaranjan Raju, Ashok Baitha and Rakesh Kumar in Bihar and Jharkhand
and Lina Das, Md. Basar Ali, Khushi Das Gupta and Surajit Adhikari in
West Bengal. We are also grateful to them and to Pramod Kumar (DPO
Jehanabad), Deepak Srivastava, Sanjeev, Ramedra Nath, B.N. Patnaik, in
Bihar, and Somen Dhar, Shibesh Das, Piyalee Sharma Das, Alok Kumar
Mukhopadhyay and Dibyendu Sarkar, in West Bengal, who worked with
us in 20002001 on the DFID-funded action research projects that grew
out of, and developed, our original research project.
In addition to these co-workers we want to thank all the people we
worked with in the field localities. We cannot list everyone by name, but
in Bihar and Jharkhand we extend special thanks to Ram Lakhan Manjhi,
Shiv Nandan Pike, Soma Munda, Ranjeet, Sudhir Manjhi, Budhwa
Lohra and Thuchuwa Lohra in Murhu Block, Ranchi District; to Ram
Ballav Paswan, Sonelal Paswan, Ram Prasad Paswan, K. D. Rai, Master
Bhogendra Rai, Lal Babu Rai, Chander Singh and Anant Singh in
Bidupur Block, Vaishali District; and to Laxman Ram, Sundershan Ram,
Lohri Ram, Ramashish Ram, Viday Paswan, Chandrageet Ram, Sriman
Narayan Singh, Kesho Ray and Baidaynath Singh in Sahar Block,
Bhojpur District. In West Bengal, we want particularly to thank Shibshankar Adhikary in Debra Block, Ratan Karmaka and family in Old
Malda Block, and Mona Mishra, all of whom made fieldwork a pleasure.
We are also indebted to Arun Mal, Katik Potar and Abani Biswas in Old
Malda, and to Narayan Sur, Parbati Sing, Manbendra Sing, Rampada
Tudu, Debi Sing, Pradip Maity and Himansu Sarkar Roy in Debra. We


realize that our questions must have seemed odd at times, not to mention
time-consuming, and that very few of our respondents will get to see (let
alone read) this book. Nevertheless, the book is very respectfully dedicated to the people whose stories are at the heart of this volume, and we
hope that we shall in time be able to repay their many kindnesses in other
and perhaps more practical ways.
Elsewhere in Bihar and Jharkhand we are in the debt of a large number
of friends and colleagues who guided our research project and helped
bring it to life. Again, we are forced to single people out from among a
diverse group of activists, NGO workers, government personnel, journalists and politicians, to all of whom we remain grateful, but special
thanks are certainly due to Arun Das, Raghupati, Rupesh, Akshay, Pawan,
Sushil Kumar, Ms Nutan, Ms Subhraja Singh, Urmila, TN Singh, Ranvindra Bharati, Sushil and Sagir Rehmani from the community of social
and cultural activists (including representatives of Lok Samiti, Bhor [literacy campaign], Panchayat Parishad, the Bihar Education Project, the
State Commission on Child Labour, and not forgetting the twenty-two
actors from Bhojpur who staged the folk drama, Dugdugi ); to Fr Jose, Fr
Francken, Fr Manthara and Ms Vizi Srinivasan from among the community of NGO workers; and to K. H. Subramanyam (Commissioner and
Secretary, Rural Development), Jayant Das Gupta (Secretary, Panchayati
Raj), and Vijay Prakash (Secretary, Higher Education). At the District
and Block levels, we would like to thank Uday Singh, Deepak Kumar,
H. S. Meena, Sudhir Kumar, Suman Kumar, the DDC of Vaishali
District, and the BDOs of Bidupur, Sahar and Murhu Blocks, from
among the community of government personnel, noting as well that without the help of a large number of Village-Level Workers, panchayat sewaks,
junior engineers, and others it would have been difficult to carry out our
work as we did.
Thanks also to Pranab Chaudhury of the Times of India, Patna,
to Mammen Matthew of the Hindustan Times, Ranchi, and to N. R.
Mohanty of the Hindustan Times, Patna, for their advice and support,
which were always welcome, and to Shaibal Gupta and his colleagues
at the Asian Development Research Institute, Patna. The A. N. Sinha
Institute, also in Patna, was kind enough to host a meeting we held in
January 1999 at the beginning of our research in Bihar. The academics
and activists who attended that meeting were collectively responsible for
pushing us to rethink the direction of some parts of the planned research,
and we remain grateful to them. Finally, we are pleased to thank Shri
Laloo Prasad Yadav (ex-Chief Minister, Bihar; currently Union Minister
for Railways), Shri Jagdanand Singh (Minister, Water Resources), Shri
Ram Chandra Purbey (Minister, Primary Education), K. D. Yadav
(State President, CPI-ML), Ram Dayal Munda (Jharkhand activist and



politician), and Ravi Shankar Prasad (State BJP leader, ex-Minister, Government of India), from among the community of politicians, for repeatedly taking time out of their busy schedules to share their thoughts with us.
In West Bengal we would like to thank Debdas Banerjee, Dwaipayan
Bhattacharyya, Indranil Chakroborti and Abdul Rafique for their intellectual support. We also benefited from discussions with Dr Surjya Kanta
Mishra (Minister, Panchayats and Rural Development), Prasad Ranjan
Roy (former Secretary, Department of Panchayats and Rural Development), Manavendra Roy (Secretary, Department of Panchayats and Rural
Development), Malai De (Department of Panchayats and Rural Development), Rajiva Sinha (now with UNICEF), Jude Henrique (UNICEF),
Dilip Ghosh and Bijon Kundu. At the District and Block levels we thank
Shefali Khatoon, Sushil Kumar, Md. Abdul Gani, Dibyendu Das, Athena
Mazumdar, Sanatan Ram, Dhiren Choudhury, Dilip Das, Abdul Malik,
Dilip Kumar Sarkar, K. N. Dhar, A. C. Sikar and Munsur Ali in Malda,
and Mamad Sahid, Jahangir Karim, Biman Bhumia and Robin Sing in
Midnapore. Grateful thanks also to colleagues at the State Institute for
Panchayats and Rural Development, Kalyani, and at the Centre for Studies in Social Science, Calcutta, who helped us with institutional support.
In New Delhi, we have benefited from discussions with Gerard Howe
and Arif Ghauri at DFID, with Mark Robinson at the Ford Foundation,
and with colleagues at institutions as diverse as JNU and the World Bank,
including Anand Kumar, T. K. Oommen and Yogendra Singh. Special
thanks also to Ronald Herring, Kuldip Nayyar and A. J. Philip for their
strong support of our action research project in Bihar.
We have also discussed our work with a number of colleagues in
Europe and North America, and would like to thank Abhijit Banerjee, Fiona Candlin, Kanchan Chandra, Sharad Chari, Partha Chatterjee,
Shubam Chaudhuri, Nicholas Dirks, Chris Fuller, John Echeverri-Gent,
John Harriss, Barbara Harriss-White, Walter Hauser, Patrick Heller,
Craig Jeffrey, Sarah Jewitt, Sudipta Kaviraj, Steven Legg, Janek Mandel, Emma Mawdsley, John de Monchaux, Tanni Mukhopadhyay, Roopa
Nair, Suppiramnaiam Nanthikesan, Ranjit Nayak, Philip Oldenburg,
Johnny Parry, James Putzel, Saraswati Raju, Sanjay Reddy, Ben Rogaly,
Nikolas Rose, Sanjay Ruparelia, the late Professor T. Sathyamurthy,
Alpa Shah, Edward Simpson, Bishwapriya Singh, Kristian Stokke, Judith
Tendler and Ashutosh Varshney for engaging critically with our work.
We are also grateful to two anonymous referees for Cambridge University Press, and to our Editor there, Marigold Acland. Above all, we want
to thank our partners and children, Pilar and Joanne, Paula and Anna,
Nina, Saagar, Shikhar and Roshini, and Lori, Lili and Alexandre. This
project has occupied us for the best part of six years, and we are extremely
grateful to them for their patience and support.


Backward Classes
garibi hatao
Gram Panchayat
gram baithak
gram sabha
gram sansad

executing agent or foreman

original people; preferred name for
Scheduled Tribes
worldly (self-)interest
the weaker sections, or the Scheduled
Castes and Tribes and Other Backward Castes
upper or respectable folk; gentlemen
ten million
big brother or political boss
Marathi word for the oppressed (the
the traditional moral order
state-directed or dominated, in the context
of economic development
an end to poverty (slogan of Indira
village council; the lowest tier of the
panchayat system
informal meeting
formal village meeting provided by
government statute
statutory village meeting; smallest
panchayat constituency
children of god; Gandhis term for the
ex-Untouchables, now Scheduled Castes
honour or dignity
caste in the sense of named birth group
scribal caste of north India, now seen as
high caste


Other Backward
Classes/Castes (OBCs)

Panchayat samiti
panchayati raj
Scheduled Castes

Scheduled Tribes

Zilla parishad


unfinished (of infrastructure); often

earthen works
one hundred thousand
headman of village, now of a panchayat
organized left insurgents
socially and educationally deprived
communities, not including Scheduled
Castes or Tribes, for whom compensatory
actions are now authorized by the state
council, official institution of local
Block-level council
official system of local self-government
President of the gram panchayat
finished (of infrastructure); permanent,
often concrete-built
political fixer
President of the zilla parishad
President of the panchayat samiti
those castes recognized by the Constitution
as deserving special assistance in respect of
education, employment and political
representation (other than the OBCs); in
effect, the ex-Untouchables
in effect, the official term for Indias adivasi
populations; those communities recognized
by the Constitution as deserving special
assistance in respect of education,
employment and political representation
(other than the Scheduled Castes and OBCs)
hamlet, or small neighbourhood within a
revenue collector and landholder under
British rule
District-level council

used here of West Bengal



All-Bengal Primary Teachers Association

Assistant District Magistrate
Assistant Engineer
Additional Executive Officer
Block Agricultural Officer
Block Development Officer
Block Education Officer
Bihar Education Project
Bharatiya Janata Party
Bharatiya Kisan Union
Commission on Agricultural Costs and Prices
Community Development Programme
Community Organizer
Communist Party of India, Marxist
Communist Party of India, Marxist-Leninist
Deputy Development Commissioner
Department for International Development, UK
Divisional Forest Officer
District Magistrate or Collector
District Primary Education Project
District Primary School Council
Development of Women and Children in
Rural Areas
Employment Assurance Scheme
Employment Guarantee Scheme
Education Guarantee Scheme, Madhya Pradesh
Eastern India Rainfed Farming Project
Enhancing Pro-Poor Governance (action research
Economic and Social Research Council, UK

List of abbreviations


Gujarat Grassroots Innovation Augmentation

Gram Panchayat
Integrated Child Development Scheme
Indian National Congress
Integrated Rural Development Programme
Junior Engineer
Joint Forest Management
Jawahar Rozgar Yojana
Key Resource Person
Kerala Sashtra Sahitya Parishad (peoples science
Maoist Communist Centre
Marginal Farmer and Agricultural Labour programme
Mazdoor Kisan Shakti Sanghatan
Member of the Legislative Assembly
Member of the Legislative Council
Member of Parliament
National Campaign for the Peoples Right to
National Democratic Alliance
Non-Governmental Organization
National Planning Committee (of the INC)
Non-Timber Forest Product
Other Backward Castes (or Classes)
Public Distribution System
Perspective Planning Division
Panchayati Raj institutions
Rashtriya Janata Dal
Rashtriya Swayamsevak Sangh
School Attendance Committee
Scheduled Castes
Small Farmers Development Agency
Swarajayanti Gram Swaroznar Yojana, a village
self-employment programme that replaces the IRDP
Sub-Inspector (of primary schools)
State Institute of Panchayats and Rural Development
Supply of Improved Toolkits to Rural Artisans
Special Area Programme
Sub-Saharan Africa
Scheduled Tribes
Tribal Development Agency




List of abbreviations

Total Literacy Campaign

Transparency International
Trinamool Congress
United Nations Development Programme
Village Education Council
Vishwa Hindu Parishad
Village-Level Worker
West Bengal


In recent years there has been a sea-change in the ways in which the
state in India has sought to present itself to its poorest citizens. To listen
to leading members of the National Democratic Alliance (NDA) government in 2004 one would think that the year 2000 (or even 2001 or
2002) was something like Ground Zero in this respect. Ministers from
leading human development departments were in the habit of swatting away criticism of their ministries on the ground that everything
was in flux. In a world of village education committees, citizen scorecards and newly vibrant panchayati raj institutions, not to mention a
new era of publicprivate partnerships, it apparently made no sense to
criticize ministers for faults that may or may not have dogged previous
This was nonsense, of course, for many of the innovations that were
being trumpeted by the NDA were first given shape by the Congress and
United Front governments of the 1990s, when village education committees and joint forest management were launched with appropriate
pomp and fanfare.1 It would also be unwise to assume that new rhetoric
about a kinder and more responsive system of government must correspond in any clear way to the perceptions of poorer or more vulnerable
people. All democratic governments are tempted by the fruit of exaggeration, and Partha Chatterjee is right to insist that poorer people in
most of the world (2004: 3) are very often compelled to meet the state
as members of social groups that transgress the strict lines of legality
in struggling to live and work (Chatterjee 2004: 40). They inhabit, that
is to say, the rough and tumble worlds of political society, where governmental agencies are met by wit and by stealth, and not uncommonly

For this reason, too, the removal from office of the NDA after the 2004 general election is
unlikely to lead to a significant movement away from what might be called the new public
administration, or that more or less consistent running together of agendas for public
service reforms, the decentralization and devolution of government activities and budgets,
and participatory development. These agendas are also said to describe a prospectus for
good governance. On the election, see Ruparelia (2005).

Seeing the State

by violence.2 Civility and pluralism are not the defining features of their
And yet something has been going on. New Delhi can now point with
pride to a significant reduction in rates of income poverty in the country,
albeit that these were sustained by a period of concerted economic growth
that began a full decade before the reforms of 1991.3 It can also claim that
Human Development in the country is getting better. A recent report
by the United Nations Development Programme (UNDP) suggests that
Indias Human Development Index (HDI) score rose from 0.439 in 1992
to 0.571 in 2001.4 And, perhaps most of all, it can begin to make the
argument that these improvements have been induced by a new regime
of governance. Notwithstanding Chatterjees claim that civil society and
the poor co-exist in India like oil and water, government spokespeople
insist that ordinary people are being listened to at the Block, District,
State and national levels. They are reaping the rewards of an electoral
system that empowers even the poorest men and women as citizens of
different territorial jurisdictions. Public officials at the highest levels of the
state can be called to account by citizens organizations or through public
interest litigation. In the localities, meanwhile, decisions are taken and
public monies increasingly are spent by elected representatives who are
accountable to villagers through gram sabhas and other open meetings.5

In his 2001 Leonard Hastings Schoff Memorial Lectures, Chatterjee is mainly concerned
with the political battles that have to be waged by groups like urban slum dwellers. In
his view, the men and women who make up these groups are not treated by the state as
if they were citizens. The fact that they occupy land illegally, and thus call into question
the sanctity of private property, means that the state cannot deal with them as members
of civil society. For Chatterjee, civil society in a post-colonial setting is limited to elite or
bourgeois groups and their forms of politics. At the same time, however, the state does
recognize a governmental obligation to populations of slum dwellers and other subalterns.
This obligation reflects a prior commitment to practices of welfare provision and social
control. The circle is squared, Chatterjee contends, in the field of political society. This is
where groups of the urban poor seek the support of political parties like the CPI-M in West
Bengal or Shiv Sena in Bombay (see Hansen 2001), or patrons (including criminals)
from outside the formal worlds of party politics. Later in the book we shall comment on
Chatterjees instructive but perhaps overstated division between the worlds of civil and
political society a division which is inattentive to some of the more hybrid forms of
statepoor encounters that we describe in part II.
UNDP (2002: 16), drawing in part on data from the 55th round of the Household
Consumer Expenditure Survey of Indias National Sample Survey Organization. Critical
evaluations of the poverty reduction thesis, and of some NSS data sets, can be found
in Chandrasekhar and Ghosh (2002), Pogge and Reddy (2003), and Reddy and Pogge
(2003). For a review, see Harriss and Corbridge (2003).
UNDP (2002: 17). In addition, the reported Gender Development Index increased from
0.401 in 1992 to 0.533 in 2001. The Gender Empowerment Measure increased from
0.226 to 0.240 over the same period (UNDP 2002: 17).
These formulations also skirt over the fact that accountability mechanisms have long
been present within the administrative services. Even in colonial times officers would
hold meetings with villagers from time to time, perhaps in the form of a janata durbar (see
chapter 3).


It will be one aim of this book to interrogate the optimism of the governments account of recent developments in the fields of governance
and governmentality. Is it really the case that poorer men and women are
coming to enjoy the status of citizens, and are being engaged as such by
government officers (and not simply as members of beneficiary or troublesome populations, as Chatterjee maintains)? And can this reasonably
be described as a national story, or are we picking up the effects of policy changes, and patterns of political mobilization, that have been put in
place in some regions and not in others? How do poorer people see the
state, and how are governmental agencies seen by the people who advise
or work for them? What would count as a convincing causal explanation,
as opposed to a suggestive narrative sequence?
In these regards, it is worth noting that even some of the main sponsors of the new public administration are cautious about what should
and should not be claimed. Agencies including the UNDP, the United
Kingdoms Department for International Development (DFID), and the
World Bank, cast their evaluation studies in the most positive light and
tread softly around points of contention or criticism. Where possible they
signal the benefits of decentralization or of enhanced participation. Yet
many of the people who work for these agencies are experienced and at
times rather cynical individuals who know that much remains to be done.
Their work in villages in central and north India cautions them against
a chorus of acclamation that mistakes promise for performance. They
point gently toward continuing problems of elite capture (of development benefits), and of the misuse of public funds by poorly trained and
poorly paid government servants. They also highlight a persistent gender
gap in terms of rates of participation in village open meetings.
Some Left critics, meanwhile, point out that the decentralized local
governance structures now lauded by New Delhi work most effectively
in a state like Kerala, and perhaps also West Bengal, where there is a
supportive political culture.6 It is one thing to provide institutions to
promote accountability and decision-making at the panchayat, Block and
District levels, and quite another to produce men and women who are
able to participate effectively in these new or revamped structures. The
production of skilled citizens is not something that happens overnight.
Men and women have to be educated, they need to develop a broad set of
capabilities, to use Amartya Sens term, and they need to be acquainted
with the costs and benefits of new structures of rule.7 Confidence has
to be built up, and it is here, say critics like Chaudhuri and Heller, that

The classic statement is that by Kohli (1987). See also Webster and Engberg-Pedersen
Sen (1984).

Seeing the State

the comparative advantage of organizations like the Communist Party of

India, Marxist (CPI-M), and associated bodies like the Kerala Sashtra
Sahitya Parishad (KSSP: the Peoples Science Movement), is most in evidence (see also chapter 7).8 To expect a similar level of success in Madhya
Pradesh is to confuse moral exhortation around the virtues of civic society
with the hard work of constructing a pro-poor political culture. (Digvijay
Singh learned this the hard way in 2003.)9
Barbara Harriss-White goes even further. In her account of India Working she dismisses views of the state that she considers to be formalistic, or too focused on statutory responsibilities. She contends that the
official part of the state has been hollowed out over the course of the
last twenty or thirty years, and has been replaced by what she calls a
shadow state (Harriss-White 2003: 77). This vast assemblage of brokers, advisers, political workers, crooks and contractors surrounds the
official state, deprives it of funds, and helps to ensure that it is run in
part for the private benefit of some of its employees. The other main winners are the largely self-employed men (and some women) who benefit
from a world of state-produced shortages and sanctioned fraud. They are
the top dogs in Indias intermediate classes. The losers are the labouring households who make up the bulk of the India of the 88 per cent
(Harriss-White 2003:1).10 They work unprotected in Indias informal
and black economies. They also need the protection of intermediaries to
chart an unsteady course through what Chatterjee has called the politics
of governmentality (Chatterjee 2004: 23).
To read Harriss-White is to be put in mind of a world that is far removed
from that described by government boosters. States fail poorer people on
a regular and predictable basis, and they will continue to do so notwithstanding recent innovations in the field of governance. Harriss-White
reaches these conclusions, moreover, on the basis of an approach to social
science that she describes as field economics (Harriss-White 2003: 9
10). Economics and political science sometimes deal with entities like
the economy or the state as if they were self-evident or transparent; as
if the conditions in the trenches, as Joel Migdal puts it, really do approximate to the textbook descriptions imposed upon them.11 Harriss-White
will have none of this, and rightly so.


Chaudhuri and Heller (2002).

The former Congress Chief Minister of Madhya Pradesh was voted from office in
December 2003. See our discussion in chapter 7.
The figure refers to the percentage of people who live outside the major metropolitan
cities, and thus at some distance from the centres of corporate capitalism (Harriss-White
2003: 1).
Migdal (2001).


We said before that a first task of this book is to interrogate official

stories about the state in India. This being so, we need to follow the lead
of Harriss-White and Chatterjee and dispense with black box approaches
to questions of government, participation and empowerment. Boosterism and ethnography are uncomfortable bedfellows and we shall want
to exploit this tension. There is much to be said for researching political society and the local state from the inside-out. At the same time,
however, as the second part of this book will make clear, we believe that
the lives of poorer people in rural India are being changed perceptibly,
and in some cases for the better, by the new technologies of rule that
we described above. An exclusive emphasis on the shadow state, or on
a relentlessly vertical political society, sometimes fails to point up the
spaces of citizenship that are being created, or perhaps widened, in the
wake of the good governance agenda and the popular mobilizations to
which it can give rise. These developments are happening more rapidly
and more deeply in some regions than in others, but the effects are real
nonetheless, including in some parts of Bihar and Jharkhand, despite
popular (or stereotypical) views to the contrary. It follows that a second
aim of this book will be to document these changes, and to explain their
differential spatial impress.
Technologies of rule, sightings of the state
We take the phrase human technologies of rule from Nikolas Rose, and
in part I of the book we link it to an idea of seeing the state.12 The book
ranges quite widely at this point, as befits a volume in this series. Our purpose, however, is to think about how we might make sense of the facts on
the ground that we report in parts II and III. We agree with Harriss-White
that states should be understood anthropologically. Instead of thinking
of them as discrete or singular entities, we prefer to follow Foucault in
speaking of dispersed practices of government. States are best thought
of as bundles of everyday institutions and forms of rule.13 In part, this is
for cultural or ideational reasons. Partha Chatterjee and Sudipta Kaviraj
suggest that it is a mistake to assume that the life-worlds of elite, Englisheducated Indians coincide with those of their subaltern or vernacular
counterparts.14 Very often they do not, and we should expect lower-level
public officials to reinterpret and sometimes significantly to change the
practices of government that are handed down to them by Indian Administrative Service (IAS) officers at the District, State or Union levels. But
states also need to be understood as consisting of diverse and not always

13 See Fuller and Harriss (2001).

Rose (1999); see also chapter 2.
Chatterjee (1997a); Kaviraj (1991). See also note (2).

Seeing the State

cohering human technologies of government. These are the quotidian

practices of rule that structure and even produce settings for the conduct
of business between the state and its citizens or subjects. They refer
not only to institutions like the IAS or the zilla parishad, but also to those
forms of knowledge, vocabularies, practices of calculation and so on, that
help to produce designated social groups (Scheduled Communities, for
example, or households that are BPL: below the poverty line) and bring
them into contact with agencies of government. They also refer to an
idea of the state as an impersonal and disinterested actor the state that
claims to find expression in the Constitution of India, for example.15
We expand on this approach in part I of the book, where we consider
some of the different ways in which governmental agencies in India have
sought to address the question of the poor. We shall argue that poorer
people very often encounter the state through technologies of rule that
help to structure a war on poverty. In New Delhi during the Emergency,
for example, many slum dwellers would have encountered the state in
the form of a bulldozer that was being driven, metaphorically at least,
by a discourse that produced urban poverty as a sin or as unsightly.16
By the same token, we believe there are good reasons to suppose that
statepoor encounters are now being restructured to some degree by
new technologies of rule that seek (or claim) to produce members of the
rural poor as clients of the government, and as active participants in their
own empowerment. These technologies are built into a new suite of participatory development projects. Some of these are operated by external
agencies like DFID and the World Bank. Others are run by the Government of India working in conjunction with non-governmental organizations (NGOs). They are also built into important new ventures that have
been prosecuted by ministries concerned with employment creation (the
Employment Assurance Scheme) and primary education provision (the
advent of Village Education Committees is said to present new opportunities for parental supervision of teachers).
Our point, of course, is not that these new technologies of rule will
always secure their stated effects. That is a claim best left to government
boosters. But nor do we contend that they are bound to fail, or that the
unintended consequences of government action must secure the further
domination of the state over its (potentially rebellious) subjects. This is

On the state idea, see Abrams (1988). For India, specifically, see Austin (1966).
The encounters would not have been as one-sided as this sentence implies. Chatterjee
is right to insist that illegal slum dwellers will call on such protectors as are available
to them in political society, as well as on their own resources and courage, to resist the
agencies of urban improvement. Sadly, however, in New Delhi during the Emergency,
these battles were often lost: see Selbourne (1977) and Legg (2005a).


one of the arguments that James Ferguson makes in his generally splendid
account of the effects of bureaucratic rule in Lesotho, and it is one that
Partha Chatterjee comes close to on occasions.17 Our argument is more
modest. We consider it unwise to assume that the agenda of the new public administration does not open up significant spaces of empowerment
for the men and women it seeks to position as participants or possible
beneficiaries. We make this argument, moreover, because we join with
Fuller and Harriss in insisting that the ways in which technologies of rule
are made flesh will depend on the manner in which they are interpreted
and put into play by lower-level government workers, elected representatives and others. We also need to see why and how they are seized upon,
understood, reworked and possibly contested by differently placed people within the population of the poor (or the rural poor in this book) in
both civil and political society.
This is why we have chosen to adapt a phrase associated with James
Scott. Instead of looking at the ways in which the state might see its
citizens, which was Scotts concern in Seeing Like a State (Scott 1998),
we prefer to enquire into some of the myriad ways that the state comes
into view. We are concerned here particularly, but not exclusively, with the
ways that governmental agencies are seen by different groups of people
within the rural poor. One of the guiding premises of this book is that
the state still matters greatly to people in rural India. It is sometimes to
be feared or avoided, of course, although this is not something we pursue
in detail here. Our focus is on the developmental state.18 But it can also
be at the heart of peoples livelihood strategies. Richer individuals know
this too, and Harriss-White rightly contends that the placing of male
members of trading households in the professions and the bureaucracy
[including the Electricity Board and the IAS] (Harriss-White 2003: 113)
amounts to a pre-emptive bid for state licences or contracts. For most poor
men and women, however, the state has only recently been positioned as
a source of social power, and then mainly by members of the political
classes. A tribal woman in rural India is more likely to turn to sarkar
(government) for an entitlement, such as a ration card or pension, or
perhaps for employment or to register a death. Or she might want to
call upon the state to enforce her right to a minimum wage, say, or for
protection against an accusation of witchcraft. She might also want to
send a child to school or to a health-care facility.

Ferguson (1990). Also Chatterjee (2004).

Roughly, those agencies of state and governmental practices that are charged with
improving or protecting the incomes, capabilities and legal rights of poorer people. Our
focus here will be on government departments, but the definition deliberately reaches
out to political parties and judicial bodies.

Seeing the State

The tangible outcomes of these encounters will matter a great deal to

the men and women involved. A job or certificate gained is very different to a job or certificate that is refused. But it also matters to people
how the encounter is structured and performed, and this brings us to a
second premise of our work. We can learn about the practices of government by attending to the diverse ways in which the state is experienced
and understood by differently placed individuals, including by its own
employees. A low-caste man who is treated with respect by a teacher or
a Block Development Officer might come to see the state in a very different way than an adivasi woman who is kept waiting for hours to see
sarkar, who sees gangs of males push in front of her in what passes for a
queue, and who is made to touch the feet of the official she finally meets
(perhaps with the help of a dalaal, or local broker) at the end of rough
and uncivil language. We need to pay close attention to how and where
these performances take place if we want to come to a more nuanced
understanding of how people inhabit and encounter the state, and how
they react to its everydayness and their senses of what it is to be a citizen,
client and/or subject.
This brings us to a third proposition. We shall argue that the sightings
of the state that poorer people make are never straightforward or unitary.
None of us sees the state (or the government, the market, even public culture) in a direct and unmediated fashion. We always see the state through
the eyes of others, and with close regard for past memories, accounts that
circulate in the public sphere, and how we see other people getting on
or being treated. And yet this is not a trivial observation, or one without consequence. Part of the attraction of the new public administration
is precisely that it expects a tribal woman in eastern India, lets say, to
see the state in terms of practices of corruption or extortion, as well as
of gender bias and a propensity for ad hoc or personalized rule. It then
follows that she can be empowered by the exit option, or by a decentralized form of rule that makes the state accountable to citizens on the
basis of their statutory (and thus in some respects equal) rights. This
suggestion is predicated very exactly on the notion that new practices of
rule will promote new and unhindered sightings of the state.
But if we do not entirely share this optimism, or partake of what
Chatterjee describes as the unscrupulously charitable theoretical gestures [of neoliberal ideology] (2004: 39, and see our discussion in
part II), we do accept that it makes sense to think of the co-production
of statepoor encounters by three main sets of actors. We believe, that
is to say, that public policy debates in India for example, on the scope
and purpose of participatory development, on the prospects for reducing
corruption can be illuminated by studies of how government works in


practice that draw on three main vantage points. These are: the sightings
of the state made by poorer people both as citizens and as often vulnerable
members of political society; the sightings made by government officers
in different line departments, and people at different levels of authority in
political society; and the sightings made by members of the wider development community, including experts from the World Bank and other
lending agencies, and senior bureaucrats in the Government of India.
Furthermore, because these sightings are mutually constitutive, it is
a mistake to suppose that development studies are simply a means by
which people in the development business look in on the world they seek
to describe and even to mend. They are not, and we should not assume
that claims on behalf of good governance are mere rhetoric, however
much these phrases are cheapened by misuse. Development studies must
rather be understood as a set of human technologies of rule that help to
structure and produce the worlds they aim to describe. They are not
without effect, and it is unhelpful to contrast development studies as
somehow bad to a more virtuous concern for post-colonialism. Matters
are more complicated than this, as we show in part III. At the same
time, it is important that we are aware of what Hirschman once called
the indirect or recruitment effects of new ideas or practices of rule.19
Even after the promulgation of new legislation and training manuals, a
panchayat sewak or his/her supervisor might not convene village meetings
in the manner that is called for in policy statements. But the fact that he
or she has to hold meetings on a regular basis, and is now required to
reach out to all the sensitive people in the area (see our discussion in
chapter 5), including poorer people who might make trouble for him or
her, suggests that real changes are happening nonetheless. In this case,
a new technology of rule has pushed a named agent to widen his or her
previous circles of engagement, and perhaps also to change the terms on
which these engagements are transacted. A sighting of the state by an
external agency finally gives rise to a revised sighting of that state (or a
slightly new version of it) by men and women who are constituted as its
The organization of the book
We can now say that the job of part I of this book is two-fold. Chapter 1
considers in some detail what it means to talk of seeing the state, and how
this lens might deepen our understanding of the politics of the governed.
We review some of the existing literature on the everyday state and society

Hirschman (1981).


Seeing the State

in India. We also draw attention to the ways in which encounters with the
state are produced by dispersed state agencies amid conditions of greater
or lesser institutional scarcity. Finally, we consider how recent debates in
development studies have sought to attend to these conditions, and provide remedies for state failure. In chapter 2 we provide historical depth
to these and other debates. A common conceit in development studies
is that everything is new, when this is rarely the case. The first part of
this chapter looks at the ways in which accounts of poverty in India have
been produced by a very diverse set of human technologies of government, including the Census, the National Sample Survey and discourses
about shame or backwardness. We also consider how, and with what consequences, certain individuals or groups have been labelled as members
of the Scheduled Castes and Tribes, or as belonging to the Below Poverty
Line (BPL) population. Poorer people very often see the state because the
state has chosen to see them. Subsequent parts of the chapter consider
how various state agencies and political parties have proposed to wage
war against poverty in India, and how important schemes for poverty
alleviation were democratized in the 1970s, 1980s and 1990s in accordance with new views about the rights of the Backward Classes, and of
the capacities and entitlements of poorer people. We also show how the
multiplication of schemes speaks to the growing importance of visuality
and presentation in the promotion of an anti-poverty agenda. Politicians
need to be seen to be active on behalf of the poor.
We can put this another way. The first part of the book provides us with
a language with which we can approach the debates now swirling around
the new public administration and the boosterism that is attached to it.
Before we can evaluate this agenda we need to decide upon a framework in
which such an evaluation can take shape. This is what we hope to supply
with our accounts of technologies of rule and sightings of the state. But
what might be described as the evaluation itself can only take place in the
field, with proper regard for all of the subtleties that this phrase should
call to mind. In part II we draw on two research projects that we carried
out in rural eastern India in 19992000 and 20001. The first of these
projects was funded by a research council. The second was supported
by DFID and positioned us for a while as development consultants. We
shall come back to this later and in chapter 9. Taken together, the projects
allowed us to investigate the income support, empowerment and protective functions of the state. Five hundred households (400 poor and 100
non-poor) in Bihar, Jharkhand and West Bengal were kind enough to
provide us with information on the Employment Assurance Scheme, primary education, and legal struggles, respectively, and on how different
groups of rural society encountered the state in these arenas. The information was collected in part from an extensive questionnaire survey, but



also from conversations held with ourselves and with seven field assistants who lived in the villages from March 1999 to March 2000.20 This
book would not have been possible without the extraordinary efforts and
friendship of Vishwaranjan Raju, Ashok Baitha and Rakesh Kumar in
Bihar and Jharkhand, and Lina Das, Md. Basar Ali, Khushi Das Gupta
and Surajit Adhikari in West Bengal, to whom we remain deeply indebted.
The village-based data sets were supplemented by more than 280 taped
interviews with teachers, Block Development Officers (BDOs), District
Development Officers (DDOs), engineers, trade unionists, contractors,
politicians, brokers and other key informants at the Block, District and
State levels. This peripatetic research strategy allowed us to collect data
from five field sites where we expected to find very different political
cultures: from Bidupur Block, Vaishali District, Bihar, an area of considerable political competition, often along caste lines; from Sahar Block,
Bhojpur District, Bihar, an area of extreme class tensions where the CPIML (a Naxalite group) and the Ranvir Sena have been active; from
Murhu Block, Ranchi District, Bihar (now Jharkhand), an area where
adivasi people have long been mistrustful of the state; from Old Malda
Block, Malda District, West Bengal, an area where politics are clientelistic and where state failure is widely remarked; and from Debra Block,
Midnapore District, West Bengal, an area where the CPI-M has enjoyed
considerable success in mobilizing groups within the rural poor. (The
location of the field sites can be seen in figure 3.1. The names of individual villages and panchayats have been held back.)
When we set up this research design we didnt know that we would
write a book along the lines of this one, and this is decidedly not the
book of the research project. Partly by luck and partly by judgement,
however, we believe that our choice of field sites has given us a platform
to speak with some confidence about the diverse structuring of state
poor encounters in eastern India, an area where rates of rural poverty
remain stubbornly high. As we hinted before, however, the focus of part II
is not on the poverty of poorer people in this or other parts of India,
although chapter 3 does provide background data on income levels and
capabilities in the study areas. Our main aim is to look at the hows of
government: how government business is transacted, how it reinforces or
undermines local ideas of hierarchy, how it constructs gender relations,
how it provides incentives for principals and agents, how it deals with the
question of participation, how governmental programmes are involved
in the production both of agents in political society and what might be

We will have more to say about the choice of households and the scope of the questionnaire survey in chapter 3. It will suffice to say here that the households were chosen
from a full census of five field localities in different Districts of Bihar (as was) and West


Seeing the State

called incomplete citizens, and how it might create intentionally or

otherwise spaces of empowerment for poorer people. Chapters 46 take
up these how questions with close regard for a set of linked debates on
participation and social capital; on rent-seeking, corruption and the quest
for good governance; and on decentralization and the slowly changing
contours of political society. It will come as no surprise that some of our
most positive findings are drawn from Midnapore. But we also found
spaces of empowerment in Bihar and Jharkhand, and we hope to provide
accounts of how the state works in eastern India that cut against the
grain of established thinking. Although chapters 46 deal with a set of
linked debates, we believe that each chapter can be read with profit in its
own right.
Finally, in chapters 7, 8 and 9 (part III), we consider some wider aspects
of the politics of statepoor encounters in India. By engaging with work
on Kerala, Rajasthan and Madhya Pradesh, for example, we can relax
the rather stiff test of good government that Bihar and Jharkhand might
be thought to pose. Given that we worked for DFID, we also consider it
proper to reflect on our own positionalities as development researchers
and even consultants. We have been prompted to reflect on what Max
Weber once described pompously perhaps, but also very usefully as
the duties of an intellectual who stands in the service of moral forces
(Dean 1999: 36, after Weber 1972).
These reflections inform our concluding observations on the state of
development theory and practice, as John Toye once referred to the
field.21 We develop an account of the work of agencies like the World
Bank and the UKs Department for International Development that is
at some remove from the more stylized accounts of their most vehement
critics. While holding no brief for these organizations indeed we are
critical of many of their activities we hope to show that they are contested institutions which do respond, in part, to the reports of their field
officers, academics and, indeed, some activists and intended beneficiaries. Above all, we intend to show that they are learning institutions, and
that, as such, they must be engaged with politically by those who want
to take the part of the poor. The poor, after all, engage with these institutions on a daily basis, even if these engagements appear at first to be
indirect and largely discursive. Their accounts of the state, and thus to
some degree of their livelihoods and politics, their senses of citizenship
indeed, cannot help but take account of the sightings of others.

Toye (1987).

Part I

The state and the poor

Seeing the state

We have said that one aim of this book is to consider how differently placed
men and women see the state in rural India. Some of these individuals will
be employees of the state, or external advisers to the Government of India
and its constituent states and Union territories, although many more will
be farmers or labourers. Some will be political fixers and members of
the Backward Classes, while others will be farmers, Class IV government
servants and adivasis at the same time. But what does it mean to talk
about seeing the state?
We are used to the idea of the state seeing its population or citizenry.
Visuality is at the heart of many theories of power and governmentality.
Michel Foucault, most notably, has shown how the birth of modern
forms of education and welfare provision corresponds to the emergence
of biopolitics as a form of politics entailing the administration of the processes of life of populations (Dean 1999: 98). Populations emerge when
changes in working practices give rise to economic government and the
discipline of political economy, and they get bounded by new exercises in
mapping and measurement, including the production of censuses, cadastral surveys and expeditions.1 Biopolitics then refers to those government
interventions that seek to improve the quality of a population as a whole,
and these procedures produce that which we name the state as the effect
of these interventions. These can be positive and related to questions
of public health and standards of living, or even to incarceration for the
purposes of reform or improvement. Such interventions might involve the
inspection of men, women and children by state officials or agencies contracted by the state. Children, for example, might be required to attend
for eye examinations or inoculations. Prison cells might be searched for
illegal substances. But they can also be negative, as when they are concerned with the purity of the group or class. These interventions might

See Foucault (1997), and also Hacking (1982). On statistical and mapping exercises in
India, see Cohn (1987), Edney (1997) and Barrow (2003).



The state and the poor

draw on discourses which see vile, corrupt or simply foreign bodies as

appropriate targets for torture, eugenics or even genocide.2
Governmentality, for its part, can continue both sets of interventions.
If government can be thought of as the conduct of conduct, at least
in parts of Europe from the sixteenth and seventeenth centuries, governmentality can be defined as the means by which we think about governing
others and ourselves in a wide variety of contexts (Dean 1999: 209). It
involves the internalization of norms, even where those norms (for example, of proper models of sexual, economic or political behaviour) are
always contested. If biopolitics involves the extension of sight from the
sovereign to the state, we might think of governmentalization (including of the state) as involving a further extension of powers to those who
profess expertise over the private body or the body public, be they aid
workers, economists, psychologists, psychiatrists, social workers, sexologists or public health workers. These persons also claim the privileges of
sight, including those of insight, foresight and even hindsight.
Foucault, of course, wrote rather little about the non-western world.3
But there is more than a hint of Foucaults arguments in James Scotts
account of Seeing Like a State. Scott contends that: The premodern state
was, in many crucial respects, partially blind; it knew precious little about
its subjects, their wealth, their landholdings and yields, their location,
their very identity (Scott 1998: 2). In contrast, the danger of the high
modernist state is that it sees too much. It engages in simplifications that
have dangerous if often unintended effects for the citizenry. The high
modernist state is distinguished not only by an administrative ordering
of nature and society, or by an ideology that evinces a muscle-bound
faith in the virtues of reason, progress and industry; it is also defined by
its hubristic ability to see a better future for all of the people, whether
this future consists of collective farms, villagization, the urban visions of
Le Corbusier, or socialism itself. It is this weaving together of sight in
its temporal and spatial dimensions that announces the high modernist
era, and which paves the way for interventions that are potentially lethal
(Scott 1998: 5). These interventions are most likely to surface when they

Nikolas Rose (1999: 26) reminds us that Foucault discussed the connections between
micro-fascism and macro-fascism in The History of Sexuality (Foucault 1979: 14950).
On biopolitics, archaeological practice and the politics of spatial exclusion, see Nadia
Abu El-Hajs (2001) important account of territorial self-fashioning in Israeli society.
We should note, too, and partly as a result, that the question of how well governmentality
theory travels has been discussed recently by scholars including Chakrabarty (2000),
Kalpagam (2000) and Mehta (1999). The focus of this book is less on the sovereign
and disciplinary nature of colonial governmentalities than it is on the different forms of
governmentality that are (or can be) embedded in a state that is ostensibly committed to
the belated production of development and a more expansive conception of citizenship.
We are grateful to Steve Legg for prompting on this point.

Seeing the state


are pushed forward by an authoritarian state that is faced by a prostrate

civil society that lacks the capacity to resist [its] plans (Scott 1998: 5).4
The significance of Scotts work lies in its capacity to link the brutality of the twentieth century to certain schemes to improve the human
condition that failed. We will later want to consider Scotts account of
intentionality in the light of James Fergusons more avowedly Foucauldian
account of state failures and successes in southern Africa.5 But Scott is
surely right to insist that much of the violence of the twentieth century
was a result of utopian visions turning into dystopian realities, and predictably so. The forcible imposition of state simplifications in place of the
practical knowledge of urban dwellers and peasants facilitated a dizzying
and sometimes terrifying descent into tunnel vision. This was horribly
on display at the time of the Great Leap Forward in China. Amartya Sen
has shown how this monstrous episode gave rise to a famine which killed
more than thirty million people between 1958 and 1961.6 The famine
was caused by an administrative culture that discouraged officials from
reporting crises of food production and exchange entitlements. The fact
that the state turned a blind eye, in other words, and put blinkers on
the news media, proved in this case to be as dangerous as those cases
of monocularity where agencies of the state focused with deadly intent
on supposed threats to the purity of the nation. The famine in China,
together with various unrealizable attempts to tame nature, should then
be seen as state-produced disasters that parallel the unspeakable acts of
state violence that were fashioned by the Nazis or Pol Pot, or by those
Hutu politicians in Rwanda who used discourses of tribalness or ethnic
cleansing to encourage the slaughter of Tutsis and to advance their own
claims over what Frederick Cooper has called the gatekeeper state.7
The scale and continuing occurrence of these tragedies reminds us that
very many people experience the state precisely and perhaps lastly as a
source of physical violence.8 Even in India or perhaps especially in India
given stereotypical accounts of pacific Hindustan it is important to insist
on the physical nature of the violence that structures many exchanges
between state agencies and the people. Stanley Tambiah and Thomas


We are mindful that a full-blown pursuit of laissez-faire can also generate damaging crises
and contradictions. Marx and Engels (1967 [1848]), Polanyi (2001 [1947]) and more
recently John Gray (2001) have all made this point. We see no reason, however, to endorse
Susan Buck-Morsss extraordinary claim that liberal democracies are as likely to produce
social catastrophes as are systems of state socialism (Buck-Morss 2002: chapter 1). BuckMorsss work on Walter Benjamin and the dialectics of seeing (1989) speaks directly to
several of the concerns of this book, but it cannot be helpful to so blithely equate what
she calls the mass utopias of the East and the West. See also Lilla (2001).
6 Sen (1989).
7 Cooper (2002).
Ferguson (1990).
Sontag (2003: 60) further reminds us that Pol Pots murderous regime made many of its
victims pose for the camera before they were executed. Stalin also used the camera in this
way, as an official eye of the state.


The state and the poor

Hansen have reminded us of the continuing role of state-sanctioned physical violence in the production of urban space and politics, whether in New
Delhi at the time of the anti-Sikh riots in 1984, or in Mumbai under the
Shiv Sena.9 Paul Brass, too, has repeatedly drawn attention to the use
of physical force by Indias police forces.10 Ayesha Jalal, meanwhile, has
focused on New Delhis claims to a monopoly over the means of violence,
legitimate or otherwise, in its dealings with its rebellious peripheries in the
northwest and northeast. In her view, New Delhi has constructed a form
of democratic authoritarianism that has a great deal in common with
the military-bureaucratic authoritarianism of Pakistan and Bangladesh,
notwithstanding the meticulous observance of the ritual of elections in
India (Jalal 1995: 249).11
This book will certainly have something to say about state-directed
acts of violence against men and women in the Indian countryside. But
the violence of the state is not at the heart of our concerns, at least not
in the sense of physical violence, or violence that involves legal forms of
coercion, such as eviction orders. In part, this is because we have not
worked extensively in Kashmir, or Punjab, or along the Narmada river
valley, and we cannot hope to write a book about rural India as a whole.12
But even in those areas of India where the army is in occupation, or where
forms of rule owe more to ideas of sovereignty than to governmentality,
it is still the case that men and women seek to engage with the state
as citizens, or as members of populations with legally defined rights or
politically inspired expectations.
In many parts of rural eastern India, as we shall see, the problem is
not that the state sees too much, which is Scotts concern, but that it sees
too little. In the familiar phrases of development studies, the countryside
suffers from state failure or a lack of good governance. And yet even in
areas where government offices are badly run down, or where the forest
guard has to watch his back when walking his beat (another indication
of the fragility of the state), some people will be aware that they have
been defined as members of Scheduled Communities. They might also


On Delhi, see Tambiah (1990); see also Selbourne (1977) and Tarlo (2001) on the city
under Emergency rule. On Mumbai, see Hansen (2002). More so than Tambiah or
Brass, Hansen draws on Lacan to make a broader argument about the constitutive and
even pleasurable role of violence in the structuring of everyday life. We return to this
argument in chapter 7.
Brass (1997, 2003). See also Varshney (2001).
See also Vanaik (1990). We shall explain later in the book why we take a less dismissive
view of the role of elections in India.
Evictions of adivasi households to make way for quarries, mines or dams have been
extensive in Jharkhand, of course, and are reasonably well documented: see Areeparampil
(1992), Corbridge (1993a).

Seeing the state


know they can make claims on reserved jobs in government or the public
sector. Others will know that they have been labelled as BPLs (households
below the poverty line), and that they qualify for employment assistance
or subsidized food. And we might presume that rather more people will
know that the government (sarkar) has some responsibility to provide
villages with schools and standpipes, and perhaps even housing under
the Indira Awas scheme, or rudimentary health- and child-care facilities.
These people, in other words, have begun to imbibe the biopolitical discourses of the state itself, and its attempts to seek legitimacy precisely
through its wars on poverty and backwardness.
This is also the terrain of development and its modern form of knowledge, development studies. It is here that we want to contribute to a
deeper understanding of how the state works and is seen in parts of rural
India. In the rest of this chapter we have three objectives. We want, first,
to develop a typology of the ways in which different groups of the rural
poor might be said to see or to encounter the developmental state. We
learn a great deal about the state by examining its changing protocols
for bounding the poor, and its plans for seeking their development, protection, empowerment or erasure (see chapter 2). For their part, the poor
in India learn to see the state through their meetings with particular government officers, and with regard for those government conventions and
policies with which they gain familiarity, and this brings us to our second
and third objectives.
When a widow goes to the Block headquarters to collect her pension
she makes contact with the state in the form of a lower-level official and by
entering a designated building.13 For example, she might be required to
sign a particular piece of paper on an officials desk. But these encounters
are rarely conducted as the rulebook says they should be. The widow will
often be kept waiting for hours in the sun or the rain, and she might
have to call on a relative or fixer (dalaal or pyraveekar) to get her business
moving. Small payments (baksheesh or ghus) might also have to be made to
the accountant and/or his peon, and sometimes the payment she receives
will be several rupees short. The widow might expect this, although her
expectations will vary from place to place according to the conventions of
political society (as we show in chapter 6). The point is that she will have
learned to see the state not just through her own eyes, but with regard to
wider understandings of government.
Recent work on the anthropology of the everyday state and society in
India has begun to question the view that lower-level state personnel share

As she would have done in Bihar in 1999/2000, at the time of our fieldwork. On pensions and the post office system, see the interesting recent article by Farrington et al.


The state and the poor

the elite understandings of government that have been internalized by

some well-placed makers of public policy. Our second objective will be
to review these debates. We shall rehearse the argument that lower-level
state officials hold vernacular understandings of government that find
little place for ideas of fairness or generalized morality. We shall also
consider more general arguments about the embeddedness of the state in
We should also note, however, that the widow we have referred to
will sometimes counter vernacular accounts of the state by advancing
a more rule-based understanding of her own. She is able to demand
her pension, and occasionally to stand her ground, precisely because the
state has defined her as a citizen with rights, and because it has given
her scraps of paper to prove her entitlement to welfare benefits. It follows that our third objective will be to sketch out the sources of these
understandings, and to link them to changing discourses on the civil and
human rights of individual subjects. Our emphasis here will be on recent
debates in development theory and policy. In particular, we will focus
on questions of governance, accountability, corruption, participation and
empowerment, all of which we return to in the more empirical parts of the
Differently poor, differently sighted
Most people do not see the state as a Weberian aggregate, but this is
not to say it doesnt happen. We will have cause in this book to report
understandings of the state that come close to this. Among the Musahar
communities of north Bihar, for example, where females suffer from
especially high rates of social exclusion, and where there is little in the
way of the political representation that one finds among Paswans, it is
understandable that the state should be defined only in hazy terms and
on the basis of a limited number of direct contacts. The state appears
to function here as much as an absence as a presence. In certain adivasi communities, too, in Ranchi District, Jharkhand, a long history of
direct rule through mankis, mundas and now mukhiyas has reinforced
an experience of sovereignty that reaches back to Agency rule under
the British and which is carried forward under the Scheduled Areas
For the most part, however, the different experiences of different
groups of poor people with different state agencies should caution us
against a reductionist understanding of statepoor encounters. The
recent emphasis upon social exclusion in the expert literatures on poverty
is one sign of this diversity, even as these new discourses themselves define

Seeing the state


new populations of the poor. The same would be true of participatory

poverty assessments. As we shall see in chapter 2, the Government of
India has distinguished itself by its insistent attention to the variegated
nature of poverty in South Asia. The extraordinary range of its antipoverty programmes is testimony to this, including as it does area-based
interventions, employment guarantee schemes, compensatory discrimination, resettlement programmes, provisions for women and children,
and group-based participation schemes. And then there are its more
general programmes of educational and health-care provision, and the
enactment of laws to protect people against encroachments on common
property resources, say, or the underpayment of wages.
The production of these multiple sites of statepoor encounters, or of
poverty itself, is something that will concern us greatly in the chapters that
follow. It is not our purpose to lump together experiences that properly
should be kept apart. Nevertheless, it will help our later accounts if we
first deepen our understandings of what we mean by an encounter with
the state. What are the stages for these encounters, and what should
we look for in the performance of these exchanges? There are several
ways to approach these questions and it is mainly for reasons of exposition that we have chosen to deal with them under the separate and
rather mundane headings of when, why and who? and where and how?
Clearly, these fields are interwoven, as our empirical materials will later
When, why and who?
As soon as we begin to think about the when and whys of poorer peoples
encounters with the state, we run into problems. These encounters will
vary to a significant degree from place to place, and over time. They also
depend crucially on which agency, or who, in the state is being seen.
In the remote interior villages of Singhbhum District, Bihar (now Jharkhand), it was common in the late-1970s for a forest-dependent villager
to see the state mainly in the person of a forest guard, or a panchayat
sewak, and perhaps also in the shape of occasional meetings called by
the Mukhiya (the elected head of the local panchayat). (The position of
village leaders, fixers and local politicians is something we pick up in
part II.) In Tamil Nadu in the 1980s a marginal farming family with children might have been familiar with village extension workers, and perhaps with local revenue officials, the Junior Engineer, and the person(s)
responsible for administering the Noon Day Meal Scheme. In Gujarat,
meanwhile, the gangs of migrant field labourers that Jan Breman has written about so movingly might have been most familiar with local policemen


The state and the poor

and the government labour officer.14 Next door, in Maharashtra, significant numbers of women might have formed an impression of the state
through their encounters with the Employment Guarantee Scheme.15
Given that one of the major arguments of this book concerns the importance of specifics, we might reasonably doubt the value of making general
arguments across the breadth of India. But this neednt stop us thinking
more purposefully about the whens and whys and whos of statepoor
interactions. For example, we can think in terms of typologies of the state
and political society, on the one hand, and what we might call the generic
basis of some of these exchanges, on the other. By typologies of the state,
we have in mind charts or tables that would aim to show the strength and
functions of the official state at different spatial scales. We will provide
such charts in part II when we look at the organization of the official
state in Bihar, Jharkhand and West Bengal, and we will try to embed this
information within a broader account of the agencies functioning in the
surrounding political society (including the CPI-M in West Bengal and
components of the shadow state). In terms of the generic basis of the
whens and whys and whos of statepoor exchanges, we can make a first
cut at the issue by considering exchanges bound up with entitlements
and economic flows (both more and less direct), and those bound up
with flows of information and people (again, more and less direct).
Entitlements and economic flows Consider, first, what we might
call the material or economic bases of statepoor relations. In a very
real sense, a poorer person sees the state most directly when he or she
registers a birth or death, receives a registration certificate, picks up a
pension or some other entitlement, takes a child to school (perhaps),
receives or does not receive electricity or clean water, is interrogated by
a police officer, and so on. Not all of these exchanges will take place
on a daily basis. To the extent that they are regularized, they can also
be weekly, monthly or seasonal. Some might be annual or even decadal
(the Census). Very often, too, they will be bunched and/or episodic (as
we explain below, when we consider the case of a tribal villager in Jharkhand dealing with the local Forest Department). Such exchanges are also
received and understood in very different ways. Johnny Parry reports that
unionized male workers in the steel plant at Bhilai (now in Chhattisgarh)
very often see the government in positive terms. What with reasonable

Breman (1985a, 1985b).

Another way of putting this is to say that different states have become identified with
different types of anti-poverty programmes employment creation in Maharashtra, for
example, nutrition in Tamil Nadu, education in Kerala, health issues in Rajasthan, and
so on.

Seeing the state


monthly wages, dearness allowances, bonuses, sick pay and paid holidays, it is perhaps not surprising that some members of this (admittedly
urban) labour aristocracy like to announce that there is no mother or
father like their public sector employer.16 We might suppose, too, that a
woman who receives her pension in full on the appointed day will form a
different view of the state than a woman denied these rights or courtesies.
By the same token, some people will form a jaundiced view of the state
precisely because members of a neighbouring family and not their own
kith or kin benefit from a system of reserved employment in the public
More importantly, perhaps, calculations about the economics of state
poor exchanges take place across a number of interlocking spheres. Some
are less immediate than others, and some depend on very different forms
of sight (from eye-to-eye contact, through newspapers, possibly even from
the Internet), which might be mediated by the comments of others or by
individual and/or collective memories. This much is evident as soon as
we consider matters relating to taxation or prices. It is a well-established
proposition of historical sociology that modern states emerge from the
need to make war, and that the legitimacy of those states depends on
their need to raise funds by taxation.17 Where rule is linked to revenues
in this way it will be linked to a broadening of the polity. No taxation
without representation, as the old saw has it. Direct taxation, in particular,
encourages a measure of scrutiny of the state by its citizens. The fact that
the bulk of direct taxes will be paid by better-off men and women also
lays the ground for discourses which urge the rolling back of the state, or
which complain about the excessive subsidization of some households
by others who have earned their incomes through hard work or risktaking. These discourses become the stuff of politics. They form the
terrain for battles over the meaning and purpose of government and its
responsibilities to its citizens. In a crude sense, too, they highlight the
tensions that exist between biopolitics (and the impulse to improve a
population as a whole) and neoliberal forms of governmentality (with
their injunctions in favour of prudence and self-reliance).
Another example concerns the terms of trade between agriculture and
industry, or the city and the countryside. Marxists insist that farmers and

Parry (1996).
Tilly (1975, 1985); see also Levi (1988) and Bobbitt (2002). In many parts of the world,
however, as in oil-rich states like Venezuela or Nigeria, the flow of funds between the state
and its population bears scant resemblance to this west European model of government
(Coronil 1997, Watts 2003). This is in large part because colonial or neocolonial forms
of rule made Europeanization difficult if not impossible, at least in the terms since
demanded by modernization theorists. See also Apter (1999) and Piot (1999).


The state and the poor

labourers have very different class interests, as in key respects they do.
But there is also evidence to suggest that labourers across India have been
successfully mobilized by richer farmers in support of the new agrarian
politics.18 This politics aims to pit an authentic rural India, or Bharat,
against a loose coalition of merchants, city dwellers and their government supporters. An urban-dominated state then comes to be seen as a
vampire that drinks the blood of the countryside, and which enforces
price-twists that damage the interests of rural producers and consumers
alike. The fact that that this depiction of the inter-sectoral terms of trade
might be inaccurate recent evidence suggests that the Commission
on Agricultural Costs and Prices (CACP) has been successful in lifting
the procurement prices of grains above the market rate, and Jan Mooij
has demonstrated that many poor people do benefit from cheaper food
through the system of Fair Price Shops is not the issue.19 Men and
women will come to see the state not simply through the prices they are
charged, but through the constructions of fairness that are imposed upon
them in contending political discourses.
Sighting is never simple or straightforward, even for people blessed
with 20:20 vision. Tom Stoppard made this point very clearly in his play,
Jumpers, where he had one of his characters enquire about how the sun
looked after the Copernican revolution. Did people still see it the same?
Did they still see it orbiting the earth? The answers, of course, were yes
and no, in that order. It did still look round and yellow, but it now seemed
more like a fix point around which the earth moved in orbit.20
The state in India can also take on this before and after appearance,
including in the realm of financial transfers. And this is not simply because
people see the state through the observations of others (politicians, media
people, NGOs, kith and kin), although these mediations are vitally important. Sight is also learned and based on past experiences, and many state
poor financial transactions do not follow the rulebook. Too many studies
of the geography of public spending in India fail to acknowledge the


See, inter alia, Bentall and Corbridge (1996), Brass (1995), Corbridge (1997), Dhanagare (1983), Hasan (1998), Lindberg (1995) and Nadkarni (1987).
See Varshney (1995) on the CACP. Mooijs (1999) account of the Public Distribution
System also makes the point that men and women see the state, in the form of the operations of Fair Price Shops, very differently in Karnataka (where the PDS is subject to a
good deal of corruption and elite capture) and Kerala (where accountability mechanisms
are more securely in place in civil and political society).
More precisely, and more elegantly: George (facing away, out front, emotionless); Meeting
a friend in a corridor, Wittgenstein said: Tell me, why do people always say it was natural
for men to assume that the sun went round the earth rather that the earth was rotating?
His friend said, Well, obviously, because it just looks as if the sun is going round the
earth. To which the philosopher replied, Well, what would it have looked like if it had
looked as if the earth was rotating? (Stoppard 1972: 75; emphases in the original).

Seeing the state


reverse financial transfers that send monies or goods from poorer people to politicians and government servants. Corbridge and Kumar have
reported the case of Polus B, an adivasi smallholder and teacher in
Ranchi District, Jharkhand, who in the 1990s sought permission to cut
down ten jackfruit trees on his homestead land.21 Polus B wanted to
sell the trees to finance a small enterprise he had in mind, and he was
legally entitled to harvest the trees once he had gained the permission
of the Revenue Circle Officer and the Divisional Forest Officer (DFO).
But therein lay the problem. Although the value of Poluss trees was
80,000 rupees, net of logging and transportation costs, he ended up selling them to a dalaal for just Rs. 20,000. Polus B knew full well that it
would cost him an awful lot of time and money to get the permissions
he required. Some of his friends had made thirty trips to Ranchi to gain
one audience with a DFO. He might also be faced with illegal demands
for money at police checkposts on the road leading from his village to the
timber depots in Ranchi. Better to let the dalaal take the risk and bear the
Corbridge and Kumar reckoned that the dalaals final share of the trade
was of the order of Rs. 26,000. The remaining Rs. 34,000 disappeared
into the pockets of officials in the forestry, revenue and police services
(see table 1.1). Some of that money would later make its way to politicians. Politicians have to be able to fund their campaigns, and government
officers in Bihar and Jharkhand need to secure their next postings. We
comment further on these secondary transfers in part II. Our point here
is that we should expect poorer people in rural India to form their
accounts of the state with regard to some complicated and crosscutting
geographies of financial exchange. These geographies must then become
the objects of ethnography in a very exact fashion. We learn about the
state about its different boundaries, about its workings, about perceptions of it precisely through case studies.
Information and people This will also be the case when we come to
non-financial exchanges. Although our typology of statepoor exchanges
cannot hope to be exhaustive, it should be clear that these exchanges
include flows of information and people. In each case the flows will be in
both directions. In the case of information, agencies of the state engage in
regular exercises to extract information from and about its populations.
The capacity of the state is defined by these exercises. Gerard OTuathail
reminds us of the fate of one of the first English mapmakers of Tyrone

Corbridge and Kumar (2002).


The state and the poor

Table 1.1 Rent-seeking in the tree trade


Dept. (Circle




Forest Dept.
wing (Range
Clerk (500)

C.I. (1,000)

B.O. (1,000)

Circle Officer

Check Posts

Forest Dept.

Police Dept.

Other expenses


Officer in



Check posts

Daily expense
and transport
for himself @
150 40
trips (6,000)
Transport for
officers @
350 3 trips
Logging and


Summary: Estimated Net Receipt from the State Trading Office: Rs. 100,000; Expected
Total Expenses: Rs. 54,050; Payment to Owner: Rs. 20,000; Expected Profits for Dalaal:
Rs. 25,950 or Rs. 650 per day of labour.
Note: the data here are based on the Polus B case, but the model should not be construed
as a direct representation of that encounter; rather, it should be seen as a generic model.
This version of the model assumes that the dalaal will log the tree himself and deliver the
timber to the FD depot. The costs would be different were he to employ the FD to do the
(a) This payment will often be in the form of a political donation or other favour.
(b) The Range Officer might use some of this money to procure genuine administrative
facilities which are not provided by the bureaucracy because of a lack of funds.
Source: Corbridge and Kumar (2002: 778).

Seeing the state


District in the northern reaches of Ireland: he was attacked and had his
head cut off. The people refused to be mapped, at least until they had
been beaten into submission.22 Similar acts of refusal have been recorded
in India, and in the United States in 2000 the rate of nil returns to the
Census mapped out a veritable geography of resistance to the state that
peaked in the black inner cities and in various fastnesses of the west and
southwest. People were more likely to make themselves known to the
Census takers in middle-class suburbs and in the Germanic states of
the north, including Wisconsin and Minnesota.
The Census remains one of the principal conduits of biopolitics in rural
areas of India. Men and women experience it very differently, even so, in
different parts of the country. In many cases, women do not experience
the Census at all, at least not directly. Whether the schedules are administered by local officials, elected representatives or schoolteachers, the
identified head of household is almost always male. Women experience
the state through the stories of their husbands or male relatives. The states
preference for dealing with poorer people on the basis of defined households leads to similar maps of inclusion and exclusion when the flows of
information are circular. As we report in chapters 3 and 4, some women
in our field sites were in possession of cards that confirmed their eligibility
for work under the Employment Assurance Scheme (EAS). Others had
certificates confirming their status as members of the Scheduled Castes
or Tribes. These cards or certificates, however, are usually made available by the state only after it has first collected information from the
household unit. EAS cards, for example, are supposed to be allocated to
members of registered labouring households, and it is households that
are defined as Below Poverty Line (BPL) on the basis of periodic forays
into the countryside by government officers.23
On other occasions the flow of information will be from the state
to the population, but here too we need to be alert to the modalities
of the exchange. We also need to pay attention to the way that information is received, translated and understood. John Reids Final Report on the
Survey and Settlement Operations in the District of Ranchi, 19021910, published by the colonial state in 1912, noted in passing that forest offences
were to be policed according to an edict which proclaimed that sakhua
[sal, or shorea robusta] trees could not be felled if they were of a girth
of twenty-seven inches or more at a height three feet from the ground
(Government of Bihar and Orissa 1912: 129). This information was duly

OTuathail (1996: 1).

The military or expeditionary metaphor is overstated, of course, but not without
resonance: see Driver (2000) for an interesting discussion of this sort of geography


The state and the poor

gazetted, in English. Even where the information was passed to local

Mundas or Mankis in Hindi, Mundari or Nagpuri, one might doubt that
a reference to twenty-seven inches meant a great deal. Villagers and forest
guards would need to translate this command into local conceptions of
girth, and in the process a space might open up for misunderstandings
or police actions. (Recorded forest offences in the Chaibasa and Kolhan
reserves, a little further south, peaked significantly in the years 1912/13
to 1915/16.)24
The edict on sakhua fellings is an example of an indirect flow of
information from the state to the population. Most of the villagers who
needed to know the new rule would probably never be aware of it. The
state made no effort to contact households or individuals directly. Matters are clearly very different when the state provides information on
crop prices to farmers across the airwaves, or when it posts bills outside the Block Development Office, or in villages or tolas (hamlets).
The experiences that men and women form of the state will be significantly different in each of these instances, but in each case the medium
of exchange allows for a more direct sighting (or sounding) of the state.
In those parts of rural India where even poorer families now have access
to TV sets, perhaps powered by a car battery where there is no electricity supply, the possibility also exists for what Rajagopal has called,
in the context of the screenings of the Ramayan by Doordarshan that
began in 1987, a collective libidinal experience.25 This mode of experiencing the state reaches back to conversations that people might have
on the basis of shared readings of a newspaper. Whether it also anticipates those sightings of the state that might be provided by Internet
access along the lines perhaps of the panchayat-level computer booths
that have been promised in Andhra Pradesh and Madhya Pradesh is
a moot point. Much will depend on who accesses the booths, and how
they share their experiences. If members of the rural poor do gain access
to computer booths, will they mainly be younger males? And what consequences might this have for information retrieval, circulation and even
It is likely that the Internet will change poorer peoples experiences
of and reactions to the state, just as new technologies have done previously. Chapter 7 will discuss the accountability campaigns waged by the
Mazdoor Kisan Shakti Sanghatan (MKSS) in Rajasthan, where social
activists have put pressure on local government officials to place more
of their budgetary information in the public domain. If villagers can see
how their (the publics) money is being spent, and on what, they can
more easily challenge unfair patterns of government funding. To this

Corbridge (1993a: 141).


Rajagopal (1994: 1662).

Seeing the state


end the MKSS has committed considerable resources to the purchase of

photocopiers. Photocopying allows for a sighting of the state that is continuous and more or less permanent. The retrieval of information about
the state does not depend on impromptu conversations, or the memories
of one or two individuals who have coaxed information from government
officials. Memory is provided as paper copy and provides the basis, in
this case, for challenges to the state that make use of the states own data.
The Internet, in principle, should allow for more of the same, particularly
where Internet access is connected to a printer. It should also allow for
more immediate responses to the state. These responses might take the
form of complaints under a citizens charter. They might also extend to
alternative websites that seek to name and shame allegedly corrupt public
Here, of course, we rub shoulders with the populist rhetoric of some
politicians and activists. Notwithstanding the potential for empowerment
that the Internet might one day offer, it is important to insist that Internet
access in India will remain uneven for years to come, and that Internet
usage is rarely the unmediated activity that some of its proponents believe
it to be. Peoples use of the Internet will continue to be shaped by the
information they receive from other sources. If we want to understand
what is happening in rural India we will need to couple an understanding
of information flows to an understanding of flows of people.
This is true in at least three respects. Strong states are defined by their
ability to set and police their boundaries. Citizens see the state through
a system of passports and visas, and with regard to the state officials that
monitor their movements. It might be thought that these geographies are
at some remove from the life-worlds of poorer men and women in rural
India, and very often this will be the case. In one of our field sites, however, in Old Malda district, West Bengal, which is close to the border with
Bangladesh, the Border Security Force looms large in local imaginaries of
the state. We can presume, too, that the states ability to command movement would have impressed itself on those thousands of men and women
from Chota Nagpur who were shipped to the tea-gardens of Assam at the
turn of the nineteenth and twentieth centuries.
The mass of the population also experiences the state at home. To
put it another way, the citizenry sees the state not just in terms of flows
of power, money, commodities or information, but also in terms of the
circulation into their domains of the men and women who represent (or
who can be made to represent) the state. When we talk about seeing the
state we need to press closely on whether and how certain individuals
are considered to be state employees (teachers, for example). We also
need to have regard for the career tracks of government servants. Some
will be peripatetic and highly trained, like Indian Administrative Service


The state and the poor

officers. Others will not be. Assistant engineers or accountants in a Block

Development Office in Bihar might remain there for many years. Most of
them will have learned their jobs by shadowing others or simply by doing
it.26 Local people can be expected to form their accounts of the state with
regard to their encounters with these and other differently placed public
They will also see the state, finally, through the movements and activities of local fixers and political leaders. We shall have more to say on
this later on. For the moment we should note that more and more of
these leaders will be women and members of the Backward and Scheduled Caste populations. Legislation to ensure the reservation of seats
in the various tiers of Indias polity is working to promote a different
composition of the political state. Poorer men and women can expect
to have exchanges with politicians and government servants from more
diverse social backgrounds. How these exchanges are staged, however,
can matter as much as the exchanges themselves, and can tell us a great
deal about peoples perceptions of the state (both as outsiders and
How and where?
The performativity of government business and politics is not a secondary
matter when it comes to considering how people see the state. Consider
the provisions attached to the receipt of workfare benefits in some jurisdictions in California. Women seeking benefits in Riverside are required
to turn up to interviews with service providers not just at the appointed
time, but according to an approved code of dress and presentation of the
self. According to Jamie Pecks description, the woman must not wear
excessive make-up. Her hair must be tied behind her head. Her fingernails must be clean. She must not wear a short skirt or spandex pants,
or jeans, shorts or tennis shoes. She must not chew gum. And she must
agree in principle to take a urine test for signs of drug use (Peck 2001:
Peck doesnt tell us whether the supervisor is also required to dress in
a particular fashion, be it lounge suit, smart-casual or uniform. But his
basic point is well made: the local state in Riverside is using its command
over money to exert fierce control over the body corporeal. Of course,
not all exchanges between states and poorer people will follow this model.

Karen Coelho also refers to engineers in Chennais Metrowater who learn the job on
the ground (2004: 6). We came across this excellent paper as we were preparing the
final version of our book. We are grateful to Lalli Metsola for drawing it to our attention.
See also chapter 5 for further discussion.

Seeing the state


There will be times when the state is meant to respond to the instructions
of a beneficiary population.27 We see this very clearly in part II when we
come to a discussion of the Employment Assurance Scheme in eastern
India. EAS legislation is written in such a way that members of registered labouring households are able to demand employment from local
government officials. Village-level meetings should also be held to select
the agents who will execute approved projects. The state is then to be
informed of the decision.
The point we wish to make here, however, is a slightly broader one
about the staging of statepoor encounters. Dress codes can matter. The
wording of exchanges certainly matters. Who gets to speak when and in
what tones? It might also matter where the business of state is staged. Is
it always in a government office, as we might expect it to be in those
countries that have developed scientific forms of government? How
might people see the state if the distinction between home and work
is blurred? How, too, might officers of the state see their own buildings,
in terms, for example, of the circulation of files, or the reproduction and
government of hierarchical relations between staff members? And how
might these sightings affect the way that government business is carried
We will deal with some of these questions in chapters 36. As we
said before, the devil is in the detail. But one way to sharpen the narrative is to think about something as humdrum as the queue, or what
Americans call waiting in line. More even than the question of dress and
self-presentation, the causes and significance of queueing (and queuejumping) are hugely neglected in the social science literature, and yet
they have a great deal to tell us about how social encounters are structured.28 Consider, for example, how men might wait their turn to jump
into the barbers chair in London or New Delhi on a Saturday morning.
On such occasions, queues express not only the scarcity of a resource (in
this case, of barbers to customers), but also an ideology of equality, albeit
one that is mediated by a shared capacity to buy a common good in the


There is some evidence, too, that government agencies and officials have increasingly
to respond to complaints from the public, especially in urban areas (including urban
slums) where there are now well-defined expectations that water and electricity, for
example, should be provided by the state. The fact that these expectations are often
frustrated not least because of plans to charge customers for privately provided services leads to intense struggles in political society and around the meanings and sightings of the state. At the same time, a culture of complaint points in the direction of the
more active citizens that government reformers have been calling for. More research is
needed on the making of complaints by different individuals and social groups, and on
the ways these complaints are handled (accepted, avoided, resisted, deflected) by public
servants. Coelho (2004) provides important pointers. See also chapters 5 and 7.
For a preliminary discussion, see Corbridge (2004).


The state and the poor

private sector.29 If someone mistakes his turn in the queue by one place,
he will be let off with a humorous rebuke. But if someone tries to jump
the queue openly, and by two or more places, he would most likely be
asked who do you think you are? His behaviour would have breached a
form of governmentality (the self-regulation of conduct) that marks out
all customers as equals.
In many other settings civility will go out of the window, and not
just in New Delhi. Take a trip to any railway station in London during rush-hour and you will see middle-aged Englishmen barging women
and more elderly people out of their way to secure a seat that they know
instinctively they are not in line for. But if claims about English civility
need to be taken with a pinch of salt, it might still be argued that the
English have a respect for queues that is not often to be found it Italy,
say, or Miami, and which is rarely to be seen when men and women
from poorer communities try to meet the state in eastern India. A sense
that might is right is far more common, and is regularly on show in and
around state buildings in rural eastern India. We learn a lot about the
state, about how it works and is seen by different people, by attending to the patterns of spatialtemporal behaviour that men and women
engage in to make contact with sarkar, and to conduct their business
with it. How many times does a person have to turn up at the Block
Development Office to see the BDO or a Junior Engineer? How long
does he or she have to wait on any given occasion? How often do they
observe others getting ahead of them? How do they respond to this? How
are they treated when they meet an official of the state? How are they
addressed? How is their use of time acknowledged or respected? Which
rooms are they allowed in? Are they allowed in as citizens, or must contacts be forged by a broker? Are they seen as a nuisance or member of
a troublesome social group? How do officials deal with one another,
or with the brokers and politicians who might exert pressures upon
These are some of the questions that need to be asked when we talk
about seeing the state. They can be added to questions we raised earlier about dress, language and the presentation of the self, all of which
are highly gendered. But there is also the matter of the geography of the
state itself (as opposed to patterns of access to it). Later in this book we
will comment on the layout and repair of government buildings, and on
the passing of files between government servants. If we want to engage
debates about the capacity of the state, or state failure, it helps to have in

Marx once described money as a great cynic and leveller, and what we observe here is
an effect of the equality that money in one sense confers.

Seeing the state


mind a sense of the physicality of the state and its resources. But there is a
prior question here as well. Just where does the state begin and end? How
should we think about the statesociety distinction where a significant
amount of state business is transacted on the verandah of a government
officers private residence? More to the point, how do different groups of
people in rural India make sense of these entangled geographies? Raising
these questions brings us face to face with a growing literature on the
anthropology of the everyday state and society in India, and with questions about how the state is seen by those who are in its employ. It also
raises questions about the territoriality of the state and the politics of
State and society: embeddedness, scarcity
and territoriality
In some parts of Africa we might want to make sense of a raucous geography of queuing in terms of a model of the absolute scarcity of the state.
In his controversial account of states and power in Africa, Jeffrey Herbst
argues that the failure of many regimes in the region is to be found in that
peculiar combination of circumstances which brought localized polities
to power at a time when the international community insisted on dealing
only with nation-states.30 The leaders of these polities were able to use
foreign aid to strengthen their control over the focal points of their newly
independent countries, but they were sometimes unable to extend their
control of territory much beyond the capital city and its environs. The
low population densities of rural Africa also conspired against the efforts
of some regimes to impose a monopoly over the legitimate use of force
within a given parcel of territory.
In India, however, notwithstanding prolonged military incursions in
the northeast and Kashmir, what might be called the scarcity of the state
is best understood in relative terms. Although we shall meet severe cases of
state depletion in parts of Bhojpur District, Bihar (a Naxalite heartland),
or in Malda District, West Bengal, for the most part the developmental
state is well entrenched and is underpinned by the All-India Services and
by the far greater number of men and women working for their state
equivalents (the Bihar Administrative Service, for example). In part this
reflects the legacy of European colonialism in India, but it also reflects
the considerable efforts at nation-building by Sardar Patel at the time
of Independence and by Nehru in the 1950s. India was made to hang

Herbst (2000).


Corbridge and Harriss (2000: chapter 3).


The state and the poor

The relative scarcity of the state in India has generally been approached
in terms of large-scale models of the contradictions of Indias political
economy. In the work both of Pranab Bardhan, and Lloyd and Susanne
Hoeber Rudolph, the state in India is said to have been captured by various demand groups, including organized labour, well-paid bureaucrats,
and bullock capitalists/richer farmers.32 It is then unable to prosecute the
politics of command that has characterized the developmental states of
Southeast and East Asia. At its best, the state in India comes to be defined
by those far-reaching mammaries of welfarism that have been satirized
by novelists like Upamanyu Chatterjee and Siddhartha Deb.33 The state
confers the blessings of consumption upon those who are able to access
and milk it. At its worst, the state simply fails to work. Unable to raise taxes
from those who should be required to pay them, elements within the federal state turn instead to deficit financing or fail to pay several thousands
of people who are in their employ. By mid-2003, many employees of the
State Road Transport Corporation in Bihar had received less than fifteen
days salary since 1994. Small wonder, then, that state officials fail to
show up for work, or make their incomes by selling their services to those
who can pay. The privatization of the state has probably gone further in
Bihar than in any state in India, but not for the reasons announced by
economics textbooks. People are bypassing the state because it is unable
to deliver the supplies of water, electricity or security that they need, and
many officials are making their incomes by providing these services by
other means. The absent teacher who provides private tuition is one case
in point. The looting of the states supplies of medicines is another. As
Krishna Ananth reports, Medicine packs bearing marks indicating that
they are supplies to the Health Department are available for sale with
chemists in Patna and elsewhere in Bihar (Ananth 2003: 13).
Examples such as these can be multiplied across India, and point us
towards a body of literature that is consistent with the models of political
economy but which is more directly concerned with sightings of the state
by government officials themselves. This tradition of writing reaches back
at least as far as F. G. Baileys work on the local state in Orissa. On the basis
of prolonged fieldwork in the Kondmals, Bailey was able to identify the
roles played by richer peasants and village faction leaders in bridging the
worlds of the state and the locality. Although most villagers preferred to
keep the state at a distance, there was by the 1950s a general appreciation
that this was not always possible, and that the village should make efforts
to draw down state funds for a local school or post office. Perhaps more

Bardhan (1984); Rudolph and Rudolph (1987).

Chatterjee (2003); Deb (2003).

Seeing the state


pertinently, however, Bailey observed that, the link between Bisipara [his
research village] and the Administration is the single thread of imperium.
No-one in Bisipara is mothers brother to the Deputy Commissioner.
The social roles of the administrators and the men of the village do not
overlap. Even caste is irrelevant (Bailey 1957: 248). He continues:
The division persists inside the Administration, as one would expect, since the
Administration is an organization and not a community. Those who are recruited
locally as policemen or messengers remain members of their village communities
and retain the outlook of a villager. Their attitude to the government (Sircar) is
fundamentally the same as that of the ordinary cultivator. Their loyalty remains
with the village, and this applies even to the headmen . . . There is, in fact, a parody
of the four castes of Hinduism. In this parody there is the Gazetted Officer caste,
the Non-Gazetted Officer caste, the Babu (clerk) caste, and the rest, comprising
the menials in the Administration and the villagers. In the Kondmals they do
not inter-dine and they do not intermarry, and it is very hard to get from one
class to the next above. There is only one [local-born] Gazetted Officer and
[he, a university graduate] is something of an outsider, since his grandfather, a
Christian, came to the Kondmals in the service of the Administration. (Bailey
1957: 2489; emphases in the original)

Although Bailey plays down the importance of caste, it is clear that his
account of the relative scarcity of the state has much in common with
a more recent literature on statesociety interactions. This is so both in
terms of the territoriality of the state and what Benedict Anderson has
called a sense of the imagined community.34 One of the great conceits
of government is the suggestion that the writ of London or New Delhi or
Islamabad reaches without interruption from the commanding heights of
the state through the agencys central offices and dispersed field offices
to the trenches that are at the bottom of the state hierarchy.35 But this
will only rarely be the case. Far more often, the men and women who
populate state agencies are mindful not only of the rulebook and their
supervisors, but also of their need to live and work with their peers and
with those they are meant to serve, as well as with their representatives.
The forest guard to whom we alluded earlier might well be charged with
responsibilities for forest management, and might indeed have coercive
powers that he (or more occasionally she) can bring to bear on villagers.
But the beat officer also has to live locally, and to this end he needs to
develop the skills of a street-level bureaucrat. Failure to do so, as Vasan
explains, can result in any number of difficulties, from problems in finding
food or accommodation to the risk of attack in the depths of a forest.36
The forest guard thus comes to see the state as a complex organization

35 The phrasing here is after Migdal (2001: 11821).

Anderson (1983).
Vasan (2002). See also Lipsky (1980) on street-level bureaucrats.


The state and the poor

buffeted by contending social forces. This sighting conditions the way

that he deals with his charges, and with how they in turn come to see the
The relative scarcity of the state, however, is not simply a matter of
local resistance to its attempts to put down roots that are independent
of political society. According to Sudipta Kaviraj and Partha Chatterjee,
the workings of the local state must also be understood in terms of a
model of embeddedness that highlights a sharp disjuncture between elite
and vernacular understandings of the state-idea. Their analysis takes on
a Marxian hue that we dont find in Bailey, drawing as it does on Gramscis ideas about passive revolution. Kaviraj and Chatterjee maintain
that the weakness of the bourgeoisie in India at Independence was sufficiently profound that it had to seek the capitalist transformation of the
country with the help of a rising rich peasantry and a supposedly progressive state. The vehicle for this model of structural transformation
was the Planning Commission, and the idea was circulated that citizens
should Place [their] prayers at the feet of the sarkar, the omnipotent and
supremely enlightened state, [where they would] be duly passed on to
the body of experts who [were] planning for the overall progress of the
country (Chatterjee 1986: 160).
The problem was that most ordinary Indians refused to play the game.
Not only was their experience of government very different sarkar being
resented where it was not seen as a source of immediate funds or as a site
for venality or of the absurd but they had few expectations that it should
behave in this modern fashion. Kaviraj maintains that the upper echelon
of the state-bureaucratic agency was infused with a colonial mentality that
separated it from the life-worlds of the social majorities and the states own
lower-level officials. The failure of the state to secure its stated outcomes
was not simply a matter of resource scarcities, at least in a pecuniary
sense; it also reflected the fact that the state had feet of vernacular clay
(Kaviraj 1984: 227). The English-speaking elites who formed the shock
troops of the Nehruvian and post-Nehruvian states found their mandates being reinterpreted beyond recognition by the ordinary Indians
who worked very low down in the bureaucracy (Kaviraj 1991: 91). A
large number of these men and women, Kaviraj suggests, would no more
have thought outside the cellular structures of Indian social life a life
structured by family, kin, caste and community than they would have
conceived of their bosses as superiors only in terms of their position in a
graduating batch or the government Gazette. As Satish Saberwal further
observes, state institutions such as the courts and bureaucracies have not
had the normative support necessary for their reliable, effective functioning because their western logic does not command much of either

Seeing the state


understanding or respect on the ground (Saberwal 1996: 150; quoted

and discussed in Fuller and Harriss 2001: 9).
To the extent that the life-worlds of the superior and the subaltern do
depart in this manner it is vital that research pays close attention to the
language and staging of statesociety interactions. The relative scarcity
of the state might have as much to do with (mis)understandings as with
the distribution of resources, even allowing that the two spheres will be
closely linked. But there is a further sense in which the seemingly unequal
and sometimes humiliating encounters between citizens and state officials, or between officers within the state, are shaped by the production
of scarcity, and this has to do with politics. F. G. Bailey recognized this,
of course. His work is closely attentive to the ways in which members of
the Congress Party articulated the exchanges that bound the Kondmals
of western Orissa to a state and did so in a manner that did justice to
each sides needs and expectations. But it is only in the more explicitly
pluralist work of Myron Weiner, or of later writers like the Rudolphs (or,
indeed, Chatterjee and Kaviraj), that one finds an account of the politics
of scarcity that is sensitive to the effects on the state of the mobilization
of ascriptive identities.37 Perhaps the signal virtue of this body of work
is that it links the study of political economy to that of identity politics.
Weiners work also triggered a greater realization that the bottlenecks that
produced so much uncivil behaviour in or around government offices was
the result of scarcities produced by an inefficient economy and an overdeveloped polity. The weakness of the private sector in India propelled the
countrys new citizens towards sarkar for all manner of benefits and safeguards that the state could not meet in full or even in large part, and
which perhaps no state could ever meet. Politics then degenerated into
a form of competitive populism that pushed voters to seek the support
of those politicians who could best deliver the resources which the state
was meant to disburse objectively and without partiality. What Bailey
later called the civility of indifference gave way to forms of behaviour
characterized by rudeness, shoving and a heightened sensitivity to group
Weiners argument has recently been extended by Kanchan Chandra, and with considerable ethnographic acuity. In contradistinction to
Baileys earlier argument about the irrelevance of caste, Chandras work
in Uttar Pradesh shows how the mobilization of horizontal ethnic groupings like the Other Backward Classes (OBCs) or the Scheduled Castes
has been the most important vehicle for the capture and reworking of the

Weiner (1962); Rudolph and Rudolph (1967).

See Bailey (1996); see also Weiner and Katzenstein (1981).


The state and the poor

state at Block, District and even State levels.39 The extension of quotas
to the OBCs in the 1990s has given further impetus to the development of a patronage democracy in which access to state resources is
fiercely controlled by ruling-group politicians acting in a discretionary
manner. The difference now is that numbers are counting, and the
Forward Castes in north India are losing out, or are required to make
new alliances. In Mayawatis Uttar Pradesh, or in Laloo Yadavs Bihar,
this argument further suggests, more and more citizens are forced to
pay homage to their MPs or MLAs, or indeed to the highest-ranking
politicians in the state, if they want to secure even the most meagre
The perversity of these arrangements can be so great that Laloo Yadav
might seek to run down the system of public health-care in Bihar in
order that he or his acolytes can provide scarce hospital beds for their
supporters. In neighbouring Uttar Pradesh the chief minister can insist
that village leaders approach her (or him) directly if they want a new handpump to be constructed. The queueing behaviour that we began to sketch
out earlier is then produced by this deepening politics of scarcity, and not
simply as result of cultural understandings about hierarchy or equality.
Just as importantly, the pressures upon politicians to contest elections on
a regular basis are so strong that pressures are in turn brought to bear
on state officials from the trenches all the way up to the commanding
heights of the Secretariat. The power of politicians to transfer government officers is just one indication of the three-sided relationships that
hold between elected representatives, citizens and public officials. If we
are to understand how the state works, and how it presents itself to various
groups within the rural poor, we need to understand that the rulebook
and the training academies at Dehra Dun are just one source of its selfunderstandings.40 If the argument of this section is right, sightings of
and within the state take shape within regimes of relative scarcity that are
produced in three dimensions: by the uncertainties of understanding and
translation that structure exchanges between elite and vernacular groups;
by the inefficiencies of Indias public sector (with its tendencies both
to rent-seeking behaviour and real capacity constraints: see chapter 5);
and by the pressures that are brought to bear on government officials by
ethnic and other interest groups and their political representatives and

Chandra (2004). At the time of writing this book was unavailable to us, but we believe
that we are correctly summarizing one of its major arguments.
Albeit, these are powerful sources for the self-understandings of all-India officials, as we
confirm in chapter 5. The culture of batchmates and seniors/subordinates is not to
be underestimated: see Potter (1996).

Seeing the state


Mending the state? Views from outside

It would be wrong to suggest that the relative scarcity of the state that
we find in some parts of Bihar or West Bengal is replicated in the same
degree across India. We shall comment in part III on recent experiments
with decentralization in states including Kerala and Madhya Pradesh. At
the same time, however, we need to pause before dismissing out of hand
Vijay Nambisans suggestion that Bihar is showing the rest of India its
political future.41 Barbara Harriss-Whites corruscating analysis of society and economy in Tamil Nadu confirms that the criminalization of
politics is well established in parts of South India, where The State is
[also] used by the intermediate classes for accumulation rather than for
legitimation (Harriss-White 2003: 47). In her view, this large grouping of
self-employed business people and surplus-producing farmers depends
for its survival on the continued production of state subsidies for water
and electricity (most notably) and state restrictions on competitive market structures. These classes thus define themselves against big business
and economic liberalization, on the one hand, and against smallholders,
the landless and the working class, on the other. In Tamil Nadu they also
engage in precisely those social practices that we have come to associate
with Bihar or UP: economic misdemeanours of all sorts (including adulteration, arbitrary deductions, tampering with weights and measures);
economic crimes such as theft, fraud and unlicensed activity (including
the tapping of electricity and TV cables); mafianization, or the pursuit
of organized crime based on the privatization of physical security measures (usually with the direct or indirect involvement of MPs and MLAs);
tax evasion on a dramatic scale; and the continual oppression of labour
through low wage rates, causalization of contracts, the use of child labour,
and the watering down of health and safety regimes.42
Harriss-White has no truck with the idea that liberalization is the solution to the problems of the state and politics in India. In her view, the
continuing de facto privatization of the state in Tamil Nadu has been
enhanced by the doctrines of economic liberalism, which have only worsened a more fundamental problem of state depletion (what we have called
the production of the relative scarcity of functioning state institutions).
For our purposes, however, it matters rather more that the state in India is
now being produced amid a competing set of discourses which challenge
many previous assumptions about the ways that ordinary people should

Bihar is developing into one of the political possibilities open to democracy which
increasingly looks like coming to fruition, Nambisan (2000: 8).
This listing after Harriss White (2003: 64).


The state and the poor

be asked to encounter the state. Some of these discourses propose only

minor changes to the existing optic, as when parts of the organized labour
movement seek a divorce between the trade unions and the political parties that have traditionally held them captive. On other occasions we can
detect a more determined tack away from the state in favour of the market or the people (or even both), and since we shall meet these ideas
later in the book it makes sense to introduce them here. These ideas are
important because they help to define the sightings of the state that are
made by many of the people described within them.
Exit, Voice and Loyalty
We can usefully begin by revisiting a famous book by Albert Hirschman,
a scholar whose work defies crude attempts to pigeon-hole the discourses of development studies in terms of Left and Right. Hirschmans
account of Exit, Voice and Loyalty was first published in 1970 and returned
him to problems he had encountered previously in the Nigerian railway
system.43 What interested Hirschman was the fact that the Nigerian
Railway Corporation performed so badly even though it was faced with
competition from long-distance road haulage companies. In his view,
this paradox was explicable only in terms of the peculiar combination
of exit and voice that he found in Nigeria: exit did not have its usual
attention-focusing effect because the loss of revenue was not a matter
of gravity for management [which could dip into the public treasury in
times of deficit], while voice was not aroused and therefore the potentially most vocal customers were the first ones to abandon the railroads
for the trucks (Hirschman 1970: 45). Nigerians were then saddled with
an inefficient and oversubsidized public railway system, and an arena of
exchanges between officialdom and ordinary citizens which encouraged
an oppression of the weak by the incompetent and an exploitation of the
poor by the lazy (Hirschman 1970: 59). The endless delays that railway
users encountered, which began with long and perhaps unruly queues
to get tickets in the first place, was caused, finally, by a combination of
exit and voice [that] was particularly noxious (Hirschman 1970: 45) and
which made recovery unlikely.
In less subtle hands than Hirschmans this combination of exit and
voice considerations is reduced to one or the other, with little attention
being given to the ways in which the two can be combined to promote
loyalty. For many neoliberals the overriding concern has been to secure
a rolling back of the state in the developing world. The failure of the

Hirschman (1967).

Seeing the state


state is here diagnosed in terms of an excess of rent-seeking behaviour by

public officials, and an absence of effective competition from or within
the private sector. Ordinary people are bound to confront the state as
a site of inefficiency and corruption as long as they are unable to exit
from it. The promotion of quotas or affirmative action is precisely the
wrong way to empower poorer people, for it further protects state agencies from effective and generalized public scrutiny. What starts as a system
of compensatory discrimination designed to bring beneficiary populations up to the level of the average in ten or twenty years, is extended
ad nauseam by the creamy layers of the special interest groups that are
thereby produced. What is required instead is the concerted promotion of
employment opportunities in the private sector, and to this end a responsible state must first put in place responsible fiscal, monetary and foreign
trade policies. Empowerment, in this discourse, need not begin in the
locality or with policies that are directly focused on the poor (in the sense
of classic poverty-alleviation schemes). The geography of empowerment
rather begins at the international and national levels with the prosecution
of economic reforms. Disempowerment is a result of economic distortions, or distortions of an economic regime that would secure the maximization of individual utilities through unfettered markets.
At its crudest, a discourse of economic liberalism encourages a view of
the state as a dead weight, or, worse, as the promoter of economic unreason, special interests and the continued impoverishment of the masses.
Evan Osborne draws on this line of reasoning when he declares that
government-dispensed rents of the order of 3045 per cent of national
income are channelled through Indias reservations system and account
for the inevitable Balkanization of Indian politics (Osborne 2001: 679).
In this specific respect his analysis comes close to that of Kanchan Chandra, but it is not clear from Osbornes paper that he accepts Chandras
suggestion that the continuation of the reservations system will empower
(give voice to) a widening circle of ethnic groups in a second-best world of
limited privatization. In any case, there is now growing recognition within
the neoliberal camp that markets cannot be promoted in the absence of
effective structures of governance. This moment of recognition falls short
of Karl Polanyis insights about the institutional, and thus irreducibly
social, nature of real markets, but it has encouraged the World Bank,
especially, to discount its earlier support for shock therapy in favour of
sequenced economic reform initiatives that pay at least some attention
to the political conditions and consequences of liberalization.44 It has
also encouraged the World Bank to fund surveys of public officials, as it

Polanyi (2001 [1944]); see also Platteau (1994). See also World Bank (2001).


The state and the poor

has done in Uttar Pradesh, in an attempt to see rather better how and
why those officials might be persuaded to serve their clients in a more
transparent manner.
Accountability, participation and decentralization
Whether or not the World Bank will get its way on public service reforms
in India is a moot point. Central government has long since declared
its support for a gradual package of economic reforms, and if Barbara
Harriss-White is right it will be the vested interests of Indias intermediate classes, as much as the compulsions of centrist politics, that will push
the state to continue with a process of economic reforms that is distinguished by its partiality and uneven tempo. What is more certain is that
the World Banks strictures on good governance have been mimicked,
joined, critiqued and rejected by a range of non-state actors (including
opposition parties) that look at the problems of poverty alleviation and
empowerment from a more heterodox stance. What then obtains is a continuum of reform proposals or political initiatives that begin on the Right
and which work their way round to the point where Left and Right are
almost joined. Support for a strong exit option is only the most obvious expression of this tendency. Although post-developmentalism shares
little in common with the Washington Consensus, the former favouring
community where the latter favours markets, there is a strong measure of
agreement in their shared disdain for dirigisme, or the idea that states can
directly empower poorer people. In both cases, an agenda of state reform
is viewed with deep suspicion.45
The anti-state agendas of the radical post-Left have undoubtedly
coloured the perceptions of the state of at least some villagers in areas
like Uttaranchal (the Chipko movements), the Narmada valley (the antidams struggles), and in and around firing ranges in Orissa or Jharkhand
(including those at Balaipal and Neterhat).46 We can assume, too, that
some parts of this discourse will have played well in those areas where
people have long expressed a strong distaste for outsiders, or where dalits
and adivasis have been mobilized by Naxalite groups. The front cover of
Ashis Nandys recent book, The Romance of the State and the Fate of Dissent in the Tropics, features a photograph by Krishna Murari Kishan which
depicts a muscular village labourer being beaten by four policeman, one
of whom stands poised to bring down his rifle on the mans head or shoulders. For Nandy, the state is an originary source of violence, and in his

See Corbridge (1998); Kiely (1999) for commentaries.

For an overview, see Routledge (2005).

Seeing the state


vision of empowerment it must be opposed by a Gandhian moment of

recovery of the self. This recovery happens through a process of psychic
cleansing which rejects The beautiful prose, the laudable sentiments,
and the languages of rationality and science [which] cover up . . . [the]
criminal enterprise [which is everywhere built into] state formation and
nation-building (adapted from Nandy 2003: x).
For the most part, however, Left political parties, activists, and NGOs
in India have been committed to an ideology of improved service delivery to the poor which makes demands of the state, rather than being
straightforwardly against it. In the 1950s and 1960s this agenda was
mainly concerned with increasing the share of total state resources that
was available for spending on the poor. More recently there has been a
concerted effort to bring pressure to bear on the state to make it responsive to the accounts that poorer people offer for their own poverty. It is
here that hands are occasionally joined with the World Bank and other
development agencies.
We see this clearly in the clamorous demands that have been voiced
recently for participatory development initiatives.47 Participation can
mean different things to different people, as we shall see in part II, and
can be more or less intensive.48 It is evident, even so, than an ideology of
participatory development sits easily beside demands for the greater voice
of men and women in the political process, and in the selection of particular development projects. As we noted previously, the Employment
Assurance Scheme is distinctive precisely because it is built around the
assumption that registered labourer households should demand employment from the local state when they are in need of work. The EAS
also calls for villagers to hold public meetings for the selection of work
schemes and executing agents (contractors). In the Eastern India Rainfed Farming Project, meanwhile, a substantial UK-funded aid project in
Jharkhand, Orissa and West Bengal, well-defined groups of poorer men
and/or women are required to be formed in order to make demands of
Project officers.49 The Project assumes not simply that poorer villagers
will learn to deal with one another by forming these groups, but that
they will gain bookkeeping and leadership skills and learn to deal with
authority figures. The exit strategy of the Project calls for poorer villagers

See our discussion in chapter 4. For a reasonably dispassionate overview, see McGee
Perhaps the most incisive treatment is by Brett (2000). See also Platteau and Abrahams
Kumar and Corbridge (2002). The EIRFP has backing from the Government of India
and we come back to it, briefly and as part of a broader discussion of participation, in
chapter 4.


The state and the poor

to strengthen their stocks of linking social capital in order that they can
make demands of the state when the aid money dries up.50
What is interesting about these interventions is that they constitute
members of the rural poor as individuals who have a right to be treated
as equals by the state and Project officials with whom they come into
contact. The supposition, indeed, is that it is members of the rural poor
who should be dictating to the state, and exercising their statutory rights
and preferences. It is this suggestion that apparently stands behind the
widows claim to her pension as of right, and it is a supposition that is
strengthened by linked demands for the greater accountability of public
servants. As we said before, the strategy of the MKSS in Rajasthan is to
make the state acutely visible to its clients and customers. In this vision,
technology (a photocopier, the Internet) can be one means for bringing
the citizen and the state into a supposedly unmediated encounter that
offers each party an undistorted sighting of the other. Politicians and
dalaals are cleared out of the way, and political society is made both more
civil and transparent. In another vision the politicians remain in the picture, but they are supposed to be accountable to their electors through the
panchayati raj institutions now in place across India. This is the agenda
of decentralization, the precise meanings of which will vary sharply from
place to place.51 (The levels of decentralized decision-making and revenue control that are to be found in Kerala are only weakly copied in
some other states, including in CPI-M-dominated West Bengal.)
All of these discourses are intent on changing the conditions under
which different groups of poorer people are coming to see the state.
Development studies must then be understood not simply as a discipline
which looks in upon different societies in the Third World, or even as the
locus of a set of policies which seek to repair the state and civil society in
some of those countries. Development studies must also be understood
as a set of contending discourses which help poorer people to make sense
of the state according to different accounts of gender, personal autonomy and the intrinsic worth of individuals. The interventions to which
they give rise including the many failures which Ferguson reports,
and which we document in part II become part of the technologies that
people make use of to see the state and to make demands of it. They


Linking social capital refers to the mainly vertical ties that poorer people have with those
in positions of power and influence. Bridging social capital refers to the ties between
people from different community backgrounds, while bonding social capital refers to
the much thicker ties that exist between people in a given family, kin or community
We comment on these agendas in chapters 5 and 7. We would simply point out here that
they make an appeal to an idea of direct or unmediated sight that we find instructive and
yet unconvincing.

Seeing the state


are as much constitutive of the process of sighting as they are a set of

observations of those sightings.
We have tried in this chapter to provide a preliminary sketch of some of the
issues that are involved in speaking about seeing the state. Sightings are
always complex and take shape against the sightings of other individuals,
communities and institutions. They also take place over the airwaves
and on computer screens, as well as in paper copy, memory, speech and
other direct interactions. The issues they engage, moreover, are very often
deeply contested, and point in the direction of diverse political agendas.
James Scott reminded us in the 1980s that peasants do not usually engage
in highly visible or openly rebellious forms of politics. More often, they
engage their antagonists by stealth and behind the scenes, mobilizing
what he called the weapons of the weak.52 This is surely what we should
expect. It is at one with what we know of rural or agrarian politics in India
(allowing for the fact that some exchanges are open and aggressive), and
quite consistent with the multiple sightings of the state and other political
targets that we have pointed towards here.
To better understand how these multiple sightings are staged it is
important that we turn our attention to specific localities and forms of
encounter. This will be our task in part II, where we will focus in turn
on questions of participation, governance and the contours of political
society. We will show how recent accounts of the merits of participatory
development are beginning to impact on at least some of the encounters
that poorer people have with state officials. By the same token, we will
maintain that demands for the increased ownership of development by
participating poorer men and women itself constitutes a technology of
rule (a structuring of statesociety relations), and one that is sometimes
radically at odds with the ambitions and capabilities of poorer people
and state officials alike. A similar argument will be advanced in respect of
good governance. Akhil Gupta is right to maintain that the discourse of
corruption, by marking those actions that constitute an infringement
of [legally defined] rights . . . acts to represent the rights of citizens to
themselves (Gupta 1995: 389). At the same time, however, as we shall
show, the practices of corruption are always complicated by the multiple
pressures which bear upon government staff occupying different positions


Scott (1985). We cannot say for sure when this phrase was first used, but Gandhi referred
to the weapons of the weak in Hind Swaraj (1997 [1908]).


The state and the poor

in a line department, say, or as a result of their links to (or dependence

upon) local politicians.
Before we turn to this task, however, or to a more general discussion
of the politics of statepoor encounters (part III), we want to consider
how several key sites for statepoor interactions have been produced by
the state itself through its longstanding, variegated, and recently changing war on poverty. In chapter 2 we consider how different governments and state agencies have sought to define, bound and even invent
the poor by means of interlocking and sometimes conflicting discourses
about depravity, demography, income levels and affirmative action. We
also report on how the war on poverty has been waged since the end of
the 1960s. At this point, too, we direct attention towards several of the
spaces for statepoor interactions that we consider further in part II, and
which make particular claims about the importance and merits of good
governance and participation. Here, of course, is our link back to the
concerns of contemporary development studies, always a focal point for
analysis, and to the work of scholars as diverse as Merilee Grindle, Arturo
Escobar and Robert Chambers.

Technologies of rule and the war on poverty

In his now-famous account of the making and unmaking of the Third
World, Arturo Escobar argues not only that an era of developmentalism was inaugurated by President Truman in his Point Four speech of
29 January 1949, but also that the aid programmes which followed were
justified by the discovery of mass poverty in the less economically accomplished countries (Escobar 1995: 21). In making these claims Escobar
directs us to the production of poverty as part of a wider (geo)political
discourse, and this is a central theme of this chapter. The production of
poverty as a failing, or as an incomplete set of capabilities, is linked to the
production of persons who can be labelled as poor, and who can either
be reproached for being the bearers of certain pathologies the illiterate
man who has to be educated, the overly fecund woman whose body has
to be disciplined and/or acclaimed as people who deserve the help of
others. Whether or not members of rural society are unaware of their
poverty before they are labelled as such by outsiders, as Lakshman Yapa
maintains was the case for him, growing up in Sri Lanka, is something we
consider later.1 But it is clearly the case that the production of poverty by
various government and other agencies creates many of the spaces within
which poorer people are bound to see the state. The designation of
households in India as Below Poverty Line (BPL) positions them as beneficiaries of developmental programmes which require them to have contact with sarkar. The same might be said of households belonging to the
Scheduled Castes, although in this case various agencies of the state are
committed to the disappearance of an entire category of persons. Members of the Scheduled Castes are to be lifted out of poverty, and spirited
away from their negative social identity as erstwhile Untouchables.
Escobars work on the production of development and poverty is a
useful corrective to accounts that seek to naturalize these social constructions. In this chapter we shall also find it useful to follow Escobars

Yapa (1996); see also Shrestha (1995) on becoming a development category.



The state and the poor

characterization of the age of development in at least one further respect.

We accept, that is to say, that an ideology of developmentalism is distinguished by its optimism regarding the malleability of internal and external
nature (roughly, human nature and the physical environment). The
productions of development and of poverty alleviation that take shape
under the Pax Americana are made possible, in part, by an insistence
on the potential equality of all human beings, regardless of their geographical location or genetic backgrounds. This is very different from the
years between 1860 and 1940, when most western accounts of progress
and backwardness were produced within discourses which emphasized
the permanent and disabling effects of race (social Darwinism, with
strong links to the White Mans Burden, eugenics and even genocide)
and geography (tropicality, or environmental determinism).2 The idea
of development is predicated on the view that men and women can be
created afresh, as modern subjects able to take their place in a world
defined in relation to the hyper-modernity of the West, and the United
States especially.3 Education and industrialization are obviously central
to this endeavour.
Escobar and his fellow post-developmentalists are less reliable, however, as guides to the complexities of social thought and action that are
to be found within what he calls the discourse of development (Escobar
1995: 4). In this book we reject the idea that it is helpful to reduce more
than fifty years of governmental interventions in the South to a single technology of rule. Nor do we think it sensible to declare that development has
produced only a nightmarish combination of debts, impoverishment and
malnutrition (Escobar 1995), or to imply that such anti-poverty schemes
as have been tried since 1950 have always been failures, or have been radically at odds with the accounts that poorer people have given of their own
difficulties. Poverty is always a social production, but it is nonetheless real
in important respects, and is generally described in negative terms by the

This is not to say that elements of these discourses do not survive, albeit in mutated form.
Andrew Kamarcks book, The Tropics and Economic Development: A Provocative Enquiry
into the Poverty of Nations, was published by the World Bank in 1976, but would not
have looked out of place sixty years earlier. Work by Jeffrey Sachs and colleagues (2000,
2001), however, on tropicality, or even by William Easterly (2001) on tropical misadventures, poses geography as a problem, but presents the economy and technology as
its redeemers. Sachss work is flawed in important respects, but his outlook on environment and development issues is more Promethean than determinist. See also, and more
widely, Drayton (2000) and Stepan (2001).
On the nineteenth-century origins of some developmental thinking, see Cowen and Shentons account of the Doctrines of Development (1996). This book offers an incisive critique
of post-developmentalism. Its major weakness is that it does not take seriously enough the
challenge posed by modernization theories to biologized accounts of social development.
See also Cooper and Packard (1998).

Technologies of rule and the war on poverty


people trapped in its clutches. We instead take a position that is more in

tune with the Foucauldian stance that Escobar wished to adopt. David
Lehmann has suggested that Escobars work was a missed opportunity
in terms of applying Foucaults ideas to the study of development.4 By
this he meant that Escobar fails to recognize that development is not a
singularity that can reasonably be described with a capital D. It is more
instructive to think of developmentalism as a set of discourses which
can combine various accounts of progress and social transformation with
elements of evolutionism and teleology, but which dont have to make this
commitment. To put it another way, it is important to think about developmentalism as a set of discourses that are united in some respects, but
which are contending in others: for example, around questions of sustainability, or the proper role of the state in managing industrial development,
or the best way to define and measure human well-being.
These discourses are not simply the bearers of particular class, sectional
or geopolitical interests, although these are important; they are also put
into play in relation to discourses about human rights, inter-generational
equity, the functioning of markets, the virtues of participating in civil society, and so on. Over time, some of these discourses will gain the upper
hand, and will take on the appearance of stability or even inevitability. We
saw this during the 1980s when the counter-revolution in development
theory and policy gained in strength amidst a more general revival of
ideas about free markets and sound money.5 But we also need to bear in
mind that an appearance of stability or coherence can be deceptive, and
that ideas emerge and contend on a more ad hoc basis. Nikolas Rose has
made this point very well in relation to the policies of Britains Conservative governments in the 1980s and 1990s. Although Thatcherism has
become known as a form of politics which mixes neoliberalism with social
conservatism, it is important to insist that some of the programmes for
which it is best known were not realizations of any philosophy, [so much
as] contingent lash-ups of thought and action (Rose 1999: 27).6 Roses
point, following Foucault, is that it is more useful to talk about technologies of government [which are] imbued with aspirations for the shaping
of conduct in the hope of producing certain desired effects and averting certain undesired events. Human technologies of government can
then be understood as an assemblage of forms of practical knowledge,

5 Toye (1987). See also Lal (1983) and Stewart (1985).

Lehmann (1997).
Rose is right in respect of privatization, but it would be a mistake to discount the role
of ideas (philosophy), in this case of public choice theory (Buchanan, 1967, 1987), in
the production of the community charge/Poll Tax. On Thatcher, see also Hugo Young
(1990); on Thatcherisms edgy combination of ideas about free economies and strong
states, see Andrew Gamble (1988).


The state and the poor

with modes of perception, practices of calculation, vocabulary, types of

authority, forms of judgement, architectural forms, human capacities,
non-human objects and devices, inscription techniques and so forth, traversed and transected by aspirations to achieve certain outcomes in terms
of the conduct of the governed (which also requires certain forms of conduct
on the part of those who would govern) (Rose 1999: 52, emphasis added).
Rose illustrates this argument with reference to Ian Hunters work on
the emergence of popular schools in European states like Prussia. Hunter
is well aware that the government school was promoted as a means of
the mass moral training of the population with a view to enhancing the
strength and prosperity of the state, and thereby the welfare of the people
(Hunter 1996: 1489; quoted in Rose 1999: 53). But the fact that states
may have wished to promote the strength of their populations does not
mean that they can simply whistle the means of moral training into existence (Hunter 1996: 149). The ways in which different systems of schooling were produced in Europe had more to do with a complex series
of exchanges and trade-offs between the administrative apparatuses of
states that were beginning to governmentalize themselves and religious
institutions, practices, knowledges and techniques for the spiritual disciplining of souls (Rose 1999: 54). (In England and Wales, there were
also exchanges with the trade union movement and with ideas emanating
from the statistical movement of the late nineteenth century.) In other
words, the technology of schooling was not invented ab initio, nor was it
implanted through the monotonous implementation of a hegemonic will
to govern: the technology of schooling like that of social insurance,
child welfare, criminal justice and much more is hybrid, heterogeneous
[and] traversed by a variety of programmatic aspirations and professional
obligations (Rose 1999: 54).
The same might be said of the technologies that have emerged to govern development or the alleviation of poverty. These technologies are not
simply the result of a class-based or imperial will to govern, which seems
to be the suggestion of some Marxists and many post-developmentalists.7
We need to understand not just why, but also how various agencies of the
state in India have produced different groups of the population as poor,
or Backward, or disadvantaged, or Scheduled, and how and why they
have proposed to deal with these conditions. Later in this chapter we
will focus on the explosion of anti-poverty schemes that emerged in India
in the 1970s, 1980s and 1990s. Far from continuing a common logic of
rule, these schemes have embodied very different assumptions about the

Which is not to say that power is not linked to interests: we will come back to this in
part III.

Technologies of rule and the war on poverty


capacities of state agencies, the voluntary sector, and poorer individuals,

groups and communities. To begin with, however, we focus on the contending and sometimes cohering technologies of rule that have combined
to produce different accounts of poverty alleviation in post-Independence
India. These technologies have made use of the Census, the National
Sample Survey, and forms of calculation designed to produce headcounts of those in absolute poverty. They have also comprised various
discourses about the obligations of a post-colonial state to its ethnic
minorities, as well as those referring to the imperatives of democratization, economic improvement and population biology. Needless to add,
these technologies have structured the encounters and sightings that have
followed between different poorer persons and their counterparts in the
From charity to capabilities
The British were certainly aware of the existence of mass poverty in
their South Asian colonies, just as they were at home. The authorities in
England had been exercised by the question of the relief of the poor since
at least the seventeenth century. Malthus wrote memorably on the topic
in 1798, and in the nineteenth century his views about the self-cancelling
effects of poor relief were confronted by an agenda for social reform that
would not have looked out of place in the writings of his principal antagonists, Mister Godwin and Monsieur Condorcet.8 The interventions of
Edwin Chadwick, and later Charles Booth and Robert Mayhew, were not
based in accounts of the perfectability of man, but they did focus attention on the threat which the poor posed not only to themselves but to
members of the higher social orders.9 The cholera epidemics in London
and Liverpool in the 1840s helped to focus the attention of the authorities on the need for better sanitation provision for the labouring poor. In
much the same vein, the threat of crime and physical violence, and later of
social and political unrest, prompted photographic essays on the slums of
Glasgow and expeditions into Unknown England, as areas including the
East End of London were sometimes known.10 It also spurred the beginnings of the social liberalism by means of which reformers like Sidney



Malthus (1970 [1798]). Godwin and Condorcet were celebrants of the French
See Porter (1999) for a lively review. Susan Chaplin (1999) has written an interesting
account of the strategies used by middle-class Indians (use of antibiotics, bottled water,
etc.) to insulate themselves from similar threats of contagion in the contemporary era.
Keating (1976).


The state and the poor

and Beatrice Webb, Maynard Keynes and William Beveridge hoped to

save capitalism from itself.11
At least some of these proposals spilled over to India, where the Webbs
were influential in designing a town plan for Tatanagar, and where Keynes
was concerned with the currency. The British were also made aware of the
condition of the poor by their own instruments of rule and revenue: by the
Reports of the Famine Commissioners, for example, or by references to
irregularities in the recruitment of free labour to Assam that were written
into some reports on Inland Emigration (see chapter 1). The need to
hospitalize victims of the plague in the Bombay Presidency in the last
years of the nineteenth century also caused the British to reflect on the
connections between disease and destitution and apparent threats to the
social order.12
But the recognition of mass poverty did not lead inexorably to the promotion of human technologies of government that would seek its abolition, and in this respect Escobar is right to point to a watershed in official
thinking about poverty since the late 1940s. In her important account of
the politics of the urban poor in early twentieth-century India, Nandini
Gooptu suggests that a minor key in British anti-poverty discourses was
focused on the ways in which poor surroundings and a lack of space
produced behavioural traits that were said to be indicative of depravity.
At its best, the social liberalism of the Webbs, or of the town planner
Patrick Geddes, resisted the biologized accounts of urban poverty that
were common among imperial officers (and some Indian academics too,
it has to be said).13 But while Geddess proposals for a garden city movement in urban India were consistent with his grand vision of civilisational
transition and civic nationalism (Gooptu 2001: 83), the reluctance of the
British to spend money on proposals that could be represented as radical,
a trifle dangerous and impractical (Gooptu 2001: 83) ensured that
Geddes returned to England a disappointed man. The authorities preferred to think of urban reform in terms of models of confinement and
zoning that enjoyed considerable support as well, among the Indian middle classes. The poor were to be kept in their place, and subjected to
regular police actions.
Away from the city, the British relied more on a model of poverty
that placed blame on the ignorance of the poor themselves, especially
the Untouchables, and on the backward customs of some members
of Indias feudal elites (and, more rarely, European landowners). Given

See Dahrendorf (1995) and Skidelsky (1992).

As discussed by Klein (1973, 1988); Harrison (1990); and Arnold (1993).
At its worst the politics of the Webbs was also strongly informed by eugenics.

Technologies of rule and the war on poverty


that India was produced at the end of economic arrangements that limited
the pace of industrialization, it is unsurprising that the British preferred
to focus on poverty in the sub-continent as an effect of Indianness itself.
Even Max Weber referred to the Hindus apparent dread of the magical
evil of innovation.14 The alleviation of poverty was thus confined mainly
to famine relief, or to urban-based interventions that mixed a fear of the
undisciplined body of the native with a growing emphasis on the virtues
of private and/or religious philanthropy. In some cases, too, the apathetic
Hindu was urged to reform him- or herself by embracing the more worldly
traditions of Christianity, especially in its Protestant incarnations.
It was largely in opposition to these approaches that the nationalist
movement began to advance its own agendas for dealing with poverty.
The drain of wealth theory that Dadabhai Naoroji put forward in the late
nineteenth century was a staple part of an account of the impoverishment of the masses that refused to locate the most fundamental causes
of poverty within India itself, or at least within an India that could be
made to rule itself.15 And later on, of course, in the inter-war period,
the need to win votes in municipal elections encouraged many nationalist politicians to make an appeal to the poor that ostensibly refused
the blandishments of the imperial power. Jawaharlal Nehru clearly had
an eye on the broader politics of the nationalist struggle when he told a
reporter for The Hindustan Times in October 1920 how much he objected
The lady who visits the slums occasionally to relieve her conscience by the performance of good and charitable deeds. The less we have of this patronizing and
condescending approach to the problem the better . . . . there are large numbers
of earnest men and women who devote themselves to the service of their fellow
creatures . . . They do good work . . . Yet, it seems to me, that all this good work
is largely wasted, because it deals with the surface of the problem only. Social
evils have a history and background, roots in our past, and intimate connections
with the economic structure under which we live. (Nehru, quoted in Agarwal and
Aggrawal 1989: 206)

The fault-line that appears here between surface appearances and their
structural determinants is instructive for another reason, too. Nehru
imbibed the distinction from the leftist texts he was then reading, but
his appeal to root causes was to become a staple of development thinking
more broadly. Post-colonial countries of all stripes capitalist, socialist and mixed were enjoined to throw off the shackles of tradition;
whole economies had to be structurally transformed, entire peoples

See Inden (1995).


Naoroji (1901); see also Dutt (1904).


The state and the poor

subjected to modernization.16 But how, exactly? This was the question

that faced India most acutely at the time of the Constituent Assembly Debates (19469), and the answers that were mapped out there
run counter to much conventional wisdom about the discovery of mass
poverty in the 1940s and 1950s. They did so, not least, because they
described an agenda for improving the capabilities of poorer men and
women that has more in common with Amartya Sens approach to development as freedom than with less expansive notions about the raising
of per capita incomes.17 They also developed an agenda that recognized
precisely those community-based rights that are not fully addressed in
Sens work. Special provisions for the Scheduled Communities helped
to ensure that the production of poverty and the poor in independent
India would be marked by crosscutting, if sometimes reinforcing, human
technologies of government.
Provider, protector and promoter
Jawaharlal Nehru gave a glimpse of his version of the new agenda for
poverty alleviation when he closed the debate on the Resolution of Aims
and Objects of the Constituent Assembly. He declared that: The first
task of this Assembly is to free India through a new constitution, to
feed the starving people, and to clothe the naked masses, and to give
every Indian the fullest opportunity to develop himself according to his
capacity (Constituent Assembly Debates: 22 January 1947).
This is often dismissed as so much cant, on a par perhaps with
Trumans rhetoric two years later, and just as lacking in real political
content. Ambedkar, after all, who shared with Nehru many of the responsibilities for constructing a new India, would later tell the Assembly that:
On 26th January 1950 we are going to enter a life of contradictions.
In politics we will have equality and in social and economic life we will
have inequality . . . In our social and economic life we shall, by reason
of our social and economic structure, continue to deny the principle of
one man, one value (quoted in Khilnani 1997: 35). But we should tread
warily before dismissing either Nehrus intervention, or various constitutional provisions, as mere rhetoric. There were certainly inconsistencies


The second article to be published in Economic Development and Cultural Change, the first
journal of development studies, made this point very clearly. Its author, Morris Watnick,
also insisted that the West would have to work hard to displace the appeal of communist
strategies for modernization in the Third World. Much like Escobar, he was acutely
aware of the geopolitical significance of Trumans plea for a bold new program of
technical aid to backward areas (1952: 22).
Sen (2000).

Technologies of rule and the war on poverty


in Nehrus accounts of socialism in India, and all manner of problems in

the translation of his ideas into policies for the redistribution of land to
the tiller, say, or for the promotion of universal literacy. We must remain
alert to the gaps that opened up in Plan promise and performance, or
between the rulebooks of the state and the actions of state officials. At
the same time, however, we need to take rhetoric seriously, the more so
where it is linked in some degree to the promotion of named technologies of rule. The fact that agrarian reforms in India were sabotaged in
the 1950s by richer farmers does not gainsay the fact that land reform is
a continuing objective of the central government, as stated in successive
Five Year Plans, and that the efficiency and equity arguments for agrarian
reform remain part of a broader political vocabulary (not least in Bihar).
Important laws remain on the statute book, just as they do in regard to
encroachments on common property resources.
If we return to the Constituent Assembly we see that the state was positioned there as a provider, protector and promoter for groups of people
who were still referred to as the masses or the millions. This perspective
was restated by the Planning Commission, which declared that the central objective of development in India was, to create conditions in which
living standards are reasonably high and all citizens, men and women,
have full and equal opportunity for growth and service (Government of
India, Planning Commission 1952: 29). The references here to citizenship, equality and service to the nation were by no means accidental, nor
were they without consequence. Nehru had declared at Independence
that, The service of India means the service of the millions who suffer.
It means the ending of poverty and ignorance and disease and inequality of opportunity (quoted in Agarwal and Aggrawal 1989: 241). And
now the Planning Commission proposed to add its weight to those provisions which had been written into the Constitution with the stated aim
of securing an adequate means of livelihood for all citizens, as well as of
the minimization of inequalities of income, status and opportunities, free
and compulsory education for all children, improvement of public health,
and social justice for the Backward classes. The Commission declared
that Indias problems were a legacy of a traditional society and static
economy in the past, petrified to some extent by colonial rule (Government of India, Planning Commission 1961: 1). It was, The evolution of
the social structure during centuries of feudalism, in regions which were
not then developed by communication, [that had] led to the existence
of large communities which suffered handicaps and disabilities imposed
by other economically and culturally dominant groups (Government of
India, Planning Commission 1952: 634).


The state and the poor

Poverty was thus defined as a broad-based set of absences, or missing capabilities, that were produced not by the poor themselves, but
by oppressive social forces (Empire, feudalism, casteism, perhaps even
religion) that could quickly be removed. Particularly in the years 1946 to
1956, the war on poverty in India was conceived in terms that proposed
a close link between the remaking of India and the making of modern
citizens. The promotion of economic growth and of household incomes
was one part of this agenda, but it was by no means the major element. To
some degree this reflected the weakness of the states financial position,
and the fact that industrial growth would take some time to promote. But
it also spoke to a real concern for social justice and the rights both of individual citizens and of corporate social units. This concern was expressed
not simply in legislation to secure the abolition of zamindars, supposedly
the main depressors of agricultural productivity in the countryside, but
also in a raft of measures designed to address the problems of Indias
weaker sections.18
These initiatives varied significantly between the Scheduled Tribes
(STs) and the Scheduled Castes (SCs). The debates of the Constituent
Assembly also revealed a division in attitudes to Indias adivasi populations that still continues. The extreme paternalism that has produced the
poverty of these people as a product of their location (remoteness), mode
of subsistence (forest-dependence) and general primitiveness (Indias
junglees), has coincided with a penchant for exoticism which has celebrated the genius of the tribal people (a favourite phrase of Nehrus)
and their right to be different.19 The Oxford-educated tribal leader from
Jharkhand, Jaipal Singh, also claimed that the republican and egalitarian traditions of adivasi society could be adopted with profit by caste
Hindus.20 These contending discourses have helped shape the particular
technologies of rule under which many tribal people meet the state. In
addition to the labour and immigration officials who have long rubbed
up against populations that were anything but sedentary, the state has
presented itself to many adivasis through the slow accumulation of Block
Development Officers and District Development Commissioners who
staff the Scheduled Areas, and who join the police and forestry services
in providing comparatively executive forms of rule. Perhaps most importantly, a significant number of STs have been brought into the state as
recipients of reserved seats and jobs.
A version of the republican ideals that Singh claimed for tribal India was
also deployed on behalf of the Scheduled Castes. Nandini Gooptu writes

Thorner (1956); see also Harriss (1992).

Singh (1989); see also Corbridge (1988, 2002a).
See Volume IX of the Constituent Assembly Debates: 6534.

Technologies of rule and the war on poverty


that a pre-Aryan identity of the untouchables as the original inhabitants

Adi Hindu of India (Gooptu 2001: 144) was constructed in the towns
and cities of the United Provinces in the early twentieth century. Elements of this ideology later informed the politics of the Republican Party
of India (set up in 1957 after Ambedkars death), and the Dalit Panthers in Maharashtra. For the most part, however, this construction was
rejected by other caste groupings and by the colonial power. The poverty
of the Untouchables was explained by the British as being a result of
their oppression by caste Hindus (hence the Depressed Classes), and
by Brahmanic scriptures as a form of punishment. They did not live in
particular regions, as seemed to be the case with most tribals, and they
generally did not own land. In addition, they were discriminated against
in terms of access to places of worship, schooling and even water.
It followed that the battle against Untouchability had to be waged in
more diverse arenas than would be the case with the tribal problem.
Access to reserved seats and jobs would be common to both enterprises,
and would ensure continuing struggles around the designation of different groups as SCs or STs, and the acquisition of pieces of paper to certify
group membership. (These struggles would be expanded in the 1990s
in the wake of the Mandalization of politics.) Members of the Scheduled Castes, however, would also need to call upon agents of the state
including schoolteachers, health workers, the police and officers of the
court to ensure their access to the public sphere, and to enforce claims
on government resources. Whereas members of the Scheduled Tribes
were thought to be open to abuse because of their innocence (which
was a value worth preserving), the poverty of the Scheduled Castes was
defined by patterns of social exclusion enforced by others. The one sure
way to remove their poverty was to remove the caste system itself: a conclusion, or an ideal, that appealed to Ambedkar as well as to Nehru, albeit
in slightly different ways.
Gandhi, of course, did not endorse this conclusion. He preferred to
look for the erosion of Untouchability within Indias villages, perhaps as
a result of social welfare efforts that would cut across caste and community boundaries. A thin version of his faith in community welfare was
later made flesh in the Community Development Programme that was
launched in 1952. The First Five Year Plan declared that, In view of
the large unutilised and under-utilised resources in the system, schemes
for mobilizing local effort for local development have to receive high
priority (Government of India, Planning Commission 1952: 45).
Appealing, again, to the idea that poverty in the Indian countryside would
be curtailed as soon as various burdens were removed (including rackrenting, caste competition, and colonial taxes), the Planning Commission


The state and the poor

further declared that, It is schemes of this type spread all over the country, more than development projects, which are likely to activate these
resources (Government of India, Planning Commission 1952: 45). Such
a view, however, with its touching faith in the cumulative psychological effects of inter-caste cooperation (Government of India, Planning
Commission 1952: 45), was set to recede when planners received word
that the benefits [of CDP] did not reach the less privileged sections of the
village community in adequate measure (Government of India, Planning
Commission 1961: 291), and as soon as funds were in place for a more
resolutely industrial assault upon the traditional structures of rural life.
Economy, demography, poverty
It is important at this point to note that recession does not mean disappearance. The technologies of government that were put in place to
deal with poverty in the 1940s and 1950s have largely survived to the
present. The Scheduled Communities have continued to be defined by
state legislative bodies, albeit with occasional changes in their numbers,
and they have continued to receive special treatment under Part XVI
of the Constitution of India. The extension in 1969 of the system of
reserved jobs to include employment in public sector enterprises marked
a significant expansion of the technologies of compensatory discrimination first enacted in 1943 for the SCs, and in 1950 for the STs. These
(largely national) technologies of rule have mandated the continued collection of statistics on the populations of the Scheduled Communities,
most notably in Indias decadal Censuses.21 The Government of India
is also required to receive periodic reports from a Special Officer for
the Scheduled Castes and Tribes. Under Article 338, the Commissioner
of Scheduled Castes and Tribes is required to review the standing of
the constitutional provisions put in place to safeguard the Scheduled
Communities, and to report his findings, via the President, to each House
of Parliament. The Commissioner is also able to push central government
to release grants from the Consolidated Fund of India for the purpose
of promoting the welfare of the Scheduled Communities of particular
states. These grants would be expected to augment the more general
Plan spending on education and health-care which has also threaded its
way through the states anti-poverty programmes from the 1950s to the

Instructively, demographic statistics on castes other than the Scheduled Castes were not
collected in the Censuses from 1951 to 2001. In the wake of V. P. Singhs decision to act
upon some of the recommendations of the Mandal Commission Report (see below),
this might soon change; certainly there are pressures in that direction.

Technologies of rule and the war on poverty


Notwithstanding these continuities, however, there was an observable

shift in the way that different agencies within the Government of India
began to think about poverty during the period of the Second and Third
Five Year Plans (195666). In some degree, this involved a narrowing of
the definition of poverty. In the First Plan period, especially (19516),
and to a lesser degree under the Second Five Year Plan, the Planning
Commission had proposed that, A comprehensive concept of living standards [should certainly] include the satisfaction of basic needs like food,
clothing and shelter, but linked this as well [to the] normal satisfactions
of family life, enjoyment of physical and mental health, opportunities for
the expression of skills and recreational abilities, and active and pleasurable social participation (Government of India, Planning Commission
1952: 613). By the time of the Third Plan, however, the government
was insistent that poverty was mainly the result of low productivity and
a lack of continuous work. Consistent with its new focus on the economy a form of practical knowledge which must itself be seen as a
human technology of government the Planning Commission urged that
there should be additional opportunities for work to enable the lowest
income groups to earn enough through productive employment to meet
their minimum needs (Government of India, Planning Commission
1961: 11).
The significance of this discursive shift is apparent as soon as we
recall the ways in which the economy functioned in the rhetoric of
the nationalist movements. For most Gandhians, the commitment to
industrial modernity that was announced by the NehruMahalanobis
model of structural transformation was threatening in itself. Gandhi saw
in large-scale industrialization the foundations of the loss of self to which
true anti-poverty programmes would be opposed. For other wings of
the movement, however, whether under Patel, Nehru, Ambedkar, Bose
or Savarkar, the weaknesses of India as a nation, and of the individual
bodies within it, were straightforwardly a result of the countrys lack of
industrial prowess. The economy functioned here as an absence, or as a
signifier of the fetters which were imposed on the country by systems of
imperial preference and agrarian involution. The fact that food production struggled to keep pace with population growth in the years 190040
was one vital, and impoverishing, outcome of this system of misrule;
another, famously, was the ruination of Indias handicraft industries, and
the deliberately stymied growth of its manufacturing industry until at
least the 1920s.22

For a considered review of the issues, including the onset of some measure of industrial
protection in the 1920s, see Tomlinson (1988). See also Blyn (1966).


The state and the poor

This constitution of the economy as an absence also allowed it to function as an extraordinary site of potential enrichment, and this is how it
came to be written in the mid-1950s. The economic case for land to
the tiller land reforms was now boosted by work which claimed to show
an inverse relationship between farm size and productivity.23 Agrarian
reform made sense for efficiency reasons, as well as for reasons of social
justice. More significantly, perhaps, at least in terms of practical impacts,
there was a potent coming together of a number of the ideas that sustained
a first generation of development studies: the importance of planning
and savings, for example, and of import-substitution industrialization.
Indeed, the new orthodoxy came to maintain that, a precipitate transformation of the ownership of productive assets was . . . detrimental to the
maximization of production and savings (Chakravarty 1987: 10). The
economy itself, suitably protected from foreign competition in the short
run, would do the job. Unemployment would decline once labouring
people were put to work in the consumption-goods industries that would
spring up in the wake of the capital-goods-based revolution. Poverty in
turn would ebb away in the 1960s, save perhaps in some parts of the countryside. It would affect those people unable or unwilling to find work in
the cities, or in the modern sector of the economy.24
This production of the poverty problem had significant implications
for how poorer people would be defined and presented to different agencies of the state. In geographical terms, there was a palpable shift in public
expenditure patterns in favour of the city. There was also a new emphasis
on the labour exchange as a site for the collection of statistics about the
working and non-working poor, and of encounters between poorer people and sarkar.25 Less obviously, perhaps, there was renewed attention to
what the First Five Year Plan had called the pressure of population in
India (Government of India, Planning Commission 1952: 23).
In the run-up to Independence most nationalists had been at pains to
deny the importance of overpopulation as the principal determinant of
Indias mass poverty. Palme Dutt noted in India Today that nine out of
ten Western readers, who have not had the opportunity to acquaint themselves with the facts were only too happy to jump to Malthusian conclusions about excessive population growth in India (Palme Dutt 1989:
48, quoted in Krishnaji 1998: 385) even though the facts suggested
that a much larger population could be fed once all cultivable lands were
brought under the plough, and once the zamindari system was abolished.

The best reviews remain those of Thorner (1956) and Harriss (1992).
A similar view was expressed by W. A. Lewis in his two-sector model of economic growth:
Lewis (1955).
For recent work that touches on this subject, see Parry (1999) and Breman (2004).

Technologies of rule and the war on poverty


Many nationalists did contend, however, that the rate of economic growth
could not be maximized in India amid high rates of population growth.
Under the Chairmanship of Jawaharlal Nehru, the National Planning
Committee (NPC) of the Indian National Congress concluded in 1935
that, the size of the Indian population is a basic issue in national economic
planning, in so far as its unrestricted increase in proportion to means of
subsistence adversely affects the standards of living, and tends to defeat
many social and ameliorative measures (quoted in Krishnaji 1998: 386).
Nevertheless, the NPC went on to suggest that: While measures for the
improvement of the quality of population and limiting excessive population pressure are necessary, the basic solution lies in the economic
progress of the country on a comprehensive and planned basis (quoted
in Krishnaji 1998: 386). Growth, in other words, would provide a contraceptive effect of its own, although the state would be required to promote
birth control on a voluntary basis.
This conclusion neatly anticipated the demographic transition model
of the 1940s. In the 1950s, however, the causal relationships assumed
to obtain between economic growth and population growth were significantly reversed, in India as in much of the developing world. One of
the authors of the demographic transition model, Kingsley Davis, now
began to suggest that rapid and excessive population growth in India
would produce social conditions favouring the rise of authoritarianism.26
As Simon Sretzer has shown, this fear expressed a deeper unease in the
United States about the massing hordes in South and East Asia, a fear
that would later be exploited by population biologists like Paul Ehrlich
in their frankly racist accounts of the population bomb.27 But causality
was also reversed for economic reasons. The new growth models placed a


Davis (1951). Sretzer makes the important point that the work of the Office of Population Research in Princeton was closely associated with the State Department, and may
have come under pressure from that Department to strike a more interventionist note
regarding the desirability of speeding up the demographic transition in Asia. In the
course of late 1948 and 1949 those in the United States still dreaming of a globe emerging from colonial servitude into a regime of liberal democratic free trade were awakening
to a nightmare, experiencing a strong sense of loss of control in a dangerous and alien
world (Sretzer 1993: 676). No less than development studies, or the mathematization
of economics (which Mirowski links to funding from the RAND Corporation and the
military in the early 1950s: Mirowski 2002), population science was constituted in part
as a Cold War technology of government.
Sretzer (1993). Ehrlichs account of The Population Bomb provides a garish rendition of
one stinking hot night in Delhi when his taxi ride through the dust, noise, heat, and
above all people, left him frightened. Since that night, he continued, Ive known the
feel of overpopulation (Ehrlich 1968: 15). His failure to note that he might have met
with still larger crowds in (then) largely white London or lower Manhattan was neatly
taken to task by Mahmood Mamdani in his cutting and often funny account of The Myth
of Population Control (1972).


The state and the poor

particular emphasis on physical capital formation, and in this framework

it was easy to conclude that rapid population growth must represent a
loss of savings to the more productive parts of an economy. This would
be especially acute in a country suffering from a scarcity of capital in general, or where poorer people had to propel themselves out of a low-level
equilibrium trap.28 Spending on dependent populations could then be
presented as a luxury that countries and families should do without, at
least until the benefits of rapid economic growth had been secured.
This presentation of the population problem was tempered in India
by a strong commitment to voluntarism when it came to family sexual
matters, aided no doubt by a measure of prudery.29 It was only during
the years of the Emergency (19757) that this commitment was suspended in favour of the savage and humiliating assaults on male and
female bodies that were sanctioned by Sanjay Gandhi and his henchmen. But the importance of demographic issues as a site of statepoor
encounters should not be discounted. In many rural areas of India, as in
many urban slums, poorer women are brought into regular contact with
health-care workers who profess concern for their bodies, and above all
their reproductive health. In some cases these interventions will be welcomed, as when women have been coerced into having large numbers of
children by their husbands. It would also be a mistake to suppose that
health-care professionals are unconcerned with a womans health, or are
simply using this issue for the purpose of population control. At the same
time, however, it would be nave to assume that anti-natalist policies are
never seen as a threat. Many Muslim families feel under pressure from
the Sangh parivar, which has repeatedly drawn attention to their supposed proclivity for large numbers of children (and which constitutes the
rapid growth of Muslim India as a threat to the body of the (Hindu)
nation).30 Still others will see anti-natalist policies as being against their
best interests, and will sometimes express a sense of puzzlement when
faced with campaigns to distribute condoms, for example, or IUDs.
Regardless of how such programmes are judged, the states expressed
concern with the body corporeal always leads to a heightened concern for


The model is discussed in Nelson (1958) and Enke (1971). See also Elvin (1973) for an
application to the Chinese past.
Such prudery, along no doubt with caste and religious concerns, continues to inform
the attitudes to HIVAIDS of leading members of the government and opposition in
India see Dube (2000); see also Farmer (2003) for an account of what he calls the
new war on the poor. For an innovative and witty account of sexual panics in the West,
see Lacqueur (2003).
See Jeffery and Jeffery (1997: chapter 6). The Sangh Parivar is that body of organizations,
including the BJP, the Vishwa Hindu Parishad and the RSS, which is committed to the
Hinduization of all politics in India see McKean (1996).

Technologies of rule and the war on poverty


the production of numbers: numbers of women aged between 15 and 45,

say, or of men and women who have been sterilized. These figures have
to be collected, and they bring the state into further and repeated contact
with its target populations. Certain members of the rural or urban poor
might also be made the beneficiaries of anti-natalist interventions which
link material incentives for the poor to performance quotas for named
bureaucrats. As Emma Tarlo has shown, the attempt to link different
agencies of the state to the disciplining of individual bodies reached a
peak during the Emergency, when thousands of victims of slum clearance
in Delhi were promised resettlement plots if male householders volunteered for sterilization. In just one colony in East Delhi, Tarlo and her
co-worker, Rajinder Singh Negi, found 3,459 personal files from 1976,
975 of which contained a DDA Family Planning (FP) Centre Allotment
Order. By this order, the Delhi Development Authority sought to collect information on the applicants: Name and Age, Fathers Name, Plot,
Number of Family Members, Date of Voluntary Sterilization and Nature
of Assistance Claimed. A Resettlement Order might then be made by the
Officer in Charge (Tarlo 2001: 79).31
This bringing together of economic and population-based accounts of
poverty also led to the production of poverty lines and poverty headcounts. Very much in line with its view that poverty resulted from low
levels of productivity and a lack of continuous work, the Third Plan document urged that additional opportunities for work would enable the
lowest income groups to earn enough through productive employment
to meet their minimum needs (Government of India, Planning Commission 1961: 11). This pushed the government to define what it meant by
minimum needs, and to specify the means by which information could
be collected (the National Sample Survey, for example), and the proper
units of analysis (villages, households, individuals, etc.).
Consistent with what was by then a more narrowly economic conception of poverty, the Perspective Planning Division (PPD) introduced
the notion of a minimum level of living in 1962. This set the national
monthly minimum level of consumption in rural areas of India at Rs. 20
per capita. Individuals falling below this line were said to be poor, or
suffering from absolute poverty. While other advisory bodies suggested
a different rupee figure the Nutrition Advisory Committee declared that
it would cost Rs. 35 at 19601 prices to provide for a balanced diet and
modest consumption of non-food items the governments preference
for a definition of poverty on the basis of basic minimum needs was

See also Selbourne (1977). For more on the authoritarianism that has often been latent
in Indias family planning programmes (outside the Emergency), see Vicziany (19823).


The state and the poor

now established. Throughout the 1970s the major effort of government

was directed to further refinements in the measurement of a poverty line
defined in terms of calorific norms, rather than to a more comprehensive
assessment of living standards. The PPDs Task Force on Projections of
Minimum Needs and Effective Consumption Demand decided in 1979
that in rural areas a person would need to have sufficient income per
month (Rs. 49.09 at 19734 prices, using the NSS Round of that year,
Rs. 56.64 in urban areas) to command a daily calorie norm of 2,435 in
rural areas and 2,095 in urban areas. And in the 1980s and 1990s these
assessments were updated on the basis of progressively more robust price
deflators, at the state as well as at national levels, as the debate on Indias
absolute poverty scaled new technical heights.
Garibi Hatao
As it turned out, the production of statistics about the state of absolute
poverty in India could not have come at a worse time than the late 1960s.
The non-foodgrains sector of the rural economy performed quite well in
the 1950s, but the rate of growth of cereals and pulses between 19523
and 19645 only just kept ahead of the rate of growth of the countrys
population. When the rains failed in India in 1965 and 1966, the country
had to be bailed out with grain transfers from the United States. The
situation in the countryside was a long way from the picture that had
been promised in the Plan documents of 1956 and 1961. Rice riots broke
out in Kerala in 1966, and in one of her first acts as Prime Minister Mrs
Gandhi announced that she would not eat rice until there were adequate
supplies of rice available in [that state] (Frank 2002: 295).
The theatricality of Indiras attempts to take the part of the poor would
become more blatant still in the 1970s, when she demanded an end to
poverty (Garibi Hatao). In the late 1960s, however, she was faced by the
more immediate problem of famine in Bihar, and the threat posed to
Congress rule in Uttar Pradesh by the decision of the Jat farmers leader,
Charan Singh, to leave the party in 1967.32 The suspension of planning
between 1966 and 1969 further symbolized the seriousness of the issues
confronting a weakened polity. It came as little surprise that Dandekar
and Raths famous article on poverty in India, published in Economic and
Political Weekly in 1971, confirmed not only that many millions more
Indians were in absolute poverty in 19701 as compared to 19601, but

Amartya Sen has famously argued that famines cannot happen in democracies, but the
famine in parts of Bihar in 1967 would seem to indicate otherwise (Singh 1975). There
was also a continuous state of famine and near-famine in parts of Orissa in the 1990s
(Sainath 1996).

Technologies of rule and the war on poverty


also that the incidence of absolute poverty had increased to 54.8 per cent
from 45.4 per cent in the countryside, while staying close to 45 per cent
in the towns. India seemed to be going backwards, and was increasingly
being seen on the international stage not as an emergent great power,
but as something of a basket case (to use the unpleasant language of the
Mrs Gandhi responded to the electoral setbacks of 1967 and 1969
by splitting the Congress Party and by seeking to reclaim the socialist
credentials of her father. Morarji Desai and the Congress-(O) replied
by joining forces with the Jan Sangh and the Swatantra and Samyutka
parties to fight the general election that Indira called on 27 December
1970; they did so, moreover, under the slogan Indira Hatao (get rid of
Indira). Mrs Gandhi hit back with the simplistic but effective battle cry of
Garibi Hatao (Remove Poverty). According to the most recent of her
biographers, Garibi Hatao was a call for the eradication of Indias worst
evil. And as a vote-winner it worked. Garibi Hatao was a thunderbolt . . .
a revelation . . . a revolution. Its impact was instant and electric. The
poor, who were the vast majority of Indias electorate, now saw Indira
as their saviour (Frank 2002: 325, quoting in turn from Narasimha Rao
1998: 6212 and Malhotra 1989: 128).
Even allowing for the exaggeration of a biographer, there are
some interesting things going on here. Those scholars who compare
Mrs Gandhi unfavourably with Nehru sometimes fail to acknowledge
that the daughter had to operate in a political landscape that was quite
different to that facing her father in the 1950s.33 Indira had to revamp
the Congress machine in the context of what Lloyd and Susanne Hoeber
Rudolph have called demand politics.34 Charan Singh, of course, symbolized the switch from command politics to demand politics very well.
His campaigns on behalf of the richer peasantries of north India were
indicative of a new political landscape in which interest groups could
force the hand of government agencies which previously had sought to
dictate to groups in civil and political society. The abandonment of Indias
obsession with capital-goods-based industrialization was one sign of this,
and was more or less announced at the time of the Fourth Five Year
Plan (196974). But that Plan also confirmed that the government had
embraced the Green Revolution, and was now paying attention to the
poverty of the countryside. The setting up of the Public Distribution
System (PDS) in 1966 was indicative of this shift, and the PDS, of course,

This would be true of Paul Brass (1994), notwithstanding his generally excellent analyses
of Indias politics post-Independence.
Rudolph and Rudolph (1987); see also Byres (1988).


The state and the poor

as Jan Mooij has shown, from then onwards would be a major site for poor
peoples encounters with the state in urban and rural India.35 With it
came yet another set of cards that defined the poor and their entitlements.
Mrs Gandhis genius, if such it was, was to ride the first waves of the new
demand politics. She recognized that the Congress-(R) would need to
develop new campaigning styles and vocabularies if it was to put together
a political coalition that would reach beyond its traditional support bases
in the Forward and Scheduled Castes.
Interestingly, Garibi Hatao emerged as a slogan before Mrs Gandhi
acquired real popularity as the liberator of Bangladesh.36 To the extent
that it did have an electric and instant impact, and positioned her as
the saviour of the poor, this is surely also because the poor had been
invented as a political constituency in the 1960s. And this in turn reflected
two major developments: the diffusion of democratic ideas and the slow
erosion of vertical voting blocs, to be sure, but also the production of new
technologies of government which defined a Below Poverty Line (BPL)
population even as that population was set to grow in size and to announce
its voices. There was a dialectical relationship between the production of
the poor and the capacity of Mrs Gandhi and others to take their part.
This being the case, we must treat very carefully the rhetoric of Garibi
Hatao and other slogans that seem to want the eradication of poverty.
Whether or not the poor must always be with us, as many conservatives
like to suppose, there are strong reasons for insisting that some politicians
would look with alarm on their diminution or disappearance. Concepts
of inequality, deprivation or relative poverty function in part to make this
We shall come back to this observation soon enough. For the moment
we should note that the Fourth, Fifth and Sixth Five Year Plans, all

Mooij (1999).
Nandini Gooptu notes that appeals to the garib janata (poor common people) emerged
at the heart of political discourse in the late 1930s, which is precisely when Indira
Gandhi was learning the grammars of modern politics. In the 1930s, the garib janata
referred to the morally superior, deserving simple folk, who were excluded from power
and denied their due (Gooptu 2001: 425). As Gooptu explains, this language of poor
but deserving (and also poor and cheated), drew on a tradition of nautanki theatre that
was deployed by proponents of Adi-Hinduism as part of a wider repertoire of nirguna
bhakti, a heterodox devotional alternative . . . to brahmanical Hinduism . . . [that
espoused] an egalitarian religious message (Gooptu 2001: 148). It is doubtful that Mrs
Gandhi paid much heed to what might today be called first nation sentiments when
she appealed directly to the garib janata. In addition, while it is true that nationalist
politicians of various stripes made rhetorical appeals to the garib janata in the 1930s and
1940s, it is not inconsistent to say that the poor emerged as a political force in their
own right as a group or set of groups with political voice only after another twenty
or thirty years.

Technologies of rule and the war on poverty


of which took shape under Mrs Gandhis leadership, were also distinguished by their continued dialogues with the concerns of development
economists and the major aid agencies. The Fourth Plan was ahead of
the game in suggesting that neither agricultural or industrial growth
would be sufficient to generate productive employment enough to do
any more than contain the problems of unemployment and underemployment, and in proposing special programmes . . . to provide for
what amounted to redistribution with growth (later the slogan of the
approach to poverty alleviation favoured by the World Bank (Corbridge
and Harriss 2000: 85). The Fifth Plan continued this theme, and insisted
that Indias national planning should not only raise the per capita income
but also . . . ensure that the benefits are evenly distributed, that disparities
in income and living are not widened but in fact narrowed (Government
of India, Planning Commission 1974: 8). And the Sixth Plan announced
that, There is . . . convincing evidence which points to the limited effectiveness of trickle down effect . . . Thus specific programmes meant
for selected target groups of population are essential components of a
strategy designed to assist in the removal of unemployment and poverty
(Government of India, Planning Commission 1981: 17).
As always, there are important areas of continuity in these proposals.
Inequality had long been a stated concern of the Government of India,
and the importance of employment provision, as we have seen, had been
a central component of Indias poverty discourses since the 1960s. But
this is to be expected. Old technologies of government rarely make way
for new ones in a one-to-one fashion. Far more often the process is gradual, and it involves a measure of additionality as well as a replacement
effect. Nevertheless, the changes of the 1970s and also the 1980s were
real and substantial, and they were produced in part by changes in the
discourses of development studies, and indeed of intellectual life more
generally. Marxism was a growing force in the 1970s, and the rise of a
feminist movement, and of feminist development studies, would slowly
push the Government of India to at least some recognition of the needs
and experiences of women, and even of different groups of women.
The setting up of a sub-scheme of the Integrated Rural Development
Programme (IRDP), in 19823, to deal with the Development of Women
and Children and Rural Areas (DWCRA), was one sign that women
were not only to be approached by state agencies in terms of their fecundity. The IRDP, moreover, which was set up on a pilot basis during the
Emergency (in 1976), and which was extended to all parts of the country
from 1980, was not only Indias most important anti-poverty programme
in the 1980s, but was also considered by many to be a model for rural
development programmes across the Third World. As with Redistribution


The state and the poor

with Growth, Indias prosecution of integrated rural development was

at least as influential within the World Bank (which proposed a sectoral
approach to rural poverty alleviation in 1975) as were the ideas that began
to flow in the opposite direction.37
The extraordinary diversity of the anti-poverty schemes which emerged
in the 1970s and 1980s has to be seen in this broader context. The
schemes that were set up in the late 1960s or early 1970s to deal with
the problems of small farmers (the Small Farmers Development Agency:
SFDA), marginal farmers and agricultural labourers (Marginal Farmer
and Agricultural Labour Programme: MFAL), or tribals (the Tribal
Development Agency: TDA), undoubtedly spoke to some very real problems that were facing these groups. The same would hold true of the
special area programmes that took shape at the same time. These were
focused on particular geographical regions that were considered to be
marked out for backwardness by dint of their location tribal areas,
once more, but also drought-prone areas, desert areas, hill areas and
border areas. (The argument can be extended to include important and
influential state-directed schemes for poverty alleviation, including the
Employment Guarantee Scheme in Maharashtra, which was made statewide in 1972/3 in response to the extended drought conditions of 1971/2.)
It is also proper to insist that these programmes be judged according to
the conventions both of project analysis and political science. We benefit
from being told that different projects have had a low or high take-up
rate, or that the leakage of funds to non-intended beneficiaries is 10, 20
or 30 per cent (see part II of this book). There is even merit in those
critiques of the reluctance of successive regimes to deal with the real
or underlying causes of Indias poverty, whether this is understood in
terms of a failure to support the institutions of a developmental state or
of a less regulated market.38
At the same time, however, an appeal to the real causes of poverty, while
it calls to mind Nehrus advice to do-gooders in the 1920s, misses something important about the importance of appearances.39 Perhaps most
of all, it misses the importance of words and images, and of the need to
take these seriously, not least in the realm of government. When Nehru
later asked the Constituent Assembly to clothe the naked masses, he
was invoking an imagery of poverty as disgrace that has continued to be

Drawing on Corbridge and Harriss (2000: 85). Redistribution with Growth was published
in 1974 by Hollis Chenery and colleagues.
See Bardhan (1984) and Bhagwati (1993) for opposing perspectives.
We take our cue, in part, from Nikolas Rose: Against interpretation, then, I advocate
superficiality, an empiricism of the surface, of identifying the differences in what is said,
how it is said, and what allows it to be said and to have an effectivity (Rose 1999: 57).

Technologies of rule and the war on poverty


powerful. The disgrace, of course, attached not only to the naked themselves and consider how differently Nehru and Gandhi saw the absence
of clothes as signifiers of value but to those who looked upon them.40
The naked masses had to be clothed, uplifted and disappeared. Much the
same view coloured the Planning Commissions account of urban poverty
at the beginning of the First Five Year Plan. Most of the towns of India
[it suggested] . . . have a large proportion of sub-standard houses and
slums containing insanitary mud-huts of flimsy constructions . . . The
disgraceful sights presented by the ahatas of Kanpur and the bustees of
Calcutta are conspicuous examples of this state of affairs (Government
of India, Planning Commission 1952: 5934).
During the Emergency, this dialectic of disgrace was shrunk so badly
that a progressive discourse of human rights (the right not to be shamed
in public) was again submerged beneath a contemptuous and frankly
punitive account of the urban poor as polluters of good taste. To make
New Delhi modern visibly modern the slums had to be bulldozed away.
Matters improved in the 1980s, but a perception of the poor as a deficient
social mass continued to dominate official discourses about poverty until
at least the end of that decade. (It still continues, of course.)41 To the
extent that real changes could be observed, they were to be found in a
transferral of the site of disgrace from the body corporeal to the asset
base of poorer households. During the Sixth Plan Period, the Planning
Commission promised that, Programmes . . . will be drawn together so
that they focus upon the level of the individual household, and raise at
least 3000 of the poorest households above the poverty line in each block
during the Plan (Government of India, Planning Commission 1981: xxi).
The war on poverty now spoke of the removal of poverty and of direct
attacks upon it. Poverty was once again conceived as being something


Gandhi tended to equate poverty (but not exploitation), with simplicity and authenticity, and thus an absence of clothes with a kind of purity or child-like innocence (see
Alter 2000). For Nehru, in contrast, nakedness was more often seen as a symptom of
extreme religious asceticism (the irrational), or, more usually, of a degree of deprivation
that hindered human development. In urban areas, of course, nakedness could also be
associated with lewdness, and with an inability on the part of poor labouring males to
avoid the temptations of the brothel and the bottle (see Gooptu, 2001: 678). Here,
perhaps, the instincts of Gandhi and the social-religious reformers (including members
of the Arya Samaj and the Hindu Mahasabha) coincided in some degree with those of
Nehru and the proponents of modernization. On clothing and politics more generally,
see also Cohn (1996) and Tarlo (1996).
Not least in New Delhi, where businesses and middle-class residents are once again
invoking images of order, cleanliness and rationality in support of their campaigns
to widen roads and displace poorer people from their (permanent or temporary)
settlements: see Baviskar (2003) on the making of metropolitan Delhi. Chatterjee (2004:
61) also draws attention to the unpleasantly named Operation Sunshine in Calcutta in


The state and the poor

like a physical object separated from social relations, and households were
to be treated to schemes that would raise them above the line. It was
only in the 1990s that poorer individuals or households, or even social
groups, were allowed to function seriously as active agents of their own
An emphasis upon roots rather than surfaces also conceals the importance of visuality in politics. Politicians have to know how to work a crowd.
Long before politicians like Laloo Yadav dreamt of attending political rallies by helicopter, or even in a Tata Sumo, Mrs Gandhi liked to descend
upon the masses from the skies, like a goddess.42 Television was her ally
in projecting this image to a much wider audience. But Mrs Gandhi also
knew the importance of reaching specific groups within the poor, and
of appearing to be active on their behalf. The multiplication of schemes
for named groups of the poor needs to be understood in this context as
well. Schemes for Tribal Development added to existing programmes of
compensatory discrimination. The Small Farmers Development Agency,
meanwhile, imposed new systems of registration and sighting, with only
those farmers working one to three hectares of land supposedly being
eligible for its dispensations of irrigation equipment, credit, supplies
and (other) technology. Schemes like this and the Marginal Farmer and
Agricultural Labour Programme (targeted on the landless and those with
less than one hectare of farming land), had the effect of disaggregating
the poor and of inventing more specific sites for statepoor encounters.
This trend was further continued in the 1980s when the BPL population
was targeted en masse, through the IRDP, but also in terms of its component groups: BPL rural youth (ages 1835) through a scheme for the
Training of Rural Youth for Self-Employment; groups of BPL women
through DWCRA; BPL rural artisans through a scheme for the Supply
of Improved Toolkits to Rural Artisans (SITRA), and so on.
The multiplication of these schemes spoke for sure to problems identified by development experts, non-governmental organizations, and even
the poor themselves. But they were also multiplied by politicians anxious to present themselves as gatekeepers of the welfare state, or of a
patronage democracy. The naming of schemes thus came to matter precisely because of its superficiality. The more schemes, the more clearly was
government seen to be working for the poor. James Ferguson makes a similar point about the extension of bureaucratic power in Lesotho when he
listed the extraordinary number of development agencies seventy-two
that were active in that small landlocked state in the years 197584.43
He makes the point that these agencies constituted Lesotho as an empty

See Rajagopal (2001).


Ferguson (1990: 67).

Technologies of rule and the war on poverty


space that needed to be filled by expert bodies, whether from abroad

or from within the state. The nameplates advertising the arrival of these
agencies then proclaim that new presence, and function as an apparent
source of comfort and perhaps even of development. Appendix 1 provides
a partial listing of the major programmes for poverty alleviation that were
operational in India in 1999, at the time of our fieldwork. In addition
to pointing up one of Indira Gandhis legacies to the country she ruled
for so long, it also helps to describe a new and expanded geography of
statepoor encounters in contemporary South Asia.
Democratizing poverty
The proliferation of poverty-alleviation schemes under Mrs Gandhi was
not complemented by significant budgetary transfers in support of the
war on poverty. During the Fifth Plan period the Plan Outlays for Social
Sector Spending as a percentage of Total Plan Outlays fell to 25.5 per
cent, having been above 30 per cent during the first four Plan periods
(table 2.1). Spending on Social Services, which includes spending on
education, health, housing, and special programmes for the Scheduled
Communities and other disadvantaged groups (including women), fell
to just 12.1 per cent of the Total Plan outlay at this time, as against
19.7 per cent during the Second Plan period. This is consistent, perhaps,
with Mrs Gandhis preference for spending on named schemes, such as
those that took shape from the time of the Fourth Plan under the Special
Areas Programmes (albeit the resources that were committed to the SAPs
were and remain relatively insubstantial). In addition, while it is true that
spending on Agriculture remained high during the first five Plan periods, there was a switch under this heading from spending on the Community Development Programme to more general support for farmers,
including better-off Green Revolution farmers. Insofar as Mrs Gandhis
governments did commit significant funds to its multiplying and highly
visible schemes for poverty alleviation, it was during the Sixth Plan period
under the heading Rural Development: this covered spending on special
poverty alleviation and employment-generation programmes. The governments of Rajiv Gandhi and V. P. Singh ensured that spending under
this heading, and that of social services, would account for more than
20 per cent of Total Plan Outlays in the Seventh Plan period.
It was not until the 1990s, however, that spending on the Social Sector
as a whole climbed back above 30 per cent of Total Plan Outlays. Several
factors account for this upward trend, but first among them would
be what commentators have called the Mandalization of politics, and
what Christophe Jaffrelot refers to as the triumph of quota politics over




Total Social Sector


















Sixth Plan Seventh Plan Eighth Plan Ninth Plan

(8085) (8590)

Source: Planning Commission data, quoted in Government of India: Ministry of Finance, 1998.




Social Services
Rural Development
Special Area

First Plan Second Plan Third Plan Fourth Plan Fifth Plan
(5156) (5661)
(6166) (6974)

Table 2.1. Social Sector Plan outlays as a percentage of Total Plan Outlays: Centre, States and Union Territories, 19512002

Technologies of rule and the war on poverty


kisan (farmer) politics in north India.44 Geography matters here because

the slow rise to power of the Backward Castes in India happened much
earlier in the south than in the north. Non-Brahmanism has been an
important force in Tamil Nadu since at least the 1920s, and this is
reflected in the high percentage of government jobs that are now reserved
in the state. In Karnataka, too, and in Kerala and Andhra Pradesh,
important Commissions were set up in the late 1960s or early 1970s
to determine who might belong to the Backward Classes and for what
reasons (they varied significantly from state to state).45 The major significance of V. P. Singhs decision in 1990 to act on at least some of
the recommendations of the Second Backward Classes Commission (the
Mandal Commission Report) was that it extended the scope of reservations at an all-India level. Singhs National Front government proposed
that 49.5 per cent of jobs in central government services or the public
sector should be reserved for members of the Socially and Economically
Backward Classes. Since 22.5 per cent of such posts were already reserved
for members of the Scheduled Castes and Tribes, this implied that a further 27 per cent of posts would be reserved for members of the Other
Backward Classes (OBCs).
Singhs own Janata Dal party gained votes disproportionately from the
3,743 castes, tribes or communities that Mandal had identified as Backward and which made up 52.4 per cent of the population. Indeed, his
party had its major sources of electoral strength in those rural, Shudra,
north Indian, OBC communities that were also about to propel Laloo
Prasad Yadav and Mulayam Singh Yadav to power in Bihar and Uttar
Pradesh. The Yadavs, of course, were the best-positioned of these communities, and were soon established as the major recipients of the new
posts that each government created, whether by expansion or by transferring out members of the Forward or Intermediate castes. In Uttar
Pradesh, Jaffrelot reports, Out of 900 teachers appointed [by Mulayam
Singh Yadavs] second government, 720 were Yadavs (Jaffrelot 2003:
380). In Bihar, meanwhile, Laloo Yadav moved to ensure that, an IAS
from the Scheduled Castes replaced a Brahmin as Chief Secretary [in
1993] and an OBC took over the charge of Director General of Police
from another Brahmin (Jaffrelot 2003: 380). He also ensured that of
the 1,427 lecturers recruited to Bihars universities and constituent colleges in 1996, most candidates [and appointees] were OBCs and, more
precisely, Yadavs (Jaffrelot 2003: 380).
The number of jobs allocated in this way probably matters less than
the broader effects that this second democratic upsurge has had on

Jaffrelot (2003).


For discussion, see Galanter (1991).


The state and the poor

statepoor encounters in India. One of Laloo Yadavs most insistent

claims is that income poverty in Bihar is not of great concern; what
matters, and what has driven his politics, is the fact that members of
the Backward Classes have been engaged in a struggle for power and a
search for honour (izzat). The major contribution of his governments
has been to ensure that members of the Forward Castes cannot mistreat
the Backward Classes in a sustained or systematic manner. Similar confrontations are in play across north India, and they are likely to get more
heated as the size of the government cake refuses to keep pace with the
numbers of people who can claim to be poor, backward or disadvantaged. This is where the logic of demand politics collides with the politics
of scarcity. It is only central government that seems to have the resources
to cope with the pressure of reservations while also funding an expansion
of the Social Sector budget, both in absolute and relative terms. Interestingly, this expansion has coincided not just with pressures from below,
or from the OBCs, but against a backdrop of economic reforms and liberalization. Despite the predictions of some commentators, the 1990s
saw a significant increase in spending on Indias social services and rural
development programmes. It is possible that this extra spending has been
made necessary by an increase in absolute poverty brought on by the
reforms themselves. This would be the view of many on the Left and it is
probably not at odds with what happened in the first two or three years
of the reforms.46 What is certain, however, is that the debate on whats
been happening to poverty in India has been recharged since 1990. It
has also coincided with a more general assault on the quota and BPL
economy by certain groups within Indias business and upper caste elites
who feel threatened by the steady rise to power of the OBCs.
This fight-back has taken several forms, not all of which sit easily
together, but at its core, significantly, there has been a concerted challenge to the ways in which poverty and deprivation in India are defined,
measured and treated. This is evident in the decision of the Jats of Uttar
Pradesh to give up their quest for Kshatriya status in favour of being
included in the states list of OBC communities. It is also evident in
the support offered by some Forward Caste communities for the idea
of reservations for women, including high caste women, and indeed for
some twice-born communities.47 It is further apparent in the fact that
many members of Indias economic elites are keen to shrink the state,
and to declare success in the war on absolute poverty, even as some

The best review is probably that by Sen (1996).

For example, recent campaigns on behalf of Rajputs in Rajasthan, and Brahmans

Technologies of rule and the war on poverty


members of locally dominant farming castes will go to great lengths to

get themselves listed as BPL.
It is at this level, we would contend, at the level of the production of
identities and numbers, that the war on poverty is now increasingly being
fought, and where the state is most often sighted.48 In a very real sense we
can say that the war is being fought over such matters as the production of
the BPL schedule, as well as over the mechanics of its collection and later
use (see box 2.1). There are clearly forces at work that want to see the
numbers of people in absolute poverty driven down, and who might want
to put pressures on the lower-level state officials charged with identifying
BPL households to act sparingly. These surveys are carried out every
few years, and the administrator might be told to look out for a bicycle or
other such item as proof of the non-existence of poverty or BPL status. At
the same time, there will be forces working in the opposite direction. The
administrator is likely to hand over the job to influential village members,
many of whom might know that bikes should be hidden, and almost all of
whom will know the advantages of getting friends and supporters listed
as BPLs. The production of BPL statistics takes shape within the vortex
of these social forces, and at once defines a major arena of statepoor
encounters and sets the scene for further encounters for those who can
get their names registered as BPL.

Box 2.1 Seeing and measuring the BPLs

When the state in India in 1992 wanted to see and measure the
BPL population in various states it did so by looking inside household
cash boxes. District Magistrates/Collectors were told to instruct their
enumerators to collect information on household earnings from land,
wages, remittances and other sources. Many DMs realized this was
an impossible task. Some of them told their enumerators to use their
eyes, and warned them if I visit your village and find out that what is
visually decisive has been left out in order to preserve the sanctity of
the schedule, you will be in trouble. Similar sentiments were doubtless expressed in 1997 when the government called for a survey to
be based on household expenditures, or the cooking pot. In 2002
the government changed its approach for a third time in ten years. An


This is also true in the United States. Forty years after President Johnsons War on
Poverty, significant and often unpleasant battles are being enjoined around affirmative
action, workfare (see chapter 1), and the responsibilities that poor people are said to
have for availing themselves of opportunities in a market-access society.


The state and the poor

Expert Group recommended that BPLs should now be counted on the

basis of a definition of relative deprivation which scored households
from 0 (very poor) to 4 (not poor) on 13 dimensions. Five of them are
listed below for illustrative purposes:
no. Characteristic

Availability of
Less than 2
normal clothing
wear: per person
in pieces

Between 2 Between 4 Between 6 10 or above

and 4
and 6
and 10

Ownership of


Any one

Status of
labour force


Female and Only adult

and no
production for other

11 Type of

For daily

latrine with

latrine with

latrine with
supply and
Two items Any three
or all

males only


All or most
items on a
long list
TVs, and

Borrowing No
only from indebtedness;
institupossess assets

We have not spoken to DMs in Bihar, Jharkhand or West Bengal about

how this schedule is completed in the field, but we can guess that similar short-cuts will be used and that richer households will again
put pressures on enumerators to be listed as BPLs. In any case, the
Government of India gave discretion to the states to decide the cut-off
points that would convert this 52-point classification into a working
definition of very poor and poor at the District, Block and even

Technologies of rule and the war on poverty


panchayat scales. Having made this concession, however, which would

seem to disable any attempt to think about poverty on a consistent
basis across a state, and far less across India, the Centre also instructed
DMs to ensure that any number they came up with should not be more
than 10 per cent higher than the poverty estimate that had already
been calculated by the Planning Commission for the year 19992000.
The DM was also instructed to compare his/her figure with the relevant figure given in the NSSO survey of consumer expenditure for
19992000. Somewhere in between these competing pressures and
definitions, it is fair to say, figures for BPL populations are produced
and designations are made. Whether and to what extent those designations or certificates are then used for public policy decisions is, of
course, another matter, as indeed is the question of whether a BPL
household becomes aware of its status.
Source: based on interviews, and Letter No. Q-16205/4/2002-AI(RD),
dated 13.9.02, issued by Dr P. V. Thomas, Economic Adviser to the
Ministry of Rural Development, Government of India, to all Secretaries of Rural Development, State Governments, providing guidelines
for the BPL survey for the Tenth Five Year Plan (20027).

Empowering the poor

The fact that some BPLs are active in the matter of their designation is also
curiously, if perhaps not intentionally, bang up to date with the thinking
which has informed the design of many poverty-alleviation programmes
since about 1990. The Right and the Left are seemingly agreed that the
state should not be in the business of dictating to the poor, or of providing resources for them in the top-down fashion that once prevailed. Poorer
people are rather to be protected against the rent-seeking behaviour of
state officials. They are also encouraged to voice their own accounts of
what it is to be poor, and what they might need from agencies in the
government and voluntary sectors. Participation, accountability, decentralization and democratization have become the new watchwords in a
discourse which promises that poverty will be reduced by good governance, and by people doing it for themselves. As the World Bank puts
it, The poor are the main actors in the fight against poverty. And they
must be brought center-stage in designing, implementing, and monitoring anti-poverty strategies (World Bank 2001: 12).
The thinking behind this new technology of government has emerged
from several quarters, and is still contending for power with more


The state and the poor

traditional accounts of the relationships that should hold between poorer

people and representatives of the state. A cynical view of the neoliberal
agenda might suggest that its insistence upon poorer people solving their
own problems is consistent with its broader assaults upon public spending. But this would underestimate the ideological power of the counterrevolution in development theory and policy. The neoliberal agenda also
mobilizes a concept of empowerment which puts particular emphasis
upon the self-worth of the individual. It suggests that the maximization
of a persons potential is held back by oppressive or simply incompetent government as much as it is by a lack of education or health care
(although these capabilities are acknowledged to be important). The key
point is to put oneself in the position of the customer rather than that
of the service provider.49 When monopoly powers accrue to government
servants the stage is set for statepoor encounters that are abusive and
inattentive to the real needs of the customer. What the customer needs
most is alternative service providers, or sufficient voice that he or she can
hold state agencies accountable for their actions. A democracy that functions properly at the local level is one way to ensure such accountability.
This is one reason why the sponsors of economic reform in India have
generally also been proponents of Panchayati Raj and the devolution of
administrative powers and budgets to local authorities.
These arguments, which we shall expand in part II, have also been
deployed by those who are mistrustful of the reforms, albeit with considerable shifts of emphasis. Robert Chambers has been a pioneering figure
in this regard. His work on participatory poverty assessments, and on
the question of whose reality counts, has been cited regularly by those
who wish to insist on the need to work with different groups of poorer
people around their own agendas for empowerment.50 But whereas the
neoliberal agenda is interested in improving the capabilities of poorer
people to work in and through markets, and in designing formal institutions for good governance, the emphasis of many on the post-Left is
on capacity-building initiatives that aim to resist these seductions. The
emphasis is on community, and on the possibility of poorer people taking charge of their lives within a well-defined locality.51 More moderate
accounts within this tradition call upon non-governmental organizations


A point made long ago, we seem to recall, by W. Arthur Lewis in an exchange with
Thomas Balogh on the nature of socialism. For a contemporary view from a very
different perspective, see the essays in Krueger (2002).
Chambers (1983, and 1988).
See, for example, Chakrabarty (2003), Friedmann (1996). Also Vyas and Bhargava

Technologies of rule and the war on poverty


to play a central role in capacity-building initiatives as well as in running

service delivery schemes. More radical accounts might express scepticism about the possibility of poorer people (especially women, and members of the Scheduled Communities) being able to voice their concerns
effectively in aam sabhas or gram sabhas, or indeed in meetings with the
panchayat sewak or the Block Development Officer. The emphasis here
might be on spatial closure strategies that seek the empowerment of the
oppressed by removing them from contacts with the state and the market alike. Participation would then no longer be oriented to the agendas of
In practice, these more radical agendas soon run up against their internal contradictions (how can states, markets or hierarchies be avoided?),
and cede ground to mainstream accounts of the merits of participation
and accountability. This mainstream, however, which enjoys strong support now from the World Bank and most development agencies, is itself
proposing a radically new account of the ways in which different groups
of poorer people might come to see and meet the state in a country like
India. As we have suggested several times already, newer technologies of
rule are rubbing shoulders with, and sometimes are jostling, older forms
of government. The picture does not remain the same.
Joint Forest Management is a case in point, although we shall have
cause to refer to it only fleetingly in this book. JFM emerged in a government circular of 1990 as a middle way between state control of the
forest usufruct and full control by local user groups. The circular called
upon state governments to implement JFM systems in order to regenerate protected forests and reduce rural poverty. The guidelines asked
state governments to devolve everyday forest protection, management
and development responsibilities to local community institutions (cooperative or committee-based) at the village or panchayat levels. These
institutions would include serving Forest Officers and would prescribe
benefit-sharing arrangements following regeneration.53 Unlike community forestry, which would take the state out of the picture entirely,
JFM proposed that villager understandings of sylvicultural practices were
extensive but not complete. Poorer people would still benefit from the
professional advice that Forest Department officials could offer, and

See Esteva and Prakash (1998); Rahnema and Bawtree (1998).

By the end of the first quarter of 2001, there were 44,943 official JFM groups (village
forest committees, or VFCs) protecting over 11.63 million hectares of governmentowned forests, or 15.5 per cent of the recorded forest area of the country, making it one
of the largest such programmes in the world (Corbridge and Kumar 2002: 767, drawing
on Borgoyary 2001).


The state and the poor

which would in time build up their own stocks of forest knowledge.

The State Forest Departments Trading Wing could also provide villagers
with information on the price of timber and non-timber forest products
(NTFPs) at different locations.
This same emphasis on the state as a facilitator and partial stakeholder
is also to be found in a number of recent initiatives in the fields of primary education provision and employment assurance, two areas that will
feature strongly in part II. Nehru and Ambedkar put particular emphasis
on education as a core component of Indias anti-poverty programmes.
The Constitution of India directed in 1950 that the State shall endeavour
to provide within a period of 10 years . . . free and compulsory education for all children until they complete the age of 14 years. This goal
was restated in the National Resolution on Education of 1968 and the
National Policy on Education of 1986. India failed lamentably to meet its
performance targets, however, and the gap between rhetoric and reality
in the education sector was exposed with particular vigour in the 1990s,
not least by those impressed with Amartya Sens account of poverty as
capability deprivation (see also table 2.2).54 The new public education
agenda proposes to deal with this learning deficit, in part, by encouraging educational provision in the private or not-for-profit sectors, which
should increase parental choice; but there are also provisions for states to
empower the parents of children in government schools through Village
Education Committees (VECs) or their equivalent.55 Given that many
poorer villagers come to see the state most often and directly in the
figure of the schoolteacher, the suggestion that they might have power
over him or her through a VEC proposes a radical reworking of this
optic. The directness of this relationship stands in sharp contrast to the
statutory provision for inspections of schools and their employees by the
Sub-Inspectors of Primary Schools employed by a states Department
of Education. When combined with mobilization campaigns (some of
which are aimed at improving adult literacy, as for example in the Total
Literacy Campaign), the devolution of powers to a VEC suggests that
parents and children might at last be empowered to challenge the power
of the teaching trade unions, and to get to grips with problems of poor
quality schoolteaching and even teacher absenteeism.
An insistent emphasis on participatory development is also to the fore
in Indias largest anti-poverty scheme of the post-reform period, the
Employment Assurance Scheme (EAS). The EAS began life in 1993

Sen (1985); see also Dr`eze and Sen (2002: chapter 5).
For example, the School Attendance Committee in West Bengal.

Technologies of rule and the war on poverty


Table 2.2. Literacy rates (age 7+) in 17 major states

of India, 1997



Himachal Pradesh
West Bengal
Tamil Nadu
Jammu and Kashmir
Madhya Pradesh
Uttar Pradesh
Andhra Pradesh








Source: National Sample Survey Organization, 53rd Round,

quoted in
(Courtesy: Institute of Applied Manpower Research)

when it was deployed by the Ministry of Rural Development in New

Delhi, to provide gainful employment during the lean agricultural season
in manual work to all able-bodied adults in rural areas who are in need and
who are desirous of work, but who cannot find it (Government of India
1993b: 1). In important respects it drew on state-level schemes to provide employment in the off-season, including the Employment Guarantee
Scheme (EGS) in Maharashtra.56 The EAS was distinctive, however, in
its ambition to provide sufficient resources (80 per cent from New Delhi;
20 per cent matching funds from state governments) to provide up to
100 days of waged employment for a maximum of two adults per household in need, wherever these household members voiced a demand for
work. The scheme would be triggered by demands from below, by men
and women who could not otherwise find work banging on the door

On the EGS, see Echeverri-Gent (1993), Herring and Edwards (1983), Joshi and Moore


The state and the poor

of their local government office and asking for assured employment.57

Ideally, this work would be provided from existing plan and non-plan
works in progress, but if these schemes did not allow EAS monies to
be used so that at least 60 per cent of funds were spent on unskilled
labour, other schemes would have to be found. The selection of these
schemes, moreover, would also bear the imprint of local people. The EAS
guidelines stipulated that village open meetings would have to be called
to decide on the sorts of schemes that villagers might like to see commissioned (the results to be passed upwards to higher-level panchayat bodies
and Block Development Office), and for the selection of the contractor
charged with executing a scheme. They also required that the accounts
of EAS schemes be presented to villagers each year through their open
meetings (the gram sansads of West Bengal, for example).
In sum, EAS schemes would be triggered by local demands, given shape
through public meetings, and made accountable in the same forums. In
terms of statepoor encounters, they would constitute members of the
labouring poor as prime movers in a large-scale anti-poverty initiative, the
major outcomes of which would also be chosen, very largely, by members
of village society. Instead of being supplicants of the state, the poor were
to be its customers and even its masters.
Very much in opposition to Escobar, we have argued in this chapter that
the production of poverty in India, and of various sites where the poor
encounter the state, have not stayed still over the past fifty years. The antipoverty agenda in India is made up of several technologies of government,
some of which have gained in strength recently as others have receded,
and most of which survive in sometimes uneasy proximity. It would be
absurd to deny that poverty isnt mainly defined by the Government of
India in income or calorific terms, or that absolute poverty isnt still seen
as a disgrace by many of those charged with its alleviation. In any case,
this definition speaks to an important understanding of poverty, and there
are good reasons for welcoming the sharp declines in absolute (income)
poverty that have recently been reported for both rural and urban areas.
At the same time, however, it is clear that agencies within and without the
Government of India have begun to rethink their anti-poverty agendas
quite radically since the late 1980s. This is partly in recognition of what
Yogendra Yadav has called the second democratic revolution in India, or

Sometimes literally, but more often through intermediaries and contractors: see
chapter 4.

Technologies of rule and the war on poverty


the fact that the Backward Classes are now making much greater demands
not just of the state but within the state.58 The pervasiveness of quota
politics is one important sign of this; another is the severity of the political
struggle that is being waged for control over the local state. In addition,
the governments thinking on poverty, and the presentation of its antipoverty agendas, has been influenced by the voices that have been raised
on behalf of civil society and the voluntary sector, and by those who have
urged that the poor should be allowed to speak for themselves.
In theory, these voices of the poor have lent considerable weight to
the promotion of the EAS, or JFM, or VECs as new vehicles for the selfempowerment of disadvantaged individuals, households or social groups.
They have done so, not least, because they have mobilized some quite
radical assumptions about the rights and capacities of poorer people,
some of which were already present at the time that the Constitution
of India was promulgated in 1950. One suggestion of this book is that
these new technologies of government cannot be reduced to a singular
discourse of development, nor can it be assumed that they are without
effect. Jaffrelot claims that India is undergoing a silent revolution, and
this is surely correct. The fact that Indias revolution doesnt share the
qualities of speed and extreme violence that we associate with revolutions
elsewhere is less important than the fact that power is leaching steadily,
and in some respects ineluctably, to the lower castes, and has been claimed
by them in terms which often resist the presumptions of a benign and
disinterested state.
A second suggestion of this book, however, is that we learn about the
state not simply through an analysis of its published technologies of rule
its guidelines for JFM or EAS, its recruitment and training practices,
its systems of reward, sanction and promotion, its means of registration
of men and women as SCs, STs or BPLs, its name-plating of visible
schemes for the alleviation of poverty but also through the ways that
it works in the trenches. The failure of land-to-the-tiller land reforms in
the 1950s showed how named agents of the state come under pressure
from competing forces in political society. (Some officers will also have
been major landowners themselves.) If we want to understand how the
state works from the point of view of the rural poor, we need to focus
in depth on certain initiatives like the EAS or the VECs which claim
to bring poorer people into contact with the developmental state in the
most enlightened fashion. These initiatives provide something of a test
case for investigating statesociety relations more broadly. We also need
to do this in different locations, so that we can isolate more clearly the

Yadav (1996).


The state and the poor

effects of particular political regimes and the technologies of government

with which they are associated.
Before we turn to these tasks, however, we need to do something else:
we need to consider to what extent the poor see and rely on the state at
all, and for which reasons and in which circumstances. We need first, in
other words, to say something about the livelihood strategies, capabilities
and contact networks of different groups within the rural poor, including
those networks that take shape away from the eyes of state agencies. This
will be our task in chapter 3.

Part II

The everyday state and society

Meeting the state

In this part of the book we draw on fieldwork in Bihar, Jharkhand and
West Bengal to comment in more detail on how different actors come
to see and engage the state in eastern India. Chapters 46 offer different
and yet complementary takes on what is necessarily an interlocking set
of issues. In chapter 4 we consider why and how (and if) people participate in a range of development schemes, including the Employment
Assurance Scheme and Village Education Committees. These schemes
make important assumptions about the construction of citizenship and
civil society in rural India. In chapter 5 we direct our attention to the
career paths of various government servants, and to the ways in which
they construct working lives and practices that may or may not agree with
the agendas of good governance now being promoted by New Delhi and
the international development institutions. In chapter 6 we focus on the
ways in which poorer peoples encounters with the state are structured
with close regard for the conventions of the political societies that operate
in our study areas. We also take up the question of corruption here, as
we do in chapter 5.
In this chapter we want to say something about the livelihoods and
social networks of poorer people in our study areas. We will introduce
some of the individuals (for example, brokers and local uppers) who
become key figures in the stories we tell in chapters 4, 5 and 6, and we
shall comment on the ways in which poorer men and women use their
non-state social networks to access the state or keep it at a distance. In
all of these tasks we find ourselves moving away from the macro views
of the state that dominated part I of the book. Our focus now is on the
micro operations of the everyday state in eastern India, and the way these
operations are understood, reshaped and contested by ordinary people.
(We recognize, of course, that the macro and the micro are not easily
separated, for conceptions of one affect sightings of the other.)



The everyday state and society

Throughout this part of the book we draw on fieldwork that was mainly
conducted in the period from February 1999 to March 2000. This work
was supplemented by studies carried out from September 2000 to July
2001 as part of the action research project we discuss in chapter 9. As
we explained in the Introduction, we worked in five Districts that were
chosen to exemplify different political regime types in Bihar and West
Bengal (see figure Figure 3.1). Within each District we identified a single
Block, and within it a village or locality, in which we could get to grips
with these political cultures in a more concerted fashion.1 It wasnt easy
to choose these localities (or indeed the Blocks), and the final choice
was made only after extensive discussions with local activists, journalists
and government officers, and after several alternative field sites had been
visited.2 We must emphasize, too, that we didnt know in advance how
the state was working in these Blocks, and certainly not from the point of
view of different groups within the rural poor; we simply hypothesized that
effective pro-poor governance was most likely to be found in Midnapore
District, West Bengal, and in Bhojpur District, Bihar, where there have
been long histories of pro-poor mobilizations.3
There are between 250 and 350 households in each locality. The
research programme began with a household census that provided basic
information on the livelihoods of every family. This was followed up with
group interviews and by detailed questionnaire work with a sample of 100
households in each field site. The samples were put together on a stratified
random basis and included 80 poor households in each case. Members
of the research team stayed in the localities throughout the fieldwork year.
We needed to win peoples trust in order to build up a depth of ethnographic understanding on livelihoods, social networks, and the everyday

The term locality is more accurate in the case of the West Bengal field sites. The three
field localities in Bihar are discrete villages. Each is made up of a number of different tolas
(neighbourhoods) that generally are defined on caste or ethnic grounds. In West Bengal,
our localities are defined by the boundaries of the electoral wards of a Gram Panchayat.
Each locality contains a number of small, discrete settlements (referred to as villages by
their inhabitants), that have neighbourhoods differentiated by caste or religion.
There is clearly substantial variation in the political cultures of even one District, not least
when it is as large in spatial and population terms as Midnapore District was in 1999.
(The District was then home to around 10 million people. Also written as Medinipur, it
has since been sub-divided.)
Our decisions were also influenced by a range of practical concerns. Most importantly,
we needed to have a degree of interest in and cooperation with our work from villagers
and local power-holders. Beyond this, our localities needed to have safe, available accommodation and be reasonably accessible to Kolkata/Patna. These demands were fairly
modest, however: for example, the Bhojpur locality was more than a two-hour drive on
non-metalled roads to the nearest rail station, and our accommodation in Midnapore was
in a room borrowed from a junior school. As we report, too, in part II and chapter 8, our
hypotheses were not confirmed, at least not in such simple terms, and certainly not in

Meeting the state


Figure 3.1 Map of the studied Districts

functioning of the state. Intensive work in each locality was also combined from the outset with interviews conducted at the Block-, Districtand State-levels. About 280 interviews were conducted across the field
sites with government officers and local politicians, as well as with the
contractors, brokers, teachers, engineers and others who are key players
in the de facto operations of the local state or political society.
We begin this chapter by reviewing the livelihood options of our sample households and the experiences of poverty that commonly are made
by poorer people. We then consider how people use their non-state
social networks to support their livelihood strategies, and to access (or
avoid) the state. Lastly, we consider how the state makes itself visible to people in our localities, and how government agencies initiate


The everyday state and society

and/or respond to encounters with poorer people. By looking at poverty

and the state from below we are not aiming to privilege the views of
poorer people. That would repeat a failing of an earlier generation of
village ethnographies and would reinforce a populist strand in contemporary development studies. Our poor persons perspective is rather
to be seen as a first cut through a range of empirical materials that we
will examine from different points of view in subsequent chapters. It also
enables us to highlight certain blind-spots within the states own accounts
of statepoor relations, and this too will be a recurrent theme of later
Experiencing poverty: livelihoods, inequality
and social networks
Men and women in rural eastern India undoubtedly learn to describe
their life conditions and opportunities in terms of categories provided
for them by state agencies as members of the socially marginalized
Scheduled Communities, for example, or as BPL households. But their
descriptions of their life-worlds are not limited to these forms of biopolitics, nor are their experiences of hardship or social exclusion a product
of discourse alone. The relationship between the word and the world, to
put it crudely (and as Foucault well understood), is hardly accidental,
and the daily lives of people are indelibly shaped by pervasive, and in several respects slowly changing, patterns of structural inequality in terms
of access to land, or what Marxists have called the means of production.
These patterns of inequality will vary significantly from place to place, as
our own field studies make clear, and are mediated by the social structures of accumulation described recently by Harriss-White by caste,
gender, religion and ethnicity, for example.
In Old Malda Block, Malda District, West Bengal, we worked in a
peri-urban locality about five kilometres away from Malda town. The
town sits astride a national highway and is a key source of casual work.
Agricultural production here, as in most of the eastern part of the
District, is based primarily on the cultivation of monsoon rice and winter
vegetables for local consumption: the commercial mango farming or sericulture found elsewhere in Malda is not present. The population of the
ward we worked in is made up predominantly of adivasis and Scheduled
Castes (table Table 3.1). It also includes a significant number of recent
migrants. These households have come from elsewhere in West Bengal
and (illegally) from Bangladesh, attracted to the area in part because of
economic opportunities in Malda town.

Meeting the state


Table 3.1. Census households by community and levels of well-being

and poverty









67 (20%)

136 (41%)

117 (35%)

10 (3%)

N = 330




187 (56%)
109 (32%)
36 (11%)

147 (44%)

22 (7%)

163 (49%)




49 (15%)
264 (83%)
5 (2%)

75 (23%)

68 (21%)

175 (55%)

N = 318




88 (27%)
223 (68%)
15 (5%)

103 (32%)

122 (37%)

101 (31%)

N = 326





32 (12%)
207 (81%)
18 (7%)


46 (18%)

167 (65%)

44 (17%)

N = 257



102 (31%)
121 (37%)
107 (32%)

N = 332

Notes: GC General Caste, SC Scheduled Caste, ST Scheduled Tribe.

Source: Initial household census.

Our other field site in West Bengal, Debra Block, is located in the
centre of Midnapore District, in a region that has benefited from the
spread of micro-irrigation and the recent growth of flower cultivation for
the Kolkata market. The locality that we worked in is made up of three
hamlets that are eight kilometres distant by non-metalled road from the
nearest small market centre on the KolkataKharagpur rail line. The
locality includes significant populations of Santal and Bhumij tribals
whose ancestors migrated as agricultural labourers from further west in


The everyday state and society

the District, and who settled permanently in the area over a generation
ago. Socially, the ward is a lot more cohesive than the Malda field site.
Bidupur Block, in Vaishali District, Bihar, is about forty kilometres
from Patna and fringes the northern bank of the Ganges. Agriculture
here is more capital-intensive than in our other Districts, and benefits
from canal irrigation and ready access to markets. The southern part of
the Block is largely given over to the cultivation of bananas. In caste terms,
the Block is dominated politically by the Yadavs, albeit in circumstances
of extensive competition, and this community, which we have classified
as Other Backward Caste (OBC) in table Table 3.1, provides 99 of 318
households in the research locality.4
In Sahar Block, Bhojpur District, Bihar, the poor quality of roads
means that our field site is relatively isolated, but even here irrigation
ensures that double-cropped agriculture (wheat and rice, with some
additional winter vegetables) is widespread. After three decades of sporadic violence, class tensions in the area have improved recently following the institution of peoples courts and the participation of the local
CPI(M-L) in electoral politics.5 Nevertheless, the very geography of our
field panchayat is structured around enduring caste and class divisions.
The land-owning Bhumihars make up almost a third of the villages
households, and the large Dusadh community, one of the lowest-status
Scheduled Caste groups in Bihar, has taken the lead in a string of labour
agitations. As will be clear from what follows, caste relationships continue
to be important in both Vaishali and Bhojpur in shaping poor peoples
social networks and their interactions with the state.
Finally, in Murhu Block, Ranchi District, Bihar/Jharkhand, we worked
in a village in which locally dominant adivasi communities are mixed with
sadans, or long-settled households from within the caste Hindu population. The adivasi communities are themselves quite diverse: Manjhi and
Munda communities each make up about 20 per cent of the village, but
there are also significant numbers of Oraons and Lohras.6 In contrast
to some other parts of Ranchi District, our field locality has a relatively
dense population that is supported mainly by rain-fed rice production. It
also has good access to the market and services of Ranchi city, some fifty
kilometres away.

For confidentiality reasons, the field localities are not named.

Despite the fact that many CPI-ML party workers still faced criminal charges and were
effectively forced in to hiding, the Block at this time had an elected CPI-ML Member of
the Legislative Assembly.
It should be noted that Mundas and Oraons generally do not consider Manjhis and Lohras
to be adivasi in this part of Jharkhand, although they are classified as Scheduled Tribes
by the state. They are not seen as original settlers or clearers of the forest.

Meeting the state


Income poverty and livelihoods

Severe income poverty is a fact of life for people in all five localities.
When we worked there in 19992000 we took account of government
definitions of poverty, as well as local peoples descriptions of their own
circumstances. But since we wanted to work in depth with members of
80 poor and 20 non-poor households in each Village-Block-District,
we also developed a working definition of poverty for our own (comparative) purposes. In effect, we developed an income poverty line that
made reference to a household with unskilled labour as its only source
of income, full employment, and a favourable ratio of earners to dependants. Households falling below this reference point were classified as
poor, and those falling significantly below this reference point, and particularly those with highly insecure incomes or showing other signs of
distress (food scarcity, dilapidated housing) were classified as destitute.7
As table Table 3.1 makes clear, a majority of the poorest households
belong to the Scheduled Communities, although in Bidupur and Sahar
Blocks a significant percentage of the Other Backward Classes (OBCs)
falls below our poverty line. Table Table 3.2 expands on the sources of
household income across the field sites. The general patterns are not
surprising: richer households have higher incomes and in many cases
enjoy a degree of security from salaried jobs and pensions, small businesses, and/or landholdings. Because of the severity of our poverty line,
some households that are fully dependent on unskilled labour have been
classified as non-poor (where most household members were of working
age and were fully employed), but dependency on unskilled labouring
work and landlessness is most highly concentrated in poor and destitute
Given the low wages that prevail in all these areas, seasonal variations
in employment can rapidly push many of the poorest households into
distress.9 Problem times vary across the Districts and Blocks, of course.

This is a particularly strict definition of income poverty, given the low wages that prevail
for unskilled labour in all five localities. A number of our non-poor households would
fall below the World Banks $1 a day poverty line (World Bank, 2000: 17). We used this
narrow definition primarily as a quick and dirty means to focus subsequent interviewbased work on poorer households, and we specifically do not assume that income and
assets are the only important dimensions of poverty.
In group interviews, poorer villagers evaluated the different income-earning opportunities
open to them. Cultivating their own land was deemed most desirable as it offered flexibility and independence: most of our respondents viewed white-collar jobs as completely
Wages for unskilled labour were generally at or below the government minimum wage of
Rs. 48 (US $1) per day in 1999/2000. Casual labour contracts vary dramatically across


The everyday state and society

Table 3.2. Household land ownership and income sources by poverty ranking

Field locality

Poverty level

Number of


labour only
































Notes: 1 Land ownership refers to agricultural land, rather than ownership of household
plots. Land ownership among poor and destitute households is high in the Midnapore
field site in part because of the government-sponsored land distribution: these were largely
marginal holdings under 1 acre.
2 Salaried employment is predominantly government jobs and pensions, but includes other
salaried jobs or businesses of equivalent security.
Source: Initial household census.

In Murhu, unemployment is high between the main rice harvest and the
following years planting; in Old Malda, flood disruption during the monsoon affects both on- and off-farm employment; and in Bidupur and
Sahar secure irrigation means that unemployment patterns are less seasonal in nature. Where distress is felt, reduced consumption is a common
first response but, beyond this, household-coping strategies vary greatly
between Districts and communities.10 Some individuals or families will
reduce their savings and/or take loans from employers at high interest
rates. Others will engage in temporary labour migration, home-based
craft work (such as bidi-making or weaving rush mats) and, particularly for
eastern India, but it is worth noting that womens wages are typically significantly lower
than mens.
10 A first response is to reduce food intake and variety significantly: a minimal diet of
rice, chillies and salt is typical, supplemented in some cases by gathered foods. Where
households face particular hardship they eat only once a day.

Meeting the state


adivasi households, the gathering of wild foods.11 In Debra, Old Malda

and Murhu Blocks, a lack of employment was deemed to be a major
constraint on well-being, and migration for unskilled work beyond the
village being forced to go outside was reported as a key indicator of
economic hardship.12
For the poor, however, poverty is felt not only in the stomach and the
pocket, but through markers of social distinction that are vital to their
experience of well-being or ill-being. In part, these markers are material:
being shabbily dressed, living in dilapidated housing, or being unable
to provide a guest with basic hospitality, amount to publicly performed
indications of a persons poverty.13 Furthermore, these experiences are
not simply the result of individual economic status, but are expressions of
deeper divisions within society. In Bihar, caste is important in structuring
these experiences of poverty: to be a chamar or a dusadh is to be of low
social worth polluting, even in the eyes of higher-caste villagers. These
prejudices are being challenged in Bhojpur, especially, where slights to
the honour of the Scheduled Castes can escalate quickly into public confrontations, but these changes are occurring in a context where caste
remains ever visible as a social marker. Elsewhere, caste or ethnic divisions are not as actively politicized, but they are far from being insignificant.14 In both West Bengal Districts, and in Ranchi District, adivasis
are marginalized for not having Bengali or Hindi as a first language, or
because of their religious beliefs and cultural practices. Stereotypes of
tribals at best as simple and reserved, but more commonly as profligate





Many of the wild foods (such as snails) gathered by adivasis are associated with social
taboos for other ethnic groups: for a fuller discussion of the use of such common property
resources, and their physical and social costs, see Beck (1994: chapter 7).
Although distress migration was widespread, not all migration was of this form. Some
households particularly in Vaishali, Bhojpur and to a lesser extent in Malda had
members working in relatively skilled jobs outside the village. In Vaishali and Bhojpur,
the extent of this migration, combined with agricultural intensification, was having a
tightening effect on local labour markets.
Seating ones guest on a stool or chair, and serving them tea and snacks, would constitute
proper hospitality in eastern India, yet even this would stretch the resources of many
poorer households in our study areas.
However much Bengali society prides itself on its indifference to caste, indigenous social
markers remain imbued with caste values. The term chhotolok (literally, little people)
is used to describe the poor, but applies primarily to the scheduled communities. Categories of munish/mazdoor, chasi/kisan and malik (labourer, farmer, landowner) at first
seem class-defined, but they are often applied to jatis as well as to individuals, and are
inseparable from notions of cleanliness. To move between categories is not simply to
change position within the labour market, but also to some extent to participate in a
process of Sanskritization (see Ruud, 1999). Ironically, it is probably only among the
bhadrolok (cultured or higher people), the urbane and primarily urban intelligentsia,
that questions of jati membership are largely irrelevant.


The everyday state and society

Table 3.3. Poverty level of female-headed households







9 (23%)
14 (37%)
15 (39%)

7 (29%)
11 (46%)
6 (25%)

0 (0%)
3 (100%)
0 (0%)

2 (15%)
6 (46%)
5 (38%)

2 (12%)
10 (59%)
5 (29%)


N = 38

N = 24


N = 13

N = 17

Source: Initial household census.

drunkards circulated unchallenged within mainstream society and in

the offices of government agencies.
Gender relations within and beyond the household are highly unequal
in all localities, and the experience of poverty is particularly harsh for
women. Income poverty rates are significantly higher among the small
number of female-headed households we interviewed (primarily widows
and divorcees) than for the locality as a whole (table Table 3.3). Even
where these households own land, they are often unable to earn incomes
from their holdings.15 Within all households, it is also women whose food
intake will be cut first and hardest at times of scarcity. Although Scheduled
Caste and Scheduled Tribe women do not face the same restrictions on
working outside the home that confront higher-caste women, they risk
harassment and abuse for the public display of ones poverty. Local
constructions of proper female behaviour also discourage General Caste
women, in particular, from participating in public events. Even among
groups where social taboos are weaker it is difficult for women to raise
their concerns directly.16
Education and capabilities
Gender, caste and ethnic differences also structure access to education
in all five field areas, much as one would expect. According to official
statistics, literacy rates for adults and children aged 7 or up were lower
in Bihar in 1997 for both men and women than in all other major states
(see table 2.2). An average literacy rate of 49 per cent, however, conceals
male/female levels of 62 and 34 per cent respectively, which are close to
the levels we found in our field sites (table 3.4a): the total literacy rate
was marginally higher in Vaishali but so also was the gender gap. Literacy

For a classic discussion of the issues, see Agarwal (1994).

One example would be in the conduct of our own research: interviews with poorer
women often attracted richer, male caste Hindu observers who took it as their natural
right to interpret the womens views for us.

Meeting the state


Table 3.4a. Literacy rates (7+) by gender, class and caste: Bihar field sites






Destitute households
Poor households
Non-poor households










Scheduled Tribe
Scheduled Caste
Minority (Muslims)
General Caste
















All households










Source: Household census, 1999.

Table 3.4b. Literacy rates (7+) by gender, class and caste: West Bengal field
Midnapore ward

Malda ward



















Scheduled Tribe
Scheduled Caste
General Caste







All households







Poor households

Source: Household census, 1999.

rates in West Bengal rank fifth among the seventeen major states, but here
too an average figure of 72 per cent conceals figures of 81 per cent for men
and 63 per cent for women. In West Bengal, too, as we quickly discovered
in the field, literacy rates among adults had been raised markedly in some
Districts by the Total Literacy Campaign (TLC) of the 1990s. This is one
reason why literacy rates (7+) were significantly higher in our Midnapore
field site (where the TLC had been active) than in Malda (where it had
not been: see table Table Table 3.4b), even though the rates of school


The everyday state and society

Table 3.5a. School attendance by gender, class and caste: West

Bengal field sites
Midnapore ward

Malda ward

School attendance (age 712)


Male Female All

Male Female

Destitute households
Poor households
Non-poor households


63% 25%
68% 50%
90% 74%




Scheduled Tribe
Scheduled Caste
General Caste


64% 37%
100% 57%
92% 83%




All households


78% 56%




Source: Household census, 1999.

Table 3.5b. School attendance by gender, class and caste: Bihar

field sites




Destitute households
Poor households

11% 10% 13% 27% 57% 0% 30% 26% 36%

31% 34% 28% 57% 63% 50% 55% 59% 51%
55% 59% 50% 36% 55% 25% 88% 93% 79%

Scheduled Tribe
Scheduled Caste
General Caste


All households

36% 38% 33% 52% 61% 43% 58% 62% 53%












Source: Household census, 1999.

attendance of children aged 712 were remarkably similar in both wards

(table 3.5a). Despite popular conceptions about gender equality in adivasi communities, gender disparities in literacy among the Scheduled
Tribe (ST) populations remained high (35 per cent in Midnapore,
19 per cent in Malda, and 23 per cent in Ranchi). In all five field sites,
the relationship between literacy and our income classes was strong. In
Bhojpur, the destitute were more than five times as likely to be non-literate
as were non-poor villagers.

Meeting the state


Of course, these figures present only a first cut at estimating school

attendance rates in our study areas. Trying to work out how often a child
attends school isnt easy. Interviews we carried out with parents in West
Bengal suggested that children are attending school on a more regular
basis in Old Malda than in Debra Block, but this was not confirmed by
our field observations. We formed the distinct impression that children in
Old Malda were often missing school to find work in the hotels and teashops of the nearby town. They did so, moreover, without their parents
finding out. Many adults were also working outside the locality. Bunking
off was less easy for children in Debra Block, where adult employment
was more often provided in the local agricultural economy.
The fact that some parents in Old Malda were not aware that their
children were missing school is something we take up in chapter 4. We
should not assume that all decisions about schooling are made by parents.
It needs to be firmly noted, however, that in all of our field areas a clear
majority of parents (over 70 per cent of respondents) expressed a desire
for their boys to be educated to Class 10 or higher. In West Bengal, it
was only among the Malpahariya (ST) community in Old Malda Block
that a significant minority of parents (22 per cent) suggested that their
sons could do without any formal education (as opposed to 34 per cent
of households who thought their daughters could be unschooled). This
was mainly for economic reasons (what job would my son or daughter get
even if she or he was educated?), or for reasons of language (teaching in
Bengali), or opportunity cost (earnings forgone). Among the poor more
generally, there was a clear feeling that boys should be educated in order
to take advantage of new jobs that were opening up in the non-agricultural
economy. Questions of empowerment/self-respect were also important:
education was thought to provide some protection for the boys and their
families against being cheated. There was little expectation that female
children would be directly productive in Midnapore (especially), and
family-related advantages were more often cited as a reason to educate
girls. This refers to the advantages that a minimum level of education
might confer on a girl in marriage and dowry markets.
The fact that children are receiving less education in our field sites
than would be considered desirable by their parents is mainly a product
of economic hardship. Economic constraints appeared to be especially
strong in the Midnapore, Malda and Vaishali field sites (table Table 3.6).
Parents also voiced concerns about the quality and utility of the education
that children were receiving in government schools, particularly in Malda
where there was widespread disquiet about the performance of two public
schools. Parents in both West Bengal localities thought it was necessary
for them to supplement (in Midnapore) or supplant (in Malda, in effect)








Source: Village questionnaire surveys, 1999.


Desired level can be

Not sure
Desired level cannot be
(No desire for education)
N. A.













Table 3.6. Ability to support desired level of education






























Meeting the state


the public educations their children were receiving with private tuition.
This generally cost Rs. 2530 per month for a child at the primary school,
which was a significant disincentive for poor families.
In chapter 4 we shall discuss the matter of loyalty to governmentprovided education, and the possibility that parents might take steps to
improve their local public schools by taking part in Village Education
Committees (or School Attendance Committees in West Bengal). At this
point, we simply note that the exit option was attractive to many parents
in our field sites. In some cases this was expressed in terms of a preference
for private tuition, or for the sort of segregated schooling that saw parents of General Caste pupils in our Malda field site opt to educate their
children in a neighbouring (and better provided) Gram Panchayat, leaving
the wards own schools almost entirely to children from the Scheduled
Communities. In other cases it was expressed in a sort of weary fatalism
that refused to accept that education was within everyones reach. In the
words of one respondent in Malda. If you cant afford the tuition, what
is the use of sending your children to school when they do not learn anything there? In this case, its better for them to stay at home and help their
parents (see also box Box 3.1).

Box 3.1 Parental attitudes to education in Malda

and Midnapore Districts
Gathered below are some of the responses that we collected from poorer
villages in Old Malda Block, Malda District, and Debra Block, Midnapore in 1999 when we asked them about the value of education.
The responses are representative but not exhaustive.
The prospect of unemployment hampers the interest in secondary and higher
education here. As we are poor, we cannot afford to pay the bribes that are
necessary to get a job.
A person who has outside knowledge has a good chance to find some job,
for example, helping out in a shop, doing accounts, etc. We should be able to
get some return from the education of our children.
Earlier, there was just agriculture. Now, there are different kinds of works. For
these jobs, one has to go outside and to be literate is more important. Literate
people cannot be cheated so easily while we illiterate people are cheated in
every sphere of life.
Our children will be able to mingle with children of more educated classes.
They will be less isolated. They will be more aware and have more general


The everyday state and society

knowledge. This will enable them to give their parents advice. They will also
be able to write [official] letters.
An educated person is less dependent on others. The boy can help his parents
with paperwork such as looking after land records, measurements, etc. He also
can teach the next generation.
An educated person cannot be cheated so easily. Being able to write and read
also means that the person can maintain certificates and deal with offices. This
will be of use for the whole family.

The present age is the age of literacy and education. Being educated, one can
achieve self-confidence and [better] mentality. [One can] build up towards
better understanding for ones own betterment.
Education for girls is needed for marriage purposes only. Education up to
class VI to VII is sufficient. We do not want higher education, because we
cant afford it. The girls wont get a job anyway, so higher education is

The non-state social networks of the poor

This contrast between the school and the home exposes something else
besides, which is that for many poorer people the state is not at the heart
of their social imaginaries or livelihood strategies. Notwithstanding the
ambitions and reach of the developmental state in India, most poor people
(indeed, most people) do not look upon sarkar as a first port of call when
they face difficulties or want something doing. The social networks of
the poor form a backdrop of spontaneous participation that will often
eclipse more formal participatory encounters with agencies of the state or
the broader development community. This is an obvious point, perhaps,
but it is one that is understated in the participation literature that we
review in chapter 4.
Given the nature of the poverty we have reported for our field sites, it
will be obvious that gaining access to employment and credit is a major
concern for many poor people. The support of local patrons and employers is key here, as it can be when it comes to getting a child admitted to
school. Opportunities for poor labourers to work as year-round farm servants are limited in all five areas, and the days of labourers being bonded
to particular employers through debt obligations have more or less ended.
Nevertheless, investing in good relationships with key employers is seen
as a practice that can provide mutual benefits. In Vaishali and Bhojpur

Meeting the state


Districts, where the labour market is relatively tight, employers play an

active role in maintaining these relationships. Members of some poorer
households stated that they would not go cap-in-hand to employers to
seek work. This would insult their honour and might weaken their wage
bargaining position. In the other Districts the opportunities for slack
season work are more limited and labourers can ill afford such pride.
Government public-works programmes are periodically undertaken in
all five field sites, but even in Midnapore, where these schemes have created the most work for local labourers, the states role as a direct employer
of the poor is by no means extensive (see chapter 4).
The importance to the poor of locally powerful individuals is further
underlined by the search for credit. Loans for poor Scheduled Caste and
Scheduled Tribe households come mainly from their regular employers,
particularly to meet seasonal consumption needs. These loans are repaid
with interest, sometimes by giving free labour instead of cash.17 Loans
from neighbours and relatives are also important, although these are generally for smaller amounts than are those obtained from employers. If
neither source is available or sufficient, known moneylenders inside or
beyond the village are approached. Government loans from schemes such
as the Integrated Rural Development Programme provide credit on softer
terms but, again, they are not sufficiently available to take the place of
richer villagers who can provide ready money.18
Importantly, throughout all our Districts, what can be described as
private, market-based relationships dominate transactions in the spheres
of work and credit. The relative bargaining power of the parties involved
varies, and in some cases the support that landlords and employers give
to their workers shows a degree of care that reaches beyond a narrowly
economic relationship. There is also a degree of mutuality in developing
these networks of support. Trustworthy labourers are valuable to their
employers, whose behaviour in respect of labour should not be construed
as altruistic. The transactions, however, remain the private business of the
parties concerned. Although no one talked about richer villagers having a
generic duty to help the poor, they emerged nonetheless as key individuals
within poorer peoples social networks.

Typical interest rates for these loans would range between 5 and 10 per cent per
Significantly, schemes such as the IRDP are specifically intended to provide productive
assets, and as such give loans of the order of a few thousand rupees to a fortunate
few households. Although many of these loans are diverted by the poor into short-term
consumption needs (cf. Williams, 1999), the design of the IRDP and equivalent schemes
does not meet the general need for seasonal credit.


The everyday state and society



workers and

civil servants



Tribal councils
(minor role in

Key providers of
work and credit:
power relatively

active, but
hold great

Little direct
contact with

PRIs weakly
instituted; near
defunct School


Tribal councils
(minor role in

Key providers of
work and credit:
land reform
slightly limits
their power

Various CPI-M
active; fewer

Little direct
contact with

PRIs strong in
most areas of
village life;
active School


Caste councils
(some dispute

Key providers of
work and credit:
also act as

active: support
divided by

VLW main
source of direct

VEC present
and partly
PRIs dissolved
in 1999,
reinstated in


through caste

Key providers of
work and credit:
power restricted
by extreme-Left
(Naxal: here
CPI-ML) actions

active: support
divided by
party lines,
with caste


Janata Durbar
active; Peace
Committee and
weak VEC
present; PRIs
dissolved in
reinstated in


Tribal councils

Key providers of
work and credit:
little role in
other spheres

other party

Little direct
contact with

VEC present,
but weak; PRIs
dissolved in
1999 and
reinstated in

Figure 3.2 Schematic account of key sources of support for the poor,
by District

Commonalities between the field sites in the provision of work and

credit dissolve when we look at the sources of non-financial support upon
which the poor are dependent (figure Figure 3.2). In Vaishali, higher-caste
individuals are able to project their power beyond the economic sphere
of work and loans. As informal leaders they still play an important role in
mediating village disputes and manage to retain a degree of respect and
standing for doing so. Their power, however, is eclipsed in the sphere of
government-funded development programmes: rich and poor villagers
alike recognize that connections to powerful political figures are central
to controlling such funds. The local Member of the Legislative Council
(MLC), BB, who belongs to Laloo Yadavs Rashtriya Janata Dal, has a
particularly important role to play, although he must have regard for other
local brokers and intermediaries. The government Village-Level Worker

Meeting the state


is also a key source of support for poorer villagers. Payments to him are
often necessary if villagers are to access those individualized aspects of
state support, such as pensions, loans or improved housing under the
Indira Awas Yojana, which are aimed at the poor. Formal government
institutions are weak in the study village. Bihars panchayats were finally
suspended in 1999, leaving unelected caste councils with some residual
influence over village affairs.19
In Bhojpur, three decades of intermittent but violent conflict between
Naxalite forces (the CPI-ML) and the Ranveer Sena had only recently
quietened down when we conducted our fieldwork. Agricultural relations had normalized in 19992000, and labourers and landlords were
more aware of the need for generalized exchange. But the legacy of the
conflict had a lasting impact on the social networks of poorer villagers.
Despite an uneasy truce over agricultural pay and conditions and the
ownership of land, the violence had undermined any legitimacy that
higher-caste villagers or traditional institutions might once have enjoyed.
Class- and caste-consciousness were high, with lower caste groups actively
pressing for their rights, particularly in relation to political and developmental activities. In this still volatile situation, the police and other formal government agencies, such as the Janata Durbar (peoples courts),
were key sources of help in resolving disputes. The Janata Durbars are
discussed further below, and in chapter 6, but they had the effect of
making higher-level government officers much more visible here than in
the other Districts: villagers were more aware of the presence of these
officers, and were sometimes able to petition them directly for different forms of government support. These occasional meetings aside, a
breadth of local political intermediaries had emerged to represent different caste groups within the village and to press their claims on the
In Murhu Block, Ranchi District, contact with government officials was
far less direct than in Sahar Block, Bhojpur District. Despite the suspension of the old panchayats, the Mukhiya (panchayat headman) of our field
site still acted as virtually the sole conduit to the government development bureaucracy. His reputation within the local Block office for being
developmentally minded had allowed the village to draw down significant resources through various development programmes (see chapter 5).
The benefits of these programmes had been broadly distributed between
different social groups within the village, and the Mukhiyas role had until

Prior to the 2001 elections, Bihar had last held panchayat elections in 1978. Not surprisingly, the vast majority of panchayats had been effectively inoperative many years before
their official dissolution.


The everyday state and society

recently been unchallenged. The village still had strong indigenous tribal
institutions for dealing with conflict resolution and other major decisions,
and a range of community-specific traditional leaders were key sources of
support for rich and poor villagers alike. Support in the search for slack
season migration opportunities usually came from immediate neighbours
or relatives. Given the dominance over public affairs that was exercised
by the Mukhiya and some other community leaders, there was little scope
for a wider range of political intermediaries to become established.
One might have expected that village social networks in the West Bengal
Districts would have been reshaped by the presence of active panchayat
institutions, and in Debra Block this was indeed the case.20 Panchayat
members and party workers dominated many areas of public life. Six male
CPI-M activists and the partys female council member lived in the ward.
As a group, their social and economic backgrounds reflected those of their
constituents, and they were active in helping households in all manner
of tasks, from school registration to settling wage disputes.21 Interviews
with villagers suggested that these people generally were trusted, and were
seen to be playing a valuable social role.22 Many villagers pointed to the
female Gram Panchayat (GP) member as a person to turn to if there was
a dispute, and this would seem to signal respect for her office over and
above ascriptive sources of power: her gender, caste and poor economic
condition appeared to be no impediment to her playing a public role.
More generally, a range of local party workers (including some less active
opposition members) provided a broad base of contact points between
poorer villagers and different branches of the state. The lowest-level government workers could often be seen within the villages or the local
panchayat office, although other civil servants would generally have to
be contacted through panchayat members. Panchayat representatives had
thus displaced the leadership roles played by larger landholders before
the 1970s, and it was only within the Scheduled Tribe communities that



Despite West Bengals good record in innovation with PRIs, the decentralization
of responsibilities to the grassroots was not matched by devolution of financial
resources equivalent to that in Keralas recent experiment with grassroots planning (see
chapter 7).
The senior party worker, NS, had completed secondary school, and owned a small
amount of land (one acre): he and two other paid party workers enjoyed some security
through their party connections, but were not significantly well off. The remaining party
members all came from households that suffered from seasonal unemployment: three
fell below our poverty line, and only two were caste Hindus.
Particular value was placed on party workers knowledge and experience of government
and their engagements with poorer households. Typical comments made about one of
the full-time party workers by poorer interviewees were that: He is the friend of poor,
knowledgeable and a good person. All people respect him, or more simply: He is my
neighbour. He knows my condition.

Meeting the state


indigenous institutions retained some importance, primarily over cultural

In stark contrast, the same framework of political institutions within
the Old Malda field area was associated with a very different pattern of
social networks. A single panchayat samiti (Block-level) member cultivated
direct and personal loyalty from people in the locality.24 We return to the
mediation of power in chapter 6, but it is worth noting here that while
many poorer households might have been aware of panchayat institutions,
and the presence of other political figures, for most such people this man
was the de facto boss of the locality. Other figures were important by virtue
of their control of Gram Panchayat programmes, but this power was also
personalized in large degree. Although these new political representatives
had displaced an earlier generation of high-caste village leaders, they had
done so without producing public support for functioning panchayati raj
institutions. Politics remained clientelistic, which is to say that services
or goods were geared very directly to political supporters of the patron.
For the poor in each of our Districts, groups of local uppers individuals with economic and/or political resources are important sources
of support. Although the social networks these individuals produce vary
greatly, a number of general points still emerge. First, the resources controlled by employers and moneylenders are key to the survival strategies
of many among the poor. Government assistance provides access to work,
credit and other benefits for some people, but it does little to diminish
the roles of the economically powerful. Second, where poorer people
do require help from government agencies, their contacts with the state
are for the most part routed through low-level intermediaries. The very
poor thus face a double burden: not only are their sources of livelihood
highly insecure, but their social networks are often less able than are those
of richer households to provide access to key resources and support in
times of crisis. From their perspective, staying on the right side of powerful
intermediaries political workers and richer villagers is an essential part
of their everyday participation in village life. Finally, the degree to which
the social networks of the poorest have been penetrated by sarkar does
not simply reflect state-level differences in regime type, or the current


Conflicts over personal relations such as marital infidelity might be turned over to
traditional leaders and priests. Nevertheless, CPI-M penetration of public life went as
far as sponsoring elements of tribal festivals.
When villagers were asked about who they would turn to for help in various fields
accessing health-care, work, loans, sorting out village disputes this individual dominated in almost all cases, and to a degree that was not matched in the other Districts.
The one exception was education: state schools were failing miserably in the locality but
he had not chosen to intervene here, perhaps because of the limited opportunities for
kick-backs and skim-offs.


The everyday state and society

institutional arrangements of local government. Important differences

within Bihar and West Bengal are also shaped by experiences of political
mobilization, as well as of localized patterns of leadership and social dominance, and past government arrangements, among other factors. Within
the uneven geography of statepoor encounters that emerges, the range
of aspects of public life within a village over which sarkar exercises some
direct influence also varies greatly: this much should be apparent from our
discussion of Ranchi, Bhojpur and Midnapore. Nevertheless, the experiences that poorer people form of the state are not simply demand-led.
In addition to being a (sometimes minor) part of their livelihood strategies, government imposes itself on villagers in other ways, as we shall now
Seeing the state from the villages
Chris Fuller and John Harriss have suggested that to enter a local administrative office, a government school, or a police station is to cross the internal boundary into the domain of the state, whose conceptual separation
from society is perhaps most ubiquitously symbolized by all its specialpurpose buildings with their painted notice-boards outside (2001: 23).
But they go on to note that the everyday state remains a difficult object
of study in spite of these official attempts to mark out its distinctiveness.
Part of the difficulty arises because the state is simultaneously a dispersed
collection of institutions of government (the state system) and a powerful myth about their imagined unity and purpose (the state idea). In
addition, when poorer people meet local state officials, they do so with
reference to their non-state social networks. They draw on the same social
and organizational practices that help them to negotiate their poverty. For
the most part the state is met through the persons who bring these practices to life, as well as through a broader range of caste leaders, brokers
(dalaals) and political fixers (pyraveekars).
We will expand on the role of mediators in shaping patterns of rule in
chapter 6, but whether the poor engage sarkar directly, or via an intermediary, many encounters with lower-level state officials will blur the boundaries between the state system and what lies outside it. Akhil Gupta argues
that, in the conduct of government business, these officials consciously
adopt the practices of traditional patrons: they are willing to take bribes,
and they are as likely to operate from their homes or tea-shops as from
their desks. He goes on to note that the performances that are required
when meeting with government officers (in official and unofficial spaces)
blur these boundaries still further: although impersonal norms should prevail in these encounters, poorer men and women may touch officers feet

Meeting the state


and present their relationship with government as one of supplication and

not of citizenship.25
In telling the story of the ways in which poorer citizens see the state from
the villages, we need to pay careful attention to the details of these encounters and their implications for both the state idea and the state system. As
will become clear below, poor people are not always passive supplicants
when they meet the state in eastern India, and their contacts with the
everyday state are important in informing their ideas of how sarkar does
and should work. The expansion of the technologies of rule described
in chapter 2 not only extends the states reach towards the poor, it also
provides poor people with glimpses of how power operates and how they
might work the system to better advantage. But we begin our discussion
with a brief account of the geography of local government institutions in
our study areas, for this is where the poor see the state in the very tangible
forms that Fuller and Harriss describe.
Geographies of the state
The geography of the local state includes a vast array of institutions that
runs from the District, where government offices fill an entire campus and
where the staff can number several hundred persons, to the Block and
(in West Bengal at least) the humble anchal office, which might be staffed
by a handful of the lowest-grade government workers. It is not possible
here to provide a full institutional mapping of government in each of
our five study sites, but figure Figure 3.3 outlines the organization of the
Employment Assurance Scheme in West Bengal from the perspective of
the official state. From a bureaucratic point of view, the range of elected
representatives and government personnel listed here is significant and
well-ordered. Institutions and people are arranged into clear hierarchies
from the IAS-cadre District Magistrate down to the lowliest peon. For
many poorer people, however, and many women especially, a geography
of the state which runs from the commanding heights of the state capital
down to the extension workers in the trenches is not at all what they see.
Figure Figure 3.4 shows the government in West Bengal as it is viewed by
two adivasi women in our Midnapore field site. For Debati and Rani, the
people who really matter in their lives are the local panchayat member and

Gupta (1995) more often refers to discourses than to performances. Fuller and Harriss
argue (2001: 13) that discourse is somewhat carelessly used in Guptas work to refer
to both practice and action. We prefer the idea of performances as this explicitly highlights the importance of practice and action. Furthermore, performances can involve the
deliberate playing out of roles for public consumption, regardless of private motivations
to the contrary (see Scott, 1990).






The everyday state and society


Key Political



Zilla Parishad

(Chair of ZP)

District Planning


Chairs of


ADMs and AEO

(MPs, MLAs,
and ordinary ZP

Public Works

(Other District
Level staff)

(Chair of PS)

General Meeting
of Panchayat


Chairs of



(MP, MLAs,
and ordinary PS

Public Works

(Other Block
level staff )


Gram Panchayat


Gram Sabha
(public meeting)

Job Assistant

(Block Council)

Gram Panchayat


Gram Sansad
(public meeting)
Committee (lay

Note: An anchal is a village cluster. It consists of about a dozen wards, and up to two
dozen villages or smaller hamlets.

Figure 3.3 The EAS in West Bengal: key actors and institutions

the panchayat pradhan. The Block- and District-scales of government, still

less Kolkata and Delhi, are notable only by their absence.
Of course, getting villagers to map the state in this way is subject to all
sorts of methodological difficulties. The visual impression given is clearly
one of absence, and this paucity of information might be at odds with
the verbal accounts that such women can offer of the state. In this case,
both women knew the names of the Pradhan and the Gram Panchayat
(GP) member of their electoral booth, and they considered both of them
to be important people. The GP member, a tribal woman, is the person

Meeting the state


Figure 3.4 Chapati diagrammes of the government: Midnapore field


they turn to first if there are problems, or if there are quarrels between
villagers about land or some other issue. The Pradhan is also seen on a
regular basis. He comes to official meetings in their neighbourhood a few
times a year, accompanied by the Panchayat Secretary and other officials,
and is often seen outside these formal settings. The former GP member of
their booth, RT, is known to them as someone well-versed in government
matters. They have also visited the anchal office, although they are not
acquainted with the purpose of the Block office. But if other government
people are occasional visitors to their neighbourhood, the women have
no idea who they are or what position they have, a situation they blame
on their own illiteracy. They know that government sees them as STs and
BPLs, but they have only a hazy understanding of how this is supposed
to feed through into concrete rights and benefits.
Accounts such as these raise a number of important issues. Very obviously, they highlight the isolation of the poorest families, and poor women
in particular, from higher-level government officials (compare with the


The everyday state and society

male CPI-M supporter in figure Figure 3.4). These women are socially
and economically active within and beyond their neighbourhood and village, and yet the Block still less the District offices of the state are
not an important part of their life-worlds. For them, the state is experienced very directly and very locally, and not very often. Their accounts
would surely resonate with those of women in the Lohra (ST) community in Murhu Block, Ranchi, where views of government were typically
limited to the person of the mukhiya himself. Many of the Lohras we
spoke to assumed that government development schemes were undertaken as a direct result of his personal work and benevolence. Once again,
engagement with the state system was limited and ideas of government or
authority are often indistinguishable from individual holders of an office.
Nevertheless, it is important that these perceptions are not merely seen
as a lack of awareness, or as the leftovers of a pre-modern consciousness. Rather, they are actively produced through current patterns of social
and political exclusion: although Debati and Rani may blame their own
lack of education, their views of the state reflect broader aspects of their
Not all villagers suffer this degree of isolation. A good number of the
people we spoke to were familiar with the Block office, and some villagers were making careers as party or social workers, or informal
brokers, who could link other villagers to government officials and services. In any case, the government centres, services and staff that are most
visible to poor households beyond the pradhan or mukhiya are primary
schools and anganwadi centres, which feature alongside the occasional
presence of a Village Level Worker (Bihar in 1999), or Gram Panchayat
staff members (West Bengal). For poor men, especially, the Block is also
an important site for encounters with the state, and in both Bihar and
West Bengal the Block Development Officer (BDO) is the best-known
figure in the civil service. He or she acts as a combination of development
planner, local magistrate, and manager of other civil servants. Alongside
the BDO, the Block is also the lowest point at which many government
departments have officers, and Indias developmental Blocks are in many
cases contiguous with that other key state institution, the thana or police
The District offices are generally more distant and only a small proportion of the poor will make this journey to visit government officials
or to participate in a court case. For the villagers we worked with in
Debra it involved a 30-kilometre train ride; villagers in Sahar were faced
with a three-hour walk and bus journey. The District offices are imposing and potentially impenetrable. Access to key figures at this level is
unlikely to occur without a mediator. This will be especially true of the

Meeting the state


office/home of the District Magistrate (DM), or collector, which in Bihar

will be guarded by jawans carrying sub-machine guns. (In West Bengal
the DM is often no more accessible to ordinary citizens, but his or her
office is more likely to call to mind that of a Chief Executive Officer.)
As in the Block Office, the browning walls of the Collectors main room
will be decorated with the occasional photograph of Gandhi and Nehru,
and perhaps also of the Prime Minister of the day. There will also be a
wooden plaque upon which will be listed the DMs who have served in
the area since 1947, or still earlier in many cases. The continuity of the
state, and of the all-India services, is clearly a source of pride.
Given the remoteness of the District- and state-levels of government,
differences between the lowest tiers of the development state in West
Bengal and Bihar can be especially important for the poor. Making up
the front line of the state in Bihar are Block-based staff. The VLWs and
karamcharis make their peripatetic rounds to the villages they are assigned
to, but villagers might need to contact them at the Block office if they shirk
their duties or if they need assistance on a more urgent basis. Although
staffing levels might be roughly equivalent in West Bengal, the panchayat
secretary and panchayat karmis are physically closer to villagers there
because of the network of anchal offices, of which there will generally
be between six and twelve per development Block.26 The anchal staff
might only perform basic tasks, and most offices will be simple tworoom buildings with the most rudimentary facilities (in 1999 computers
were unheard of, and telephones by no means universal), but this at
least provides a convenient site at which government officers and Gram
Panchayat members can be contacted. The transaction costs of meeting
the state are therefore lowered, and the physical linkages between the
bureaucracy and elected panchayat members are demonstrated on a more
regular basis.27
As we will note further below, it would be wrong to infer that poor
peoples greater proximity to the local state leads directly to a more active
sense of citizenship, as some of the good governance literature perhaps

Anchal offices will generally be within 6 km of most villagers. Block offices can be up to
25 km away.
West Bengals anchal offices typically contain a room for the pradhan and one for the
office staff. At the Block level too, panchayat samiti members and Block civil servants
will often share the same building. In the panchayat rooms, walls are again adorned with
a roll-call of past chairpersons, mirroring the list in the BDOs chamber. The symbolism
is probably not lost on the majority of those visiting these offices: elected members are an
official part of sarkar, working alongside civil servants, and they are there as a permanent
fixture. Although the infrastructure of the state is lacking below the Block level in Bihar,
the times at which the village extension staff are present in the Block offices are at least
meant to coordinate with local market days, thus widening villagers opportunities to
contact public servants.


The everyday state and society

suggests (see chapters 4 and 5). Placing the structures of a decentralized

state on the statute book in West Bengal does not always mean that offices
are fully staffed and functional on the ground. Even when they are, their
presence does not automatically empower women hurt by high levels
of social exclusion like Debati and Rani. Conversely, at the time of our
fieldwork in Bihar, it was in part the very distance between villagers and
the lower reaches of the state (the BDO, say, and the MLA) that made
poor peoples agency in targeting the state so visible in terms of journeys
made or brokers used. (Recall Polus B in chapter 1.)
Encountering the developmental state
To understand what these geographies of the state mean for poorer people
we need to move on from mappings of state institutions to a consideration
of how poor people encounter government personnel. The range of peoples encounters with the state across our field sites is vast, and extends
far beyond the developmental institutions described above. Alongside
these officers and offices, the poor will have day-to-day encounters with
schools, health services and various arms of the states disciplinary apparatus, all of which will contribute to their sense of how, and for whom,
government operates.28
We shall hope to touch upon some of these encounters later in the
book. Nevertheless, our focus here is on the developmental state, and
we conclude the main part of this chapter with three brief vignettes from
our field sites in Ranchi, Bhojpur and Midnapore. We consider how poor
people access the states developmental institutions, and whether and
how their access to state-sponsored development opportunities rubs up
against that reliance on local uppers that we described earlier. We also
focus on the proactive roles played by some individuals within a very
heterogeneous population of poor people.
Challenging the Mukhiya in Murhu
We begin with our Ranchi field site, as it was here (and in Malda) that
the developmental activities of the state were most clearly monopolized

In Malda, in particular, the state was experienced through the location of a Border
Security Force barracks adjacent to our field site an important presence when the traffic
of people and goods across the border with Bangladesh is commonplace. In Ranchi, the
disciplinary arm of the state would be felt more often through the presence of forest
guards and the multiple army bases that fringe Ranchi city. In Bhojpur the police and
various paramilitaries have been an occasional but important presence in the area during
recent conflicts between the Ranveer Sena and the CPI-ML.

Meeting the state


by a single person. Block officers in Murhu described the Mukhiya of

our panchayat as a trusted individual with a reputation for being developmentally minded. For most villagers, his past record of drawing down
benefits from the state meant that his expertise in these matters went
largely unchallenged. As we noted above, this record of achievement led
some Lohras to conflate the position of the Mukhiya with sarkar more
generally. Notwithstanding the fact that the District Magistrate (DM)
had visited the village to inaugurate the completion of an Indira Awas
housing scheme, many STs thought these houses had been provided by
the Mukhiya himself.
Not everyone in our field locality was prepared to accept this interpretation. Sudhir M, an educated and unemployed young man, used the
opportunity of the SDOs visit to press for the construction of a girls toilet
for the village school. The SDO encouraged him to carry out the scheme,
and Sushir approached the Block office to formalize things. This wasnt
easy: initially, the Block staff were not willing to pass schemes to someone
other than the Mukhiya, and it was only Sudhirs threat of going back to
the SDO that enabled him to be handed the project. Sudhir later prepared
and submitted the pension applications of a group of Lohra households
to the Circle Officer (CO). The CO happened to be a college friend of his
who had recently been recruited to the Block. Sudhirs chance meeting
with the SDO, along with his frustration at seeing the Mukhiya control
affairs within the village, was beginning to turn him into a fixer in his own
Encounters with the state can be more difficult for those without
Sudhirs education or contacts in government. Ranjit M, another unemployed tribal youth in the village, told us that he had repeatedly visited
the Block office over the course of a year to question the dominance of
the Mukhiya. The Block officers taunted him at first. He was curtly told
Chappal ghis jayeaga chakker katete per tumko scheme nahin milega (You
can wear your sandals out coming here, but you wont get a scheme).
But Ranjit persisted. Eventually, after talking to Block-level Congress
party workers, he found out about the existence of the official roster that
detailed when aam sabhas the village open meetings that are meant to
decide upon the selection of executing agents for development schemes
should be held. A proper aam sabha had never previously been held in
the village. The Mukhiya and members of the Block staff had simply concocted minutes of meetings to keep the official records straight. Armed
with this information, Ranjit pushed for a proper meeting, and was successfully elected as the executing agent of the next development project
in the village.


The everyday state and society

Neither Ranjit nor Sudhir acted out of altruism. Sudhir collected Rs. 50
from each Lohra household to get their pensions processed, and both men
expected to earn more than prestige from managing a small development
scheme.29 Given the limited means of their families it would be wise to
skim off some financial benefits. We return to the motives and careers of
such small-time fixers in chapter 6, but our point here is rather different.
For some among the poor with drive and insight, chance meetings with
powerful outsiders can be a first step in targeting the state.
The politicized poor in Sahar
Sahar Block has seen a long history of collective actions on the part
of the poor, mainly around land and labour issues. It is perhaps not
surprising, then, that poorer people have more champions than would
commonly be the case in Ranchi District, including Scheduled Caste
members of the CPI-ML. Nevertheless, the connections between a career
in agrarian politics and confidence in encountering the developmental
state are far from direct. The violence of agrarian politics means that
many grassroots activists have been driven underground after being listed,
falsely or otherwise, on police First Incident Reports for charges such as
rioting or murder. In any case the ideological thrust of the extreme Left
has been towards overthrowing the state rather than engaging with its
developmental agendas.30
One way in which the legacy of agrarian politics has assisted state
poor encounters is through the practice of Janata Durbars. These public
hearings have been widely used in Bhojpur and other troubled Districts.
They are a means of diffusing sources of tension before they break out
as episodes of public unrest or organized violence. These meetings bring
higher-ranking officials, including the DM, closer to the villages, and
they have given the most astute members of the rural poor a degree of
understanding of how the state system works. On occasions, too, poorer
villagers have been able to appeal directly to a DM for the redress of
certain grievances.31



The money Sudhir collected was described as money for chai pani literally a payment
for tea and water for the Block staff. These would have been small bribes, of which
Sudhir would have most likely retained some amount for himself. (See also table 1.1.)
The areas MLA was a case in point. Although a competent leader of agrarian struggles
in the area, he appeared to be coping less well with the task of using the developmental
state in favour of the poor.
At one such meeting, Laxman, a Scheduled Caste leader, managed to ensure that a
community building was placed inside the harijan neighbourhood of the village. He had
convinced one of his Scheduled Caste colleagues to donate the land on which the building
was to be built. The DM approved the deal and granted money for its construction under

Meeting the state


The confidence gained by using the developmental state in this way can
lead to more direct challenges to the power of landowning Bhumihars.
We saw this in a public meeting to select an executing agent to run a
Rs. 70,000 irrigation tank repair project under the Million Well Scheme.
A Scheduled Caste leader (neta), NR, led a campaign to put himself or
another Musahar forward as a rival to the Bhumihars chosen candidate.
NR said he didnt want someone to be elected who would be a stooge for
the landowners. It was important that labourers on the scheme received
the full minimum wage. By arranging for a large number of women from
his own community to attend the meeting, NR helped to ensure that
the vote went in favour of his community. Although many Bhumihars
complained that the election was unfair, the poorest and socially lowest
group in the village gained control of the tank repair project.
In this and other less dramatic examples from Bhojpur we see that a
section of the poor is trying hard to get the state system working in its
favour. Placing polite requests for community assets in front of the DM,
or knowing how to gerrymander public meetings, suggests a nuanced
understanding of the rhetoric and practices of the official states descriptions of itself. The success of such tactics, however, ultimately relies on the
understanding, won through years of agrarian conflict, that the Scheduled Castes must protect themselves against the Bhumihars when the
state is not around to act as a referee. There is a saying in Bhojpuri that
only when every head is broken will everyone see sense, and the threat of
broken heads is precisely what has forced some members of the Bhumihar community to concede ground to the Scheduled Castes when dealing
with the developmental state.
The power of the panchayats in Debra
Our final snapshot of poor peoples encounters with the developmental
state is in the more mundane form of a gram sansad meeting in Midnapore. These meetings have been mandatory since the early-1990s in West
Bengal, and the Left Front Government has seen them as one important
mechanism for extending poor peoples engagements with the panchayat
system. They are also meant to enhance the accountability of elected
the EAS. Unknown to the DM, Laxman had pulled off quite a coup: the donated land
was about to be the subject of a court case, and potentially was going to be lost by
its owner after what would inevitably be a difficult and costly legal battle. Both the
donation and the building had been given the DMs public approval, and would thus
be the unquestioned property of the harijan community. It would also provide a public
meeting place away from the control of Bhumihars. Laxman would get some personal
profit from arranging to have the building constructed.


The everyday state and society

members and increase the transparency of panchayat decisions. Meetings

should be held twice a year in a public place a school building or a patch
of open ground in every ward of every panchayat.32
We observed a meeting in May 1999 that was attended by a little under
eighty villagers in our ward, most of them from poorer households, and
various local CPI-M party workers. The meeting was chaired by the
ward Gram Panchayat member, with the support of the panchayat secretary. Representatives of the Block-level panchayat and government staff
were also present. The proceedings were formal and business-like: the
Gram Panchayat member acted as chair, and the Block panchayat member reported on general progress in village development. He reported
a decrease in development funds from the government over the past
few years, and gave exact figures about how much money flowed to the
panchayat under various government schemes. He also mentioned the
Gram Panchayats efforts to raise its own money (through afforestation
schemes, a toll tax, and other measures), and told the villagers to be
more health conscious and make use of the GP sanitary scheme. The
meeting then moved on to discuss development priorities in the area.
Lay participants, and particularly women, were expressly invited to air
their views and make suggestions. The elected GP members and other
party workers remained surprisingly silent throughout this discussion:
as we found out later, they had already primed villagers to put forward
suggestions for development work (see chapter 4).
From the perspective of poorer villagers, this everyday contact with
the state acts to reinforce a set of norms and expectations of government. Meetings can be ritualistic, and the contributions made by poorer
villagers might indeed derive from prior prompting by party workers,
but the discussions we observed still managed to break away from a
scripted performance. Figures for government spending were contested
and the failings of panchayat members were pointed out. Alternative plans
were also proposed, particularly by wealthier, educated villagers, and
members of the opposition parties. These and other disruptions were
handled in a professional manner by the elected members who chaired
the meetings. Debate helped to entrench certain norms of democratic
governance: of civil behaviour in meetings (where opposing views are
dealt with respectfully), of the idea that peoples input into development

In Midnapore and Malda in 1999/2000, panchayats were being monitored by the Block
office to ensure that meetings were held and were fulfilling their proper functions. Draft
agendas for the meetings were produced centrally by the government of West Bengal
to ensure that core business was attended to: these were altered to fit locally relevant
subjects by the Block office, and villagers themselves could (in theory) add additional
business for discussion.

Meeting the state


decisions should be rational and needs-based, and of the duty of citizens

to contribute to local government through panchayat-level taxes.
Most poor people remain passive participants in these performances,
but we should not forget that grassroots party workers are also drawn
from their ranks. Just as importantly, even the people who stayed quiet
could see their fellow labourers and sharecroppers presenting opinions
on an equal footing with their maliks, and the latters views did not always
prevail. As a piece of social theatre, this may be less dramatic than the
arrival of the District Magistrate at a Janata Durbar, but it reinforces the
normality of the developmental states active presence in village life.
Several of our respondents in eastern India would tell the story of these
different encounters in terms of an evolutionary model of statepoor
relations that moves from tribalism through confrontation and on to some
sort of social-democratic settlement.33 In our view, this model is mistaken. If the Mukhiya is a major power holder in our Murhu field site
this is less the result of his traditional role within adivasi society than it
is a function of his power to monopolize the funds and information that
arrive at the Block office of a developmental state. What we see in all
three Districts is the playing out of power relations between government
officials, local power holders and the poor a power-play that has important implications for the states reproduction, both as a system and as an
As poor people retell, compare and discuss their sightings of the state
within the villages, it should come as no surprise that the lessons learned
from each encounter will be carried forward into fresh ones. The personal
characteristics of key individuals a pro-poor labour officer, a corrupt
sub-inspector of police, a politically neutral BDO will often loom large
in the minds of poorer people, and as such their reported views of the
quality of government are likely to be highly discontinuous or fractured.
Encounters with the developmental state also build up a dynamic picture of it, both as an idealized set of values and practices (the state as
it should work), and also as its flawed but more commonly experienced
counterpart (the state as it does work). The sense of what Akhil Gupta
calls translocality is important here: The state requires us to conceptualize a space that is constituted by the intersection of local, regional,

In Ranchi, Block officials repeatedly referred to tribal communities as simple, innocent

or naturally disinclined to political activity. Midnapore, in contrast, was often described
as being mature in its government and politics.


The everyday state and society

national, and transnational phenomena . . . Bringing the analysis of public culture together with the study of the everyday practices of lower levels
of the bureaucracy helps us to understand how the reality of translocal
entities comes to be felt by villagers and officials (Gupta, 1995: 392).
Experiences of these everyday practices (of officials who are democratic,
discriminatory, or corrupt) and of public culture (of the heterogeneous
local values that penetrate the state) contribute to an imagining of the
state among the poor that is far richer than is conveyed in the chapati diagrams of Debati and Rani. Women like this see the painted signboards
that mark out the state from society, but they also have a fairly shrewd
idea of how these distinctions break down in practice.
What is now being proposed in India is that understandings such as
these will be deepened as poorer people come to meet the state through
participatory development schemes and the agendas of good governance.
As we shall see, however, the states attempts to present itself as a new
agent of participatory development are read by poorer people against
their previous and possibly negative experiences of participation, and
with close regard for the vitality of their non-state social networks. This is
one reason why poorer people do not always greet participatory development projects with the sort of acclaim that government boosters would
like to see or hear. This is certainly true of the Employment Assurance
Scheme (EAS), and it is largely true of attempts to set up Village Education Councils (VECs) or their equivalents. Exactly how this is so, and
for what reasons, is something we discuss further in chapter 4.


The idea that ordinary people should have a say in the ways in which
government programmes in the United Kingdom are run on their behalf
is so well established that it has become a commonplace. Parents there
have long been afforded some of the rights of scrutiny and decisionmaking which are now being extended to parents in rural Bihar through
Village Education Committees. In principle, too, they are able to exercise
some control over local planning and budgetary decisions through their
participation in local government elections again, just as in India. We
know, however, from the UK and other richer countries, that turnout
rates in local elections are often quite dismal, and that the democratic
process can be undermined by backdoor deals with commercial interests,
including property developers. It would be odd, then, if we didnt expect
similar patterns of disinterest, subversion, and/or elite capture to hold in
rural India. And yet even a cursory reading of some of the more ebullient
texts of the Government of India, or of development agencies like DFID
or the World Bank, suggests that such recognition is often played down
or is simply missing. Participation, according to Bill Cooke and Uma
Kothari, is threatening to become a new tyranny.1 It is a discourse that
wishes away conflicts of interest and power, and which promises the poor
not just direct sightings of the state but powers of oversight as well. It also
protects itself from critical comment by daring people to speak against
Tyranny or not, we shall have cause in this chapter to disrupt at least
some of the claims made on behalf of institutionalized participation. We
shall also want to consider why it is that so many poor people fail to
engage directly with the Employment Assurance Scheme, village education committees, or even with the social groups that are at the heart of
many non-state development interventions (not excluding those run by

Cooke and Kothari (2001).



The everyday state and society

NGOs or similar bodies). Some of the reasons for non-participation are

well known and have been rehearsed in the general literatures on participation that we review in the next section. They include supply-side issues
like information circulation and uptake, and the expectation that benefits
will be creamed off by government officers, politicians and richer villagers
(something we consider further in chapters 5 and 6). They also relate to
the assessments that people make of the appropriateness of the good or
service on offer, and of the costs and benefits of getting involved in the
proper fashion. All too often, the projects that are trumpeted by New
Delhi are not working effectively for poorer people. They are not central
to their political or economic concerns as we began to outline them in
chapter 3.
At the same time, however, we see no reason to speak against participation as if it was a singularity. More people are engaging with educational
issues in parts of north Bihar and Jharkhand than we might have reason
to expect, and a version of participation seems to work quite well for
some poor households in Midnapore (see below). In any case, the fact
that most people in the UK dont vote in local council elections doesnt
mean they are unconcerned about town or village affairs. Some people
will buttonhole their representatives directly, or make their views clear in
a letter to a local newspaper. Others will clean up litter in a local park
without being asked to do so by a formal organization. They participate,
that is to say, informally, at a distance, or behind the scenes and the
same holds true in parts of eastern India. We shall contend that in all of
our research areas, with the possible exception of Midnapore, it has been
the indirect effects of a discourse of participation that have been most
effective in carving out spaces of citizenship for poorer people, however
small and disappointing these spaces might seem to be. The fact that people in Vasihali and Bhojpur are not generally active in school oversight
committees does not mean they are happy with the standards of their
local schools or that they are unprepared to voice their discontent. People participate in public life in more varied and less predictable ways than
either proponents or critics of participatory development seem prepared
to admit. At times, indeed, their participation is expressed in forms of
resistance to government. On such occasions, it is common to find people
taking aim at official programmes for participatory development in the
name of a broader technology of rule that is meant to endow them with
certain rights and expectations. In some cases, too, these acts of resistance force a reassessment of the merits of participation in the broader
development community, a point we take up in the conclusion to this



Pathways to participation
Participation is not a new idea in the context of statepoor relations in
rural India. As we remarked in chapter 2, the notion that poorer people
should contribute in some way to community development was enshrined
in the programmes that took shape under that name in the 1950s. The
communitarian strand of Indian planning failed to achieve parity with
the top-down structures of rule associated with the Second and Third
Five Year Plans, but it was still assumed that villagers would cooperate to
maintain free-flowing canals or functioning grain stores.2
This assumption rehearsed a typically Gandhian view of the Indian
village as virtually free from social conflicts (Jaffrelot 2003: 35). This
was also the case, as Jaffrelot reports, when a committee appointed by
the Legislative Assembly of the United Provinces called in 1948 for the
abolition of zamindaris and for the revival of the village republic and
a theory of trusteeship (Jaffrelot 2003: 35, citing Government of the
United Provinces, 1948: 519). To the extent that Nehrus wing of the
Congress Party engaged this viewpoint it was to suggest that the introduction of universal suffrage, land-to-the-tiller land reforms, and compulsory
primary education would reduce the need for trusteeship. Poorer people
would be empowered to participate on a more equal footing in the organization of village life. The passing of an Untouchability Offences Act in
1955 came too late to dissuade Ambedkar from his view that the Congress
Party, much like Mr Gandhi, was Tory by birth and by faith, but it was
doubtless seen by Nehru and the Congress socialists as a further nail in
the coffin of high-caste power in the countryside.3 Nehru understood,
as Gandhi apparently did not, that spaces for popular participation in
Indias political economy had to be sustained by politics and institutional
change, and not only by moral exhortation or appeals to the better nature
of members of the landed elites.
We will come back to this distinction soon enough. It bears repeating that the rhetoric that surrounded community development in the
1950s generally did not extend to the programmes that emerged in subsequent decades for the alleviation of poverty in either rural or urban
areas. The slum clearance programmes that Sanjay Gandhi unleashed on
Delhi during the Emergency (19757) made this point with unpleasant
clarity. Half way around the world, in Central and South America, urban

For discussion, see Frankel (1978); Hanson (1966).

Ambedkar had made this jibe against Gandhi in the 1930s after Gandhi had refused the
British offer of separate electorates for the Depressed Classes: Ambedkar, quoted in Dube
1998: 83.


The everyday state and society

planners were taking note of the work of the British architect-planner,

John Turner, as well as that of Charles Abrams and William Mangin.4
Their research had suggested that poorer people used their limited capital
to build settlements that corresponded to their existential needs. When
access to jobs was a priority they would live in city-centre slums. But when
a concern for safety and security became more important they moved to
a peripheral settlement where they could slowly improve their dwellings.
The poor were cast as authors of their own fortunes, and were applauded
for the ways in which they committed human and other capital to the
building of appropriate urban and peri-urban settlements. The job of
government, Turner argued, was to provide these pioneers of participatory development with legal titles to land, and with urban services like
water, electricity and transport. In Delhi, meanwhile, and quite tragically, the authorities were encouraged during the Emergency to code the
urban poor as threats to disorder, or even to the aesthetic of the newer
parts of the city. Many of their settlements were trampled out of sight.
If ever high modernism visited post-colonial India it was during this
period, and it was further evident in the plans of the city authorities to
build public sector housing to specifications that put it beyond the reach
of the labouring poor. A World Bank report of this time, authored by
Oscar Grimes, estimated that the cheapest public housing in New Delhi
was beyond the reach even of most working households in the formal
Matters were rarely this bleak in the countryside, but even here the
notion that poorer people could contribute effectively to the shaping of
government programmes on their behalf took several decades to dawn.
The preference, rather, as in the raft of schemes pushed forward by
Congress governments in the 1970s and 1980s, or in Tamil Nadus Noon
Day Meal Scheme, for all its successes, was for a model of governance that
constituted beneficiaries as recipients, and which assumed that expert
knowledge remained the preserve of government servants.6 It was only in
the 1990s that this began to change in significant ways. Pressed in part by
Robert Chambers and his many followers in the NGO community (and
latterly in the major lending institutions), the Government of India began
slowly to decentre the role of government experts. Chambers called for


Turner (1967, 1968), Abrams (1964) and Mangin (1967). It should be noted that this
work had little to say about the organized nature of land invasions in many cities in Latin
America; it ignored, that is to say, the ways in which individuals and households were
situated in local political societies.
Grimes (1976).
Successfully so, in some cases. Dr`eze and Goyal (2003) have argued that a version of the
Noon Day Meal Scheme in Madhya Pradesh has fared quite well, though see also Singh



a new attitude of mind on the part of public officials.7 He reached back to

Gandhis injunction to connect public politics to changes in the personal
morality of people in authority. In particular, he reminded his readers
that few presumptions should be made about how the poor experienced
poverty, or wished to see it addressed. An attitude of humility had to be
struck that would encourage outside observers to engage in participatory
assessments of rural poverty, or exercises that would transfer leadership
and voice to members of poorer households themselves.
These exercises were supposed to lay the basis for a new generation
of project interventions. They would involve beneficiaries in the choice,
design and implementation of schemes that would address their most
pressing needs. Suppose poverty was experienced mainly in terms of
coercive mobility. Poorer households might then plump for social and
economic interventions that would protect the rights of migrant labourers
in their new places of work. More generally, the impetus of Chamberss
work, and that of Michael Cernea, Norman Uphoff or Anil Gupta, was
to reclaim (or reposition) the targets of government (and other) interventions as knowledgeable and skilled agents in their own right.8 This being
the case, participation could be phrased as a positive end and not simply as a means to something more tangible. In seeking to change whose
reality counts proponents of participatory development have argued that
the world-views of the lowers (in this case, the beneficiaries/subjects of
development programmes) should dominate those of the uppers (development practitioners and other important outsiders). Participation would
then have the potential to change power relationships dramatically within
mainstream development practice.
Some critics of Chambers have complained that his focus on individual
moralities takes the place of a more sustained commentary on the production of power differences within village communities. Others suggest
that the participatory development paradigm is a natural bedfellow for
neoliberalism.9 Both stand accused of a form of ontological voluntarism

Chambers (1983, 1994a, 1994b, 1994c).

Cernea (1985), Uphoff (1996, 2002); Gupta (1981, 1987). We mention here three influential and in key respects very interesting individuals, who draw variously on the work of
Freire, Berry, Gandhi and others to make their arguments. Cerneas work was and remains
influential in World Bank circles. Uphoff and Gupta have also tried to put their ideas into
action. Uphoff has a longstanding commitment to participatory irrigation management
in Sri Lanka and Nepal. Gupta, meanwhile, set up the Honey Bee network in Gujarat in
the 1980s to collect and disseminate the insights of best-practice or simply odd (that is,
innovative) farmers, and, later, with the government of Gujarat, the Gujarat Grassroots
Innovation Augmentation Network (GIAN). We come back to the important question of
academics getting involved in these and other fashions in chapter 9.
See Guijt and Shah (1998) and Welbourn (1991) for the basic arguments; see also Henkel
and Stirrat (1999) and Wright (1996).


The everyday state and society

that privileges the supposed autonomy of individual agents. Be this as

it may, it has been Chamberss consistent reply that modern accounts
of development cannot reasonably be prosecuted without close regard
for the importance of popular participation. Participation might not be
a panacea for development, but Chamberss more important suggestion,
as we understand it, is that systematic and repeated efforts to involve
members of the rural poor in the execution of development projects and
programmes are likely to be empowering. The major argument in favour
of participatory development is that it forms part of a broader process of
social capital formation that works slowly but surely to undermine, and
perhaps even to overturn, existing hierarchies of power and rule.10 Put
another way, it is a process of development that should bring the state
more clearly, and more evenly, into the sightlines of citizens.
Rates of participation
Assessing the effects or effectiveness of participatory development
schemes is no easy matter. Some of the outcomes that proponents associate with participation, including more balanced sightings of the state,
might be caused by factors external to specific development schemes, or
take several years to mature. Even the assessment of rates of participation
can be difficult. There is a world of difference between the simple fact of
attendance at a meeting, and the ability to contribute effectively to that
meeting or to shape its conclusions. Participation can be more or less
active, and more or less passive. It also matters a great deal who gets to
set an agenda, and who is able to call a vote or structure a discussion.
Nevertheless, it makes sense to start our discussion of participatory development schemes with the issue of attendance, or physical presence and
absence. We do so, moreover, with our eyes firmly on development-cumempowerment schemes such as the Employment Assurance Scheme,
school oversight committees, and Joint Forest Management initiatives. In
this chapter we are not concerned with rates of attendance in the broader
sweep of local government meetings. The very different experiences of
men and women in terms of participating in panchayati raj institutions
are something we discuss in later chapters.

This would seem to be the position of agencies like the World Bank (1995) and the
Government of the United Kingdom (1997). What this position neglects is the need for
concerted political actions to aid patterns of participation that can begin to challenge
entrenched hierarchies of rule. We discuss this later in the chapter; see also Williams
(2004). A sophisticated account of how the participation agenda is beginning to move
on and does respond to criticism can be found in Booth et al. (1998); see also Holland
and Blackburn (1998).



Table 4.1. Proportion of male and female members of sample households

gaining one or more paid labour days from the Employment Assurance Scheme



Field site: Poor Non-Poor Poor





Non-Poor Poor

25% 0%
2.5% 0%


Old MaldaMalda

Non-Poor Poor Non-Poor Poor Non-Poor

20% 15%
2.5% 0%

60% 0%
41% 0%

21% 14%
14% 0%

Source: Village surveys, 1999 (n = 100 for each village: 80 poor, 20 non-poor).

The Employment Assurance Scheme

The Employment Assurance Scheme is an excellent place to begin a discussion of sightings of the state in relation to participatory development.
We said in chapter 2 that the EAS took shape in 1993, and was at the time
of our fieldwork the largest employment provision programme in India.
It was also a major component of the countrys anti-poverty policy. In
1999, each of our study Blocks was meant to receive between Rs. 7 and
9 million per year under the EAS (that is, close to $200,000), which was
a significantly greater sum than was committed to other programmes,
including the IAY, JRY or IRDP (see Appendix 1). The EAS guidelines
called for at least 60 per cent of these funds to be spent directly on the
wages of employed persons. This was to ensure that non-poor households
had few incentives to capture EAS funds. Remuneration was tied to the
performance of unskilled labour, and it was the unskilled labourers,
of course, who were meant to trigger the flow of funds in the first place.
The EAS guidelines required poorer men and women who were suffering
from a lack of paid work in the lean season to demand employment from
officers of the local state. The Block Development Office, in particular,
was required to be responsive to these demands from below.
We return later on to the question of peoples participation in the matter of scheme selection, and the system of village open meetings that
was required to support this process. But what of the more straightforward question of gaining access to work under the Employment Assurance Scheme? As table 4.1 suggests, the pattern of uptake of employment under the EAS varied significantly across our five field sites,
although in only one case (Debra Block, Midnapore) did a majority of
poor/destitute households find work for one or two people from this
quarter. In Old Malda, Murhu and Sahar Blocks, about a quarter of
poor/destitute households received work from the EAS, while in Bidupur


The everyday state and society

Block, Vaishali District, not a single person from our sample households
received employment in this way (at least not within their own locality).
This pattern of uptake cannot be explained in terms of a lack of demand
for lean-season employment in the study areas, or a greater need for such
work in Debra Block. Off-season unemployment was seen as a major
problem in all five areas, including in Blocks where double-cropping was
common. If the problem was reported to be less intense in Murhu Block
this was because local people had long been migrating for work in the
lean season. The low rates of take-up had more to do with patterns of
information supply and circulation, and demand management, as we
shall see in the next section. By the same token, the fact that more EAS
work was carried out in Debra Block by members of the most destitute
households (90 per cent of which received some work), when compared
to other poor households (69 per cent), reflected the efforts at targeting
that were made by party workers and local government officials. These
factors also help to explain why poorer households in Old Malda Block
that were in receipt of work from the EAS gained on average only 5.7 days
of paid work in the period 19959. (Put another way, this amounts to 1.4
days over the same period for the average poor household, whether
or not any work was received from the EAS.) Even in Debra Block, the
corresponding figures of 12.8 and 9.6 days were well short of a maximum
of 200 person-days per year per unskilled labouring household. The EAS
did not employ many people for very long.
Participatory school management
An approach to rates of participation that is focused on direct beneficiary
outcomes works less well for school oversight committees. In this case
benefits do not always accrue to named households, or even to the poorest or most excluded families. But even here we can make a preliminary
assessment of rates of participation. In Bihar, we have data on the number
of Village Education Committees that were formed in each of three successive years (19978 to 19992000) in all of the Blocks in our research
Districts. A total of 5,234 VECs were set up in these Districts by 1999
2000, out of a planned total of 5,503 VECs. This implies that more than
50,000 villagers would have been members of educational committees in
Bhojpur, Vaishali and Ranchi Districts by the year 2000.
Impressive as these figures are, however, the government of Bihars own
data reveal that many VECs have met only rarely, and in some cases not
at all (see table 4.2). In 19978 two-thirds of the VECs of rural public
schools in Bhojpur District failed to meet at least once, and in 19989



Table 4.2. Number of meetings held by VECs of rural P.S. schools, Bihar
Study Districts, 19982000
Number of
19978 19989
More than 0.2%






19992000 19989






n = 2143 n = 2113 n = 1328 n = 1338 n = 1368


n = 1723 n = 1718

Source: Bihar Education Project, State Office, Patna.

this figure rose to nearly three-quarters. In both Vaishali and Ranchi Districts, in contrast, it was common to find VECs that met more than six
times, although in Vaishali there was a noticeable tail-off in interest in
19992000 as compared to 19978. The Block-level data we have seen
broadly confirm these District patterns, although in Sahar in 19989 fully
80 per cent of VECs failed to meet even once. We discuss these patterns
later in the chapter. At this point it will suffice to say that our village-level
data and ethnographies are largely consistent with these officially reported
patterns. They also reveal that such meetings as did occur in our Ranchi
field site were dominated by local teachers and the Mukhiya, while in
Bhojpur they were run by two upper-caste landlords and their kinsmen.
The VEC of our Vaishali field site was more balanced and included members from the Rajput, Yadav and harijan communities. The vice-chairman
of the committee was an influential harijan leader.
In West Bengal the picture is rather different. As in Bihar, a formal
system of schools inspection reaches down to the field-level in the form
of Sub-Inspectors of Primary Schools (SIs). We shall have more to say
about them in chapter 5. At the community level they are complemented
by School Attendance Committees (SACs). We are unable to present
data on these Committees for the generality of our Districts and Blocks.
We can say, however, that in the Midnapore field site we observed SACs
that were active and well integrated into the local education system. The


The everyday state and society

Figure 4.1 Sources of help in solving problems with education, schools

and teachers, West Bengal

committees comprised the headmaster as secretary, the Gram Panchayat

member of the respective ward as chairperson, the local SI, two educated
villagers and eight guardians. The thirteen-member committees met six
times a year, and ten of the members were selected annually in parents
meetings. Although the committees worked well, it is important to note
that many parents turned directly to their panchayat member when they
had an issue to voice (see figure 4.1). They did not confine their participation around educational issues to the SAC.
In Old Malda Block the performance problems that we reported in
chapter 3 were not being addressed by active SACs. The SACs that were
meant to be attached to the primary schools we studied were defunct,
and the District Primary School Council (DPSC) chairperson of Malda
District suggested this was the case in most parts of the District. We
were told that Maldas SACs had last been rejuvenated in the late 1980s,
when headmasters acted on directives received from the Department of
Education. In 19992000 we could identify the chairs and secretaries of
the SACs in our wards, but we could find no evidence of poorer people being involved in the committees, or, indeed, of current guardians of



Table 4.3. Awareness of the existence and principal objectives of the

Employment Assurance Scheme
Bidupur- SaharMurhu- DebraOld MaldaVaishali Bhojpur Ranchi Midnapore Malda

Field site:
Heard of the EAS

Non-poor 25%


Aware of the major

provisions of the










Source: Village sample surveys, 1999 (n = 100 for each village; 80 poor, 20 non-poor).

school-going children being committee members.11 Ensuring high participation rates among poor households is a significant and widespread
problem, and is it is one that is accentuated by a lack of direct or quickly
realized benefits.
Information circulation
The patterns of participation that we have just described are hardly surprising, even if they are in some respects disappointing. The World Bank
likes to maintain that People are the means and the end of development,
but it is wise enough to acknowledge that they have different amounts of
power and resources (World Bank 1997: 110). Some people also know
more than others and are able to control, in some degree, how information circulates across a space-economy. This was apparent in all our field
areas when it came to the Employment Assurance Scheme.
Table 4.3 provides some basic and rather striking data on patterns of
knowledge about the EAS. As will be quickly apparent, it was only in
Debra Block, Midnapore, that more than 15 per cent of poor households
reported knowledge of the Employment Assurance Scheme (30 per cent,
with 45 per cent of non-poor households). And even in Midnapore only
6 per cent of poorer households (15 per cent of non-poor households)
could offer an informed account of the major provisions of the EAS (in
the sense of knowing about its demand-led components, or its provision

The fact that most villagers named the school secretary as the primary source of support
for the resolution of school problems is indicative of how little notice panchayat members
and local politicians pay to educational issues in Malda; it should not be read as evidence
of the efficacy of the local SAC.


The everyday state and society

for up to 100 days paid employment). In Old Malda Block only one poor
household reported knowledge of the EAS, and in Vaishali and Bhojpur
Districts, Bihar, the figures were scarcely higher. These figures, moreover,
including those for Midnapore (though not Vaishali), were actually lower
than those suggested by our first index of participation. In Debra Block,
that is to say, we were able to ascertain that 75 per cent of poor and
destitute households had gained work from the EAS, while only 30 per
cent of these households were able to name the scheme as such. For
the most part, the beneficiaries of the EAS simply knew the scheme as
government work, and were largely unaware that it was not provided as
part of the JRY or any other programme. Interestingly, the highest level
of awareness of the EAS, but not of its specific provisions, was among the
sample non-poor population of Old Malda Block, where people linked
to local contractors were well informed about it.
We need to be careful, then, before jumping to a conclusion which
suggests that a lack of awareness leads to low levels of uptake of work
under the EAS. The supply of information is a serious issue, and the
broader development community would be right to point to it here, as
in many similar cases. But the fact that people are getting work from
a scheme they cannot name suggests that something else is going on.
Government is being sighted in some cases, but not as the governments
own rhetoric would lead us to expect. Very few of the households that
had gained work from the EAS were in possession of an EAS card, for
example, although some non-participating households were able to show
us their cards.12 In most cases, the government was not being sighted
at all. Labourers were recruited and paid by contractors, much as they
always had been.
A similar disjunction between participation and information supply
and usage was apparent in the VECs we looked at, as well as in some
Village Forest Protection and Management Committees with which we
are familiar. The proposition that participation can or should lead to
development, or even social capital formation, depends upon a conception of informed involvement. In the case of the VEC in our Bhojpur
field site, as we have reported, a number of Scheduled Caste men and
women were listed as members of the Committee without being aware of

The ability of a labourer to demand work from an EAS scheme was supposed to be
linked to possession of an EAS card, which was provided by the Block office. Among
our Bihar field sites, EAS cards were most in evidence in Vaishali, notwithstanding the
fact that not one labourer there had gained work on a local EAS scheme. In West Bengal,
53 per cent of our respondents in Midnapore, all from poor households, had EAS cards,
as against 0 per cent in Malda.



the fact, and far less of the remit and purpose of the institution. But this
in turn spoke to the fact that an enormous number of VECs had been
formed very quickly in Bihar. In the Ranchi field site, for example, we
discovered that the VEC had been formed in one day, and then without
prior notice being given to villagers. Members were nominated by the
Mukhiya. This was completely at odds with the guidelines that were supposed to be followed by officials of the Bihar Education Project. These
called for trained catalysts (utpreraks) to work in each village for at least
a week before a VEC was constituted and members were elected. We
will explain in chapter 5 why so few BEP staff were minded to take their
responsibilities seriously. (Capacity constraints were a factor, but more
important was the fact that BEP officials were insulated from political
pressures at the Block level.)
In the case of Joint Forest Management, meanwhile, Manish Tiwary
has reported that the best organized Village Forest Protection and Management Committee (VFPMC) in his Bihar (Ranchi District) research
area, Vanadag, had been started by a local Oraon leader, and was at the
time of his research (2000) unregistered by the Forest Department.13
Elsewhere in Ranchi District he found very few examples of VFPMCs
that met on a regular basis, or whose members were properly informed
about their purpose, responsibilities and possible future gains. In contrast,
several of the Village Forest Committees (VFCs) he studied in Midnapore
District worked rather well in all these respects. Sarah Jewitt has presented
similar findings from Bero Block in Ranchi District.14 The VFPMC that
the Forest Department had registered in Nehalu Kaparia had a mixed
record when it came to holding open meetings, or involving villagers
actively in planting decisions. Although most households were aware that
Nehalu jungle was under something called Joint Forest Management,
very few had clear ideas about the benefit-sharing arrangements that the
legislation called for; nor did a majority of participating households supply men or women for forest guarding patrols, as they were supposed to.
Matters were very different in Jamtoli, another village in Bero Block that
is just ten kilometres from Nehalu as the crow flies. The residents of this
village could see the benefits that forest protection had brought them over
a period of forty years. They were surrounded by mature stands of sal
(shorea robusta) forest. According to Jewitt, they were also well informed
about their forest protection responsibilities. Here, too, an Oraon leader,
Simon Minz, had formed an unregistered forest protection committee,

Tiwary (2001, 2004).

Jewitt (2000); see also Corbridge, Jewitt and Kumar (2004) and S. Kumar (2002).


The everyday state and society

and knowledge about environmental issues had been circulated through

traditional village meetings and at local fairs.
The point we are making, of course, is not that information circulation
is unimportant, but that it is (and needs to be) actively managed. To be
fair, this is a point that Chambers and his supporters would be keen to
endorse. Chambers might argue that the sorts of information gaps that
we have just described are likely to arise when projects impose forms
of knowledge on local people, when what is required is a less extractive
process of knowledge production. This would help to explain why information loss was less of a problem within a non-state development project
like the Eastern India Rainfed Farming Project (EIRFP), a joint initiative of DFID and the Government of India.15 According to Kumar and
Corbridge, most households in the village clusters where they worked
had come to a reasonable understanding of the aims of the Groups that
Community Organizers were trying to form. Most participating individuals understood that they were required to join one or more groups to
make demands of the project, whether for a bee-box (hive), a checkdam, improved seeds, or something else.16 We would argue, even so, that
this sort of position, while it highlights the connections that can be forged
between information and participation from the bottom-up, is strangely
silent when it comes to the management of information by members of
the different social groups that must rub shoulders with one another in the
organization of any large-scale social intervention. It also tends to constitute the problem of information loss in a passive voice, or one that blames
the designated information provider (usually a government department). The possibility that men and women might sometimes choose
not to collect the relevant information is not seriously considered (see
chapter 4).
Managing projects and project management
We have said that the Employment Assurance Scheme calls for poorer
people to have direct interactions with officials of the state, and to make
demands of them. But we have also seen that this can be difficult when
people are poorly informed about the scheme, and this difficulty is


We first mentioned the EIRFP in chapter 1. Although it is not a standard state-run

development scheme, it does have backing from the Government of India and was active
at the time of our research in Bihar (Jharkhand) and West Bengal, as well as in Orissa.
We believe that it provides a useful comparative perspective.
Kumar and Corbridge (2002), drawing on fieldwork carried out in Bero Block, Ranchi
District, in 19992000.



compounded by the fact that sightings are only rarely one-to-one. We

didnt come across a single case of an unskilled labourer demanding
work directly from a panchayat or Block Development Office. For the
most part, poorer people gain their knowledge of a programme like the
EAS from local village leaders, labour contractors, party officials and so
on (see chapters 3 and 6), and it would be unwise to assume that these
people have no personal interest in the scheme. They plainly do: even if
EAS schemes are run as per the guidelines, they can provide a considerable sum of money to builders and other contractors, and the durable
assets that they are meant to provide can be a source of social power. It
matters who claims credit for bringing a check-dam or a school building
to a village. As we saw in chapter 3, this can develop or maintain someones standing as a key player in the locality. Even supposing, then, that
government officers have the time or the inclination to act as community
organizers, patiently explaining the aims and objectives of the EAS to a
beneficiary population (something which presumes a level of state capacity that is nowhere met in the field: see chapter 5), an ideology of direct
participation further assumes away the problem of conflicting interests,
just as Chamberss critics have maintained.
Of course, these conflicting interests come in different shapes and sizes,
much like participation itself. It would also be misleading to suggest that
public meetings are not sometimes held to identify and prioritize projects
that the public deem important within their own area. In the Midnapore
and Ranchi field sites it was clear that new forms of learned behaviour
had arisen around repeated acts of participation. Beneficiary Committees
have been formed in Midnapore to oversee the work of a Job Worker
who is selected in a public meeting. He or she is required to act as a
foreman responsible for hiring labourers for the scheme and directing
their work. Some Job Workers here and in Ranchi have also come from the
ranks of registered unemployed labourers, and some have tried to ensure
that registered unemployed labourers from the area in which the project
takes place carry out the designated work. At the same time, however,
some Beneficiary Committees have found it hard to verify the quality of
the physical assets produced in more complex schemes. In Midnapore
District, the Committees were able to use the services of a Sub-Assistant
Engineer and Gram Panchayat staff to help them in this task. They then
joined with the Job Worker and elected members of the local council,
as required, to sign off a projects accounts, or to publicize individual
project budgets at the time of implementation. In Vaishali District, these
guidelines largely went by the board, in a context (see chapter 5) in which
the District Magistrate and the de facto MLA of the area responded to a


The everyday state and society

series of bids from their supporters (and those they had to keep on-side)
in the local political society.17
Even in Midnapore, however, an apparent willingness to run EAS
schemes as per the guidelines concealed a high degree of political management, and this is the point we want to develop here. There are some
interesting comparisons to be drawn with Vaishali, not all of which work to
Midnapores advantage (see chapter 5 on the physical outputs produced
by EAS projects in Bidupur). It was certainly the case that a higher percentage of poorer people attended EAS village meetings in Debra than in
the other field areas. Seventy-five per cent of poorer respondents in Debra
had attended a gram sansad, and panchayat members who we spoke to
confirmed there were no difficulties in ensuring that meetings were quorate. Of those attending, women made up about a fifth of all participants,
at least in the meetings we observed directly. For many poorer villagers,
however, attendance at meetings was generally passive. Some told us that
they were attending a meeting primarily because party activists had told
them to. A larger number had attended on a voluntary basis but felt unable
to speak up, mainly, they said, on account of their illiteracy or uneducated status. Furthermore, a number of key decisions regarding the EAS,
including the prioritization of schemes, the selection of the Job Worker,
and the election of a Beneficiary Committee, had effectively been taken
out of the control of the public meeting.
Villagers told us that the main purpose of the gram sansad was to
endorse the suggestions of panchayat members, or the party, which made
the important decisions. Significantly, they expected little else, or perhaps
we should say no less. Elected panchayat members confirmed this perception. They maintained that a good gram sansad was one in which people
were able to suggest several development projects that could be considered further for possible action. Much like Kenneth Arrow, however,
they rejected what might be called the hard-line populist or participationist fantasy that the individual preferences of attendees could be simply
aggregated into a social welfare function that all would agree upon.18


We comment further on this in chapters 5 and 6. To anticipate: the EAS in Bidupur was
largely redefined as a road-building scheme, and most of its immediate benefits went to
non-SC households (chapter 5). That said, the leading power brokers in the District and
the Block, including the District Magistrate and BB, the local Member of the Legislative
Council, took great care to respond to a series of demands that were placed upon them
by brokers, contractors, influential caste leaders, teachers, village leaders and others
lower down in political society. The distribution of EAS benefits was also made to
respond, in large part, to competitive processes of participation that took shape outside
the EAS guidelines. As we shall report in chapter 6, these processes extended to a series
of discussions on the location and type of specific EAS projects.
For the ArrowDebreu theorem, see Arrow (1951) and Debreu (1959).



Keeping a firm eye on their standing as politicians and party members,

a number of panchayat members told us that a bad meeting would be
one in which too many schemes got proposed, or where villagers spent
time raising other, more contentious issues, such as arguments about the
quality of past development works or the performance of elected members. Higher-level cadres were concerned that false expectations should
not be aroused in the broader population, and especially among party
supporters. The party wanted to cultivate a reputation for efficiency, as
well as for taking the part of the poor, a theme we develop further in
chapter 6. Senior members were thus properly sceptical of the suggestion
that all labouring households could be found work let alone 100 days
of paid employment in a short period of time, even if New Delhi did
make available fresh tranches of funds to the state government.
Similar noises about knowing what is best for the poor were made by
key power brokers in Ranchi and Vaishali Districts (see also chapter 5).
Before rushing to condemn this judgement, however, it is important to
consider how and why it might be deployed. We hold no brief for the
CPI-M, and we shall later be critical of some aspects of its stance on
educational issues in Midnapore. Party members are also not above corruption. But the CPI-Ms stance on the EAS makes good sense for a
political party that has to set priorities, and which refuses the rather
nave assumption that people always know what is best for them. This
assumption, which features strongly in the most strident prospectuses for
participatory development, ignores the asymmetries of knowledge that
inhere in any large-scale system of social production and exchange, and
which can make a space for expertise and a measure of delegated powers. In any case, the CPI-M has not refused a measure of participation
around the EAS. In Debra Block, it organized a series of party-run gram
baithaks. These allowed poorer individuals who were not opposed to the
party to take part in public meetings where they could be sheltered from
the ego-clashes that typically break out between political big men. The
gram baithaks also enabled the party to prepare for, and perhaps even
rehearse or stage-manage, official village open meetings. For example, the
baithak would decide which individuals should propose EAS schemes in
forthcoming gram sansads. This gave poorer men and women a measure
of confidence that they might not otherwise have had, and, importantly
for the CPI-M, it ensured that not all schemes were proposed by party
activists themselves.
In a sense, of course, this stage management can be seen as something akin to the training in participation that is offered by community
organizers in the EIRFP, or by NGOs working for joint forest management. Whether the CPI-M needs to be as fearful as it is of unrestricted


The everyday state and society

participation is a moot point, and an important one, but there can be

little doubt that its agenda in Debra Block was carefully thought out, and
in its own terms pro-poor. In our Murhu Block field site, too, most villagers looked with approval upon the Mukhiyas (rather successful) efforts
to draw down resources from the state on their behalf. Adivasi families
required their traditional leader to play this role. They expected to have
their say about matters of concern through a system of rule that provided
for regular inter-village councils that would be presided over by a Parha
Raja.19 This was assuredly not the case in Vaishali District, where BB, the
local Member of the Legislative Council (and de facto MLA), took care to
direct control of EAS projects to a large and diverse group of his political
supporters, and to some antagonists, rather than to the Scheduled Castes
or members of the labouring poor (for more on this, see chapter 5). And
it was not the case either in the Nehalu and Boda clusters of the EIRFP
in Bero Block, Ranchi District. The structuring of participation there by
members of local (tribal and non-tribal) elites had close regard for status
and power, and for what might be called the languages of engagement.
These languages, as David Mosse has reported for the Western India
Rainfed Farming Project, work to forge a connection between lay villagers and development professionals in two main directions.20 Consider,
first, the case of the community organizers, or, analogously, the lowranking (non-gazetteered) staff who are entrusted with holding EAS public meetings. Many versions of participatory theory call for the state to be
sighted on something close to equal terms, and meeting non-gazetteered
staff can be less intimidating for labouring men and women than meeting


The chief (king) of a group of villages, usually eleven to thirteen, that traditionally
has been ruled by a council of elders. By the same token, some government officers
delighted in working in a Block or District that exhibited fewer apparent signs of political
competition and thus legal or illegal pressures on government officers than would
be likely in north or central Bihar. The DC of Ranchi District had worked previously in
Vaishali and Nalanda Districts in Bihar. He told us that: There is a marked difference
in the working environment here [Ranchi]. We hardly feel any pressure from politicians.
In fact, most of them know very little about funds and schemes, and rather I have been
educating them as to how much they could claim for their areas under what programmes.
Interestingly, a female BDO in a Block closer to Ranchi city than Murhu, put matters
rather differently, in the process confirming how difficult it is to generalize about the
nature of political society across a District. There is so much of pressure. It is difficult
to work. Higher officials do not also cooperate much. We have to find our own ways,
which is very difficult. You come to Block on a working day and you will find hundreds
of people around, many of them claiming to be representative of this or that political
party demanding for award of schemes to their henchmen. Few of them come with
letter-heads with their names printed along with the designation as student leaders! If
you dont accommodate their request, they would send frivolous complaints to DDC.
Higher officials consider this to be headache, and instead of appreciating our efforts to
counter such pressures, feel we are tactless.
Mosse (2001).



a BDO or District Magistrate. We might also agree to the suggestion,

which is implicit in many accounts of the formation of linking social capital, that villagers will gain the confidence to deal with the latter after first
engaging the former.
The problem, however, is that forging this connection requires very
particular forms of performance from lower-level government officers or
their equivalents. In Bihar, it was the Block-level staff who were charged
with organizing the public meetings to select both the EAS schemes and
the foremen (abhikarta) who would oversee the work. A number of these
officials found it difficult to run public meetings in such a way that they
could make space for the voices of the poor in anything approaching an
orderly fashion. They simply didnt command the respect, or even the
fear, that can attach to offices associated with higher and more spectacular levels of the state.21 Similarly, the community organizers of the
EIRFP tend to be distant in social and educational terms from the adivasi communities they are mainly required to work with and for. The
COs of Nehalu and Boda clusters in Ranchi District needed the support
of local village elites, and they chose to work through such persons when
establishing project groups. Knowing they would be evaluated, in part,
on the basis of targets (number of check-dams sanctioned, number of
livestock loans distributed), the COs also expressed a need to work with
more educated villagers, including those from non-adivasi communities.
The idea that they might make progress with more obviously disadvantaged communities, like the Mahlis in Boda or the Lohras in Nehalu,
was dismissed out of hand. One CO told Kumar and Corbridge that:
yeh log apne khane-peene me hi mast rahta hai, group ke kaam se inko koi
matlab nahi hai (these people are only interested in drinking and eating among themselves, they dont have any interest in group activities).
Try as the CO might, it was difficult to instil a sense of cooperation in a
community known mainly for its drinking, quarrelling, dirty houses and
pig-keeping. He (or she) had to work instead with a jankar (village motivator) from the dominant groups within the village, and to offer periodic

While working in our Ranchi field site, we heard reports that rival factions elsewhere
in the District had mobilized several thousand people for the selection of their chosen
abhikarta. This had caused some EAS meetings to be cancelled. Block-level officers
lacked the authority and training to deal with such instances of disorder. Under these
circumstances, it is perhaps not surprising that many EAS meetings existed only on
paper. We were also told by the Circle Office of Patepur Block in Vaishali a Block
which we visited several times before settling on Bidupur that he would run away if
he was transferred to Bidupur Block, a locality that he considered to be dominated by
non-official (and sometimes armed) members of political society. Exaggeration aside, we
note that neither proponents nor critics of participatory development pay much attention
to the problem of an excess of participation.


The everyday state and society

training to that person (this section after Kumar and Corbridge 2002:
These acts of translation are matched by the efforts of villagers, and
especially better-placed villagers, to learn the languages of development
projects, and of the wider paradigm of participation in which they sit.
Mosse contends that, in regions where there are significant inequalities in
terms of access to land or the public space, it is farmers [not those without
land] who acquire planning knowledge and learn how to manipulate
it (Mosse 2001: 21). The formal exercise of participation then serves to
represent external interests as local needs, dominant interests as community concerns (Mosse 2001: 22), as better-off households quite understandably seek to pass themselves off as poor, or to claim that an entire
village is poor or undivided. In adivasi areas this is not difficult, since it is
commonly assumed that tribal societies are not stratified (including by
many community organizers, who are themselves mainly caste Hindus).
The more powerful households also take steps to control and perhaps
contain what can be construed as a new source of social power in their
locality. They learn the languages of the project, that is, not simply to
draw down resources for themselves and their favoured contractors, but
also to ensure that no other group is well placed to play this role.
A project does not come innocently into a locality. To the extent that it
brings with it a promise of new resources, individuals and social groups
who are not its intended beneficiaries will see it as a source of social power
or funds. Participation cannot then be confined to the poor. Participation is always met with counter-participation, and the proper role of
project managers, as several of them now recognize, is to work as effectively as they can in the world of the second best.22 Participation describes
a spectrum of social actions, and it is unrealistic to assume that it can, or
even should be maximized in all circumstances.
Costs and benefits
Proponents of participatory development can also be reticent about the
costs of participation, whether these are financial or non-financial, or
are direct or in the form of opportunities forgone. In our experience,
however, the intended beneficiaries of participatory projects often have a
good sense of the costs and benefits of participation, even when they are
unable to name a scheme correctly or spell out its statutory provisions.
Lack of information isnt the only reason why they fail to engage the
state or a development project.

See note 10; see also our more general conclusions in chapter 8.



Table 4.4. Evaluation of Malda village meeting: selected questions and




1) Did you like the village meeting on 12.11.00?





2) Did you come to know anything new from this village





4) Do you remember who are to be called for the gram sansads?

All voters (right answer)
Wrong answer
Dont know




4a) Before coming to this meeting were you aware of this?





5) You were told that at the Block level there is an elected

panchayat institution. Can you name it?
Yes, panchayat samiti
Dont know
Wrong answer




6) Do you remember what is the minimum wage rate fixed by

the government for panchayat work?
Right answer
Wrong answer
Dont know




6a) Did you know of the minimum wage rate before the village




Source: Village workshop (2), December 2000.

Consider the Employment Assurance Scheme from this perspective.

The acquisition of robust information about the scheme would clearly
have helped it to function better. As part of the action research project
that we carried out for DFID in 2000/2001, we ran a workshop in our
Malda field site on 12 November 2000 where we fed back information
on such things as the minimum wage rates for panchayat work, and rights
of attendance at gram sansads. We then returned to the village one month
later and asked about the information we had presented. As table 4.4


The everyday state and society

makes clear, recall rates were high among the fifty-four respondents. More
than 80 per cent of females and 90 per cent of males correctly remembered
the minimum wage rate fixed by the government for panchayat work.
The internalization of this information did not mean, however, that
these men and women would be strongly placed to demand work from
the panchayat in the next lean season. Information has to be carefully
digested, and set against a persons understandings of their broader social
needs and networks. As we explained in chapter 3, labouring households
might need to make demands of party members at times of crisis, and
some of them will not exercise a demand for panchayat work until it has
been clearly advertised. In any case, the costs of providing information
are not trivial. When they fall on government servants, party workers or
panchayat members, they impose tariffs (on time, mainly) that will not
always be welcome or even manageable (see chapters 5 and 6). Even
assuming that a meeting can be called at a time and date when most
interested people can attend, and when the appropriate public official(s)
can be present, the opportunity costs of that meeting can be considerable.
One of the gram sansads that we attended lasted for four hours, and was
dominated by perhaps five or ten of the one hundred people who turned
up. Participation has to promise something tangible if people are going
to consider this a good use of their time.
Whether they do or not depends on how they estimate the likely benefits of participation. In the case of the Employment Assurance Scheme
there is something akin to a chicken and egg situation in many areas.
Some of the poorest villagers in Vaishali District might value a chance
to participate in local planning decisions, but as things stand they are
unable to do so. For the EAS to work in anything like the terms it is
meant to, the initial push must come from above, from government servants, politicians or perhaps even from NGOs. Before a beneficiary can
hope to calculate his or her likely costs and benefits of participation, he
or she needs a prior set of participants to have decided that participation
makes sense for them. Chambers is right in this respect, and we return
to this issue in chapter 5.
In Midnapore District, meanwhile, the CPI-M has made this prior
commitment, and this is one reason why the EAS appears to work so
much better there (at least in terms of the indicators that we introduced
earlier). By the same token, many of the CPI-Ms poorer supporters in
Debra Block had already learned to see the state in ways that would be
unusual in Bidupur. The EAS took shape in Midnapore in a context in
which the ruling Left Front government had done considerable work in
preparing the ground for poorer people to participate in public meetings.
In Midnapore, too, notwithstanding the fact that many gram sansads were



carefully staged by party members, there was a commitment on behalf of

the political elite to the public display of a progressive state regime. The
benefits for the CPI-M of advertising their commitment to what might be
called due process, or even the rule of law, were not inconsiderable. The
cost of not doing so would be to suggest a form of political clientelism,
or simple bad governance, that party leaders were keen to avoid.
The differential performance of the EAS across our field sites would not
be that difficult to predict. When it comes to the constitution and work
of school management bodies, however, matters are more complicated.
It is true that the SACs we observed in Debra Block work quite well.
They are properly constituted, meet regularly, and help local people to
raise issues about the quality of their public schools. Set against this, the
VECs and SACs that we observed in Sahar and Old Malda Blocks were
performing poorly. In our Sahar field site the VEC of one school was
run by an imperious upper-caste landlord. He had dealt with a case of
teacher negligence by slapping him. Unsurprisingly, the teacher put in for
a transfer. The VEC of a second school was chaired by the son of another
upper-caste landlord, a graduate in this case, who was related to one of
the teachers at his school. The teacher even lived in his house. Several
parents told us that the teachers were negligent in their duties, but few
were prepared to take the matter further. In part this was because they
were aware of the social and economic power of the landlords family.
But it also spoke to the fact that most villagers were unaware of the full
composition of the VEC (as well they might be, given that all the SC
and female members of the VEC were unaware of their status). As in
Old Malda Block, the idea that ordinary villagers could take some control
over educational issues, and in a sense therefore over the state, was grossly
The one exception to this generalization might have been the VEC for
the tola school in our Sahar field site, or the school that was used by
children from the harijan or SC population. In this case a VEC had been
constituted at a mass meeting after several days of mobilization activities
involving senior officials of the Bihar Education Project (BEP). Its first
chairman, moreover, Raghav, was an active leader from within the tola.
He tried hard to voice the concern of many villagers about the lady
teacher who ran the school and who was often absent. The committee
fell into disrepair, however, when Raghav relocated to Dhanbad (now
Jharkhand) to look for work. District and state education officials also
failed to respond to the VECs demands for a second teacher to be posted
at the school. Not unreasonably, villagers took the view that attendance
of VEC meetings was not worth their while when the committee could
not secure this most basic goal.


The everyday state and society

But what of our Ranchi and Vaishali field sites? We saw before that a
significant number of VECs in these Districts, and indeed in Murhu and
Bidupur Blocks, had met six or more times. This would seem to indicate a
strong and continuing interest in educational issues. And this was partly
the case, including in our specific field sites. What was also apparent,
however, was that costbenefit calculations were working against the longrun effectiveness of the VECs. In Ranchi District this was partly because
the BEP had failed to follow up on its early work (which was in any case
pretty minimal). There was also a continuing problem of teacher transfers
in Murhu. Villagers were rightly confirmed in the view that key decisions
about education not least about the supply and quality of teachers
were being taken elsewhere.
In the Bidupur field site, meanwhile, we discovered two issues that
speak to wider debates about participation and seeing the state. On the
one hand, and perhaps predictably, we found that the vibrancy of local
VECs (in terms of meetings held and diversity of membership) was more
apparent than real. Meetings were often being held to give life to castebased conflicts. The representatives of the Yadavs, the Scheduled Castes
and the Rajputs each used the VEC to block initiatives from other quarters. On the other hand, and perhaps more significantly in the longer run,
we discovered that a group of Yadav farmers had spontaneously formed
a committee to deal with teaching issues. The children of these mainly
poor households went to a school on the north side of the village that was
without a formal VEC in 19992000. What concerned them most was
the question of teacher attendance. Committee members agreed to visit
the school on a regular basis to put pressure on the teacher. They set up
a roster to share this load and kept to it. This effort at participation,
however, had a perverse but not atypical consequence: the teacher was
said to have paid a bribe to a DSE officer to secure his transfer out. In
addition, he put the word out among his fellow teachers that the school
was not a good place to work. The result, according to villagers, was that
several more teachers paid bribes to the DSE in order not to be posted
to the village, and the school was forced to close for several months.
To the extent that some families were not inconvenienced by this closure, or closures like it, it would mainly be families from among the
Musahar communities that we met in Bhojpur and Vaishali. Many of
these families felt it made little sense to educate their sons and daughters when the parents were desperately poor and unschooled, and when
skilled jobs were scarce. Even in a village immediately adjacent to our
Sahar field site, where the local school was in the Musahar (SC) tola and
was staffed by a Musahar teacher, a large majority of households saw the
costs of educating their children (mainly the opportunity costs of wages



forgone) but not the potential benefits. The idea of participating in a Village Education Committee seemed far-fetched, and perhaps even absurd,
to members of families with almost no assets to their name.23
A similar regard for costs and benefits is apparent, finally, in a development project like the EIRFP, albeit in reverse (that is, from the point of
view of active participants). As we have explained, community organizers
are often reluctant to reach out to some of the most excluded individuals
in a cluster. It is much easier to work with literate men and women, and
people who reside close to the village school or post office. Yet it would be
a mistake to assume that differential rates of participation in the EIRFPs
groups are a function of information supply problems alone, or of a lack
of effort on the part of COs. This is to constitute non-participants and
active participants alike as little more than ciphers, or perhaps even as
We would prefer to suggest that poorer people sometimes choose to
avoid the project for good reasons. It would take a considerable investment of time and effort for such people to learn the ways of the project,
and even then they could have little expectation of leading the groups they
might join. In any case, the tangible benefits of project innovations seem
small as they are: this is a no-subsidy project, supplying in many cases
low levels of input to beneficiary populations or of little relevance, such
as a check-dam for a non-farming family, or a household cultivating only
uplands in the season when they are resident in the village. (Interestingly,
some of these families had availed themselves of information in the case
of the Indira Awas housing scheme, where they saw that BPL status could
translate into a claim on a tangible and very visible asset.) In contrast,
Kumar and Corbridge (2002: 86) found that the project was of great
interest to those families which ran the unlicensed stone quarries that
supplied boulders for the governments Indira Awas and jaldhara (irrigation well) schemes, and which were well placed to meet the projects needs
for building materials for check-dams. As ever, and quite predictably, an
informed if always imperfect appreciation of future costs and benefits

Craig Jeffrey and Patricia and Roger Jeffery have reported that some Dalit families in
western Uttar Pradesh are acquiring a formal education in part to gain a sense of distinction (in Bourdieus sense of that word), or improved self-worth and possibly also social
capital (Jeffrey, Jeffery and Jeffery 2004). To this end they have been keen to enhance
their stocks of literacy and numeracy. Sadly, we did not see evidence of this among the
Musahar communities of north (and to a lesser extent, central) Bihar. In part, this may
have to do with the poor quality of public education available to them, which in turn
is partly reinforced by the weakness of the private sector (at least when compared to
western UP). But it probably reflects, rather more, the hard costbenefit calculations of
Musahar adults. Most of the adults we spoke to could not imagine their children gaining
the sorts of jobs for which literacy would be an advantage.


The everyday state and society

had a great deal to do with observed patterns of participation, and these

departed significantly from those promised (or demanded) by a projects
The tyranny of participation?
Initiatives like the EAS and VECs are meant to help poorer people gain
paid employment and empower themselves. But they also propose a reengineering of established relationships between government and the
governed, and this is what we want to consider further in conclusion.
Has participation become a new technology of rule that substitutes a
fantastical projection of a benign and responsive state for real actions
that would disturb enduring structures of hierarchy and social exclusion?
Has it become a new form of oppression?
The new tyranny argument is not without merit. Kumar and Corbridge report that the Eastern India Rainfed Farming Project is in danger of being coded as a failure by its major funding agency, the UK
governments Department for International Development. The projects
logframe demands that improvements in livelihoods in the cluster areas
should be the direct result of villagers participation and enhanced skills
and capacities (Kribbhco et al. 1999: 5), but poorer villagers are generally
not active in project groups. This doesnt mean, however, that poorer villagers havent sometimes gained from project interventions, or that some
of them wont find work in fields watered by a project check-dam. Kumar
and Corbridge suggest that the EIRFP has worked extremely well as a
conventional (livelihoods-based) aid project. The elements of tyranny,
such as they are, reside in the fact that villagers are required to join in
groups to get inputs that they would prefer to receive directly, and in the
implicit suggestion that development without the participation of the
poorest is no development at all.
A group-based approach to participation was not a feature of the
Employment Assurance Scheme or Village Education Committees, or
Joint Forest Management for that matter. But here too we can see how
the tyranny argument might be broached. The notion that poorer people

Sanjay Kumar (2002) has published a path-breaking paper on JFM in Jharkhand which
suggests that villagers might be right to assume that the long-run costs of forest protection
(in terms of time and non-timber forest products forgone) will not exceed a future stream
of benefits from regenerated timber. He does not argue that villagers are making the
same calculations as he has been able to make (using different timber yield tables and
social rates of discount, for example). Nevertheless, the fact that they are coming to
the same conclusion through a different process of reasoning confirms that an apparent
unwillingness to participate in JFM activities is not only or simply a function of a lack
of information supply.



should join in forest patrols to help regenerate the timber stocks of richer
farmers or the Forest Department can be read this way, and the cooptation of Scheduled Caste men and women to VECs in Bhojpur can
likewise be dismissed as so much fluff, or even camouflage, when one
recalls that many of their children are unable to attend local schools.
There is clearly a measure of disguise in the rhetoric of participation,
and it is right that social scientists ask questions about whose interests
are served in this way.
That said, we cannot bring ourselves to endorse Cooke and Kotharis
critique in full. We need to recognize that participation is not a singularity, and that there are cases in eastern India where poorer women and
men have been able to engage development interventions in ways that can
be more or less fruitful, and more or less empowering. In our case, these
success stories are to be found mainly in Midnapore District, much as
Kohlis work would predict. Tiwary uncovered VFCs that did call regular
meetings and which kept records of them. He was also able to document
cases where VFCs had distributed thinnings and cash to poorer families that had participated in forest patrols, precisely as the law demanded
(Tiwary 2001: 5963). There are also reasons to believe that some hitherto excluded people, including some women, have come to see Forest
Guards and DFOs in a new light as a result of JFM. Tiwary inspected
the record books of some of the FPCs that he studied in West Bengal.
He found evidence of villagers engaging more forcefully over time in the
design of forest micro-plans, and of poorer people pressing a demand for
particular types of trees or ground-based crops.
Pressing demands, of course, is not the same thing as exercising power
or taking decisions and it would be wrong to read too much into these discussions. Nevertheless, JFM has been successful in parts of West Bengal
in bringing the state to the people, in the sense that forest officers,
including some quite senior officers, are now required to sit with members of forest-dependent households in the villages themselves. The signing of a minute by both parties, while apparently trivial, also hints at a
reshaping of a relationship that has traditionally been tense and coercive.
In Midnapore, too, we found that the CPI-M had sought to engage
people who were not its active opponents in a structured dialogue around
EAS projects. Party activists did not go out of their way to advertise the
EAS to labouring households, nor did they hold village open meetings in
a way that gave villagers full and direct control over key project decisions.
Many senior cadres cling to the view that the party knows best, and it is
well known that the CPI-M is reluctant to confront its many supporters
in the teaching profession. Nevertheless, it would be a mistake to ignore
the efforts of the CPI-M to hold gram baithaks in which poorer men


The everyday state and society

and women can make their views known to party activists. Our evidence
suggests that less powerful villagers have used these forums to make a case
for Project A or Project B, and for the most part their views (demands
in EAS terminology) have been heeded by party members. This is partly
because the CPI-M has to compete from time to time in a reasonably
competitive democratic framework. It has to behave like one among many
political patrons. But this is not the whole story. Senior cadres of the
party have long been supporters of a form of managed participation in
the institutions of panchayati raj, and at least some of the demands of the
EAS can be comfortably accommodated in that framework. The EAS
has presented an opportunity for the party to again take the part of
the poor, and it is likely that such participation as has been fostered in
this way has further encouraged the poor to see state functionaries (or,
here, party workers) as possible conversationalists. In other words, there
is some blurring of the boundaries between civil and political societies.
(We return to the implications of this remark in chapter 8.)
Matters are very different in Bihar and Jharkhand, and indeed in Old
Malda Block. The picture is not uniformly bleak, and it remains to be
seen whether the reintroduction of panchayati raj institutions will widen
such spaces of empowerment as can sometimes be observed in erstwhile
Bihar. The fact that VECs were not especially active in our field sites
might also speak to local particularities. It is possible that some VECs
in Murhu Block are more active than the one in our particular locality
because villagers have more reason to be active (because a local school is
failing, for example). As we have seen, the governments own data suggest a reasonable and continuing level of activity in Murhu and Bidupur
Blocks taken in the round. We are not convinced, however, that this is
the case, and we did not find confirming evidence in the villages that
surrounded the panchayats where we worked. What we did find, and have
reported here, are significant examples of participatory institutions being
formed and resourced by villagers outside the formal frameworks of state
programmes. These ranged from autonomous forest protection committees in Ranchi District to the (mainly) Yadav education committee that
was formed in our Vaishali field site. In addition, as we shall report further in chapters 6 and 7, ordinary citizens, including large numbers of
poorer men and women, are also participating in political movements in
these and other Districts.
Why, then, are the governments own vehicles of participatory development failing so badly (as they are also in Malda District)? The answer,
of course, is that Nehru was right all along. People will be reluctant to
jump through hoops which is how these exercises in participation are
sometimes seen if they are poorly informed about the schemes that are



on offer and the benefits that might be attached to them.25 To expect

Musahar children boys as well as girls to go to school in Bihar, or,
still more optimistically, to expect their parents to take part in VECs, is
to miss the very obvious point that these families lack even the most basic
assets: land, of course, but also a sense of self-worth and the prospect of
secure and properly paid employment. This in turn leads to something
of a vicious circle, for many Musahar households are then poorly placed
to deal with agents of the state or to demand employment from it under
the EAS. The high levels of social exclusion which they endure also contribute to their much lower levels of political awareness and involvement
when compared to another SC community like the Paswans.
The obverse side of social exclusion is pervasive state failure, particularly from the point of view of the very poor in Bhojpur (for reasons to
be explained further in chapter 6) and in Malda. In our Ranchi field site
this was countered to some degree by the effectiveness of the Mukhiya
in drawing down resources from the state. But here too the exit option
was widely favoured (or had become necessary) for poorer men and some
women. To speak of the tyranny of participation in Bihar or Malda is to
miss a more important point. It is hard to credit a discourse of participation with being a technology of rule when schemes like the EAS are
run by one or two key individuals, or where less than 5 per cent of the
intended beneficiary population is aware of the schemes existence, or
where poorer people disregard many of the rules that are supposed to
reposition them vis-`a-vis the state. Participation rather functions as an
absence, and it is equally hard to credit the idea that this might be legitimizing state power in some more subtle or perhaps unintended fashion
(as James Fergusons work might lead us to suppose). For reasons that
we explore further in chapter 6, most MPs and MLAs dont care greatly
about education or forest issues, and they see the EAS as yet another
source of patronage that they must control. The key battles in Bihar and
Jharkhand tend to be fought between villages or panchayats, rather than
between the people and the state in some more abstract sense.
In any case, as we go on to explore in chapter 5, many Block Development Offices are simply not set up to deal with the EAS in the way that
New Delhi and the major development agencies would like them to be.
One of the main lessons of this chapter, which will surprise only the most
committed advocate of participatory development, is that participation is
costly. Many of the intended beneficiaries of the EAS, or JFM, are poorly

Tiwary refers to villagers in Ranchi District, and indeed in Midnapore, complaining of

JFM, that too much demand is made on their time. There is, for example, an obligation
to attend meetings, which is an obsession with the leaders of the NGO-run forums
[especially] (2001: 259).


The everyday state and society

informed about these schemes, although we suspect that participation

rates would improve markedly if villagers thought that the expected benefits would significantly exceed their costs. But this is only part of the
story. As things stand, most poorer men and women in Bihar, and indeed
in Malda District, West Bengal, are continuing to see the state as they
saw it before the advent of participatory development programmes: at
a distance, episodically, often roughly and through intermediaries. And
this will only change, the World Bank and other donors now suggest, if
a discourse of participation is allied to reforms of government itself. For
poorer people to see the state more evenly and directly, the state first needs
to provide incentive systems for its employees to see the poor as clients
or citizens. What is needed, in short, is an agenda of good governance.


Michel Foucault once told an interviewer that it was important to be
humble in the face of apparent social irruptions.1 We should be properly
alert, he said, to continuities of history and geography, and not constantly
on the look out for markers of the new or what today might be called
the post-. This is surely good advice, and we need to bear it in mind
when discussing issues like participation and good governance. The idea
that states in the past have not been concerned with good government is
clearly wrong. The emergence of biopolitics is one strong indicator of the
responsibilities that governments are meant to have to their populations.
Nevertheless, there is a strong perception in the development community
that state failure and bad governance have become important issues since
the 1970s, and this perception has been linked to a broader critique of
rent-seeking behaviour, simple predation, and dirigiste development.
In the next part of the chapter we review some of the debates that
have attended the rise of the good governance agenda. We shall also follow Adrian Leftwich and Rob Jenkins in drawing attention to the ways in
which the agendas of good governance can be said to depoliticize accounts
of development and rule.2 They do so, not least, by refusing to pay close
attention to questions of state capabilities, and the incapacity of some
regimes to secure control over their territories. This matters most in parts
of post-colonial Africa, as we indicated in chapter 1, but it is of significance too in parts of eastern India.3 They also make quite far-reaching
assumptions about the self-regulating properties of the economy and civil


I think we should have the modesty to say to ourselves that, on the one hand, the time
we live in is not the unique or fundamental or irruptive point in history where everything
is completed and begun again. We must also have the modesty to say, on the other hand
that . . . the time we live in is very interesting; it needs to be analysed and broken down,
and that we do well to ask ourselves, What is the nature of our present? (Foucault
1983: 206).
Leftwich (1993); Jenkins (2002).
For example, in Sahar Block, Bihar and in some of the more remote Blocks of Jharkhand.



The everyday state and society

society, and thus about the production of modern subjectivities. Roberto

DaMatta suggests that in Brazil the individual is considered an unpredictable and unstable force that must be contained through incorporation
into the collective social body (after Nelson 2003: 1213). The individual is someone who is not incorporated in a home and who does not enjoy
the protection of membership in a social group. Such a person, DaMatta
suggests, belongs to the masses, and finds him or herself at the mercy of
the police and the law. For everyone else, the laws seem not to apply,
a concept captured by the Brazilian saying, To our enemies, the law;
to our friends, everything! (Nelson 2003: 13, quoting DaMatta 1991:
168). Should we assume that matters are very different in India?
In the middle parts of the chapter we consider how the new governance
agenda might look from the point of view of some of the key individuals
at the Block and District levels who are charged with making it work.
Instead of thinking of government actors in impersonal terms, we want
to give a stronger sense of their embeddedness in local society. We can
then begin to describe the contrasting and often competing pressures that
are brought to bear upon them: pressures from their seniors and communities of experts, for example, and pressures from important actors in
political society, and perhaps even from some poorer clients, citizens and
family members. We do this first in regard to the Employment Assurance
Scheme and discourses that circulate about corruption and development.
We show that the good governance agenda presupposes the construction
of a type of individual that is uncommon in our field areas, and the manufacturing of which is hampered by a lack of state capacity. Government
officers will find it hard to behave like a Weberian bureaucrat when they
lack the support of a Weberian bureaucracy. At the same time, however, it
is not clear to us that subaltern officers subscribe straightforwardly to the
tenets of vernacular society (as described by Chatterjee and Kaviraj, in
unconscious recognition, perhaps, of DaMatta), or that they are opposed
in principle to some of the developmental concerns that are voiced by
higher-ranking public officials. We will maintain that neither perspective
on governance (or rule) pays sufficient attention to the diverse ways in
which government officers come to see the state. We make this argument,
moreover, not simply in relation to the EAS, but also in regard to education. To a lesser degree, we are also able to report some of our findings
about the protective function of the state, or the ways in which the legal
system is constituted, understood and occasionally put to use by different
We conclude by returning to the critique of the good governance agenda
that has been voiced by scholars and activists who prefer to concern
themselves with politics (and political society). We have already said that



we share this preference in some degree, and we shall elaborate on the

assumptions about the codes of conduct that are central to the new public
administration. There are financial costs associated with what has been
called the deepening of democracy. But we also want to make the point,
which we develop further in chapters 8 and 9, that a degree of romanticism
about politics is present in the claims of those who shout loudest about the
dangers of depoliticization. These claims speak to an account of politics
that makes reference to what we shall call an ideal outside, or a world
where politics can be properly constituted and made to secure its desired
effects.4 Our perspective is less enchanted. It is geared to thinking of politics as a continuum of practical and not always additive actions around
the construction of social and economic relationships and forms of
Good governance
Governance is generally understood as the prevailing patterns by which
public power is exercised in a given social context (Jenkins 2002: 485).
As such, it has been a concern for all ruling parties and states. It is equally
the case that all governments have concerned themselves with questions
of good governance. Since there is no clear definition of the good, it
is possible for totalitarian regimes to make claims about efficient rule,
just as corrupt and inept governments can make it look as though they
are addressing important issues of statecraft. As Jenkins reminds us, it is
the chameleon-like quality of the good governance agenda that makes it
so attractive to stakeholders.5 This is certainly the case when it comes
to the Bretton Woods institutions, which are bound by their articles of
agreement from intervening in a countrys politics. Politics is the preserve of sovereign countries. Institutional change and good governance
supposedly are not.

We will comment in chapter 8 on the analysis of political society developed by Chatterjee

and others. Romanticism is less evident here than a prevailing pessimism. This is the
counterpart, in some degree, of a particular reading of Foucault on power, albeit one less
imbued with the agentless and non-intentionalist reading of history that led Edward Said
to criticize Foucault in the 1980s for [his] refus[al] to consider the space of existence
beyond the power of the present (Legg 2005a: 10, summarizing Said 1984: 2457).
We are being polite. Jenkins refers to development agencies drawing promiscuously
on ideas that have sometimes emanated from the Wests largest consultancy firms, and
which are then applied to constrain the performance-inhibiting instincts of southern (and
northern) governments by subjecting them to multilateral policy disciplines (Jenkins
2002: 486). The involvement of firms including Arthur Andersen (as was), Price Waterhouse Cooper and others, in the restructuring of forms of government in southern (and
northern) countries, has not yet received the attention from academics and activists that
it deserves.


The everyday state and society

This is a fiction, of course, albeit a necessary fiction, for it was precisely a concern for politics that prompted the most recent invention of a
good governance agenda. This first became apparent when the counterrevolution in development theory and practice began to speak about
the economy, and indeed of political economy. Writing in 1974, Anne
Krueger suggested that trade distortions in India and Turkey were practised on a massive scale by governments that were content to abuse the
privileges of sovereignty. The total value of rents in India in 1964 was
estimated to be Rs. 14.6 billion, or 7.3 per cent of national income, of
which more than two-thirds came from a system of import licences that
favoured only a small minority of the population (Krueger 1974: 295).
This in turn encouraged a political vicious circle [wherein] people perceive that the market mechanism does not function in a way compatible
with socially approved goals because of competitive rent-seeking. A political consensus therefore emerges to intervene further in the market, rentseeking increases, and further intervention results (Krueger 1974: 301).
Krueger would later serve as the Chief Economist of the World Bank
during the Reagan years. She would then press Bank directors to wean
countries away from inwardly-oriented development policies. She was
helped in this task by the debt crisis of the 1980s. This paved the way for
structural adjustment programmes that sought the liberation of the economy from the public sector and government misrule. The World Bank
also declared that oppressive marketing boards and overvalued exchange
rates were responsible for a growing crisis of food production in subSaharan Africa. This pattern of urban bias, however, could not survive
rising food prices. Food riots might break out in the cities, and regime
change, to use the modern parlance, would create a space for external agencies to impose adjustment programmes.6 Devaluation could be
pushed through, and marketing boards or their equivalents broken up.
The market would then do its work, and development would resume
what Michael Beenstock called its normal course, undisturbed by the
mistaken or simply venal claims of politicians.7
The idea that sovereign misrule could be turned so rapidly into governmentality (as expressed in the self-regulating capacities of the economy) was shaken for a while by the experience of Russia with shock
therapy measures. The major architects of the Washington Consensus

The economic analysis can be found in the so-called Berg Report (World Bank 1981); the
politics of urban bias was explored more assiduously, and from within a rational choice
framework, by Robert Bates (Bates 1989).
Beenstock (1984) made this remark in the context of the debt crisis, where he was concerned to argue against proposals for greater regulation of the international economy.
Nevertheless, the remark was clearly intended to have wide purchase.



now insisted they had been misunderstood and took care to distance
themselves from the utopian project put forward by Jeffrey Sachs and
his colleagues.8 Jeffrey Williamson declared that his original account of
a Washington Consensus had not intended to prescribe a set of correct economic policies for all countries. He was mainly concerned that
countries should avoid policies that were straightforwardly wrong, like
running consistently high trade or budget deficits, or failing to deal with
sick industries.9 But a more significant change to the governance agenda
was brought about by the end of the Cold War. Confident that socialism
had been defeated, and yet mindful of the need for effective government
to support free or freer markets, an agenda for good governance now
took shape which married respect for the markets power to create modern
subjects to an insistence that those subjects would flourish best in vibrant
civil societies. Good governance, then, came to define those patterns of
rule which protected the individual citizen from political society (politics
still being seen in essentially negative terms) and from unrepresentative
In sub-Saharan Africa and the ex-USSR this led to a particular concern
for multi-party politics and democratization. In Latin America and South
Asia attention was focused more often on the institutions of democracy,
and on the need to strengthen civic associations. Side by side, then, with
policies that were put forward to privatize or at least deregulate some parts
of the economy, measures were proposed to transfer some of the powers of
the central state to local government institutions and non-governmental
organizations. As James Manor has shown with great clarity, the agenda
for decentralization describes a range of policies that may or may not
be pursued in tandem. It is important, at a minimum, to distinguish
between: (a) deconcentration, or the dispersal of agents of higher levels


Even so noted a supporter of free trade and free markets as Jagdish Bhagwati was unimpressed with Sachss activities in Russia. Alarmed that he might offer similar advice to
his new interest, India, Bhagwati wrote that: The last time that technocratic full-speed
ahead advice to a reforming government backfired badly was when shock therapy for
macro-stabilization was prescribed for Russia, with a backlash that gave Russia much
political turmoil and little economic progress while returning Jeffrey Sachs unceremoniously . . . to Harvard. [He would later join Bhagwati at Columbia University.] I am
reminded of his famous line: You cannot cross a chasm in two leaps, to which Padma
Desai replied: You cannot cross it one leap either unless you are Indiana Jones; so you
drop a bridge instead (Bhagwati 1993: 37).
I can see no advantage to democracy having major parties spouting economic nonsense
(Williamson 1993: 1330. See also Williamson 1990).
In the World Banks formulation it defined, political institutions such as constitutional
rules, the division of power among levels of government, independent agencies, mechanisms for citizens to monitor public behaviour, and rules that inhibit corruption . . . . [all
these] succeed in restraining officials of the state from arbitrary action (World Bank
2001: 115).


The everyday state and society

of government to lower level arenas; (b) fiscal decentralization, or the

transfer of budgetary resources to lower level arenas; and (c) devolution,
or the transfer of democratic and administrative powers to lower levels of
government (after Manor 1999: 5, citing Rondinelli 1981). Perhaps more
important, however, is the fact that one or more versions of this agenda
are now circulating widely in both northern and southern countries. They
have the imprimateur not just of the Bretton Woods institutions but also
of the worlds leading consultancy companies (keen to make money from
applying the lessons of the new corporate governance) and academic
units like the Harvard Center for International Development.11
In a striking demonstration of one of this books central theses, indeed,
it is clear that Merilee Grindle at Harvard has played an important role
in challenging the state (the title of one of her books) to do better. She
has acted as a consultant to a number of countries, helped teach senior
officials in short courses at Harvard, written widely on the need for public
sector reforms, and become a champion of the view that a new generation
of public servants must have a sense of ownership of the new governance
agenda. (This might be done by linking pay more clearly to performance
and by sensitizing officials to their responsibilities to those they serve.)
We would also maintain, however, that the agendas of good governance are constantly in flux, and are open to at least some degree of
critical reflection by their proponents. Judith Tendlers book, Good Government in the Tropics, provides a case in point. Her study of a preventive
health project (Health Agent Programme) in Ceara state in Brazil in
the early 1990s suggested that very significant health-care improvements
had been secured by thousands of highly motivated field agents, including mainly female nurse-supervisors. Vaccination rates soared and infant
mortality rates fell. Although poorly paid, these agents showed a strong
commitment to the programme because they had been engaged through
a remarkable process of merit hiring (Tendler 1997: 28).12 More so
than some other writers, however, Tendler was reluctant to focus only
on questions of cost or the fact that the state had surrendered one of its
traditional functions to a non-state actor. She also noted that the project
has worked well in Ceara because higher-level state functionaries had
taken actions to marginalize the role of local mayors, or political actors
who might otherwise have seen it as a new source of patronage. Politics
might remain a source of intrigue or even damage, but the suggestion

Jenkins (2002: 485) refers to inputs from the fields of corporate governance and network governance, but neglects the role of academic centres and the courses they provide.
Although Tendler does not give it the same emphasis, she also points out that healthagent jobs offered full-time work year-round, in an agricultural economy where employment was highly seasonal (Tendler 1997: 29).



that a strong (central) state might be required to make a space for nongovernmental activities begins to point away from the middle ground of
good governance studies. Whether it also begins to point towards a more
robust understanding of political society is less clear, and is something
we shall come back to.
Good governance in India
Significant elements of the new public administration are now recognizable in India and are being pushed quite vigorously by New Delhi. It is
important to understand, even so, that the good governance agenda in
India has taken shape with reference to a long tradition of concern about
statesociety relationships. The leading lights of Indias anti-colonial
struggles all spoke about the importance of good government, and suggested that power had to be used for and by the people. The debates
which surrounded this platitude related mainly to the site of empowerment. Gandhi pointed to the village and the panch, while Ambedkar
looked to District and state-level institutions to break the power of locally
dominant castes.13 By the 1960s, moreover, the question of corruption
was forcing itself onto the national political agenda. The setting up of
the Santhanam Committee in 1963 was one indication that the postcolonial state was not living up to its lofty ideals. Its report on the prevention of corruption recommended the setting up of a system of Chief
Vigilance Officers (CVOs) to review existing opportunities for corruption and malpractices (Government of India 1964: 289). The CVOs
were also required to maintain proper surveillance on officers of doubtful integrity. Gunnar Myrdal later generalized some of the Reports conclusions in his account of the soft state in India (and other countries
in Asia).14 His calls for less government, for government that avoided
administrative delay, and for government that reduced the scope for personal discretion on the part of its officers, anticipated some of the concerns of the new public administration. The relevance of these calls was
highlighted again in the 1970s and 1980s as concerns grew about the
abuse of executive power and the criminalization of politics.15 By the end


Ambedkar referred to Indias villages as dens of ignorance, narrow-mindedness and

communalism (Constituent Assembly Debates, quoted in Galanter 1991: 3). He looked
to strong, rational and centralized rule as a means of empowerment and enlightenment
for the rural poor.
Myrdal (1968).
The offices of the Indian Express were raided in 1987 precisely and ironically because
the newspaper was seeking to expose the government of Rajiv Gandhi over the Bofors
affair. On the criminalization of politics, see especially Bardhan (2001), Brass (1997)
and Kohli (1990).


The everyday state and society

of the 1980s, Prime Minister Rajiv Gandhi was sufficiently convinced

that decentralization would be a vote-winner that he built it into his electioneering slogan: We have given you power, he took to saying.
In fact, nothing much changed under Mr Gandhi, and it was left
to Narasimha Rao in the 1990s to turn promises into policies.16 The
economic reforms that were introduced by his government were at first
intended to deal with Indias balance of payments problems, but a broader
programme of administrative and economic reforms was also meant to
address what Atul Kohli had called Indias growing crisis of governability. It did so, however, in a manner that might have surprised Kohli. Kohli
had suggested that widespread administrative failure in India was being
produced by an absence of enduring coalitions and a structural incapacity to accommodate political conflict without violence (Kohli 1990:
23). It followed that a solution had to be found in the field of politics.
Narasimha Rao, on the other hand, and the prime ministers that followed him in the 1990s, preferred to follow the logic of the new public
administration. They looked, that is to say, to a slowly evolving mix of
deregulation, privatization, civil service reform and decentralization to
produce the administrative and political effects that Kohli had deemed
Embedded officials and the Employment
Assurance Scheme
It is not possible here to comment on all or even most aspects of this new
agenda (see box 5.1). In particular, we cannot discuss the local consequences of macro-economic policy changes, although we recognize that
these might be significant for poorer people. Much will depend on how
state governments amplify or dampen these changes in terms of their
public spending decisions. We do note, however, that some very severe
objections have been raised against the good governance agenda, several of which turn on the assumptions it makes about state capacity and
the world-views of public officials. In the context of sub-Saharan Africa,
especially, Rob Jenkins follows Robert Jackson in talking about the quasistates that are often to be found there.17 Unable to enforce even minimal levels of social penetration across the territories they are meant to


This, at any rate, is the judgement of James Manor (1999: 445). It would perhaps be
fairer to Mr Gandhi to point out that it was under his leadership that six Technology
Missions were set up (to explore, for example, C-DOT connections to remote locations),
and that the 73rd and 74th Constitutional Amendments were prefigured by the PM-DM
workshops that he instituted.
Jenkins (2002: 486), citing Jackson (1991).



Box 5.1 The new public administration and poverty

alleviation in the countryside
In addition to a new emphasis on privatization, deregulation of labour
markets, a slimmed down and more responsive civil service, and new
initiatives in the fields of education and health-care, the Government
of India has redesigned some of its poverty alleviation programmes,
not least in the countryside. Very much in line with participatory development philosophies, and with ideas to the fore about the formation
of self-help groups and social capital, the Ministry of Rural Development has launched two initiatives that have taken shape since the time
of our initial fieldwork.
Indias wage-employment schemes, the EAS and JGSY (formerly
JRY), have been reformed as two streams of the Sampoorna Grameen
Rozgar Yojana (SGRY), which was launched in September 2001.
The SGRY is self-targeting and has the primary objectives of providing additional rural employment and ensuring food security at
the household level. Wage payments now include an in-kind component; that is, workers under the SGRY are partly paid in food grains
from the Food Corporation of India. The scheme is decentralized and
demand-led. Only those districts that demonstrate a popular need
for the programme, by swift utilization of funds and a capacity to
administer it, receive a second yearly instalment. SGRY projects are
put forward by the panchayats, and the second-stream village-level
schemes require the approval of the village assembly (gram sabha).
The panchayats are responsible for the implementation of SGRY
schemes, but the execution of works can also be entrusted to wellestablished Self-Help Groups of the Swarnjayanti Gram Swarozgar
Yojana (SGSY) (SGRY Guidelines, Ministry of Rural Development,
The SGSY is Indias other main poverty-alleviation programme.
This self-employment scheme was launched in April 1999, and
replaced the Integrated Rural Development Programme and a few
other self-employment schemes. The SGSY is implemented in partnership between the District Rural Development Agencies, the panchayats, banks and NGOs. The programmes objective is to lift and
keep swarozgaries (self-help families) above the poverty line by organising the rural poor into Self Help Groups (SHGs) through the process
of social mobilization, their training and capacity building and provision of income generating assets. The expectation is that the SHG
approach will help the poor to build their self-confidence through
community action, establish a large number of micro-enterprises,


The everyday state and society

and ultimately lead to the strengthening and socio-economic empowerment as well as improve [the] collective bargaining power [of the
rural poor] (SGRY Guidelines, Ministry of Rural Development, Although the SGSY had given loans to over
two million individual swarozgaries by 2004, the discursive emphasis
has been on SHGs, of which almost 1.3 million have been formed since
1999. These groups have between ten and twenty members and at least
half of them should be women-only SHGs. The groups are supposed to
go through four stages: (1) formation: where groups are mobilized and
motivated by facilitators (including NGOs); (2) stabilization: where
groups begin with savings and intra-group borrowing; (3) micro-credit:
where the SHG gets access to revolving bank loans; and (4) microenterprise development: where the SHG takes up income-generation
activities in a sector that has been identified as economically suitable
for the region by the Block SGSY committee (SGRY Guidelines). The
new paradigm of development is also reflected in the evaluation of
the SGSY. This tends to highlight the number of SHGs formed rather
than the disbursal of funds or the value of assets created. The Ministry of Rural Development only notes the following regarding SGSY
on its Achievements at a Glance website (
Since the inception of [SGSY] a holistic programme for promoting
self-employment in rural areas 11.45 lakh Self-Help Groups (SHGs)
formed till date and about 1.00 lakh Groups have been constructed.
(Sources: as in text.)
command, these quasi-states are ill-equipped to govern, let alone govern well. Instead of strengthening local infrastructures of rule, however,
the Bretton Woods institutions are now demanding that the governments
of these quasi-states surrender such economic powers as they have to
external agencies. Ever neglectful of political realities, these institutions
fondly assume that structural adjustment programmes will pave the way
to good governance, even when the evidence points in the opposite direction.18 Barbara Harriss-White, meanwhile, writing of India, argues that

The World Bank has acknowledged that The state cannot provide costly public goods
without the power to tax individuals and companies to raise public revenues (World Bank
2001: 99). What it neglects is the difficulty that many states face in raising taxes when
a climate of opinion has been created which favours tax and spending cuts. Nowhere
is this more evident than in the US, where George W. Bush exploited this climate to
push through a fiscal regime which cut the taxes of the super-rich (and especially those
trousering $300,000 or more) before discipline was restored by attacks upon the surpluses which working- and middle-class Americans had built up in the social security
system. The most effective critic of Bush, by far, has been Paul Krugman, writing in the
New York Times (see Krugman 2003, 2004).



The World Banks project for the State is the opposite of what is needed
(Harriss-White 2003: 100). She also contends that the World Bank is
missing two more important points when it talks of corruption and the
need for new and more efficient forms of government: firstly, that tax
evasion is far more disabling of government than is corruption, and, secondly, that the so-called state in India has already been privatized and
turned into a vehicle (the shadow state) for the accumulative projects of
local capitalist classes (Harriss-White 2003: 1001).
We shall come back to these broader observations at the end of the
chapter. We shall repeat here only the rather cryptic remark that we made
earlier on; namely, that while we are sympathetic to significant parts of this
analysis, we are not fully in agreement with its political or policy-related
undertow (see Conclusion). For the moment, however, we want to focus
on larger issues by means of a much smaller lens. We want to consider
the question of good governance at the local level, and with close regard
for the ways in which government officers understand some of the new
initiatives they are meant to preside over. This will take us back to the
Employment Assurance Scheme and village education councils, both of
which advertise the concerns of the new public administration for flexible
bureaucracies and the responsiveness of officials to the broader public. Is
it the case that the local state is inefficient, overstaffed and fundamentally
venal, as the new governance orthodoxy so often assumes, and if so, why?
Is this likely to change as government servants are required to respond
to citizens as clients? And is it sensible to assume that directives from
New Delhi or state capitals will have the impacts at local level that are
intended (or supposedly intended) by senior-level bureaucrats? Might
there not be other pressures which affect the actions of members of what
Kaviraj has called the vernacular state and society? Might it be the case,
as Ronald Inden has argued even more strongly, that men and women
from this quarter are unimpressed by ideas of generalized morality, and
are opposed in important ways to what he calls the grandiose discourses
of planning, development and nation-building?19
These questions are not easily answered, for much will depend on the
characteristics of local political societies. But consider how the Employment Assurance Scheme was redesigned in several of our Districts,
and how this reworking speaks to questions of corruption and possible

Inden suggests that these discourses are underpinned by a concept of Reason that is
every bit as transcendental as the religious progresses that endowed the medieval king
of kings with the luminous will of Vishnu (Inden 1995: 271). In each case, particular
pilgrimages or progresses (small-scale developments) are sublimated into a grander idea
of Progress or Modernity, a grander idea that turns its back on the lives and wishes of a
majority of Indians who are expected to heed their masters voice and respond dutifully.


The everyday state and society

gaps in understanding between elite and subaltern officials. We know

that the first aim of the EAS was to set up a demand-led system of
employment provision. But an important second objective was to make
sure that any schemes sanctioned under the EAS guidelines would be
labour-intensive. Thus, while the Government of India insisted that the
provision of employment under this scheme results in the creation of
durable productive assets in the Block area (Government of India 1993b:
3), it also required that: Only labour-intensive works of a productive
nature which create durable assets should be taken up for sanction under
EAS and included in the shelf of works/projects (Government of India
1993b: 4).
In practice these guidelines gave District-level officers considerable
room for manoeuvre, as we shall see later, but the bias to employment creation was clear nonetheless. Section 3.10 of the guidelines states that: All
works started under EAS should be labour-intensive works only. Labourintensive works are defined as those which have a ratio of unskilled labour
to equipment, material, and other skilled work of not less than 60:40.
Works requiring a larger component of materials like cement, steel, etc.,
should not be sanctioned under the EAS unless the excess cost on material
components is provided from other sectoral programme funds (Government of India 1993b: 5). The guidelines further suggested that this ratio
of 60:40 could be met if new works under the EAS were distributed as
follows: 40 per cent on water and soil conservation measures, including
afforestation, agro-horticulture and silvipasture; 20 per cent on minor
irrigation works; 20 per cent on link roads (as per the Districts Master Plan guidelines); and 20 per cent on primary school and anganwadi
(health-care) buildings.
Now consider table 5.1. This makes it clear that the schemes that were
run under the EAS in our Bihar (and Jharkhand) research Districts (from
19934 in Ranchi, from 19967 in Bhojpur and Vaishali) were designed
mainly to run as material-intensive projects. In Murhu Block (Ranchi)
156 schemes were sanctioned in the years 19934 to 1998, and of these
schemes only seventeen could be described as labour-intensive (the pond
or ahar schemes). A further ninety schemes worth Rs.18.5 million generated a demand for unskilled labour that cannot have amounted to more
than 3035 per cent of spending, and this percentage declined over time
as individual road-building schemes became more expensive. In Bidupur
Block (Vaishali) just fifty-seven schemes were sanctioned in a three-year
period, and 90 per cent of the funds was spent on road projects that were
often all-weather and black-topped. This did not deter the relevant officials from declaring that 58 per cent of the budget had been spent on the
wages of unskilled labour. (In Sahar Block the figures were shown to be












Number of





Share of total

Bidupur Block, Vaishali 1996/798/9:

Average EAS spending per year
Rs 5.0 million

Notes: PCH = primary health centre; ICDS = integrated community development scheme = all totals are rounded
Source: District and Block statistics.


School building
Community hall
Bridge/ culvert

Share of total

Number of

Number of

Share of total

Sahar Block, Bhojpur 1996/799/00:

Average EAS spending per year
Rs 3.6 million

Murhu Block, Ranchi, 1993/499:

Average EAS spending per year
Rs 5.2 million

Table 5.1. Sector-wise breakdown of EAS schemes actually implemented by the Blocks, Bihar


The everyday state and society

exactly 60:40; a District-level officer in Vaishali told us that he favoured

a 58:42 split on the ground that it appeared more credible.)20
What accounts for this bias towards big projects, and what does this
tell us about the local state and the good governance agenda? It might
be thought that a bias of this sort reflects a lack of understanding among
Block-level workers. And it is true that none of the Block Development
Officers to whom we talked had a good understanding of the EAS, despite
the fact that they had the assumed the responsibilities of the Pramukh (the
elected chief of the Panchayat Samiti) in the wake of the dissolution of the
panchayats in Bihar in 1998. The BDOs knew that the EAS was about
employment provision, but they were poorly informed about the demandled or guaranteed nature of that provision. The BDOs rather understood
the EAS to be a sister programme to the Jawahar Rozgar Yojana (JRY)
programme, with the former taking up big schemes and the latter small
schemes (interview with BDO, Murhu). Levels of understanding were
still less clear lower down the government ladder, and we were assured by
a Block Agricultural Officer in Sahar that EAS mein sunishchit samay per
scheme ko poora ker dena hota hai (EAS means that we have to complete
the scheme in the assured time frame).
Misunderstandings of this sort should indeed set off alarm bells when
it comes to the assumptions that are currently being made about information flows in decentralized government structures. But they cannot
explain the preference for big schemes that is evident at the District
level. The Block and village-level workers, after all, were receiving their
information from their superiors at the District-level; they did not have
access to the same manuals as their bosses. So why did District-level officers act as they did when sanctioning material-intensive schemes? Part of
the answer is to be found in the technical imperatives of the Employment
Assurance Scheme. The District-level officers who we spoke to in Bihar
were well aware that the Block was the main implementing agency for
the EAS. They also knew that many Block offices were overburdened.
In the 1970s the Block office was required to deal with relatively few
poverty-reduction schemes. In the 1980s and 1990s these schemes were
significantly expanded, and, increasingly, were supposed to be run on a
decentralized basis. By 1999 a typical Block was receiving six to eight
times the funds flow it would have received in 1979, and the Block office
might be asked to run 100130 schemes under JRY and EAS, as well
as providing 1,000 houses under the Indira Awas scheme and 500 wells
under the Million Wells scheme.

Wherever possible, we prefer not to state the name or the precise job title of our interviewees, many of whom talked to us on condition that they would not be identified. In
some cases, however, as with well-known politicians who are easily identified, our rule
had to be relaxed and this was explained to the respondent.



In a Block like Sahar these schemes were supposed to be spread across

fifty-five villages in twelve panchayats, many of which are very difficult to
access. The Block office in Murhu is responsible for 141 villages in twentyfour panchayats, while in Vaishali there are 133 villages to serve in twentyfour panchayats. Despite repeated and sometimes justified claims about
government overstaffing in India, these workloads had to be shouldered
by about the same number of workers as would be found in a Block
in 1980. In Murhu, for example, the complement of civil staff in the
Block office ran to the Head Assistant, the Nazir (accountant) and two
assistants. On the technical side there were four Junior Engineers and
one Assistant Engineer (who also worked for the neighbouring Khunti
Block), while the complement of field staff ran to twenty-four Panchayat
Sewaks and nine Village-Level Workers. The BDO of Murhu despaired
of the situation. He told us that the better able and connected of his
workers were trying to find work in urban areas, and that his accountant
was not up to the task of handling cash transactions in the sum of Rs. 34
The situation was not much different in Bidupur or Sahar. In Sahar
there were only two Junior Engineers to deal with between 300 and 400
schemes. These men were expected: (a) to prepare estimates for proposed
schemes (prior to any approval being granted at the District level); (b)
to prepare the layout for a scheme and to work with the chosen contractor; (c) to supervise in person the most crucial stages of construction,
which might include foundation casting, the fixing of a lintel, and roof
casting; (d) to inspect and measure the progress of a scheme (in part to
secure future fund flows to the executing agent); and (e) to assist senior
engineers (including the Assistant Engineer) in their inspection visits.
Naturally, these tasks could not be completed properly, even where the
Junior Engineer was working to the best of his ability. If there is state
failure and significant leakage of funds it is at least in part because the
local state is underdeveloped in relation to the tasks set for it. And this
will be true even when supervisory staff from the non-technical side are
deployed to help the process of scheme inspection.
But it is not just technical or capacity problems that incline Districtlevel officials against the sanctioning of a large number of labour-intensive
(and kuccha) schemes. A Block-level officer in Murhu told us that: We
tried to include proposals of morum [unmetalled] roads and irrigation
ponds in our proposal as they generate maximum employment. But, during the scrutiny at the level of the DDC [Deputy Development Commissioner], those were struck off the list. In fact, in 1997, DDC gave an
oral instruction that henceforth no earthwork schemes should even be

About $800,000 at 1999 prices.


The everyday state and society

proposed. When pressed on why he thought this was, the officer replied
that: They [District-level officials] worry about misuse of money if more
kuccha works are executed . . . [This was] not simply because they want
to save skins [but because] they genuinely believe that kuccha works will
always be subject to siphoning off of government money, and they wish
to safeguard against that.
This understanding of the motives of a (senior) District-level officer
proved to be extremely accurate. In our conversations with District Magistrates and District Development Officers we found that political considerations loomed large in their decision-making. Contrary to the assertions
of some neoliberals, it is a mistake to assume that government officers in
India are intent on maximizing the rents they can extract from the misuse
of a public office. It is clear that a scheme like the Employment Assurance Scheme does lend itself to a system of institutionalized cuts and
commissions, much as Robert Wade has described in the case of a south
Indian irrigation department (Wade 1982). Junior Engineers regularly
mark up the costs of a scheme, and there are significant opportunities
for BDOs, JEs and others to receive a cut from the monies that are then
spent. Nevertheless, we estimated that the magnitude of fund leakage
from EAS schemes in Bihar was of the order of 3035 per cent of the total
flow of funds, and that (or rather but that, given received views about
Bihar) we found no cases of outright looting where a scheme existed
only on paper. (Matters were worse in Malda District, West Bengal.)
Even corrupt officials find themselves in a more complex set of relationships than is acknowledged in a simple theory of predation. Perhaps
especially in Bihar, given the publicity that has been given to the cattle
fodder scam, government servants are mindful that they might be found
out if they engage in corrupt behaviour, or that they might be informed
upon by one of their colleagues.22 This cautions them against excessively
predatory forms of behaviour. In the case of District-level officers, moreover, there are strong pressures not just to exact rents (which might be
needed to maintain close links with sympathetic politicians, or to help
secure better postings), but also to clamp down on the corrupt activities
of their subordinates.23 Thus, while it is clearly the case that some bribes
(ghus) are channelled up the hierarchy of Bihar Administrative Service
and Indian Administrative Service officers to the District and state levels,
it was equally clear to us that District-level officials were pressing for
pucca EAS schemes as a way of guarding against what they saw as the


A scam that allegedly involved the looting of state budget funds (possibly $200 million)
that were meant to pay for livestock improvements in Bihar. On this and other scams,
see Nambisan (2000).
For a more nuanced account of corruption and rent-seeking, see the collection of essays
edited by Mushtaq Khan and Jomo Kwame Sundaram (2000).



twin evils of kuccha projects: the fact that they provide so much scope
for corruption (on account of being difficult to inspect), and the fact that
some of these schemes will be built to fail (or to fall down), thus denying
any visible evidence of development.
What we observed here was a lack of trust in Block-level officials by
their District-level bosses. But this lack of trust also extends upwards
from the District to the governments sitting in Patna and New Delhi.
When we pressed District-level officers on their efforts to reshape the
aims and objectives of the EAS efforts that were freely acknowledged
they focused on what they saw as the lack of credibility that surrounds
the issue of the flow of funds. All of our respondents challenged the idea
that New Delhi or Patna could ever hope to fund sufficient schemes to
employ two adult household members for up to 100 days each year
throughout India, and then mainly in the lean season. The DDC of
Ranchi was adamant that the Centre simply didnt have the resources
to direct more than two instalments of funds to any District of Bihar in
a given year, a view based, no doubt, on his difficulties in acquiring a
third tranche of funds for Ranchi District. When we put it to him that
some Districts in Andhra Pradesh were reputed to have received five or
six instalments, he countered by saying that he had visited New Delhi to
press for a fresh round of funding, and had been told by the Secretary
of the Ministry of Rural Development that this was so much rhetoric:
no District had received more than three instalments of EAS funds in a
Whether or not this is true is not really the issue. The point is that DMs
and DDCs in Bihar dont trust the authorities in New Delhi or Patna to
provide sufficient funds to check the out-migration and exploitation of
akushal mazdoor [unskilled labour] which is how, in 1995, the Government of Bihar described the main purposes of a programme that would
be demand-driven and [with] no financial limit (Government of Bihar,
Secretary Rural Development, Letter No.3248, 17 June 1995, sent to
all DMs and DDCs, Bihar).24 They also doubted whether their own
Blocks and Districts, or indeed the government of Bihar, had the means
or the drive to make demands of New Delhi at the right time. The DM of
Bhojpur told us that even if the EAS could draw down unlimited funds
from the Centre, those funds were still budgeted on an annual basis and
would be taken up on a first-come, first-served basis. In his view, Bihar
was too slow in making its demands for further instalments of funds. By
the time it was ready to claim a third round of funds it was too late in the
financial year: other states had got in ahead.

It is significant here, as Ben Rogaly and colleagues (2002) have emphasized in their
research, that the state is coding migration as a problem, even as a pathology.


The everyday state and society

Whatever the truth of the matter and the fact that some Blocks always
have EAS funds unspent suggests that local capacity is key the fact
remains that District-level officers are wary about advertising the EAS
too widely. In their view, it is better to plan for a small number of wellcosted and at least partly monitored pucca developments than to plan
for a large number of kuccha schemes that will generate kickbacks and
local conflict, and which cannot hope to soak up the local demand for
paid labour. As the DM of Vaishali summed up: The fund that we get
now, two instalments, can hardly generate 100 days employment. In
fact, with these limited resources available under EAS, the approximate
labour days generated are around 200,000 man-days, and, the surveyed
number of labourers being close to this figure, approximately one manday for each labourer has been created in this District. If one looked at
the man-days generated in the entire State, and the figure of the Statewide registered labourers, then by and large the same ratio would be
Counting on development? Politicians and the rural poor
The problem of trust that we have just described is bound up with the
classic problem of principals and agents, and it is perfectly reasonable to
maintain that the new public administration has addressed itself to this
issue. Indeed, it would be a disservice to the good governance agenda to
maintain otherwise. Proposals for performance-related pay, and for regular information-sharing meetings between senior and junior officials, are
geared in part to reducing the importance of inadequate and sometimes
deliberately blocked information flows between the juniors (the agents)
and the seniors (the principals). What this agenda seems to neglect, however, is that both principals and agents must also have regard for their
relations with key actors in political society. Their calculations can never
be confined to the circuits of government rule. Moreover, as we have
seen (chapters 3 and 4, but see also chapter 6 for extended discussion),
the structuring of political society in eastern India is often at odds with
the descriptions that are put forward in the literatures on participation
and good governance. In our Bidupur field site not a single person had
taken up work under the Employment Assurance Scheme within the village, and yet there were few complaints about the workings of the scheme
there, nor were there in Sahar or Murhu.25 Part of the reason for this has

Some work was obtained by a group of Harijans on another (non-EAS) road scheme
connecting the village school to the western Yadav tola. They were promptly dismissed
when it came time for black-topping the road, and skilled labourers were brought in
from outside.



to do with lack of information about the scheme, as we have reported,

but this is only part of the story. There were also pressures from within
village communities, and from their political representatives, to see the
EAS as a scheme that would create durable assets, and which would bring
development to rural communities. In an important sense, people were
more interested in outcomes than in processes.
These pressures played themselves out in different ways. In Sahar
Block, contrary to our expectations, there was only limited evidence of
political mobilization around the benefits of EAS schemes. Such activities as there were took the form of direct petitioning of senior government
officers through the institution of the Janata Durbar. There were efforts
too by some Musahars in an adjoining village to make sure that someone from that community was appointed as an executing agent of an
EAS scheme.26 But in the poorest tola of the village where we worked
no benefits from major government schemes had been received over the
course of twenty years, despite this tola (with its high population of Scheduled Castes) being a stronghold of the sitting Member of the Legislative
Assembly (MLA), himself a member of the Communist Party of India
(Marxist-Leninist). In this case a lack of mobilization did spring from
a lack of information, as well as from reluctance in some quarters to
engage with the state. The local MLA was unfamiliar with the demand-led
provisions of the EAS guidelines, and had ceded responsibilities for the
selection of schemes to District-level officers. They in turn ensured that
every panchayat in Sahar Block had received some EAS funds by 1999,
even if nearly half of all the tolas within the villages of these panchayats
(including in the locality where we worked) received no funds at all (see
figure 5.1). As we intimated before, the concern of the District-level staff
was to ensure that development was visible in the field, and, in Sahar
especially, to make sure that development did not stoke up political tensions or rivalries.27 In practice, though, at least in the panchayat where
we mainly worked, there were almost no efforts to ensure that schemes
were placed in the poorest tolas, or to inform poor villagers of their right
to demand work.


We do not wish to underestimate the significance of this challenge. The fact that the
Musahars successfully challenged the power of the Bhumihars in an open village meeting
(see chapter 3), in the process mobilizing their women in an effort to secure the election
of a managing agent, is remarkable, and is testimony to the concerted strength of CPIML activities in the area. It also suggests a deepening of civil society in the locality (see
also chapter 8).
Here, as elsewhere, District officials were wary of schemes that demanded the active
participation of groups of villagers (or village factions, as they might be described).
Many officials preferred to work with and through established village leaders or local
politicians, the better, they said, to get the job of development done.


The everyday state and society

Resource Distribution by Panchayat















Rs. in thousands



Resource Distribution by Village




% of amount
% of villages


Figure 5.1 EAS spending by panchayat and village, Sahar Block,

Bhojpur District, 19967 19989

A lack of awareness of the demand-led nature of the EAS was also

apparent in Murhu Block (Ranchi District), but in several of the panchayats in this Block the Mukhiya (elected leader) was active in pressing
for a flow of funds from government. In the panchayat where we worked
(see figure 5.2), the Mukhiya was an especially strong representative of
his community, and most of the (mainly tribal) villagers looked to him
to access the state on their behalf. The Mukhiya had been responsible
for getting EAS cards for a small number of villagers (admittedly only
two from our sample population), and he worked closely with the Panchayat Sewak to draw up an annual list of possible schemes. One of his
close associates told us that: Panchayat Sewak comes and tells us that we
should list out important schemes: check-dams, schools, roads, sitting
platform, community buildings, drainage and so on. He also tells not
to list more than 56 schemes, sometimes more when the panchayat is
big. This is done after a few of us collect and deliberate [the Mukhiya
and his inner circle], but usually we give 1015 schemes. Ultimately, 24



Resource Distribution by Panchayat

Rs. in thousands




Resource Distribution by Village





% of amount

% of villages


Figure 5.2 EAS spending by panchayat and village, Murhu Block,

Ranchi District, 19934 19989

schemes come to the panchayat, and, he might have added, the Mukhiya
largely takes responsibility for placing them and for carrying out the work.
Interestingly, the only opposition to the Mukhiya came from a group of
unemployed youths who wanted to control some of these schemes. In this
case, though, and in several other cases which came to our attention, no
argument was raised against pucca development projects. To the contrary,
the youths wanted a check-dam to be built in their part of the village. Like
the Mukhiya, they were happy to define the EAS in terms of the creation
of visible and durable assets. If some work came from the scheme, that
was well and good. But work was not always the main concern: that could
sometimes be had in the brickfields of Gorakhpur, or at building sites in
Calcutta. Even the poorest villagers understood development (vikas) in
conventional terms, and welcomed tangible signs of it.
A commitment to development was still more apparent in Bidupur
Block, Vaishali District, but in this case, unusually, the leading local


The everyday state and society

politician was extremely knowledgeable about all government schemes,

and had partly taken control of them in his constituency. Unlike his counterparts in Murhu and Sahar Blocks, the de facto MLA of this area (and
de jure Member of the Legislative Council (MLC)), BB, was well aware of
the demand-led provisions of the Employment Assurance Scheme, and
he knew the details of those guidelines that specify the proper balance of
unskilled labour and materials in different schemes. Indeed, he provided
us with a critique of those guidelines, arguing that it would be absurd
to make afforestation a priority in Bidupur, given its location next to the
Ganges: what was needed was all-weather roads, and if their construction
required the importation of skilled labour, so be it.28
This attitude to employment generation is surely linked to the fact
that BB, and RR (at the time the MLA for the Hajipur Assembly Constituency, into which six panchayats of Bidupur fall), are both members of
the ruling Rashtriya Janata Dal, a political party that caters strongly to the
interests of Yadav farmers. Those most in need of employment, including
the Dusadhs and Paswans, are mainly supporters of the Dalit leader, Ram
Vilas Paswan, and his acolytes.29 But if caste and class antagonisms help
us to understand why BB and his supporters are in key respects insensitive
to employment issues, they do not gainsay the fact that his preference for
using EAS funds to build all-weather roads is widely supported in the
Block, even though it reduces the (still considerable) scope for skimming
by his favoured contractors. Indeed, in terms of the more general arguments of this book, what is truly significant is the extent to which BB, as
a leading member of vernacular society in Kavirajs terms, buys into the
discourse of development as modernization, and works closely with his
counterparts in elite society to bring this about.
We received information on these transactions from BB himself, and
from an ex-District Magistrate of Vaishali. The ex-DM told us that during
his long tenure in Vaishali he adopted a quota system wherein local MPs
and MLAs were asked to recommend and decide upon the schemes that
would operate in their constituencies. His formula was simple, if quite
at odds with the instructions he received from Patna or New Delhi. All
programme resources that came to his District were to be divided up on


The instructions direct that one should employ resources on various sectors. One of the
sectors is forestry. If we implemented the scheme in [neighbouring] Raghopur Block, all
the plantations in a year would get washed off next year due to erosion that Ganges causes
in this area! How insensible to impose such restrictions from above without appreciating
the ground conditions. Interview with BB.
The Musahars are less inclined to be strong supporters of Ram Vilas Paswan. Given
that most Yadavs will not work as unskilled labourers, it follows that members of the
Scheduled Castes are gaining work from BBs schemes, even if those schemes are not
designed with employment provision as their main aim.



Resource Distribution by Panchayat

















Daud Nagar


Resource Distribution by Village



Rs. in thousands



% of villages
% of amount


Figure 5.3 EAS spending by panchayat and village, Bidupur Block,

Vaishali District, 19967 19989

a 70:30 basis between the MLAs and the MP. If a Block fell between
two constituencies, the resources would be divided in proportion to the
number of panchayats in each constituency. The representatives could
then use the money as they saw fit. In the case of Bidupur Block, the
records show that resources have been distributed almost exactly on a
19:6 basis, reflecting the fact that BB represents nineteen of the twentyfive panchayats. They also show that BB has been active in making sure
that schemes are set up or are about to be set up (see figure 5.3) in
all of the panchayats where he has supporters.30 From his perspective,
the building of all-weather roads not only makes sense in developmental
terms, but also ensures that resources will quite literally follow a road map

In the course of one conversation, BB produced a copy of the release letter of the second
instalment of funds from the Government of India, as sent to the DM of Vaishali for
19989, and said: Our people in Bidupur are patient. I tell them that when the next
instalment would come, I would get the schemes in their area too [showing the letter].
Now the second instalment having been received, I would cover other areas as well. This
is why you find only limited schemes in few areas until now.


The everyday state and society

through his constituency, in the process dampening down inter-village

rivalries. Naturally, his key supporters become the executing agents of
these schemes. The DM, meanwhile, having ceded power to the MLAs
in recognition of their local dominance a labour contractor exaggerated
the power of the Mukhiyas when he told us that: JRY is Mukhiyas scheme
and EAS is BBs scheme still made an effort to ensure that some funds
went to those tolas or villages which were ignored by BB or RR. For those
areas, and also areas that are of interest to important political leaders from
opposition, I used the interest money accrued to the development funds.
If any MLA or ruling party leader questioned as to why I was sanctioning
projects in areas of their political rivals, I would say that that was being
done out of the discretionary DMs funds and did not encroach upon
their quotas, hence they could not have any grievance nor any locus standi
to object to this.
In Bidupur, then, as in Sahar and Murhu Blocks for quite different
reasons, a combination of social forces came together to redefine the aims
and objectives of what was intended by New Delhi to be an employment
assurance scheme. The fact that black-topped roads were built in Bidupur
also tells us something of importance about the good governance agenda.
It is true that the EAS works poorly in Bidupur from the point of view
of providing employment to local people holding EAS cards. It is also
true, as we showed in the last chapter, that ordinary people have had
very little say about the choice and running of EAS schemes. Levels of
participation and employment provision were higher in Debra Block,
Midnapore District. Most distressing of all was the fact that the roads
built in Bidupur were mainly intended to benefit the Yadav communities
and the inhabitants of what might be called the main villages in the
Block. By and large, the roads that were built did not make their way
into the tolas of the Musahars or even the Paswans. And yet, importantly,
tangible assets were built in Bidupur and a good deal of employment
was provided to labouring households outside the formal structures of
the EAS. Most of all, perhaps, this outcome was considered a reasonable
one by most of the parties that were active in Bidupurs political society
(with the possible exception of the Paswans).31
Many poor non-SC households were pleased to see evidence of government spending in their villages. Philip Oldenburg has reported that
in a village near Varanasi, Uttar Pradesh, perhaps 2530 per cent of JRY
funds were eaten and didnt make it to the villages. A number of villagers expressed their concern about corruption, but for the most part

For elaboration, see chapter 6.



what matters to villagers is perhaps how much reaches them, not how
much is siphoned off, which sum they are unlikely to know or be able to
find out (Oldenburg 2003: 1819). Similar sentiments were expressed
in Bidupur and Murhu. Where the state is seen mainly as a provider of
funds rather than as a collector of taxes, as it is in much of rural India, it is
perhaps understandable that villagers would come to such a conclusion,
and that they would like to see evidence of government spending. Roads,
schools and houses built count for rather more than meetings held or
Many of the public officials charged with running the EAS also knew
this. They were mindful that their own desire to eat the state had to
be tempered by the need to provide physical evidence of government
at work. Officials also worried about the consequences of being seen
to be corrupt. Some proponents of good governance and participatory
development might claim this sighting of the state for their own agendas,
and not without reason.32 Nevertheless, it is clear that some accounts of
the scale of rent-seeking behaviour in India, and in Bihar in particular,
are absurdly high, and we say again that we see no reason to endorse Evan
Osbornes claim that more than 45 per cent of Indias national income
takes the form of government-disposed rents.33 In addition, it needs to be
said that the Mukhiya in our field site in Murhu, and BB in Bidupur, are
providing at least one model of good governance for their supporters and
even for their localities. The outcomes of EAS spending were certainly
more visible in Bidupur and Murhu than in Debra Block, Midnapore
(where closer adherence to the guidelines had led to the proliferation
of kuccha schemes of limited value), and both men worked closely with
District and Block-level officials to diffuse fears about civil disorder. Just
as importantly, both men had provided leadership in a context in which
the official state is more fragile than it is assumed to be in the mainstream
literatures on good governance. They did so, moreover, not so much by


Up to a point. The good governance agenda at times comes close to the view that
there was only bad governance previously, and that unreformed governments were
unconcerned with corruption. This is an exaggeration, to say the least.
See also chapter 1. Osborne estimates that government-disbursed rents of the order of
3545 per cent of Indias national income are channelled through the system of reserved
jobs alone (Osborne 2001: 679). His figures draw on data provided in Mohammed
and Whalley (1984). We do not share his underlying assumption that the reservation of
government and public sector jobs is necessarily evidence of the distribution of government rents. Nor do we agree that this distribution system must promote or reproduce
the inevitable Balkanization of Indian politics (Osborne 2001: 679). For a more subtle
understanding of the ways in which the sectoral and spatial distribution of government
funds (and rents) can help to bind India together, see the outstanding essay by Dasgupta


The everyday state and society

playing the part of elected local delegates, as by playing the role of the
provider, or the traditional patron or dada (elder brother or bossman)
figure who gets things done.
Educating the state
Many of the same strengths and weaknesses of the good governance
agenda are apparent in the fields of education and the law, although
in Bihar they come with fewer reinforcing pressures from political society. The main strengths are as we began to report them in chapter 4.
The government of Bihar has since 1995 been committed in principle
to the formation of Village Education Councils that would be presided
over by a chairperson elected directly by a village open meeting. The
chair is supported by an elected vice-chair, and one of them is meant to
be drawn from an SC community. The committee would be supported
for a while initially, and then periodically, by officials of the Bihar Education Project (BEP). This initiative replaced an earlier notification of
1988 which suffered from the basic problem of the members [of the
VEC] being nominated by the Block Education Officer (BEO) and of the
mukhiya being the president (Kantha and Narain 2003: 134). The state
also initiated a Total Literacy Campaign in Bhojpur District in 1992, and
this later spread across the state. Kantha and Narain report that When
the literacy programme began . . . Madhubani and Madhepura produced
such exciting results that these soon began to be cited as model examples
in Bihar (Kantha and Narain 2003: 131). They go on to note, however,
that the enthusiasm generated soon declined and the second stage of the
campaign could not take off in these Districts (Kantha and Narain 2003:
If there was an exception to the law of diminishing returns it was in
Dumka, a District dominated by the Santhal and Mal Paharia (ST) communities that is now in Jharkhand. There is evidence to suggest that
significant improvements in male and female literacy rates have been
sustained in Dumka.34 At the same time, however, it is widely agreed
that the main reason for this success story is that the Collector himself
spearheaded the literacy campaign, and this poses problems for the good
governance agenda. If the role of the Collector is acclaimed as a sign of
committed officialdom this comes close to saying that good governance
will be achieved where there are good governors, a proposition that is

Srivastava (1998).



true but not very helpful.35 The same argument could be made of good
teachers or good BDOs, with just as little gain.
Many proponents of good governance will recognize this danger, and
will want to pay close attention to failing Blocks and Districts, and to
those many BDOs, BEOs and BEP officials who do not show a proper
commitment to their jobs and those they serve. But what they would find
would not always be of great cheer. As we reported in chapter 4, very
few officials of the BEP carry out their duties as they are meant to. It is
common for mid-ranking officials and the village catalysts (the utpreraks)
to spend less than a day in a village. VECs are often set up on an ad hoc
basis, on the basis of quick conversations with those village leaders who
can be met in an afternoon. Return visits are a rarity. But we should be
careful about accounting for this behaviour in terms of the laziness of
some utpreraks, or even with reference to the lack of a well-functioning
system of performance-related pay. Again, there is a lurking danger of
tautology, or the assumption that if the utpreraks are well motivated, well
informed and well paid the formation of active VECs will be more easily
achieved. In important respects this will be true, but it begs important
questions about where the money will come from, and about the utpreraks
own understanding of their subject positions and their responsibilities.
Our conversations with BEP officials showed they were put under considerable pressure by the target-driven nature of the new governance
agenda. Measurable outputs had to be shown in short order. In Bhojpur
District, about 200 utpreraks were given the responsibility for forming
over 2,000 VECs in the years 19978 and 19989. This might not seem
like a large number, but the VECs had to be formed in 1,794 villages in a
District where many roads are impassable in the rainy season, and which
can be dangerous and slow at the best of times. Getting petrol for scooters
and motor-bikes can also be a problem in Blocks like Sahar, where the
infrastructures of rule and development are often in retreat. And then
there is the question of how the petrol is to be paid for, and whether
the utprerak (or another quasi-governmental official) will be reimbursed.
Some rent-seeking public officials use their own resources to carry out
the official business of the state. Officials must also have regard for public safety issues and for their standing in the community. In the case of
lower-level officials this often translates into acts of obeisance, not least
to members of village elites and the schoolteachers they are meant to be

We would also need to know more about how the Collector fired up his staff, and took
steps to nurture womens organizations, and what the opportunity costs were of this
commitment to education.


The everyday state and society

looking after (or over). Higher-level officials more often express a desire
to return to an urban area before nightfall. Ever mindful of bandits and
extremists, they also find it difficult to meet the expectations of their
own families in the countryside. Better rates of pay, although welcome,
would not address these broader issues of state capacity and the provision
of rural infrastructure.36
Proponents of good governance are on stronger ground when they
point to the power of vested interests in the education sector, and on the
need to rethink the roles of some members of the teaching profession. We
noted in chapter 4 that the teaching unions are especially strong in West
Bengal, where teachers are part of the core support base of the CPI-M.
(Most teachers we met were members of the All-Bengal Primary Teachers Association (ABPTA), which is affiliated to the CPI-M.) Even where
school attendance committees function well, as they seem to in Debra
Block, and where there is a well-established system of school inspection,
it can be difficult for parents (and Sub-Inspectors [SIs]) to remove or
even rebuke well-connected teachers. We also saw that VECs in Bihar
failed to sustain interest once a teacher was able to exert his or her power
over committee members, or where the government refused to transfer a
teacher or pay for the addition of another. This had significant repercussions for how poorer people came to see the state, for the teacher was
often near the top of their list of known state functionaries.
Nevertheless, the question of how to bring teachers to book is a thorny
one. If VECs might in time be part of the answer it is clear that not
much progress will be made until politicians are brought more squarely
into the picture. But here too there are problems, and not simply those
that derive from the mistrust of politicians that is a feature of the new
public administration. Politicians in Bihar and Jharkhand are not much
interested in education. They are occasionally interested in the funding
of new schools, for they can then pose as providers of resources. For the
most part, though, the politicians we spoke to were agreed that there was
neither money nor votes in education. The flow of funds was not sufficient
for them to consider it a significant source of patronage. Nor did it create
many opportunities to provide contracts for dalaals who would turn out
the vote in their favour. The idea that individual men and women would

We should emphasize that regular staff members of the BEP had few problems with
payments, reimbursements and getting access to fuel. The BEP was well-funded by the
World Bank and staff members could fill up their bikes or scooters with petrol at the
District headquarters, Arrah. They could also access small amounts of petrol in jars in
Sahar, if the necessity arose. The problems we have described here are more serious for
staff such as Junior Engineers and panchayat sewaks, who can often be seen riding on the
back of the motorbikes of important villagers or contractors out of necessity.



Table 5.2. Schools with only one teacher, selected

Districts of Bihar, 19961999 (percentage)








Source: BEP State Office, Patna.

cast their votes on an issue like education also struck them as mistaken,
and out of touch with local realities. Votes were not cast on this basis,
and electors would not give up the security of a bloc the protection of
a social group in DaMattas terms for a more abstract right to press the
case for better schooling of their children. It is not surprising, then, that
officials of the Bihar Education Project came under few pressures from
politicians in a Block like Bidupur. The laziness that we observed in some
BEP officials can largely be explained in terms of the lack of competition
we observed in political society around the issue of education.
The lack of proper funding of education in all three states also impacts
on parental perceptions of the costs and benefits of schooling, and it is
here perhaps that we see the most glaring weakness of the new public
administration its suggestion, implicit or explicit, that money is not
the central issue, and that what matters is the quality of governance in a
cultural or institutional sense. Our work suggests that a more commonsense view is in order. Parents that we spoke to were agreed that public education for their children would be more attractive to the extent
that several teachers were provided in each school. The quality of the
school building(s) should also be adequate, at the very least. BEP data
for the research Districts suggest, however (see table 5.2), that between
18.2 per cent and 30.2 per cent of schools had only one teacher in 19989,
and less than 5 per cent of schools provided a separate toilet for girls (table
5.3).37 The condition of many of the schools that we visited was rudimentary at best, and in some cases called to mind the semi-deserted Block
development offices we had observed in Lapong Block, Ranchi District,


We say suggest because the BEPs data are not always consistent on a year-by-year
basis: compare the figures for presence/absence of girls toilets in Bhojpur from 96/7 to
97/8 in table 5.3. There is no doubt that such facilities are underprovided. At the same
time, however, the ability of the state properly to sight itself must be open to question.
This too might relate to questions of state capacity, or to the monitoring of the officials
charged with data collection and presentation.


The everyday state and society

Table 5.3. Schools with a toilet for girls, selected

Districts of Bihar, 19961999 (percentage)








Source: BEP State Office, Patna.

Jharkhand (while searching for a Ranchi field site) and in Sahar. Good
governance is all but impossible when the state is stretched this thin.
Matters were much the same in Malda District and no better even
in Midnapore. In Malda just over half of all school buildings could be
described as pucca. The District is hit regularly by floods, and the administration finds it hard to keep up with the major repairs that are often
required. The DPEPs Perspective Plan for Malda, 1999, reported that
430 (of 1,877) schools needed major repairs and a further 941 needed
minor repairs. Only 13 per cent of schools had toilets, and only 14 per
cent had enough rooms to allow the teaching of four classes at the same
time. (According to official data, 40 per cent of schools in Malda District have only one room or building, and 8 per cent have none at all). In
addition (see table 5.4), the teacherpupil ratios in Malda District and
Old Malda Block were reported to be higher than the called-for figure of
1:30. The ratios were better in the two primary schools in our Old Malda
field site, indeed considerably better than in Midnapore, but as table 5.5
shows there were still significant capacity problems in School No. 1 (five
teachers but only two rooms), and very large numbers of children were
being taught in one room.
Money matters, and it is important to recognize the truth of this when
we consider the educational state from the point of view of some of its
employees. Even allowing for the power of the teaching unions in West
Bengal and erstwhile Bihar, and for the legitimate nature of many of the
complaints that are levelled against teachers, it is important to get a sense
of how they the educators see matters. Two points are worth making
here. First, it is clear that most teachers are unsympathetic to the idea of
parent power. Unionized teachers express resentment at the idea that
they should be accountable to parents, and particularly to parents who
are barely educated or from the lower classes.38 Well-educated teachers

In reverse: If we made a suggestion to the teachers, they would laugh at us: interview
with a group of poorer villagers, Old Malda, 1999.






Gram Panchayat

Population per 1991 census. All ratio figures will have worsened by 19989.
Sources: DPEP School Survey, Malda; Midnapore District Profile; Information from SIs.

No. of primary schools

No. of child education centres (Shishu Shikha
Kendras [SSKs])
Primary schools per 10,000 population
Average no. of households per school (incl. SSKs)
Average no. of school-going children per school
No. of primary schoolteachers
Average no. of teachers per primary school
Teacherpupil ratio



Table 5.4. School infrastructure in Midnapore and Malda, 199899










Gram Panchayat


The everyday state and society

Table 5.5. School infrastructure in field site primary schools, Malda and


Field site Field site District Field site Field site District
School 1 School 2 average School 1 School 2 average
Schoolchildren per school 109
No. of rooms
Schoolchildren per room
No. of teachers per school
Teacherpupil ratio






Sources: DPEP School Survey, Malda; Midnapore District Profile. Survey, Malda; Midnapore District Profile.

who hail from towns and cities find it hard to treat poorer people as equals,
and in adivasi areas they are more likely to think of STs as junglees than
as citizens or clients. Educated or not, it would be a mistake to think of
such teachers as rational or disinterested actors in the Weberian sense.
Second, most teachers, and perhaps as many as 90 per cent in our
Districts, look after their families with sources of income from outside
the public school system. They have a business, perhaps, or land, or they
teach on a for money basis. This insulates them from political pressures.
In West Bengal, too, teachers receive a reasonable wage from the state.
Monthly salaries in 1999 were Rs. 3,3506,325 for teachers with junior
basic training.39 Set against this, it was an open secret in the late 1990s
that recruitment into the teaching profession was expensive. This was
true in both states, but in West Bengal, where the cost of a first job ran
from Rs. 10,00050,000 depending on the strength of a persons political
connections, it is also led to a flood of litigation. Disappointed applicants
charged that the selection system operated by the District Primary School
Council (DPSC) gave too much weight (40 per cent) to an interview, as
compared to examination results (60 per cent). This allowed for a lot of
flexibility (corruption) on the part of DPSC officials and their counterparts among local employment exchange personnel. In Midnapore, no
recruitment of teachers could be carried out between 1983 and 1996,
partly as a consequence. (Some few appointments were made on compassionate grounds.) Recruitment began again in earnest in April 1999,

By comparison, a fully-employed labourer working at the government minimum wage

would not earn more than Rs. 1,200 per month. It is worth noting that the salaries of
teachers are generally paid on time. In Bihar, problems of non- or late-payment of wages
are far more acute in state transportation concerns, or in services like refuse collection.



when 2,600 teachers were taken on, but at the end of that year about
2,500 posts, or 10 per cent of the total number of all teaching positions,
remained unfilled.
In Malda, meanwhile, a senior official of the DPSC pointed to an
almirah full of court cases, and declared that: This is the District of court
cases. He was not wrong, although he might have added that it was the
same in many other Districts. By 2000, a legal system that was overloaded
and open to corruption, and which was poorly regarded by many citizens
in consequence, was contributing significantly to the underperformance
of the education system in both states. The fact that many teachers and
potential teachers saw the state as a supplier of reasonably well-paid and
secure jobs jobs worth going to court for had significant consequences
for the ways that poorer people would see the state in the educational
sector. Very often, they saw it is as remote, perhaps even as foreign, and
simply not worth their while. The schooling of boys and girls suffered
hugely in consequence.
The politics of good and bad governance
The dismal state of school buildings in Bihar and West Bengal can certainly be read as evidence of poor government, but whether this lack of
spending is an indictment of the agendas of good government is much
less clear. Those agendas are remarkably flexible, as we noted at the start
of this chapter, and arguing against good governance in general is about
as helpful (and welcome) as arguing against parenting and chola chai. Proponents of good governance would look at our stories about education in
eastern India and make at least three points in quick response.
First, the rules governing the recruitment of teachers in West Bengal
have changed in recent years. A government circular of 1991 announced
a new method of recruitment that is weighted more to assessed merit
(65 per cent from marks in the final school exam, 20 per cent for teacher
training and 5 per cent for co-curricula activities) than it is to the interview
(10 per cent). It is true that no teacher had been recruited under the
new rules in Malda at the time of our field studies (19992000), but
the new rules were employed in Midnapore in 1999, and by mid-2000
no allegations of corruption had been lodged against the decisions that
had been reached, and no court cases were pending. At the very least,
this would seem to be a step in the direction of better government, or
government that is more rule-based and transparent.
Second, the established powers of the teaching establishment can be
challenged by a process of circumvention. Both the Conservative and
Labour Parties have tried this route in England and Wales, where they


The everyday state and society

have pushed strongly for the recruitment of additional teachers who have
not completed a full course of studies at a teacher training college. Teachers in the private sector are also hired directly from universities or from
other professions. Accreditation is not necessary in this sector in either the
UK or the USA. In Madhya Pradesh, too, steps have recently been taken
in this general direction. Under the Education Guarantee Scheme (EGSMP) introduced by former Chief Minister Digvijay Singh, the state now
has three classes of teacher. About 200,000 teachers are employed by the
states Education Department in 80,000 formal primary schools. They
are paid according to a pay scale that has been agreed with the teachers unions. A further 20,000 schools are staffed by shiksa karmis under
the terms of the EGS-MP. The first batches of shiksa karmis received
rigorous training and were paid Rs. 1,000 per month while they were
placed on probation in the formal school system. They are more likely
than the formal school teacher to be a local person and are recruited
and posted by the janpad panchayat (Noronha 2003: 107). Finally, there
are the EGS-MP Gurijis; they are nominated by the village community
and the sarpanch to teach fellow villagers for a stipend of Rs. 1,000 per
month, but with no tenure. These jobs have proved attractive to local
youth leaders who might otherwise remain unemployed.40
Third, the importance of careful political management can be demonstrated at the national level. In her account of the contentious politics of
education reform, Merilee Grindle argues that substantial progress has
been made in Latin America despite the odds.41 She agrees that most
politicians are lukewarm about reforming the educational system in their
countries, and that powerful forces are ranged in support of the status
quo. Grindles focus, however, is on the efforts of the counter-forces
which are pressing for educational reform. She seeks to establish both
the means by which they have set national policy agendas, and the steps
that were taken to defuse opposition to these initiatives. Her wider argument is that change is possible, and (or but) that it comes in small and
unexpected ways. Progress is rarely continuous, but when it is achieved it
is because skilled reformers have been able to exploit general sentiments
about efficiency and transparency to effect specific policy changes. These
policy changes in turn begin to change the terms under which succeeding
debates about education (or health-care, or poverty alleviation schemes)
are discussed.


In addition, while the Gurujis posts came without tenure, the hope and perhaps expectation was that they would in time be absorbed into the government cadre and made
permanent. This is one reason, Srivastava reports (2003: 25), why many gurujis performed their jobs with more dedication that at first might have been expected.
Grindle (2004).



These specific rejoinders also point to a more important set of claims

that must be weighed when we consider the agendas of good governance.
These claims have to do with politics, but in a much broader sense than
is intended by some critics of good governance. We stand beside these
critics in important respects. We have tried in this chapter to show that
money really does matter. It will be difficult to improve education or
health-care in eastern India when financial issues are so easily discounted.
We have already mentioned the lamentable state of many Block offices in
our Districts, and we shall have more to say about the late or non-payment
of public servants in chapter 7. We might add that Sub-Inspectors of
Primary Schools in West Bengal currently have responsibilities for circles
of between 60 and 110 schools, and that their job description requires
them to make at least ten inspections per month. (In other words, they
are required to visit each school once or twice a year, on average.) The SIs
we spoke to, however, said they could only manage five or six inspections
per month because of serious capacity problems. There was a shortage
of clerical staff and vehicles, and SIs could only borrow the DPSC jeep
in Malda on an occasional basis.
This is small beer, perhaps, but it is telling nonetheless. Critics are
right to suggest that an agenda of administrative reform does not always
sit easily with a strong bias away from the state and towards the market.
There are doubtless efficiency gains to be made in government, and special interests to be confronted, but in eastern India the absence of a wellresourced public sector is what is most often noticeable. In some cases,
the resources of the state have been looted by private interests, much as
Harriss-White describes in her account of the formation of shadow states
in Tamil Nadu. Government can then barely function, and ordinary men
and women see the state mainly as an absence, or perhaps as a phantasm.
And when they see government officials they often see men (or women)
who are very far from being the disinterested public servants in which
the new public administration invests so heavily. In the lower reaches of
government, especially, the pressures that are brought to bear on officials
come far more from family and community, and from brute economic circumstance, than they do from abstract models of the law or due process.
Moreover, the fact that some schoolteachers in West Bengal, including
some members of the CPI-M, are sub-contracting their jobs for Rs. 2,000
per month, suggests that money alone is not the issue, and that the sort
of calling that Tendler has described for health workers in Cerea State,
Brazil, remains a long way off.42

We do not dispute that performance-related pay might help here, along with a more
clearly defined career path. At present that path runs directly from teacher to headteacher.


The everyday state and society

We agree, then, with Leftwich and Jenkins that the agendas of good
governance should be regarded critically, and at times even with suspicion. It is certainly possible for these agendas to deflect attention from
what many on the Left would insist are the real issues: the need to build
a committed left-of-centre political party, for example, or to campaign for
a redistribution of assets towards the poor (including land), or to ensure
that land can be owned and controlled by women.43 We are also in agreement with Leftwich when he suggests that the good governance agenda is
inclined to substitute wishful thinking for the hard task of building political forces capable of putting into place the institutional reforms that it
But we also detect a measure of wishful thinking on the part of the
critics of good governance. This romanticism is to be found, in part,
in the suggestion that the agendas of the new public administration are
depoliticizing, when it would be more accurate to say that they have
changed the terms of debate about government reform. In other words
(and Leftwich would surely recognize this), the movement from politics to
the economy and public administration is itself a political act. We should
also recognize that a shift in the direction of governmentality, for all its
evident flaws, is not simply an act of partisanship or class warfare, as some
radical critics maintain. Corbridge and Harriss have described the process
of economic reform in India as an elite revolt, and we agree with this
assessment.45 The reforms in the 1990s were clearly meant to advance
the private interests of some of Indias urban and industrial elites. But
when it comes to questions of accountability, or decentralization, or even
participation, the lines of intent are less easy to discern. An intelligent
defence of good governance is that it is meant to widen those spaces of
empowerment that can be found in a world of the second-best. In other
words, in a state like Bihar, where land reform is not on the cards, and
where its advocates have not shown how (and with what consequences
and at what cost) political forces might be built up to press for it, it illbehoves some critics of institutional reform to talk about political navety.
To put it more positively, one possible defence of some parts of the good
governance agenda is that it refuses a Jacobin conception of politics which
depends upon an idea of perfectibility, or an ideal outside. It recognizes



We fully support these normative claims, and we would underscore the importance of
providing poorer men and women with land. Studies clearly show that the ownership
of land translates into greater employment opportunities, an increased ability to borrow
money, greater self-respect, and less vulnerability: see Herring (1983); Singh (1990).
[T]he [World] Banks analysis is nave . . . because it entirely ignores that good governance is not simply available on order, but requires a particular kind of politics both to
institute and sustain it (Leftwich 1993: 607).
Corbridge and Harriss (2000: chapter 7).



that the world is imperfect, and yet still open to contestation, and advances
a politics of the possible which is expected to broaden the canvas on which
a more committed pro-poor politics can be played out.
We shall come back to this general argument in chapter 8. Before we
get embroiled in an assessment of the morality of politics, however, we
need to return once again to Bihar and West Bengal, but this time with
a view to saying something more concrete about the nature of political
society in our research Districts, and especially in Midnapore, Malda and
(again) Vaishali. If the new public administration is not wholly opposed
to politics, it is largely silent on the matter of political parties and political
fixers. As we shall see, though, in chapters 6 and 7, there is much to be said
for the view that it is in political society, and not in civil society, that many
of the most important questions relating to participation and government
will be decided and fought out. For poorer men and women, especially,
and indeed for many government employees (as we saw in Bidupur), the
state is sighted in large part through the lens of political society. We need
to pay close attention to those sightings. We also need to review some
of the incentive regimes that structure real democracies, for these often
cut against the grain of the good governance agenda. Competitive politics
can be an expensive business and it is by no means clear that it must lead
in the direction of a decline in corruption. Foucault reminds us to be
humble in the face of the past, but we should also have regard for the
messiness of the world around us.

Political society

Some of the most pressing debates in development studies have concerned the relative merits of states and markets, or the means by which
markets might be regulated by a range of public institutions from the local
to the global scale.1 These debates have taken shape, most famously, in
the contrasting cases of sub-Saharan Africa and East Asia, and they have
an obvious and continuing relevance in countries as diverse as Brazil,
Nigeria, India and China. Yet if debate on these issues continues to be
fierce, there appears to be general agreement that strong states or free
markets need to be kept in check by vibrant civil societies. Indeed, it is a
common proposition in development studies that this hazy zone of freedom between the family and the state is a source of unparalleled strength
for ordinary men and women, and a source of development itself and even
economic growth.2
Robert Putman has made this claim as strongly as anyone.3 His suggestion that economic growth is promoted by a prior build-up of social
capital of peoples engagements with a dense network of civic associations has become a staple of World Bank thinking since the mid1990s.4 Even where the causal propositions of Putnam are refused, it is
clear that the virtues of civil society are widely admired. Arturo Escobar
looks to civil society as a breeding ground for oppositional movements
and experiments. It functions for him, and perhaps also for Ashis Nandy
in India, as a potential zone of resistance to the dehumanizing claims of
developmentalism.5 Meanwhile, the claims of participatory development


For illustrative contributions, see Beenstock (1984), Wade (1990) and World Bank
(1993). The best recent review is by Akyuz et al. (1998).
Francis (2002) offers a sensible overview.
Putnams name appears in the singular on the front cover of Making Democracy Work
(1993), but authorship is listed inside the book as Putnam with Leonardo and Nanetti.
See also Putnam (1995). For a critical review, see Tarrow (1996).
See the World Banks Social Capital website; see also Woolcock (1998).
According to Escobar, Development is the last and failed attempt to create the Enlightenment in Asia, Africa and Latin America (1995: 221). What he calls the unmaking of


Political society


and good governance are predicated very precisely on the apparent or

assumed strengths of civil society. Governments are kept honest by the
free and informed association of people in institutions that run all the
way from self-help groups to social movements and decentralized forms
of democracy.
There is evidence to suggest that civil society is deepening in a number of poorer countries, including in India (as we have seen in the past
two chapters).6 Nevertheless, it is a mistake to suppose that civil societies are fully formed in developing countries, or that they can be made
the bedrock of public policy. Partha Chatterjee has argued to the contrary, and we shall want to follow him in some degree in this chapter.
In Chatterjees opinion, the concept of civil society, with its emphasis on equality, individual rights and freedoms, and autonomy from the
state, is poorly placed to capture the realities of social exchange in postIndependence India (or, indeed, in other post-colonial countries).7 A
modernizing and formally democratic state arose there, he says, not in
response to impulses from within civil society, as was perhaps the case in
some Western countries, but rather as the means by which civil society
could be called into existence. In addition, if the Nehruvian project of
modernization can be understood as a project . . . to create a civil society
(Chatterjee 1998: 1011), it is important to recognize that this project
was reshaped by elements in political society. No matter how hard New
Delhi tried to build a developmental state which [sought] to relate to
different sections of the population through the governmental function
of welfare (Chatterjee 1998: 15), and by means of well-crafted and carefully targeted development programmes, all of which were presented in
the neutral language of objective planning, the fact remains that these
programmes were reinvented at the District and Block levels by politicians
and lower-level bureaucrats who did not always share the world-views of
their English-educated metropolitan superiors.
We have seen that this is indeed the case, although not quite in the
directed manner that Chatterjee suggests. But what is perhaps more
important about Chatterjees work is the suggestion that political society
that loose community of recognized political parties and their operatives,
local political brokers and councillors, and perhaps even lower-level public servants who depend upon the grace and favour of politicians bridges
between the government and the public in a manner that often refuses the


the Third World will supposedly be achieved out of hybrid or minority cultural situations [that will promote] other ways of building economies, of dealing with basic needs,
of coming together in social groups (Escobar 1995: 225). See also Nandy (1989, 2003).
For critical reviews, see Beteille (1999), Chandhoke (2003) and Mahajan (1999).
The literature on Africa is instructive here: see Mbembe (2001), Bayart et al. (1999),
Chabol and Daloz (1999), and most especially Mamdani (1996).


The everyday state and society

optimism of civil society models. This community has links upwards and
outwards beyond the narrow concerns of the development projects that
we have focused on so far. It is a community that draws its strength from
the ability to exercise control over events in the locality and to link these
to wider political discourses that emanate from Kolkata, Patna, Delhi and
elsewhere. As we shall see, the concerns of an increasingly national political society the status and role of leaders such as Atul Behari Vajpayee
or Sonia Gandhi, the politics of economic liberalization, or questions of
religion and national identity do make themselves felt in the research
villages.8 These concerns are addressed as subjects in their own right in
the broader literatures on South Asia, but in this chapter they appear
more as a set of resources on which local politics can draw. Our main
interest is in the creation and maintenance of different patterns of rule:
how politicians carve out spaces for themselves within the operations of
the local state, what they do within these spaces, and how this impacts
upon poor peoples experiences of government.
By adopting this focus we want to distance ourselves from the assumption that the presence of political society is always a negative one. We
recognize that elements in political society are fighting for a cut of the
spoils, and that their circumvention of forms of bureaucratic rationality can be socially regressive. Ordinary people accept that politics is often
dirty, and is based on compromises, alliances (sometimes with old enemies), deals and power equations rather than [we would say in addition
to: see chapter 8] principles (Ruud 2001: 134). That is why they judge
politicians in terms of their ability and capacity to get things done (Ruud
2001: 130). It also helps to explain why corruption and violence are often
part of the political process, no matter how much they are sniffed at in
the literatures on civil society.
But if the darker side of political society is an important concern for
the poor, it is not a problem that can be solved by wishing politics away.
Sunil Khilnani maintains that, In a fundamental sense, India does not
merely have politics but is actually constituted by politics (1997: 9),
and there is much to be said for this observation. Development studies
makes a serious mistake, wittingly or otherwise, when it seeks to separate
out already complex questions of development management from the even
messier business of development politics.9 For our part, we prefer to see

Indira Gandhis role in forging this national political society (see also chapter 2) was also
brought home to us in Malda where she remained a forceful presence for many villagers,
and where one woman thought she was still the Prime Minister.
As Moore and Putzel (1999: 5) succinctly note: There is a tradition in aid and development agencies of bringing in political analysis, if at all, in terms of problems and
difficulties. Politics is why desirable things may not happen. Politics is messy. Political
analysis is used only to explain and to try to fix things that have already gone wrong.

Political society


local political society as a set of institutions, actors and cultural norms that
is often constructively engaged in providing links between government
and the public, as well as in brokering deals and forming patterns of
authority that hold these deals in place.
Where we depart from Chatterjee is in the emphasis we place upon the
possibility of political society serving as a medium within which aspects of
civil society can grow and gain support. One reason for linking the study of
political culture to an analysis of political institutions is to rethink what
democratic pro-poor governance might mean in the political societies
of eastern India. It is now widely understood that democracy is always
an incomplete project, much like modernity, and that it is unwise to
think of a singular model of democracy being extended from the West to
the non-West. The traffic is much more even, as Christopher Bayly has
recently confirmed.10 By the same token, the hybrid understandings of
democratic norms and practices that we shall report in eastern India can
be read as attempts to resolve the tensions that exist between politics and
the political, where the former is understood to refer, following Chantal
Mouffe, to the ensemble of practices, discourses and institutions that seek
to establish a sense of social order and organization, and where the latter
refers to the antagonistic dimension that is inherent in human societies
and which is located in the struggles of diverse social groups for power
and resources. As David Slater explains (and we have borrowed from him
here), In this context, politics can be seen as the attempted pacification
of the political, or as the instillation and maintenance of order (Slater
2002: 257, summarizing also Mouffe 1995: 2623), howsoever this is
The decentring of political science that we are proposing is not simply,
then, about recognizing the empirical richness of eastern India, or other
geographical contexts: it rather calls for a re-examination of the categories that we use to imagine politics and the political. We want to see
how technologies of development are reworked by agents in local political society. These agents are not relicts waiting to be made redundant by
waves of reform. Moore and Putzel highlight the skills and competences
that actors in political society can bring to bear on development problems
and the construction of a sense of citizenship.11 These include the ability
to spin or represent anti-poverty measures as being in the wider public
interest. Political leaders can translate the underlying sentiments of externally designed schemes into terms that have local resonance. They also
hold in place alliances of support that will extend beyond direct beneficiaries. Equally, politicians can exercise an exit option from a development
project. Intentional non-involvement will happen when they believe their

Bayly (2004).


Moore and Putzel (1999).


The everyday state and society

energies are better spent enhancing their political capital elsewhere, or

where they are happy to cede power to other agents, or where they perceive that key elements of a development project lie outside the normal
scope of their power. Our field evidence suggests that many smaller development schemes in Vaishali and Malda Districts are sub-contracted by
MLAs and District Council members to lower-level politicians or political
In the next part of the chapter we look at the careers of some of the
political leaders who operate in our field areas. We then consider how
these careers are forged amid networks of power that operate over more
extended spatial scales, and we review how these networks are embedded in long-standing patterns of governance. Throughout our discussion
we highlight some of the clear and significant differences that structure
experiences of rule and development in parts of Bihar and West Bengal,
and most especially in Midnapore and Vaishali Districts. Poorer people
experience a political society in Midnapore that is more structured and in
some respects more civil than would commonly be experienced by their
counterparts in north Bihar. At the same time, however, we draw attention to the ways in which the spectacles of electoral politics in both states
can be linked to episodes of violence, or the production of an unrestrained
political. The costs of more civil forms of engagement are greater than
many models of good governance allow. We also show how poorer people
resist the disruptions they associate with electoral politics in Bihar, just
as some poorer people in Midnapore contest the models of representation
that are provided for them by members of the CPI-M, the political party
that fills much of political society in that District. As ever, their sightings
of the state are messier than the sightings that might be imposed upon
them by experts, government workers and party activists. Their engagements with political society are partly informed by their understandings
of civil society and the idea of what Thomas Blom Hansen has called the
sublime state.12
The constitution of political society: fixers and leaders
The 1990s saw not only the rise of militant Hinduism in India but also
a steady leaching of power to the backward classes. We shall have more
to say on this in chapter 7 when we discuss the political careers of Laloo
Yadav and Mayawati and their accounts of how the state should be made
to work. But how is this leaching of power playing out at the grassroots?
Who is being recruited into political society, and what are the implications

Hansen (2001: 35).

Political society


of this second democratic upsurge for structures of rule in the Indian

We reported in chapter 3 that the lowest-level entrants to political
society are the political fixers or pyraveekars. The figure of the pyraveekar
appears in the work of Ram Reddy and Haragopal almost entirely in
negative terms: he acts as a middleman between the government and
the people, and benefits enormously from the opportunities for largesse
provided by the growth of the developmental state.14 He also exploits
his ability to monopolize information that is not properly distributed by
agents of the official state. While the pyraveekar with one hand applies
pressure, pushes the files, lubricates the process, and extracts the benefits
from the system, with the other hand he passes on incorrect information,
misleads the target groups, and makes a private fortune. It is the negative
and exploitative dimension of the institution that speaks against it and
calls for its elimination through appropriate measures (Ram Reddy and
Haragopal, 1985: 1161).
This reading of the pyraveekar will resonate with anyone who has experienced the effects of state scarcity, but it is also prone to exaggeration.
Parrys careful ethnography of corruption suggests that brokers play only
a small role in controlling jobs and contracts in the Bhilai region of
Chhattisgarh.15 In any case, it is misleading to expect that development
interventions will run smoothly in the absence of the pyraveekar and his
appetite for extracted benefits. Inconsistencies are bound to open up
between the vision and the reality of state interventions and this creates a space for actors other than bureaucrats to provide a service to local
The presence of explicitly political intermediaries within a village is not
a new feature in Bihar or West Bengal. We saw in chapter 5 how BB stands
atop the comparatively thick political society of Bidupur Block, Vaishali
District, and we shall have more to say about the Rashtriya Janata Dal
in chapter 7. Needless to say, many of the pyraveekars who roam around
Bidupur social workers, as they like to describe themselves are linked
to BB or his leading rivals, and we shall come back to their activities later
in the chapter. In Bhojpur District, meanwhile, political society has been
deeply fragmented for several decades. It is only recently that some poorer
communities have begun to re-engage the state in its income support

The phrase is Yogendra Yadavs (1996).

The gendered language is deliberate and reflects local realities.
Parry (1999) also notes that the suggestion that corruption is universal and inevitable
has important power effects in itself, not least in playing up the importance of brokers,
and, as a result, increasing demand for their services. See also Oldenburg (1987) for a
nuanced account of middlemen in Third World corruption.


The everyday state and society

and empowerment functions. These engagements continue to be shallow

and episodic, and in Sahar Block they are generally brokered by members
of the CPI-ML. The production of political workers or dalaals who are
acceptable to members of the Forward and Scheduled Castes is some
way off, and progress in this direction is interrupted by periodic bouts
of violence between cadres of the CPI-ML and landlord armies like the
Ranvir Sena. Political society is also rather thin in Ranchi District, where
many adivasis expect to be connected to the state by their Mukhiyas (as in
our field locality) or by figures like Simon Minz, the Parha Raja of Bero
Block, whose forest protection activities we reported in chapter 4. The
Jharkhand Party and its offshoots were a presence here from the 1950s
until the formation of the state of Jharkhand in November 2000, although
that presence waned in the 1990s.16 Some villagers will also come into
contact with Congress and BJP activists, and members of the Rashtriya
Swayamsevak Sangh (RSS). The RSS was not strong in our research area
in 19992000, but it has sought to engage adivasis around education and
cultural issues in other Blocks of Ranchi District.
In West Bengal, the history of the CPI-Ms spread from the coffee
houses and campuses of Kolkata to the interior villages of the state is also
well documented, and forms an important part of the partys presentation
of itself.17 In this narrative, young leftists inspired by a combination of
Marxism-Leninism, the Maoist revolution in China, and West Bengals
own history of peasant insurgency, embedded themselves within village
societies and provided leadership for the often violent land struggles of
the late 1960s and early 1970s. After the momentum of the mass tenancy registration campaign, Operation Barga, had died away in the early
1980s, the partys power-base remained firmly rooted in the countryside.
The reinstatement of panchayat elections in 1978 also ensured that the
state would help to lead a national trend towards the decentralization of
development initiatives. A generation has now grown up with the activities
of panchayat members and their political rivals being a part of everyday
The panchayat reforms have had important impacts on the composition
of political society in the state. The five-yearly ritual, begun in 1978,
of electing over 60,000 gram panchayat members through intense partybased competition at the village and neighbourhood levels could not fail to
draw more people into the domain of formal politics. Amendments were
also passed ahead of the 1993 and 1997 elections to reserve panchayat
seats and leadership positions for women and members of the Scheduled


For a review, see Corbridge (2002b).


See Nossiter (1988), Kohli (1987).

Political society


Communities. These have ensured that an extended political society is

now formally inclusive of all sections of the states population.
Marvin Davis portrayed local politics in Midnapore in the 1970s as a
contest between rival village elites, but these days have largely passed.18
The spread of panchayat institutions in West Bengal, the active politicization of rival caste groupings in Bihar, and the universal spread of state
developmental work, have all played their parts. These processes of political inclusion are far from complete, of course, and their outcomes will
vary greatly according to the norms and values that hold together the
resulting networks of power. Furthermore, by no means all of the activities of these local political actors are expressly targeted at the state. We
noted in chapter 3 that these individuals play varying roles in village life.
The ability to give voice to the grievances of groups or individuals, and
to intercede on their behalf in the Block Office or in a local dispute, is an
important mark of a persons leadership qualities and a means by which
an actors power is displayed, augmented or contested.
Most of these fixers are small-time operators who will not have a political career in any significant sense. They may enter the public life of a
locality only briefly and then return to obscurity. This is often the fate
of minor figures in West Bengals panchayats, and particularly of women
or members of the Scheduled Communities holding reserved seats. They
are sometimes thrust into the limelight at election time only to find they
are dropped when their seats are returned to open competition. Within
Bihar, too, many small-time brokers are unlikely to progress further than
the panchayat, whatever ambitions they might hold to the contrary. Their
contacts with key figures in the Block Offices, and their connections to
higher-level politicians, might earn them a reasonable living and give
them standing in their own community, but in most cases this is as far
as it goes. They see the state episodically and through the narrow lens of
the Block Office or the office/residence of an MLA.
To progress further up the political hierarchy requires time and energy
and a degree of cultural capital. Fixers at this level need to be literate.
They need to have a reasonable knowledge of the local political landscape
and its rules, and a sense of how government works and is meant to work.
Success in political society also depends on a persons ability to perform
his (or her) power appropriately, and to acquire a reputation for getting
things done. In normal times contractors in Vaishali District are required
to function as political go-betweens who show their worth by the way
they manage government contracts. Work needs to be organized for at
least some labouring families, and the appropriate cuts and commissions

Davis (1983).


The everyday state and society

must be passed on to officials and politicians with minimum fuss. These

activities afford some contractors a reasonable living and a reputation
for being so-and-sos man in the locality. They also involve them in a
series of obligations. These favours are called in at election times when
contractors are expected to deliver votes to the MLAs and MPs who
helped embed them in political society. Elections engage contractors in
a range of work, from the spreading of rumours about rival candidates to
the organization of rallies for visiting dignitaries. Some will also engage in
acts of booth capture, as we report later on. In appendix 2 we comment
on the 1999 general election in Hajipur Constituency, Bihar: in the runup to voting, contractors in the area had to work night and day. They
also had to lay out significant sums of money for the hire of vehicles and
equipment, and to supply meals and hospitality for their political bosses.
Performing these tasks in an effective manner can lead to progression up
the political ladder, but it also comes with a price and this is something
else we shall come back to.
In Midnapore District, career progression for CPI-M workers is
through long apprenticeship within the party. This is more hidden from
public view than the careers of contractors in Vaishali, which are widely
and candidly discussed. The various stages of involvement with the
CPI-M, from joining one of its mass fronts through to the acquisition
of full party membership, are meant to ensure the proper ideological
education of initiates.19 In practice, however, they are also used to forge
increasing levels of trust between leaders and aspiring lieutenants, and to
act as filters between those potentially less desirable elements drafted in
to the party for electoral purposes and the holders of real power within
the organization.20
Progression in this set-up is also bound up with the internalization
of rules and values that are not simply derived from Marxism-Leninism.
Ruud has argued that to rise within the CPI-M in Bardhaman District it is
important for party workers to take on key cultural markers of austerity,
cleanliness and restraint (1999: 269), which themselves relate to the
CPI-Ms own bhadralok roots and longstanding efforts at social reform.
Similar ideals appeared to hold true in Midnapore, where workers were
at pains to demonstrate their hard work, commitment to the poor, and

The West Bengal Krishak Sabha (Peasant Union) is by far the most important of these
in the countryside, in terms of its numbers and political influence.
For a detailed description of the CPI-Ms election machine, see Chatterjee (1997b). In
rural areas, especially, there is often a need to engage influential and effective people from
outside the party (Chatterjee 1997b: 144) in the booth committees that are specially
constituted in the run-up to a vote. People of influence are often respected community
figures: as a later section will make clear, they can also include men who are able to
organize acts of violence.

Political society


simplicity.21 In Malda District, in contrast, these attributes were less

important. Both the CPI-M and the opposition parties have a weak institutionalized presence in Malda, and political progression depends far
more on an individuals reputation for efficacy or even for violence, the
two being linked in many cases.
It is important to note that grassroots CPI-M workers in Midnapore are
often viewed with a degree of warmth by their poorer supporters. They
are people who can be turned to in a crisis, or simply people who will
understand the problems of ordinary families. In Ranchi District, too,
the Mukhiya of our Murhu panchayat was applauded for his ability to
access a series of government schemes for the betterment of the village.
Particularly among the poorest Lohra households, these achievements
were seen as evidence of the Mukhiyas power, and not as the independent
outcomes of an impersonal, if developmental, state.
The position of the local political worker is also viewed with a measure
of ambivalence or even disdain. Ruud argues that this disdain stems from
a series of cultural separations between politics as artha, or the pursuit
of worldly interest, and the assumed moral order (dharma) that should
dominate within the home or community. He claims that in contemporary eastern India politics is perceived as a shamefaced but very real
field of activities with its own rules and logic, which are normally, albeit
not invariably, in conflict with moral ideals governing social conduct
(Ruud 2001: 133).22 As a result, popular engagement with politics and
local politicians is characterized by cynicism coupled with participation
(Ruud 2001: 116).
It is possible that Ruud overplays the separation of artha and dharma
when he suggests that a politicians behaviour is constrained not by moral
considerations, but mainly by the right circumstances and by power,
clout, influence and contacts (Ruud 2001: 134). Most villagers do not
see politics as being part of the moral high ground, but this doesnt mean
they give carte blanche to their representatives. Local politicians draw tactically on a range of ideas about moral duty to justify their behaviour, and
the mobilization of a discourse about corruption is, of course, as Gupta
reminds us, one important means by which the idea of a sublime state is
communicated to a wider public.

Bhadralok is often translated as gentlemen, and refers to an upper-middle-class urbane

or educated elite.
In doing so, Ruud echoes the work of Marvin Davis, who makes the distinction in these
conflicts between gramer kaj (lit.: village work), or local politics aiming to uphold or
reinstate the traditional order of dharma, and sorkari kaj (lit.: government work), or
local politics based around the language of the modern state, including ideas about the
equality of citizens. Ruud would argue that it is the pursuit of artha, rather than highminded Nehruvian ideals, that most Bengalis would associate with sorkari kaj today.


The everyday state and society

Nevertheless, Ruuds account of dirty politics in Bengal is useful in

uncovering the hidden transcripts of the dealings of lay villagers with
politicians, and the languages within which these interactions are conducted.23 He notes that villagers can undermine the official performance
of public meetings through misbehaviour or disinterest. They also try to
portray themselves as wily and strategic players when they ask politicians
for help. Villagers frequently use fictitious kinship terms to describe their
patrons, and in doing so they try to tie their relationships into binding
ideas of moral duty and order.24 The description of contractors/political
agents as chhoto bhai neta encapsulates both this ambivalence towards
politics, and the attempt to tame or capture it within local frames of reference. As little brother leaders, contractors are tolerated as mediators
between the village and outside sources of help, but the term is also a public statement of the perceived limits to their abilities. Villagers are in no
doubt that the little brothers are dependent on their own big brothers
and that real power lies outside the locality. These political dadas are
in turn seen as more distant and more dangerous figures, which is one
reason why villagers use intermediaries to intercede on their behalf.
To sum up: political fixers are important in a number of ways in our
field localities. Caste matters in Bihar, but more in terms of a persons
ability to mobilize resources for a named group than in terms of abstract
ideas about moral hierarchy. Political society has diversified and even
democratized from its previously narrow base, and politics has effectively
penetrated the villages of eastern India as a field of activity. The depth
and breadth of what constitutes this field varies across our localities. In
Midnapore it encompasses many aspects of social life. Elsewhere, local
grandees are still respected and politics is more often confined to struggles
for the control of state resources. Despite this variation, politics is well
established as a field of activity in and through which contests over status
and other activities can be fought out. Finally, it is important to note that
political society has become domesticated in the minds of many villagers.
The careers of political activists are linked from the outset to ideas of
reciprocity, obligation and proper conduct which are constructed in part
through vernacular idioms. This brings in to being understandings of
politics that are most definitely not about ideology or policy preferences.
Perhaps we can begin to understand the high degree of participation in
local government affairs in these terms.25 Voting, or showing up to a


The idea of hidden transcripts is taken from Scott (1990).

This contrasts with the rather self-conscious use of the English word comrade when
CPI-M workers address each other in public or as part of official business. Senior party
workers use this term in a similarly fictitious display of equality within what remains a
strictly hierarchical institution.
Especially at election time, as was evident too in the 2001 panchayat elections in Bihar.

Political society


public meeting, is in part a social obligation. It is an important public

performance which, for all its hidden transcripts and backstage denials,
confirms the ongoing ties between lay villagers and the chhoto bhai neta or
party worker, and between both of them and the world of politics beyond
the village.
Engaging the state: networks of power
We can now consider the ways in which political careers are located within
and are partly constitutive of local networks of power. Central to our
concerns here are the ways in which political society engages with the
institutions of the developmental state. The state and its development
programmes are key resources for actors in political society. They are
also important in fuelling the careers of members of political society.
A recent movement towards larger and more decentralized budgets for
anti-poverty programmes can be expected to shape local politics significantly. This will be especially the case in Bihar and Jharkhand, where the
panchayat elections of 2001, the first in nearly twenty-five years, seem to
have led to new or heightened levels of conflict between actors at different
levels in political society.
From the outset we should reiterate that the involvement of politicians
in the operation of these programmes is not wholly parasitic. It is to some
degree a valid expression of their social power, and demonstrations of
control over these resources might be as important to these actors as
crude attempts to line their own pockets.26 Furthermore, the state is not
a passive target. We saw in chapter 5 that some District-level officers in
Bihar had tried to head off attempts to capture the EAS, and had insisted
on projects that could be scrutinized in some degree. In the empirical
analysis that follows we turn again to the role played by the EAS, but this
time in West Bengal. We do so not because we believe that leakage from
the EAS is particularly rampant, but rather because a scheme like the
EAS is meant to check the nexus of power that often exists at the Blocklevel between bureaucrats, politicians, contractors and builders.27 The
contrasts we observed in the schemes outcomes in Malda and Midnapore
are particularly instructive in this regard as they emerged in conditions of

We offer some preliminary comments on the funding of political society later on.
In our research sites, levels of everyday corruption in education, in terms, for example, of
teachers not showing up for work, were more costly than the amounts diverted from more
occasional EAS projects and other poverty-alleviation schemes. This observation tallies
with a recent survey by Transparency International (TI) which suggests that Indians pay
an annual total of Rs. 267 billion (about US$ 6 billion, or US$ 6 per person) in bribes,
and that the health, education and power sectors account for most of this (Times of India,
17 December 2002). Because the TI study only assesses bribery in line departments, a
comparison of programmes implemented by local governments cannot be made.


The everyday state and society

identical local governance where there were no major differences in the

official capacity of the state at the District and Block levels.28
The operation of the EAS in West Bengal was effectively decentralized to the Block-level by making the panchayat samiti (the elected Block
council) the main implementing agency. On the recommendation of West
Bengals Department of Panchayats and Rural Development, District and
Block councils could if they wished devolve the implementation of EAS
projects to the gram panchayats, again with the aim of strengthening local
accountability. The bureaucracy would then facilitate implementation
with the help of its organizational and technical expertise. The Block
Development Officer (BDO) acts as the official executive agent and is
supported by the Block Cashier and Sub-Assistant Engineers, and by
the gram panchayat-level Secretary and Job Assistant. The accountability
structures of the EAS are complex, and in West Bengal they are intended
to engage a range of government employees, as well as the beneficiary
community and elected political representatives (figure 6.1). The executing agent, or job worker, is meant to come from the community in
which the scheme is placed, and the beneficiary committee and elected
local councillors check the job workers activities. While the Sub-Assistant
Engineer, and for smaller works the Job Assistant, take the measurements
of completed works, their technical documentation has to be accompanied by a completion letter from the beneficiary committee. Only then can
the BDO and Block Council prepare and submit the utilization certificate
necessary to draw down further EAS funds from central government.
This, at any rate, is how the EAS is meant to work. On the ground
matters can be very different, as we have already seen in Vaishali District,
Bihar. The checks are so complicated, and so demanding of the states
limited capacity, that members of political society are inevitably drawn
in to fill the gaps, and to deal with the internal inconsistencies of the
schemes own rules.29 The particular settlements or resolutions of these
gaps are in turn critically dependent upon the nature of the local political
society. In Debra Block, there is a sizeable pool of grassroots political
workers. These actors and their constituents are also located within a


We have no evidence to suggest that state capacity in Old Malda Block was significantly
inferior to that in Debra Block, Midnapore. Both had their full complement of Block-level
officers, whose training and levels of experience were comparable, and Old Maldas periurban location should have reduced problems of staff absenteeism. Furthermore, Old
Maldas BDO was an able and committed civil servant, who wanted to tackle corruption
in her Block (but was unable to).
As an example of the former, the distribution of EAS cards to labouring households was
a task that local CPI-M workers took upon themselves to complete in Midnapore, as the
panchayat staff were too overworked to complete this.

Political society








supply forms

of completed



(Muster Rolls)


Members of the public

Job worker



Elected politicians

Block council
Civil servants
Job Assistant / Sub-Assistant Engineer
Block Development


Notes: P = Prepare; S = Sign

Source: Field observations.

Figure 6.1 Local monitoring of the EAS in West Bengal: main actors
and responsibilities

system of functioning and active panchayat bodies behind which lies a

well-integrated structure of party committees, stretching all the way from
the local branch office to the District and State committees.30 These
well-established political networks are used to control the EAS in three
important ways. First, peoples expectations of the scheme are limited by
the coordinated actions of party members, with the passive support of
local civil servants (see chapter 4). Neither local government employees
nor party members had any expectation that national government would
provide unlimited funds for the EAS. As a result, the CPI-M did not
promote the idea that work provided under the EAS was in any sense a

The areas covered by the different tiers of CPI-M committees vary considerably across
the state, and are a rough indicator of the partys local strength and depth of organization.


The everyday state and society

right around which labourers could agitate. The EAS was not going to
be another Operation Barga for the Party.31
The CPI-Ms control over the EAS also extended to managing participation in the scheme, as we have already reported. The selection of job
workers and beneficiary committee members was effectively controlled by
the party in areas where they held power, albeit on the basis of discussions
with party supporters. The local party was also effective in managing (but
not erasing) corruption. The execution of EAS schemes was divided into
road schemes (undertaken by the Block office, and usually using contractors) and smaller schemes (undertaken by the gram panchayat, and
using the local job worker). Information about all schemes was given to
the general public through the gram sansads, but it was at the level of
small schemes that the beneficiary committees and the public at large
were best able to make use of this information. We have evidence of a
small scheme that ended up a few hundred rupees under budget: another
project was found to take up the shortfall, and provided extra work. The
increasing strength of the CPI-Ms main rival in Debra, the Trinamool
Congress (TMC), has meant that some TMC supporters have managed
to gain places in the beneficiary committees, and these have become more
effective bodies as a result.32 Contractors gaining commissions are also
expected to make party donations, a proportion of which is recycled to
the local community in different ways.33
What emerges from this system of political control? We have already
compared the material outcomes of the EAS in our research Districts,
but here we wish to draw out the political importance of the scheme.
Certainly, implementation of the EAS in Midnapore was not entirely



Operation Barga, the tenancy registration campaign of the late 1970s and early 1980s,
was a key moment not only in solidifying the CPI-Ms political support in the state,
but also in demonstrating the effectiveness of a different expression of its power from
the openly violent politics of the late 1960s and early 1970s. Specifically, the strategy
used in Operation Barga was to mobilize party workers and supporters around the full
implementation of existing legal rights to secure tenancy, rather than challenging property
relations through the illegal land grabs that had characterized the earlier period (and
which led to the imposition of presidents rule). The partys decision not to use the EAS
in a similar way certainly cannot be explained by a lack of ability to mobilize, as this still
exists in most parts of the state.
This is also the conclusion reached by Bardhan and Mookherji (2004) in their study of
the poverty alleviation efforts of panchayats in West Bengal. We do not want to maintain,
however, that political competition is necessarily good for governance; much will depend
on the type and cost of the competition (see also the next section).
Paying stipends and honoraria to grassroots political workers would be one case in point,
and in Midnapore these individuals did take their responsibilities seriously. More directly,
in one of the adivasi neighbourhoods of our locality, a local contractor was asked to
sponsor the CPI-Ms contributions to the Hul festival.

Political society


clean, but it did involve rituals of accountability and popular participation that were performed on a regular basis. Peoples participation in
the gram sabhas had little direct impact on major spending decisions in
the Block, but some information was allowed to trickle down. It was
often educated middle-class individuals, or rivals of the local CPI-M,
who would use this information to challenge minor infringements or acts
of embezzlement, and the party took care to respond to at least some of
these challenges.34 Purges of corrupt party members amounted to occasional but important performances of the partys commitment to probity.
The chairperson of the Debra Block Council was kicked out of the party
for embezzling funds in 1993. His fall from grace was still talked about
six years later, and served as a warning to other party bosses.
Partha Chatterjee has suggested that commitments to (self-)discipline
and development are key watchwords for the CPI-M, and they are regularly restated in Midnapore. Effective control of EAS and other local
government resources is key to these public performances, and allows
the party to be seen to govern effectively and even graciously.35 Even
when admitting the misconduct of some of its members, the CPI-Ms
control of the public transcript of their correction is used to enhance its
political standing in a wider sense.
The EAS worked much less well for poorer people in Malda than in
Midnapore. We believe this difference has more to do with the composition of political society in Malda than with the backwardness or
illiteracy of the population, which is how some local politicians prefer to
account for things. Maldas more restricted political networks are bound
up with a form of political clientelism that fails to manage the spaces of
participation opened up by panchayati raj institutions, and which supposedly are widened by government programmes like the EAS. District party
leaders enjoy the support of powerful outsiders (the national Congress
Party, and the more critical support of the state-level CPI-M), but they
do not have sufficient reach to incorporate the generation of chhota bhai
netas that has sprung up to fill these spaces. In part, this has to do with
institutional density: CPI-M local committees in Malda cover six times
the area of their counterparts in Midnapore, and there are no equivalent
committees in the Congress Party. Rather more important, however, has

Where complaints were substantiated, the CPI-M responded by getting the individuals
involved to return money, or put works right out of their own pockets.
Gram panchayat chairs who were in complete control of their councils were often at pains
to show proper respect to the political opposition, a state of affairs widely described
as political maturity among party members and civil servants in Birbhum (Williams


The everyday state and society

been the failure of grassroots party workers and the public more generally
to connect to vertical structures of political accountability.
This failure was made explicit in the operation of the panchayats and
the development schemes run through them. While formal party organization extended as far down as the Block level in Malda District, it
did so without a commitment to either broadly developmental priorities or a sense of party discipline.36 Tight-knit relationships had emerged
between key members of the Block council and government employees
charged with a role in overseeing development work, and these relationships allowed individuals of all parties to loot significant chunks of
Block-level EAS funds.37 The politicians benefiting from these corrupt
networks use these opportunities to build up their personal wealth and
prestige rather than the institutional position and standing of their parties. Block-level political bosses also seemed to have little control over, or
interest in, the activities of the gram panchayat members who were below
them. State-level politicians and civil servants were well aware of some of
these failings, and in 1999 they ordered Block Councils to devolve 50 per
cent of the EAS scheme budget to the gram panchayat. They also gave
fresh orders for gram sansads be held on a proper basis. In the absence
of sustained political supervision of gram panchayat members, however,
most of the leaders of political society that we observed at this lower level
behaved in precisely the same ways as their counterparts at the Block
level. Corruption was not so much reduced as decentralized, and possibly
Political parties in Malda do not have the power to keep control over
ward representatives and development schemes in the way they do in
Midnapore. Commitment to the cause of the poor has become a secondary concern in a context where political workers expect public offices
to be exploited for private gain. Rent-seeking strategies of this sort in turn
lead to efforts to close down the democratic and participatory spaces that
are present within panchayats and in schemes like the EAS. Politicians of
all parties in Malda are keen to leave ordinary villagers in the dark about


CPI-M leaders with a background in land struggles were few on the ground, and came
mostly from Gazole Block in the northeast of the District. Rather more recruits and
leaders had come from the student movement in Malda.
In the course of our research we came across examples of muster rolls for labourers
where the numbers of workers had been grossly exaggerated. Various other means to
divert money were widespread, such as the underpaying of labourers who were reported
as receiving the minimum wage. It was in Malda, however, that some of the most blatant forms of corruption existed. Elsewhere in the Block a scandal had arisen because
government money had been spent on a phantom road-building scheme. In Vaishali, in
contrast, where the cuts diverted to politicians and civil servants are institutionalized,
no one would have considered taking them without there being at least some physical
evidence of a scheme.

Political society


development schemes, the duties of panchayat officials, and the rights of

the poor. Panchayat members are then free to take up projects that are
hard to audit and which can support a massive diversion of government
Of course, there are always a few people who are aware of the misconduct of those above them. When these people begin to speak up, they
are generally paid off which is what they are hoping for, by and large.
All told, then, transferring a well-defined portion of schemes to the gram
panchayats has not disrupted wider patterns of rule in Malda: the gram
panchayat members use their autonomy to make profits for themselves,
and are careful not to interfere in the affairs of their seniors. Ward representatives in turn pay off (or threaten: see below) well-informed villagers
who begin to speak publicly about corruption. Poor people with weak
social networks find it hard to make complaints about the corruption
that affects them so badly. In Malda, they expect lower-level government
officers to be involved in the corruption, while higher-level officers are
not easily accessible and may not know precisely what is going on. They
also know that local politicians from all the leading parties depend on
their share of the loot to help finance their election campaigns.38
In sum, looser party control in Malda does not lead to more democratic panchayat institutions than in Midnapore. The Government of
Indias plans for an Employment Assurance Scheme that would be based
on full disclosure of information, and the full participation of citizens as
workersbeneficiariesombudsmen, were never very realistic. In Midnapore, however, the CPI-M was able to distribute some work and material
benefits to families most in need, and in the process it reproduced a sense
of its vanguard role in political society. If complaints were muted about
corruption, or about a lack of popular control over scheme resources,
it was because this resolution of politics and the political worked in
an orderly and relatively inclusive manner. The contrasting situation in
Malda shows that it was the relative thinness of political society that contributed to the failings of the developmental state. Corruption was particularly venal because leading members in political society (officials and
politicians) were largely unchecked by functioning structures of vertical

Regrettably, we do not have detailed data on the funding of political parties or politicians
in our localities. Subventions from bureaucrats (payments to secure transfers or stave
off transfers) are clearly important, along with rake-offs from development spending.
The CPI-M in West Bengal, while not above this pattern of political funding, is probably unique in eastern India in its ability to collect a small annual fee from members
of its associated mass fronts (for peasants, workers, students, etc.). In the mid-1990s,
membership of these fronts in Birbhum alone was reported to exceed 600,000 persons.
The collection of what at that time was a one-rupee fee can thus generate a modest but
significant income.


The everyday state and society

accountability. In this atomized political world the self-interest of public

representatives was paramount, and this extended to keeping ones nose
out of other peoples corrupt practices. External attempts to clean up the
District through decentralization were able to reshape the political terrain, but their ultimate effect was to displace corruption from the Block
offices to the anchal and ward levels. To put it another way, poorer people were provided with sightings of new state officials and new sites of
corruption, but not with sightings of the new model state that has been
mandated by New Delhi.
The price of rule
Finally, we turn to the wider question of the embedding of these networks
of power within more longstanding patterns of rule. There can be no
question that both Bihar and West Bengal have been at the centre of
the second democratic upsurge that Yogendra Yadav has described for
India more generally. By the same token, there can be no doubt that the
political cultures of Vaishali District and Midnapore District are distinct
from one another in key respects and map out very different models of
government. In both Districts, the party in command locally at the time
of our fieldwork was the party in power in the state, which is one reason
for focusing on them here. The ambitions and tactics of those parties,
however, the Rashtriya Janata Dal (RJD) and the CPI-M, were often very
different. Pranab Bardhan suggests that, In the infinitely layered social
structure of India a group, say the Jats in Uttar Pradesh or the Yadavs in
Bihar, can go on clamoring for equity in the sense of parity with the upper
castes, while not doing much about their customary oppression of the
Chamars [or Musahars], who are placed below them in the village society
(often taunting them for special government favors bestowed on them and
calling them sarkar ki jamai or the governments sons-in-law) (Bardhan
2001: 227). This rings true for Vaishali District in several respects, not
least in terms of the assumptions this discourse makes about the capture of
state institutions by and for defined communities. In Midnapore District,
in contrast, as we have already begun to suggest, democratic politics is
more often phrased in terms of an appeal to class, and to the claims of
reason, social justice and rural development.
This appeal to the classical categories of western politics is doubtless
one reason why West Bengal is so often contrasted to Bihar in wider
accounts of Indian politics: it is almost as if the two states personify good
and bad, reason and unreason. But if there is some basis for this claim,
albeit one that is overstated, it should not be allowed to obscure some

Political society


very real similarities of political style between the RJD and the CPI-M.
Both parties operate in a political landscape that is increasingly expensive,
something that is neglected by some advocates of decentralization and
democratization, and both parties routinely use forms of physical and
verbal violence to make the costs and benefits of democratic politics more
to their liking.
Midnapore: politics as development
We have already noted the density of political society in Midnapore, and
its dominance by party members. The CPI-Ms version of political society
fills many of the spaces where independent civil society groups might
expect to be active. It is here that the activities of the developmental state
can effectively be brought under the management of the party. This helps
to establish networks of power within which self-discipline and attention
to the needs of poorer villagers are important to the advancement of
political careers.
This linking of discipline and development a highly developed hierarchical set of party institutions with a pro-poor reformist agenda has
underpinned many accounts of West Bengals political exceptionalism in
the period since the Emergency.39 It would be a mistake, however, to
assume that the most lasting success of the party has been achieved in the
field of service or benefit provision to the rural poor. Partha Chatterjee
reminds us that:
The point is not, as is sometimes supposed, whether the Left Front government
has done a great deal to meet the demands of the people in the countryside . . .
[it] is rather that a field of political transactions has been opened which is within
the reach of most villagers and where matters of local interest can be negotiated
and sorted out on a day-to-day basis. It is in that field that the CPI-M, with its
permanently mobilized corps of workers, enjoys an advantage in the matter of
the daily renewal of the legitimacy of power. Party politics in the West Bengal
countryside today is not something which arrives along the campaign trail once
every five years; it is everyday business and goes hand in hand with government
work. (Chatterjee 1997b: 1601)

If we are to take seriously claims of Bengals success we need to ask what

values underpin this field of transactions, and what holds it in place. The
linking of the CPI-Ms political institutions, and especially its electoral
practices, to the apparatuses of the developmental state has provided a

Among others, see Kohli (1987), Lieten (1992, 1996), Crook and Sverrisson (2001).
More critical is Mallick (1993).


The everyday state and society

degree of macro-political stability almost unique in independent India.

But the values that are expressed in and through the CPI-Ms rule are
equally important in the shaping of West Bengals political society, and
the role its poorer citizens play within this.
Chatterjees work suggests that the linking of discipline and development is important not only in the practice of the CPI-M, but also in its
representation of what constitutes proper political behaviour. We have
already seen that it was in Midnapore, of all our field sites, that public
meetings concerning the conduct of the EAS and other schemes were
most regularly held, and most widely attended. In these meetings, clear
norms of good political conduct are regularly performed: CPI-M chairs
of local government bodies publicly insist on their need to behave impartially, and due process is observed. Villagers also, to some degree, accept
a duty of attendance. This theme of political pedagogy is repeated in the
vision of West Bengals history that is continually presented in a range of
public meetings from election hustings to cultural events. The accepted
story of the last thirty-five years in West Bengal contains a number of key
elements. First, it portrays the land campaigns of the 1960s and 1970s
as a time of popular awakening and struggle against an unjust feudal
order. Second, it portrays the period since the CPI-M gained power as
one of rural development, in the process linking the activities of CPI-Mled governments to a discourse of progress. The material benefits of Left
Front rule electrification of villages, irrigation improvements, or simply
rural roads are contrasted with an earlier dark age under Congress
rule. Third, the constraints on the CPI-Ms abilities to act are stressed:
people are encouraged to expect the rational and ordered use of limited
resources, nothing more.
The skill of the party and its local workers lies in linking this general
narrative both upwards to events of high politics, and downwards to peoples direct experiences. The closeness of the party to the poor allows
it to speak with authority about the injustices of previous generations:
landlords public beatings of labourers, and sexual abuse of lower-caste
women, are still important issues in the consciousness of many labouring households, and the partys role in preventing them is stressed with
regularity. Chatterjee notes that it is important as well for the party to
explain the game of national politics at election time; through its campaigns, the position of West Bengal in the overall pattern of Indian politics
is, as it were, rediscovered and reaffirmed (1997b: 161). The theme of
constraints is also important here: limited resources for local development are linked to an idea of central government conspiring against West
Bengal, and the picture of an embattled, but principled party is thereby

Political society


Like all political discourses this is a partial and partisan narrative, but
rather than comment on its veracity, or its relationship to what a communist party should be doing, we are interested in the patterns of rule
it helps to reproduce.40 The heavy element of pedagogy is indicative of
wider patterns of vanguardism within the party: despite the close links
of grassroots workers to the rural poor, there remains a degree of unease
about the behaviour of the electorate in general. The strong retention
of party control over key panchayat decisions was justified by one of our
respondents with reference to a fear (or suggestion) that the public could
be led astray by richer members of the village. The heavy stamp of party
control was thus seen as necessary, even if it sometimes had the effect of
undermining the public standing of the very council chairpersons who
are the partys official public representatives. The representation of the
party as the sole custodian not only of values that are rational and modern, but of Bengali interests more generally, is meant to justify this highhandedness: far better to have frustrated pradhans, or cynical voters, than
lose sight of these ideals.41
The CPI-M is also willing to use force to secure its version of a better and more legible society. In the run-up to the 2001 state Assembly
elections, a contiguous Block to Debra saw some of the worst political violence in West Bengal in recent years. The violence was most likely started
by supporters of the opposition TMC, but it was quickly met with deadly
force by the CPI-M. The killings in Keshpur were followed by seventy-six
political murders across the state during the 2003 panchayat elections.42
Force has always been important in the politics of West Bengal, just as it
is (and is more widely recognized to be) in Bihar. Arild Ruud notes that
groups of Bagdi (SC) males have been used as lathityals (cudgel bearers or
toughs) for and against the CPI-M in Bardhamman (Ruud 1999: 259),
and Ben Rogalys work on the management of labour relations has shown
how local CPI-M bosses use the threat of violence on a consistent basis
to demonstrate their power (Rogaly 1998).
In some of their more candid moments, members of the CPI-M old
guard will describe the organized use of force as a valid expression of


For a description of the partys rhetorical and ideological gymnastics over the question
of agrarian class relations, see Bhattacharyya (1999a).
The representation of its political opponents as outsiders is an important strategy of the
CPI-M that deserves greater attention than we can give it here. It can be seen both in the
partys attempts to portray the BJP as a party that belongs in the cow-belt (Chatterjee,
1997b: 179), and in its marked difficulty in dealing with the political career of Mamata
Banerjee, whose populism has always pandered to specifically Bengali concerns.
Mathew (2003) notes that almost half thirty of the dead were members of the CPI-M.
The point here is not to position any one political party as perpetrator or victim, but to
indicate that violence continues to be an important part of the way in which politics is
conducted in West Bengal.


The everyday state and society

lower-class revolutionary consciousness. But if there is a class war in

the countryside, the large landlords, the enemy named most often in the
CPI-Ms own pronouncements, are conspicuous by their absence. What
is more evident is continual sniping over wages, tenancy arrangements
and the control of land, in which the management of conflict becomes
a key watchword for the CPI-M. The partys efficacy in resolving these
minor disputes remains important, both to keep the memory of its past
struggles alive and to demonstrate its continuing power in the countryside. But it is hardly credible to interpret these struggles as being aimed at
a further radical shift of power in the countryside. The disputes described
by Rogaly and others are primarily those between richer peasants, and
conflicts are always fought out with an eye to their electoral significance.
What these interruptions to the norm suggest is that the political
project of the CPI-M in West Bengal is far from complete, and far from
unchallenged. We would argue that such outbreaks of violence are not
merely isolated events, or flaws to be corrected by the reinstatement of
party discipline.43 Rather, they are constitutive of a particular form of
decentralized democracy, where the possibility of political violence and
the correct following of committee procedures are equally part of village
life. We return to the consequences of this ordering of political conflict
in our conclusion, and again in chapter 8.
Vaishali: caste politics and political competition
Violence is also constitutive of politics in many Districts of Bihar, including Vaishali. The general view of politics in Bihar, which in important
respects is not wrong, is that the Forward Castes (Brahmans, Bhumihars
and Rajputs, especially) who once dominated there have been replaced
over the past thirty years by political parties that have claimed the state
for the rural OBCs.44 These acts of replacement, moreover, which have
erupted in a series of bloody confrontations, have had the effect of reinforcing a pattern of rule which refuses to deal with the collective action
problems that are supposed to define everyday politics. To the extent
that ideology has entered politics it has been through an effort to empower
the (mainly Other) Backward Castes against their erstwhile tormentors.
In some areas this has promoted a Naxalite politics focused on land and


Any suggestion that political violence is centrally organized or sanctioned is strongly

resisted by the party. Its representation as local action by unruly members allows the
higher echelons of the party to distance themselves from these incidents, however useful
they might be for short-term electoral gain.
At first under Karpoori Thakur, and later, in a more sustained fashion, under Laloo
Yadav. Useful guides to the agrarian and political struggles of the 1970s and 1980s can
be found in Blair (1980, 1984), Frankel (1989), Hauser (1993), and Januzzi (1973).

Political society


labour issues, but empowerment has more generally been conceived in

terms of honour (izzat). Part of the appeal of Laloo Yadav is that he
personifies a muscular form of subaltern politics which trades an eye for
an eye. The idea that the state should be run for the common weal is
refused in this model. Politics is conceived as a Darwinian struggle for
survival, in which the strongest politicians take strength from the often
tightly defined groups which they represent (the Yadavs, say), or with
whom they seek to ally (the Muslims for the RJD).
What is missing from this account, however, or from overdrawn contrasts between Bihar and West Bengal, is that competitive politics is
deepening in many parts of Bihar, and with many of the same effects as
can be observed in its eastern neighbour. Consider, for one last time,
the nature of political society in Bidupur Block, Vaishali District. We
were concerned in chapter 5 to show how an externally imposed agenda
for good governance was reshaped by the activities of key players in
this locality, including, most notably, the de facto MLA of the area, BB.
Government officials in Bidupur and Vaishali come under enormous
pressures from local politicians and their main supporters. Laloo Yadav
was the MLA for the surrounding Raghopur assembly constituency from
1995 to 1998, and BB is one of his leading supporters. We also showed
how the network of contractors and supporters that surrounded BB was
penetrated only to a small degree by members of the poorest communities
in the Block, including the Paswans, Dusadhs and Musahars. These communities were cut off from the metal-topped roads that were built under
the EAS, largely at BBs direction (see figure 5.3). We discovered exactly
the same situation in nearby Patepur Block, where two of us drove to a
possible research village in January 1999 with the local Circle Officer.
As we approached the main tolas of the village, the CO pointed out the
neatly built brick roads that led to the settlements of various Yadav families: roads built with EAS funds. When the brick road ended, and we
spilled on to a deeply pot-holed earthen road that ran for half a mile to
the furthest tola, the CO told us, as the Yadav tolas end, so does the
kharanja [brick road] too. Now you will see the settlements of Musahar
tola ahead.
The wretched poverty of the Musahars of this tola offers clear evidence of the failure of political society in Vaishali to deepen sufficiently
to empower the poorest. Government was not working for the Musahars
and their preferred means of advancement, or simply of survival, was
long-distance migration to Punjab and Haryana.45 Nevertheless, even in

We were told that 3040 per cent of households sent at least one member to Punjab or
Haryana. People were well-informed about wages in other states (that is, they accessed
information that was valuable to them), and one villager told us that, as members migrate
this brings down the employment-seeking population here.


The everyday state and society

this community, it was apparent that the very absence of sarkar was a
subject of reflection, and even action. Villagers said they were reluctant
to approach Block officials directly, because we are illiterate. But it also
transpired they had a tola mate who goes to the Block and speaks on our
behalf, as he is the only matriculate in the entire tola. In this particular case, the villagers maintained that the tola mate had swindled them
of funds on another project, but this is perhaps less important than the
fact that a sense of entitlement to government funds, and thus a sense of
citizenship rights, was widely felt in this disregarded community.
In the ranks of the Other Backward Classes this sense of entitlement
is equally well formed, but it is underpinned by a sense of achievement.
The state has clearly failed these and other people in all sorts of ways.
We have already commented on the poor state of educational provision
in the public sector, and the failings of the power sector and hospitals
in Bihar have been widely remarked, alongside the criminalization of
politics (see chapter 7). In the specific case of developmental funds,
however, including those distributed under the EAS, it is important to
underscore a point that we began to sketch out in chapter 5: namely,
that if BB sits at the apex of political society in Bidupur, he is required
to be responsive to a broad range of actors in civil and political society. In some respects, indeed, the high level of political competition that
can be observed in Bidupur poses greater problems of democratic management for BB than would commonly be faced by his counterparts
among the CPI-M in Midnapore. His enthusiasm for the conversion of
the EAS into a road-building scheme must be seen in this context. Instead
of dictating to middle-ranking groups in Bidupurs political society, BB
was required to acknowledge the remarkable fragmentation of traditional
authority structures that have occurred there over a thirty-year period.
Along with the local MP and DM, he had to decide on who, in what ways,
in what order, and to what degree (including financial), a wide range of
contractors, households, fixers and communities could be made to share
in the bounties of developmental spending. He had to do so, moreover,
in the knowledge that most of these individuals or groups have a very
good grasp of how the state works at the level of the Block office, and
of who precisely must be bribed to get a particular job done or contract
If BB failed to perform his tasks satisfactorily he could be punished
at the ballot box, or possibly be shunned by his own patrons within the

For many poorer people, small payments to the VLW are the first step into a government
system marked both by the professionalization of corruption and by a greater degree of
social levelling in its incidence.

Political society


Rashtriya Janata Dal. Lower-level fixers and politicians also run the threat
of physical violence being used against them. Again, this is not so very
different from some other states in north India. The deepening of democracy in Bihar is bound up with the power of numbers and with the threat
of physical force. This is also where a sense of citizenship is slowly being
forged. BB is seen as a boss and a patron, but he is also seen as someone who is responsible to the middle orders, and who might yet be
brought under some measure of control or simply be opposed by the
Scheduled Castes. Demands are brought to bear upon BB by fixers in
political society, and these fixers must have regard for the wishes, and
even demands, of ordinary people who are becoming more aware of their
rights. At the same time, however, the deepening of democracy in Bihar
is occurring alongside forms of political mobilization that call to mind
New York City under Boss Tweed. That is to say, people are learning
about democratic values in the ways that many people did in the West,
as opposed to in the textbooks of the West or the international development agencies. They understand that rules emerge from conflicts, and
that politics can be costly for those who are keen to control the power
and resources that flow from them.
The political settlements that we have described for Midnapore and
Vaishali would doubtless be found to some degree in other parts of India.
Political societies are structured by an uneasy and slowly shifting set of
relationships that are governed by patronclient relations and the threat of
violence, on the one hand, but which are also responsive to the demands
of ordinary people who put pressures on lower-level fixers and political
operatives. These demands, moreover, are sometimes articulated with reference to a discourse of rights, or a sense of peoples legitimate claims
upon the state (whether as an individual or as a member of a population). In West Bengal, the panchayats are operating in many Districts to
ensure that the poor are incorporated within state-sponsored mechanisms
for the redistribution of some assets and even some voice. Poorer people
are familiar with the practices of panchayati raj but in some areas there has
been little in the way of learning about rights or even struggle. As a consequence, the rural poor are still largely dependent on party elites rather
than being agents in their own right. It is mainly through participation
in gram baithaks that a broader sense of citizenship is being learned. In
Bihar, meanwhile, caste tensions keep social conflict out in the open, but
militate against development and the empowerment of the most marginalized social groups. Set against this, a high level of political competition


The everyday state and society

in some Districts, as for example in Vaishali, is creating new spaces of

participation for the Other Backward Castes, including the Yadavs and
the Kurmis.
Whether these acts of participation describe the making of modern citizens is a moot point. Partha Chatterjee might argue that caste or ethnic
politics stand in opposition to the civility of a politics that is focused
on individuals and the prosecution of their statutory rights. We agree,
up to a point, but we would also want to insist that pork-barrel politics
is hardly unknown in the West. The distinction that Chatterjee maintains between political society, on the one hand, and civil society on the
other, can more reasonably be thought of as a set of interlocking political
practices that are arranged along a continuum. Elections in Midnapore
and Vaishali illustrate this very well. They are certainly occasions for
the great anonymous performance of citizenship (Chatterjee 2004: 18).
High levels of voter turnout among the poor indicate a growing sense
of recognition that politicians and governmental agencies must take the
preferences of ordinary people into account. The act of voting, moreover,
feeds into a broader range of transactions between ordinary people and
their points of contact in political society. It is through such transactions
that a sense of citizenship begins to develop. We saw this in Vaishali. At
the same time, however, as Appendix 2 graphically illustrates in the case
of the 1999 general election in Hajipur Constituency, Vaishali, members
of the most excluded castes (Bihar) can be effectively barred from voting
by the use (or threat) of violence. Appendix 2 also makes clear that the
(supposedly) anonymous act of voting is bound up with intensely visual
and highly personalized performances of political power that appeal to
ordinary people as men or women, as members of caste groups, and as
the clients or supporters of named authority figures.47
Again, this is not so very different from what happens in the West, or
at least in parts of the United States. Voter fraud was widely reported in
Florida in the presidential election of 2000, and the use of ex-felon lists
to disenfranchise black voters in the same state in 2004 was also discussed
by news organizations. Problems of good governance are much greater
in Vaishali than in Florida, of course, not least because people have better means of redress in the US, but the civil/political society opposition is
less clear-cut than we sometimes imagine. Chantal Mouffes description
of politics as the attempted pacification of the political is perhaps more

In West Bengal, acts of booth capture and other forms of directly forceful exclusion
from politics may be more rare, and generally are not played out in caste terms. Nevertheless, as we have noted above, violence remains an important constitutive part of these
personalized performances of power for the CPI-M and its political opponents alike,
however free and fair the voting is on polling day.

Political society


instructive in this regard. In eastern India pacification does not imply

the absence of conflict, but its attempted regularization within particular forms. Aggressive competition is present as a threat and as a practice across all our field sites. In Vaishali, however, this takes the form of
competitive caste mobilizations for the capture of governmental resource
streams. In Midnapore it involves a political settlement that links a space
for the empowerment (and voice) of the rural poor to a continuing high
level of party control. Each pattern of rule opens up and closes down
certain options for the rural poor to engage with the political process.48
These political settlements in turn expose the shallowness of measuring
government performance in eastern India against an idealized model of
good governance or democratic politics. Judgements of this sort will
continue to be made, of course, and it is part of our wider argument
that they can have constitutive force when they are backed by the policing or funding powers of central government or the leading development
institutions. When we say that such measurements are not useful we do
not mean that democratic practices cannot be improved in each of our
field sites, or that the agendas of participatory development and good
governance are irrelevant in eastern India. Of course not: we mean to
suggest that the incentive systems that structure political cultures can be
difficult to change. A focus on political institutions can be misleading if
it encourages the view that decentralization, say, or even democratization, will promote development (or even freedom) in a straightforward
fashion. Political institutions take shape in political cultures.
Just as importantly, and this applies to the Left as well, it is a mistake
to suppose that political cultures that might be admired in one locality
can be transferred with ease to another. This was an implicit suggestion
of Atul Kohlis important and generally positive analysis of the state and
poverty in West Bengal. We will argue in chapter 7 that we need to have
close regard for the very different trajectories by which poorer people can
(and cannot) hope to make the state work better for their families. This
chapter also finds us moving back to some of the broader issues that we

Importantly, these patterns of rule, and the spaces of empowerment they open up for
the poor, are dynamic. The reinstatement of panchayat elections in Bihar soon after the
main period of our fieldwork could begin to restructure the patterns discussed above.
It is possible that the panchayats will bring a new range of personnel, including social
and charity workers committed to more positive and inclusionary forms of politics, into
political society at the grassroots level. In West Bengal, the 2003 panchayat elections
saw some potentially important reversals for the CPI-M-led Left Front, including the
loss of Malda District Council, one of the first Districts the party has lost control of in
twenty-five years and six rounds of elections. Whether this prompts a more significant
shift in the partys style of managed participation remains to be seen. In both cases,
the political trajectories remain uncertain, and any gains experienced by poorer and
excluded individuals are clearly reversible.


The everyday state and society

raised in part I. Our work in eastern India has provided us with a platform
from which to research in some detail questions relating to governance
and governmentality. We have done this mainly through a study of the
Employment Assurance Scheme and primary education provision and
uptake. We have tried to show how these sites of statepoor encounters
give rise to, and are structured by, the sightings of the state made by poorer
men and women, urged on by ideas about participatory development.
We have also looked at those sightings made by government officials and
members of political society, which are informed to some degree by ideas
about good governance and democratization. We now want to widen our
lens and consider some of the very diverse ways in which poor people
speak back to the state. Very often, as we shall see, this is done in ways
which refuse (or which seem to refuse) the blandishments of the new
public administration.

Part III

The poor and the state

Protesting the state

In the past four chapters we have tried to say something about the spaces
of empowerment that open up for poorer people in their dealings with
government officials and other authority figures around the EAS and
primary education provision. In some cases these spaces can be enduring and quite extensive, as we saw in Debra Block, Midnapore. For all
that the CPI-M attempts to fill the political society of this Block, poorer
men (and some women) are given opportunities to work on government
schemes, and they have some say, too, about the running of those schemes
and their local public schools. Political society is also quite thick and
competitive in Bidupur Block, Vaishali, although as yet the Scheduled
Communities have not managed to make much headway against the
Yadavs and other OBCs. In Old Malda Block, political society is less open
to the voices and interests of the poorest, and spaces of empowerment
are harder to detect. Poor levels of literacy and information circulation
conspire against the agendas of participatory development and good governance, and such successes as we could report tend to be episodic and
sometimes short-lived. Poorer households tend to fare worse here than
in Bhojpur or Ranchi, where patrons are often more responsive to their
In the final part of the book we want to extend our terms of reference to include a broader range of political encounters across India.
We also want to review the agendas of the new public administration
in the context of political societies, such as can be found in Kerala, that
might be thought to be more receptive to their concerns than a state
like Bihar or even Jharkhand. Once again, this returns us to James Scott,
but not to Seeing Like a State, which has surprisingly little to say about
the ways in which people talk back to their overseer. This time we are
directed to Weapons of the Weak, where Scott looks at the diverse strategies employed by members of peasant households who fight back against



The poor and the state

the depredations imposed upon them by landlords, traders, merchants

and government officials. Weapons of the Weak has been criticized for playing down the gendered nature of peasant politics, and for ignoring such
large-scale peasant rebellions as can be observed.1 But Scotts work forces
us to think through both the conditions of anti-state and anti-landlord
protests, and the highly variegated forms in which these politics are made
flesh. Faced with the greater power of the landlords and/or the state, the
rational peasant falls back on more Brechtian or Schweikian forms of
everyday resistance, among which Scott numbers foot-dragging, deception, desertion, joke-telling and non-compliance (Scott 1985: xvi).
In this chapter we want to present a similar take on the politics of
the poor in contemporary India. We should straightaway say that we
cannot hope to do more than scratch the surface here, at least in terms of
identifying and commenting upon specific political acts or movements.
There are several excellent books that take up these issues in detail.2
Our purpose, rather, is to direct attention to some of the generic forms
of resistance by means of which poorer people talk back to the state or
advance their agendas for good governance. More exactly, we want to
suggest that these forms of resistance/promotion very often take shape
with reference to the different sightings of the state that are made by
poorer men and women, and indeed by some government officers and
members of the international development community. Chapter 6 began
this task, but here we return to the broader canvas that occupied us in
chapters 1 and 2.
We begin with questions of accountability and some of the different
ways in which poorer people have demanded a right of inspection of government information. We then turn to the politics of decentralization,
and in particular to recent experiments with panchayati raj institutions in
Madhya Pradesh and Kerala. These experiments are bound up with the
construction of citizenship and involve more direct sightings of the state.
Next, we turn to questions of izzat, or the politics of honour and empowerment. Our focus here is on Bihar, and to a lesser extent Uttar Pradesh,
two key states in the Hindi heartland where a dominant politician from
the Backward Classes presents him- or herself as the embodiment of a
new type of rule. Finally, we consider how the power of the state might
be sighted along class and sectoral lines, and from the perspectives of
reformers who are active within the Government of India or the international development community.

Hart (1991), Brass (1991).

See, inter alia, Baviskar (1995), Hasan (1998), Omvedt (1993), Pai (2002), Skaria (1998),
Webster and Engberg-Pedersen (2002).

Protesting the state


Information circulation and the question

of accountability
When we invited villagers in Sahar Block, Bhojpur, to think aloud about
pathways to enhanced pro-poor governance, they quickly pointed to
issues of information circulation and usage.3 Our respondents complained about the lack of information they had on government schemes.
They also made the point very firmly that they needed information in a
form they could make sense of and which provides the basis for specific
Although no one in Sahar Block cited the example of the Mazdoor
Kisan Shakti Sanghatan (see also chapter 1), these are similar to demands
that have been made by the MKSS in Rajasthan. The MKSS has been at
the forefront of a right to information movement that has been gaining
ground in India more widely. The organization was formed in 1990 and
its initial concerns had to do with the enforcement of minimum wages
legislation. As Aruna Roy and Nikhil Dey explain, in fighting for minimum wages [for example, on government drought-relief programmes],
the MKSS first understood the significance of transparency and the right
to information. It was necessary to access records, to prevent corruption,
to try and obtain the minimum wage, and to ensure that infrastructure
actually got built. Every time the workers made a demand for minimum
wages they were told that they had not done the work, as proved in the
records. [But] when the MKSS made a demand to see the records, they
were told that these records are government accounts and therefore secret
(Roy and Dey 2001: 3). The MKSS came face to face, in other words,
with a frankly colonial bureaucracy which hid behind some of the provisions written into the Official Secrets Act of 1923.
Interestingly, the tactics that the MKSS evolved to press for a right to
information, a demand that we might consider to be modern or rightsbased, as in key respects it must be, made use of a repertoire of vernacular
understandings of the ethics of public life that reached back to pre-British
days. Christopher Bayly has argued recently, and more generally, that the
particularities of Indian nationalism have to be understood in the context
of Indian forms of social organization and ideologies of good governance
that pre-date the full western impact even if they, in turn, had been modified by colonial rule (Bayly 1998: vii). He suggests that we should tread
with care when dismissing nationalist or caste-based discourses in India as

As part of a DFID-sponsored exercise that we report in chapter 9.


The poor and the state

merely derivative, or as the bastard children of imperial categories.4 This

is to deny Indians agency in their own history (Bayly 1998: 3). It also
forgets that Punjabis and Bengalis, for example, had a sense of territory
and territorial identity before the British imposed modern boundaries.
Extending this argument somewhat, it is important to note, as well, that
what Scott has previously called the moral economy of the peasantry
(Scott 1977), has many times been invoked in different parts of India to
check uncalled-for behaviour on the part of kings and their retainers or
officials. Ramachandra Guha reminds us that what became known as the
Chipko andolan in the 1970s was anticipated in Uttaranchal in important
respects by the institution of the dhandak: a form of rebellion as custom
which combined individual and collective resistance to tyranny by officials with a simultaneous call to the monarch to restore justice (Guha
1989: 67). What was challenged here was not the idea of the state, but its
corruption by individuals who had gone wrong, or who failed to obey
the tenets of natural justice.
The MKSS sights the state in a not dissimilar fashion, but with due
regard for the conventions of law and right at the national scale. In
its first incarnation, the MKSS made use of village-based public hearings, or Jan Sunwais, to place the local state firmly in the glare of public
scrutiny. The first Jan Sunwai was held in Kotkirana panchayat in Pali
District, Rajasthan, in December 1994. Members of the MKSS had been
approached by a poor, middle-aged man about underpayment of wages.
They had used their links to a cooperative bureaucrat in temporary
charge of the Block Development Office to access relevant documents,
the contents of which were copied out by hand under the disgruntled
gaze of the office staff (Roy and Dey 2001: 4). Facing down opposition
from local goons and politicians, the MKSS pressed ahead with a Jan
Sunwai which was attended by over a thousand people (Roy and Dey
2001: 4). Using an informal idiom of conversation and exchange, but
with all the seriousness of court proceedings, the people of Kotkirana
were turned into a large jury which heard evidence from fellow villagers
about what they had discovered in the copied records. According to Roy
and Dey, again, Person after person came to the mike, to say their name
was fictitiously recorded on the muster roll; that they were away on migration; that they did not do manual labour; that the names of their dead
fathers and relatives had been entered [and so on] (Roy and Dey 2001:
5). Much as we found in eastern India, the MKSS discovered through

Bayly is taking aim at Chatterjee (1986, 1993) and Dirks (1992), among others. Dirks
replies in the final chapter of his book, Castes of Mind (2001). See also OHanlon and
Washbrook (1992).

Protesting the state


a string of Jan Sunwais that more than 30 per cent of government funds
marked for poorer people were, on average, being diverted to government
officials and local politicians. Unsurprisingly, perhaps, the biggest rakeoffs were in the poorest Districts and panchayats, where poverty (and its
alleviation) were big business (Roy and Dey 2001: 8).
In the second half of the 1990s the MKSS was able to use the system of
Jan Sunwais to expose the links between corruption and the electoral system in rural Rajasthan. The sarpanch of Kukarkheda, Basanta Devi, was
pressured in a public forum to return 100,000 rupees which had been
proven through the documents and the public depositions as having been
defrauded from the panchayat . . . The whole area was buzzing with the
news. Laloo or Jayalalitha did not return stolen money, but Basanta
Devi has (Roy and Dey 2001: 15). Like many sarpanches across India,
Basanta Devi had assumed that she had the right to dip into, or simply
loot, the contracts she had received from local politicians in return for
helping to get the vote out for them at election time. Now that right had
been challenged, and the MKSS felt emboldened to scale up its demands
for information circulation and accountability to the state level. Partly in
response to pressures from the MKSS, the Rajasthan State Legislature
passed a Right to Information Law on 1 May 2000. Perhaps more significantly, the state Assembly moved in the same month to pass an Amendment to the Panchayati Raj Act which created the legal entity of the ward
Sabha and vested it with powers of social audit.
It is still too early to judge how these social audits will fare. Ward
Sabhas will be as time-consuming as Jan Sunwais, which typically last a
day or more, and there is no reason to suppose that government officials
will make information available freely or on time unless they are placed
under pressure. Some bureaucrats are already fighting back against the
MKSS. They argue that its actions are likely to choke the arteries of
government, and will provoke a rash of false charges against government
officers who lack the means to defend themselves.
There is doubtless some truth in these complaints, just as there is a
pressing need for independent assessments of the activities of the MKSS
and the effectiveness of its Jan Sunwais.5 It is hard to escape the conclusion, even so, that the MKSS has enjoyed considerable success in
Rajasthan, and has done so for a variety of reasons that speak directly
to the concerns of this book. To begin with, and perhaps most importantly, the MKSS has encouraged the people . . . to concretely perceive
the links between their personal lives and the political processes of democratic functioning. They saw the links between the check dam and the

See Jenkins and Goetz (1999).


The poor and the state

debate over State allocations, the planning process, and the implementation machinery (Roy and Dey 2001: 5; emphases added). The MKSS
facilitated these sightings, moreover, not simply by its efforts at procuring and/or copying government records, especially at the panchayat level;
it also dramatized its quest for accountability by means of rural juries
armed with little more than microphones and perhaps a video recorder,
as well as by hunger strikes, dharnas (sit-ins), and such innovations as the
Ghotala [scam] Rath Yatra (a play performed in a dharna tent), and the
declaration of pakhand divas (hypocrisy day) and kala divas (black day),
both of which led up to victory day when the Panchayat Raj rules were
finally amended. The state was sighted by low-level technologies that produced a version of that same state in highly visual local spaces theatres,
in effect, that involved an audience in a partly scripted deconstruction of
the states descriptions of itself.
In addition, and also very significantly, the MKSS took steps to scaleup its campaigns by joining forces with the National Campaign for the
Peoples Right to Information (NCPRI) in New Delhi, and by working
actively alongside committed politicians and journalists, including Kuldip
Nayyar and Nikhil Chakravarty. By this means, especially, the grassroots
campaigns of the MKSS were made to rub shoulders with demands
for open government that were being raised in metropolitan areas, and
which have come to focus on the fourth estate (the press and media)
and the Supreme Court of India and various High Courts. As Lloyd
and Susanne Hoeber Rudolph have recently shown, the supreme courts
judicial activism helped to repair and correct the Indian state [during] the
era of unstable, short-lived coalition governments in the 1990s (Rudolph
and Rudolph 2001: 132). Helped by a more engaged presidency, and by
an Election Commission brought to life by T. N. Sheshan, the Chief
Election Commissioner from 1991 to 1996, the courts laid claim to the
primacy of the chief justices views against those of the political executives . . . in the appointment and transfer of high court and apex court
judges (Rudolph and Rudolph 2001: 139). They also played a critical
role in approximating a framework of lawfulness and predictability that
has had some success in protecting citizens rights, limiting malfeasance
and safeguarding environmental and other public goods (Rudolph and
Rudolph 2001: 132).6

It should be noted that Rudolph and Rudolph are fully aware of the enormous backlog of
cases facing the court system in India at the highest levels (765,426 cases in the Allahabad
High Court alone in 1995: Rudolph and Rudolph 2001: 137), and of the fact that judicial
activism has often been in response to pressures that first emerged in civil society, and
with environmental and human rights activists in particular (as in the cases of opposition
to environmental degradation and big dams (Narmada, Tehri), [and] child and bonded
labor and demands for Dalit (ex-untouchable) empowerment, and historical and cultural
preservation: Rudolph and Rudolph 2001: 137).

Protesting the state


Decentralization: getting closer to the state?

A more activist judiciary in India might in time prove to be a major source
of empowerment for differently placed individuals within the rural and
urban poor. Niraja Gopal Jayal is right to warn that, the state manifests an
uneven relationship with society: combative towards some forces and collusive with others (1999: 99). Even in Orissa, however, where at the local
level, the relationship between state and society becomes less paternalistic
and more directly exploitative as state functionaries browbeat the illiterate poor into submission (Jayal 1999: 99), there is evidence that the food
crisis in Kalahandi District was addressed, and partly mitigated, by the
use of institutions of parliamentary democracy, such as adversarial politics in the legislature, a vigilant press, and a judiciary receptive to public
interest legislation (Jayal 1999: 238; emphasis in the original).7 In neighbouring Jharkhand, meanwhile, the womens movement has been active
in pressing the issue of female land rights in the High Court, and indeed in
the Supreme Court in New Delhi, where a judgement is still awaited.
Such legal challenges tend, however, to be brought on behalf of poorer
people, and there can be no guarantee that a judgement in their favour
will be properly enforced. Partly in recognition of this, the Government
of India has argued that poorer people must be brought into a more direct
relationship with the state, and indeed with the political process. To this
end it passed in 1993 the two constitutional amendments that we have
discussed previously (see chapter 4), which devolved greater powers and
responsibilities to local government. In addition to arguments about the
merits of decentralization (that people will take a more active interest
in government if they have some say in how decisions are reached and
monies are spent), the thinking behind the constitutional amendment
Acts had regard for the principal-agent problem. That is to say, there
was a stated recognition of the fact that many lower-level government
officials, most of whom would be recruited from the state cadres, could
not be supervised properly by their principals. Even assuming that District
and state-level IAS officers were well-intentioned and honest, they were
simply unable, for the most part, to keep their subordinates in check.
Muster rolls could be falsified and false accounting could be practised on
a widespread basis. One way to address this question, central government
now reasoned, with support from many external actors, was to institute
both surveys of the attitudes and practices of public officials, and also
public service reforms.8 New managerial systems could be put in place

Amartya Sen would see this as supporting evidence for his thesis that famines do not
occur in open or democratic societies (Sen 2000 and cf. our discussion in chapter 2).
Including the survey of public officials in Uttar Pradesh that was commissioned by the
World Bank in 2000. See also World Bank (2003).


The poor and the state

that would reward honest and innovative behaviour, and which would
deal harshly with corruption or ineptitude. But another way to deal with
the same question was to reduce the powers of the agents. This could be
done by turning over some of their responsibilities to the people through
an invigorated system of panchayati raj institutions.
We have no desire to go over old ground here. We have seen that decentralization is no guarantor of empowerment. The costs of participation
often remain high for poorer people, and for women especially, many of
whom continue to face enduring structures of social exclusion.9 In most
states, too, the decision-making powers that come into view remain limited, at best, and are sometimes unsupported by devolution of funds.10
Nevertheless, it will be useful to consider in more detail how decentralization has played out or has seemed to play out: collecting detailed
evidence is always difficult in Madhya Pradesh and Kerala, two states
that are pointed to with some regularity in discussions of decentralization. What lessons, if any, can we learn from the experiences of people
in these states about the political possibilities that might inhere in closer
inspections of the state?
Madhya Pradesh
Madhya Pradesh is one of the poorest states in India and forms part of
the Bimaru group of backward states, along with Bihar and presumably
now Jharkhand from those among our research area. It was formed in
1956 from parts of modern-day Uttar Pradesh, Gujarat and Maharashtra,
and goes by the cutting nickname of the leftover state (Shesh Pradesh).
This changed when Digvijay Singh came to power as Chief Minister of a
Congress (I) government in 1993. The state now began to make headlines
for its experiments with pro-poor governance. Among the reforms that
Singh supported were the following: the decentralization of funds and
powers to PRI institutions for the purposes of rural development, social
welfare and some revenue matters; the promotion of an Education Guarantee Scheme (EGS-MP) to help promote universal literacy; a Participatory Watershed Development Programme that would link anti-poverty
agendas to environmental regeneration; a District Poverty Initiative
Programme that provides grants instead of loans to poorer people identified by means of participatory wealth rankings; a Citizens Charter
and Right to Information Act; the countrys first sub-national policy for
womens development; and improved IT facilities in rural areas.

Brett (2000).
For extended commentary, see Crook and Manor (1998); Manor (1999).

Protesting the state


There is some evidence to suggest, as well, that Singhs push to involve

greater numbers of women and members of the Backward Classes in participatory government did meet with a measure of success. The reservation of posts in government from the level of panch through sarpanch
on up to the post of Zilla Adhyaksh (Chair of the District-level council) helped to ensure that slightly more than a third of all panches are
women and that members of the Scheduled Communities make up about
40 per cent of elected representatives at the Block and District levels.11
As Srivastava reports, too, on the basis of extensive fieldwork in two
Districts in different regions of MP, Vindhya and Mahakaushal, the act
of bringing thousands of actors from the poor castes into the institutions
of panchayati raj has not been without impact. . . . [The process has
helped] in the demystification of the state, bringing into the open how
the powerful in society and the erstwhile middlemen build up connections with and influence on Sarkar (government). . . . [they also learn]
of networking strategies, the machinations of cuts and commissions,
routes to higher mobility, and so forth, all leading to what Hirschman
would call improved exit and voice options in the longer run (Srivastava
2003: 23).
Nevertheless, as Srivastava also reports, and as we should perhaps
expect in a state where decentralization is being gifted from above, largely
in the absence of a sustained mobilization of the poor, the workings of
the new PRIs are far removed from the rhetorics in which Digvijay Singh
once sought to surround them. Srivastava found that what he called the
constitutional sarkar was being widely nudged aside by the permanent
sarkar, or that loose assembly of the landed, the higher castes, males,
contractors, MLAs and MPs, and District-level bureaucrats. In some
cases, the control of the permanent sarkar over the constitutional sarkar
was direct and very open, as where the Brahman husband of the female
sarpanch continued to manage the business of the panchayat including
signing checks, running meetings and so on while his wife looked after
the education of their children in a small town 35 kilometres away. In
other cases, control was exercised more indirectly, as when a 70-year-old
sarpanch from a Backward Caste turned up for work, but was little more
than a frontman for a local Rajput landowner who had fought the election against a rival Rajput with his own frontman. And in still another
case, power was exercised over an elected, Backward Class sarpanch by
means of a more educated panchayat secretary who was put in place by

We are grateful to James Putzel and John Harriss for allowing us to quote from work
that Manoj Srivastava has carried out for them as part of a DFID-funded programme
on Crisis States.


The poor and the state

members of the permanent sarkar. In this case, principal-agent thinking

was turned upside-down. Instead of the principal (the elected panch or
sarpanch) exercising control over his or her agent (the panchayat secretary), the latter person was using an educational advantage to reproduce
the power of the (minor) bureaucracy.
The results of these forms of elite capture were at once predictable
and depressing. Srivastava found that panches and sarpanches were taking
advantage of a new rule that allowed for no government audit of schemes
budgeted at Rs. 50,000 or less. They used this rule to siphon off funds
from the JRY allocation, either to line their own pockets or to fund their
political ambitions. The government of Madhya Pradesh had intended
that spending of this order should be scrutinized by members of the
panchayat directly, through a social audit. But this provision was unknown
to most villagers, and was not acted upon in the form even of statutory
gram sabhas. In the Vindhya region, Srivastava reports, a simple rhyme
has been composed to express popular feelings about the enormous rakeoff of funds (up to 90 per cent on some estimates) from the JRY budget:
J-R-Y, matlab jeb kharch aayee (roughly, when JRY funds arrive, then the
pocket expenses thrive).
Srivastava came across one case where an up-sarpanch carried out
minor roadwork to the value of Rs. 800, but was asked to sign for a much
inflated figure. The up-sarpanch later found out that a sum of Rs. 53,000
had been entered in the register. He was also able to report in some detail
on the pressures under which many sarpanches laboured: it generally cost
Rs. 10,00025,000 to fight an election; there were various small costs
associated with moral patronage (helping people in distress) and status
enhancement (refreshments for visitors); and the sarpanch needed significant funds to bribe such inspecting officers as there still were, and to make
contacts with MLAs, MPs and ministers if she or he wanted to climb the
political ladder. (Interestingly, and again predictably, the transfer of some
budgets to the sarpanch from those of an MLA had not only caused many
MLAs (and MPs) to exert pressure on Digvijay Singh to tone down his
decentralization agenda, but had improved the bargaining position of the
sarpanch). Sadly, in the eyes of many villagers in MP, decentralization has
become little more than a recipe for the greater looting of the state. To
the old thieves were now added some new thieves, some few of whom, of
course, would have come from poorer families.
An optimist might say that things will change in time in Madhya Pradesh,
and that at least people from the weakest sections are now standing for

Protesting the state


election. Assuming that not all of these people are fronting for more
powerful individuals, we must suppose they find it rational to engage
with a more participatory form of democracy. But we might equally
make use of ODonnells term, low intensity citizenship, to describe
statepoor relationships in MP, and we should certainly have regard for
Chibbers charge that India suffers from an excess of democracy without association.12 Poorer people in the Bimaru states still learn to see
the state through vertical and highly mediated exchanges with members
of the permanent sarkar. Even when some of their numbers gain command of some state agencies or funds, there is no reason to suppose, a
priori, that they will take the part of the poorest or abide by the rule of
Of course, matters might be very different in Kerala. The so-called
Keralan model has been widely applauded on the Left, and it is undoubtedly true that levels of literacy and basic health-care provision in Kerala
are excellent, and put richer states to shame. Doubts continue to be voiced
about the long-run capacity of the private sector to generate jobs in the
face of high levels of unionization and government regulation of labour
markets, but even these concerns have been displaced amid renewed
excitement in Keralas experiments with participatory democracy.13 In
1996 the government of Kerala introduced The Peoples Campaign for
Decentralized Planning. Important planning and budgetary functions
that previously were left to state-level agencies were now devolved to
urban municipalities and rural gram panchayats. The government also
took steps to ensure that village assemblies (gram sabhas) would be held
twice-yearly in all gram panchayats, so that ordinary men and women
could engage with the planning process.
Thus described, there is little perhaps that sets Kerala apart from many
other states which have sought to enliven local (self-)government. In
Kerala, however, the Left Democratic Front which was returned to power
in 1996 has sought to build upon a long tradition of popular mobilization that had brought poorer people into political society through land
reforms, highly organized political parties, and mass-based organizations
of workers.14 To this end, the government committed a large number
of key resource persons (KRPs) to work with each gram panchayat at
the time of the first gram sabha, when the planning cycle was initiated.
The KRPs were trained by the State Planning Board, and were given the
job of distributing publicity in advance of the gram sabha. They also
had to work with the local government official who would preside

ODonnell (1993), Chibber (1999); this sentence after Chaudhuri and Heller, 2003.
14 Kannan (1988).
Dr`eze and Sen (1995), Veron (1999). See also Parayil (2000).


The poor and the state

at the village meetings. According to one observer, the training exercise

in Kerala amounted to the largest non-formal education program ever
undertaken in India (unnamed person, quoted in Chaudhuri and Heller
2002: 6).
In the first year alone, in seven rounds of training at state, district
and local levels, some 15,000 elected representatives, 25,000 officials
and 75,000 volunteers were given training. About 600 . . . Key Resource
Persons received nearly 20 days of training. Some 12,000 . . . District
Resource Persons received 10 days of training, and at the local level more
than 100,000 persons received at least five days of training (Chaudhuri
and Heller 2002: 6). In subsequent years, training was targeted to members of the Scheduled Communities and women, and the governments
efforts were supplemented throughout by the activities of civil society organizations, including, most notably, the Kerala Sashtra Sahitya
Parishad (KSSP: the Peoples Science Movement). In many ways, the
KSSP was the progenitor of the training effort. Observers are agreed that
the State Planning Board has depended heavily on local-level experiments
that have been initiated over the course of twenty years by the KSSPs
mainly white-collar volunteers.
Whether these efforts have truly empowered the poorest is hard to
say. We lack studies that describe in detail how the gram sabhas have functioned: whether and how those people in attendance take part in substantive discussions; and whether and how the budget is spent in a manner
that improves the capabilities of poorer people (or the designated recipients). Nevertheless, we do now have the first reports of a programme of
research that has sought to interpret a remarkably complete set of attendance registers from gram sabhas in all 990 of Keralas gram panchayats.
Data sets are available for each of the first two years of the campaign, and
they provide information on the attendance rates of the Scheduled Communities and women. According to Chaudhuri and Heller, the data sets
suggest a very high level of participation it is possible that close to a third
of the households had an adult member who attended [a] gram sabha
(Chaudhuri and Heller 2002: 8). They also point to a dramatic increase,
from the first to the second year, in the social depth of participation. Not
only did the participation rates of two traditionally subordinate groups,
women and members of SC/STs, increase substantially, in the case of
SC/STs participation rates in the second year were higher than those of
the general population in 80 per cent of the panchayats (Chaudhuri and
Heller 2002: 8).
Chaudhuri and Heller interpret these base findings to mean two things:
firstly, that members of the subaltern communities are finding it privately

Protesting the state


rational to engage the local planning process; and, secondly, that the
increased participation of these two groups in 1997 needs to be understood both in terms of the increased budgetary allocations that were on
offer to them, and with regard for the concerted efforts at mobilization in
these communities that were made by the State Planning Board and the
KSSP in 1997. They also point out that those panchayats that recorded
levels of attendance that were significantly below the average were generally those: (a) where a larger geographical area discouraged some women
from making the journey to the meeting-place; and (b) where previous
levels of popular or labour-based mobilization were low.
Taken together, these findings would seem to confirm that regularized
sightings of the local state can lead some members of the weakest sections to find it worth their while to engage the planning process. These
sightings, however, have to be tied to a visible set of rewards or payoffs (direct or indirect), and they are most likely to persist when they
are situated within a sustained programme of political activism at the
grassroots level. This was missed in the first writings on the new public
administration, and to a large extent the politics of good governance is
still neglected.15 The data also suggest that high levels of literacy have
helped people to see the state in less restricted terms than would be
common in Madhya Pradesh. In Kerala, there is a greater understanding
of, and possibly sympathy for, the idea of the state that is put forward
on a regular basis by left-leaning, secular political parties. Whether this
translates into a real and lasting redistribution of power and resources is
unclear. The first stage of Chaudhuri and Hellers study does not address
this question, although they seem cautiously optimistic that such is the
case. What is clear, however, from elsewhere in India, is that when the
voices of the poor are treated with contempt, they are increasingly likely
to seek redress by capturing the state and turning it into an instrument
of sectional power.
I am the state
The title of this section is taken from Sankarshan Thakurs (2000) engaging biography of the erstwhile Chief Minister of Bihar (The Making of
Laloo Yadav: The Unmaking of Bihar). Laloo Yadav has become a figure

Tendlers (1997) work is more attuned to political questions (see chapter 5), if not
perhaps to political struggles, and it would be misleading to suggest that some of
the leading development agencies are not beginning to engage critiques of their work
which have made the charge of depoliticization. We shall have more to say on this in
chapters 8 and 9.


The poor and the state

of fun for the chattering classes in Delhi, and a hate figure for many in the
Forward Castes. In late-January 1999, when two of us were trying to gain
an entry to a suitable research village in Bidupur Block, Vaishali District,
a Laloo heartland, we chanced upon an elderly Rajput farmer who led
us to the banks of the Ganges in a village not far from the one we settled
on. When the topic of Laloo came up in conversation, he exclaimed with
real vehemence, and in a mixture of English and Hindi, that the man
was worse than Hitler, worse even than Sikander.
Farmers like this gentleman, of course, along with many Brahmans
and Bhumihars, have long been used to exercising economic and political (including cultural) power in their villages.16 Particularly in central
Bihar, which we discuss shortly, the upper castes had been able to convert
their control over land, labour and credit markets into control over the
political system and the use of public space. From an upper-caste point of
view, life was orderly and the political had been pacified. Members of the
Backward Classes were expected to defer to their employers or masters,
just as wives were compelled to obey their husbands. (An extraordinarily
high level of dowry deaths continues to speak to this pattern of oppression in the Rajput and Bhumihar areas of our field locality in Sahar.)17
The police and other local-level state employees were expected to do
the bidding of the Forward Castes, and these communities supplied key
members of the political class in Bihar. The higher reaches of the civil service, moreover, were staffed overwhelmingly by members of the Forward
Castes, including Kayasthas, and they had a reputation for running the
state in a technically competent manner that lasted until the late 1960s.
It is sometimes forgotten that Bihar is the state which provided India with
its first president, Sri Rajendra Prasad, and that the state was ruled for
the first fourteen years after Independence by just one Chief Minister,
Dr Srikrishna Sinha.
Seen from this comfortable perspective, the arrival on the political
scene of Karpoori Thakur and Laloo Yadav could only be seen as a
descent into chaos, or the dark ages (Kali Yuga), almost literally so
given the failings of the power sector in Laloos Bihar. In some respects
Laloo Yadav was Karpoori Thakurs protege, although he lacks the latters roots in Lohiaite socialism.18 Thakur was one of the first politicians
in Bihar seriously to challenge the power of the Forward Castes and the

Das (1983), Sinha (1991).

It was suggested to us that there might have been as many as thirteen deaths over a
period of ten to fifteen years.
A follower of Ram Manohar Lohia, the main leader of the socialist opposition to Congress
in the 1960s.

Protesting the state


Congress.19 He also gave Laloo Yadav his start in parliamentary politics.

He first gave the erstwhile JP activist and student leader a Lok Sabha
ticket in 1977 and then . . . a Lok Dal seat in the Bihar Assembly (Thakur
2000: 53). But the two men were also very different, and could be considered to pose very different threats to the established order. As Sankarshan
Thakur makes clear, Karpoori Thakur had a reputation for simple living
and a basic level of honesty. It was said of Karpoori Thakur that once
he was chief minister of Bihar his brother-in-law approached for some
financial help, perhaps a job in the government. . . . Having heard [his
case], Karpoori Thakur produced a fifty-rupee note from his pocket and
handed it to him saying, Of course I will help you. We are barbers by
caste, go and buy yourself a barbers tool set with this and begin earning
money (Thakur 2000: 183). Laloo Yadav, in contrast, has been charged
with corruption on a grand scale, with his antagonists pointing in turn
to the fodder scam, to his supposed acquisition of properties overseas,
and to his wifes (Rabri Devis) declaration that it was the profits from her
herd of cows that paid for the familys shopping complex in Salar Kalan
and four rather large houses in and around Patna (Thakur 2000: 184).20
There is a more important difference between the two men, however,
which has been pointed up by Shaibal Gupta. Whereas Karpoori Thakur
represented a more Sanskritised section of the backwards, those with a
certain level of education and economic power . . . Laloo Yadav [brought]
the cockney backwards to the forefront, a non-Sanskritised, earthy, rusty
section that spoke the local dialect and existed on the fringe of the market
(Gupta, quoted in Thakur 2000: 97). As is well known, Laloo Yadav first
put together a coalition of Yadavs, Kurmis and Muslims, and Bihar has
been a notably safe state for Muslims under his leadership.21 By the time
of the 2001 panchayati raj elections in the state, however, the first in almost


This is not to discount the contributions of Jaipal Singh in Jharkhand (the Jharkhand
Party provided the main opposition to Congress in Bihar in the 1950s), or of Jayaprakash
Narayan (whose JP movement is still celebrated in Bihar, including at the museum/shrine
to JP in Patna). Nor is it meant to marginalize the opposition provided by extraparliamentarians, including peasant leaders like Swami Sahajanand in previous decades
(Hauser 1994). It is simply to record that Karpoori Thakur brought the Backward Classes firmly and forcefully into electoral politics in Bihar in the 1970s and
For more on the fodder scam, see Nambisan (2000).
Laloo Yadav gave orders for the arrest of L. K. Advani, later Home Minister in the BJPled National Democratic Alliance government, when he entered Bihar in October 1990
on the penultimate leg of his Ram Rath Yatra from Somnath (Gujarat) to Ayodhya (Uttar
Pradesh). He also deployed the police and other forces to protect Muslims in the wake
of the disturbances that followed the Mumbai bombings in February 1993. Ashutosh
Varshney (2001) has written an important account of the geography of communal violence in urban India, arguing that it is most likely to occur where civic associations do not


The poor and the state

twenty-five years, Laloo Yadav was seeking to link the Muslim and Yadav
vote blocs to communities and leaders from within the lower Backward
Classes. Although the elections were not fought on party lines, it is clear
that his Rashtriya Janata Dal was under pressure from his erstwhile allies
Nitish Kumar and Ram Vilas Paswan,22 and that, much as Kanchan
Chandras work would predict (see chapter 1), Laloo Yadav was having
to offer a small share of the spoils to allies from within the Scheduled
But there were problems here, too, as we suggested earlier. The very
scale of Bihars patronage democracy was turning the state into an empty
shell, barely able to function in key respects. Some of the funds that could
be drawn down from New Delhi were being spent on tangible projects like
road-building, as we reported in part II, and were the subject of extensive negotiations in local political societies. Yet no determined efforts
were made to come to terms with the states pressing financial problems.
Thakur writes movingly of the awful state of the Dharbanga Medical
College and Hospital, where a lack of money and security allowed Cats
and dogs [to] reign in the labour room, [and where] open-stomach
surgery is performed next to open drains with Erasmic shaving blades
(Thakur 2000: xviii). He also goes on to quote a doctor, who notes, very
much in line with Chandras thesis, that Only two kinds of patients come:
those who are well-connected and hope to get free and special care, and
those who have nowhere else to go (Thakur 2000: xviii). This is a state,
too, whose Chief Secretary in 1998 received an open letter from N. C.
Saxena, then Secretary, Rural Development, Government of India, lambasting corruption at all levels, and declaring that, The State of Bihar is
being treated like a private property by those at the top . . . [and run by] a
lower-level bureaucracy [that] has no work ethic, no feeling for the public
cause, no involvement in the future of the nation . . . they have only a
grasping, mercenary outlook, devoid of competence, integrity and commitment (quoted in Thakur 2000: 146). He went on to note that while
his ministry had set aside more than 1,000 crore rupees for rural development schemes in Bihar in 19978, Not a single rupee has been sanctioned
by our department for drinking water schemes this year because the Bihar


transcend the HinduMuslim divide. It is not clear, however, that this general thesis can
account for the specificities of the Bihar situation, a state which moved from gruesome
killings of Muslims at Bhagalpur to relative peace in the mid to late 1990s. Political will,
rather than civic association, would seem to be the decisive variable here. See also Brass
(2003), Breman (2002).
Nitish Kumar jumped ship to the Samata Party, a close ally of the BJP in Bihar. Ram
Vilas Paswan founded the Lok Dal Shakti Party to mobilize mainly among Dalit groups.
The two men formed an alliance at the time of the 1999 Lok Sabha elections, but could
not sustain this unity against Laloo Yadav in the state Assembly elections in 2000.

Protesting the state


government has not been able to finalise procedures for buying pipes for
the last one year (Thakur 2000: 1467).
It is difficult not to share Saxenas outrage, or that of Sankarshan
Thakur. But we also need to understand the conflicting rationalities that
have produced this state of affairs. One of the reasons why Laloo Yadav
and his lieutenants enjoy popular support among the Yadav community,
and among the Backward Classes more broadly, is precisely because their
sighting of the state is so sectional. If there is no sense of the public cause
or the future of the nation this is largely because these terms have meant
little to the historically subaltern classes in Bihar. The state in Bihar has
too often presented itself to the poor or the socially excluded as an oppressor, or as the preserve of the forward castes. Part of Laloos appeal, or
the appeal of Mayawati in Uttar Pradesh, is bound up with the declaration that now it is our turn. No apologies are made for recruitment
policies that staff the state with Yadavs, Kurmis and even Muslims. Nor
are apologies made for making hospital beds available to supporters of
the extensive corps of political fixers on whom senior members of the
RJD must rely. For many of the low-caste men and women who now get
beds in Darbhanga Medical College and other hospitals, it might be the
first time they have gained admission to an important state institution.
So long as some doctors and bandages remain, their sighting of the state
might be positive, at least compared to what they knew before. In a situation where state resources are scarce, and when there is a recurrent
expectation that the Forward Castes might be returned to power, it can
be privately rational to pursue a form of politics, in the widest sense of
that word, which targets the state not as an abstraction but as a favour, a
bed, or a job.
In the longer run these modes of economic and political management
will not be sustainable, and it is possible that the government of Bihar
will have to pay closer attention to funding sources in New Delhi, within
the state, and even abroad.23 But this is not yet of great concern to Laloo
Yadav, or even some of his supporters. When two of us interviewed the
Chief Ministers husband in Patna in May 1999 he embarked on a lengthy
diatribe against development, which at one point he declared was leading to holes in the ozone layer. The more serious import of his remarks,
however, was that development had more to do with patterns of social
exclusion than with economic growth. Laloo Yadav took pleasure in asking Professor-ji to take a look at the ten or so framed photographs that
were hanging on the walls of his study. What was the common theme?

Laloo Yadav made a visit to Singapore in the late 1990s, ostensibly to drum up support
for foreign direct investment in Bihar.


The poor and the state

The answer, which sadly eluded the academic, was that in each case he
was being taken off to prison. Laloo Yadav makes a good deal of this.
He is the anti-establishment figure, the lowly cowherd who embodies the
authenticity (rusticity, perhaps) of rural India, and who speaks back to
those who look down on the common people.
Nandini Gooptu reminds us that: The extension of the scale of celebration of Holi [in UP in the 1920s and 1930s] reflected the concern
of the shudra poor, first, to resurrect a supposedly lost kshatriya power
and glory, secondly, to uphold Lord Krishna both as an icon of martial
prowess and as a patron deity of the poor and powerless, thirdly, to celebrate the triumph of the good, and, finally, to enact the symbolic inversion
of the established order of power and hierarchy (Gooptu 2001: 213). The
political appeal of Laloo Yadav needs to be understood in this context.
It is too easy to write him off as a rogue or buffoon, as some are inclined
to do. Laloo Yadav rather needs to be understood as a Yadav leader who
dismisses the idea that the shudra poor in any sense deserve their present
fate. In his rough prose and flamboyant political performances, he calls to
mind both the idea of the beer, the brave but ascetic hero figure, and the
Dada, the muscular protector or boss of a neighbourhood. When Laloo
Yadav declared that he was the state, he was reminding his supporters
and his enemies of what Gooptu calls the courtly culture of shudra,
especially yadav, kings (Gooptu 2001: 217), and their associated displays
of strength. His tacit support for the Maoist Communist Centre (MCC)
in central Bihar (see next section), and his hurried visits to the sites of
upper-caste massacres of the Backward Classes, help to consolidate an
image of a political leader who refuses to bow to Kshatriyas or Brahmans.
Many Backward Caste men and women support Laloo Yadav precisely
because he articulates and personifies their aspirations to speak back,
to be virile. He is a charismatic leader in the secular and popular sense: not
a man who has divinely given talents or powers, but a man who projects
and writes into public culture the fantasies of an oppressed majority.
Sublimation becomes part of the sighting of the state that ordinary people make, and with that projection comes a barely concealed threat of
violence. The registers of this violence, moreover, which run from allbut-sanctioned revenge killings to the visual and aural symbolism of the
lathi yatra and rough speech, are registers which speak back to the violence done to the poor and excluded by the forward castes.24 Laloo Yadav
may have been less effective than Mayawati in writing public space with

Laloos use of language is neatly dissected in Thakur (2000); see Frontline for an account
of the Lathi Yatra of June 2003. A lathi is a bamboo stick or staff, a yatra is a march or

Protesting the state


statues (of Ambedkar in her case), but like Mayawati he has positioned
himself as an avenging leader. When his supporters in Bihar see the state,
they very often see him as well, and in him they might aspire to see a
stronger version of themselves. Less opaquely, perhaps, they have also
been empowered to draw the line when it comes to their engagements
with government officials and some employers. Laloo Yadav has built
successfully upon Karpoori Thakurs attempts to change the terms of the
poverty debate in Bihar. Instead of an emphasis upon jobs and services,
he has emphasized the issue of social exclusion and a persons sense of
self-worth or honour (izzat). And in this specific respect his politics can
be judged a success. Poorer men and women in Bihar might not expect to
gain much from the state, but some of them at least (perhaps the members of the creamy layers) no longer expect to be treated with disdain, or
abusively. In any case, for the Yadavs and Kurmis, especially, many of the
people they now see in the state are members of their own communities.
If Laloo Yadav is Sikander, they are his lieutenants and foot soldiers. We
are the state might be the new slogan for these communities.
Combating the state
The successes of some members of the Backward Castes in taking possession of the state in Bihar have not yet been matched by the Scheduled
Communities (although matters are different in Uttar Pradesh). Some
members of these communities prefer to keep the state at a distance, and
some have joined political movements the so-called Naxalite movements that see the state as an enemy. These movements, which began
in the Naxalbari area of West Bengal in 1969, and which are active also
in south India, particularly in Andhra Pradesh, position poorer people
and government officials as necessary antagonists in an ongoing struggle
over land and forms of rule. Government officials who are targeted in
this way are often defined in very broad terms. As we write, The Hindu
is reporting that four members of the Peoples War group have thrashed
a transport official in Gunter District, Andhra Pradesh, for amassing
wealth through corruption and getting his transfer orders cancelled. They
also torched his vehicle.25 Meanwhile, in central Bihar, Bela Bhatia has
reported several attacks on schoolteachers in MCC-dominated areas.26
The MCC has also been active in leading attacks against the armies
(senas) of leading Bhumihar and Rajput landlords, including the Ranvir

The Hindu, 27 May 2003, downloaded from the Internet on 30 September 2003.
Albeit in the course of an analysis of Naxalism in Bihar that is far from unsympathetic
to its subjects: Bhatia (2004).


The poor and the state

Sena in Patna, Jehanabaj and Gaya Districts. Some of its cadres have
also helped to enforce extensive areas of no cultivation, where organized landless labourers are under strict orders not to work the land of a
named employer, although in Bhojpur the lead in this respect was taken
by the CPI-ML.
Violent attacks on landlords or state officials are hardly new phenomena
in rural India. Certain fantasies persist about pacific Hindustan, encouraged perhaps by some Gandhian writings on the soul of India residing
in the countrys villages, but Paul Brass is right to remind us that many
villagers perceive the police, for example, as marauders, equivalent to
dacoits, not their protectors, but rather an additional, more powerful,
and more dangerous band of robbers than those for whom robbery is a
vocation (Brass 1997: 374). In his view, most villagers in north India
experience the external world as a Hobbesian state of nature in which
resort to force is at once natural and expected. Murders of state officials
are not uncommon in this world, and have long defined statesociety
relations. The dhandak (see above) is one example of this, albeit on an
individuated scale, while the so-called forest wars in Singhbhum District, Bihar (now Jharkhand) in the late 1970s and 1980s generalized
such killings over an extended spatial scale.27
What is perhaps specific to the Naxalite movements, however, is the
matter of their organization and ideology.28 Notwithstanding important
differences between them, the movements led by cadres of the CPI-ML,
or the MCC, or Peoples War, are infused with at least some degree of
ideological training which positions men and women from the Scheduled Communities, mainly, as foot soldiers in the class struggle against
feudal landlords and mahajans (money-lenders), and their protectors in
the government. In central Bihar, the activities of these various groups
have had three major effects, as the outstanding recent work of Kalpana
Wilson and Bela Bhatia helps to make clear.29
To begin with, the revenge killings that the MCC and some other
groups have organized have undoubtedly helped to dent the power of
local landlords. The flaming fields of central Bihar are dotted with
gruesome reminders of upper-caste attacks upon dalits and other poor



Areeparampil (1992); Corbridge, Jewitt and Kumar (2004: chapter 4). We should say
again that we do not agree with Brass in all respects. Political life in rural north India is
less anarchic than Brass suggests. The institution of the dhandak, indeed, which is not
one of Brasss concerns, is an attempt to re-establish a moral order of sorts, or to reclaim
politics from the political.
It is necessary to insist on the plural here, and to distinguish the tactics and changing
policies of the CPI-ML, say, in Bhojpur, from those of the MCC or a group like Peoples
War in Andhra Pradesh.
Bhatia (1998, 2004); Wilson (1999).

Protesting the state


people, as at Pipra, Arwal, Nagri and Shankarbigha, but recent attacks

on the Bhumihar-dominated villages of Chauram and Rampur-Aiyara
have demonstrated the willingness of labouring groups and the organized
Left to fight back.30 These killings, moreover, have been complemented
by insistent and often well-targeted attacks upon government officials,
including District Magistrates (DMs) and high-ranking police officers.
The DMs of Gaya and Jehanabaj Districts now travel in armed convoys,
hoping thereby to avoid the landmines that might be placed across their
paths. Fear of the MCC is now so great that various units further south
in Jharkhand, in Hazaribagh and Palamau Districts, have attempted to
demand money from aid agencies working in the area, including the Eastern India Rainfed Farming Project.31
The second objective of the ultras, or the extremists, as the government likes to refer to them, is to drive the state out of areas they
seek to control.32 The Peoples War Group (PWG) has been effective in
this regard in some of the forested areas on the borders between Andhra
Pradesh and Tamil Nadu. But in central Bihar, too, there are zones that
resemble a battleground. In the area of Sahar Block, Bhojpur, where we
worked in 19992001, the state was sometimes present in the form of an
occupying army of soldiers and policemen.33 The developmental state,
in contrast, has faded from view, and the few petrol stations left along the
heavily pot-holed road to Arras have been turned into virtual fortresses,
such is the lack of state control over some parts of the Block.
At the same time, however, this very part of Bhojpur District has been
the site of long and difficult, and in key respects successful, struggles by
the CPI-ML to raise the wage rates of agricultural labourers. Working
in the face of appalling violence and intimidation, cadres of the CPI-ML
sought periodically over the course of thirty years to organize labouring
households in strikes that removed a significant amount of cultivable
land from the effective control of its owners. Some Naxal groups have
also been able to declare the formation of Liberated Zones where order
is generated (or imposed) by the groups themselves. It would be wrong,


For more information, see:

Project officials in 1999 found it hard to tell whether it was the MCC making demands,
or whether gangs of disaffected youths were acting in its name. Nevertheless, the threat
posed by the MCC was real enough, and was one reason why the EIRFP pulled out of
Palamau District before the second stage of the project began in 2000.
Significantly, the government of Bihar refers to infestations of extremists.
It should not be assumed that these armies were always made unwelcome by poorer
people, or that they straightforwardly took the part of landowners. The connections
between Laloo Yadav and the MCC, in particular, remain deliberately open to interpretation, but there is no doubt that the ruling RJD regime has deployed paramilitary forces
to keep the landlords armies in check.


The poor and the state

however, to refer to these zones in uncritical terms. The fact remains

that most villagers are required to fund the rebel groups who live amidst
them, and some will see the rebel occupiers as little more than a quasistate, with all the latters powers to dispense summary justice. It is
also true that many villagers in the Liberated Zones remain as targets
for the landlords when the rebel units move on, as they very often do.
Thakur reminds us that the upper caste feudals have always been better
off for the simple reason that they are better entrenched. They have better
resources, better networking and their targets are always within their sight:
their own landless poor. . . . On a rough count, the Ranveer Sena, or its
variables, killed ten Harijans for every single Bhumihar life lost in the
so-called caste wars of central Bihar in the last decade (Thakur 2000:
166). Unsurprisingly, perhaps, for those families that can move to Delhi
or Punjab, to labour elsewhere, migration remains a more favoured route
for escaping the state in central Bihar.
Damaging the (urban) state
Migration, of course, is no guarantee of better treatment. Wage differentials between Punjab and Bihar are sufficiently high to induce a westward
flow of labour, but low-income migrants the world over tend to get a raw
deal from the state.34 This was and remains the case in many parts of
southern Africa, where the flow of black labour to white-owned mines
and plantations has helped reproduce outposts of urban-industrial capitalism amid predominantly non-capitalist modes of production.35 It is
also the case in California and Florida, where the gracious living styles
of the white middle and upper classes, including many Latinos in southern Florida, are made possible in part by the legal and illegal flows of
Mexicans, Cubans and Haitians into household and gardening labour
pools, as well as to the sugar fields and orange groves.36 And Jan Breman
has provided marvellous documentation over many years of the ways in
which agrarian and industrial capitalism in the Surat region of Gujarat
has been supported by the importation of cheap and unprotected adivasi
labour streams from the dryland interiors of the state, or from neighbouring Maharashtra and Madhya Pradesh. Some of these labourers dont get
to see the state at all, at least not in person. They are generally housed in
compounds attached to the houses of their employers, or in hutment


The state very often being in tow to ideologies of territory and the nation even as its
ruling classes demand the entry of migrant labour forces. For a speculative account of
the politics of the migrant hordes, see Hardt and Negri (2000).
See the essays in Wolpe (1980).
Davis (2001) and Gonzales (2001) provide useful introductions.

Protesting the state


colonies [where they] remain deprived of mutual contact with other

members of the labouring poor (Breman 1985a: 377). In any case, there
were only three labour officers employed by the Assistant Commissioner
of Labour to work in rural areas of Surat District in 1977, and these individuals liked to accept payments from [farmers] in return for neglecting
to register complaints and track down infringements of the regulations
(Breman 1985a: 303). As Breman points out, These state servants would
not be averse to helping the agricultural labourers but these, of course,
cannot afford the price of protection (Breman 1985a: 303).
It is unclear whether any (or many) of these labouring men and women
will have joined in some of the marches or demonstrations that have taken
shape under the banner of the new agrarian politics a politics which
claims to pit an authentic rural India (Bharat) against the neocolonial
powers of urban India. On the face of it, their participation would seem
unlikely, and quite possibly against their best (economic) interests. Most
of them will have an interest in lower foodgrain prices, not the higher
prices being called for by farmers leaders like Mahendra Singh Tikait of
the Bharatiya Kisan Union (BKU). This will also be the case for local
pools of agricultural labour and yet there is certainly evidence of men
and women from these communities joining in the new politics. The very
scale of some of the marches on Delhi that were organized by Tikait in the
1980s and 1990s, or by Sharad Joshi of the Maharashtra-based Shetkari
Sanghatana, suggests that labouring men and women must have jumped
on trains and buses to swell the crowds to several hundred thousand or
Ashutosh Varshney has sought to explain this seeming contradiction
within the framework of a rational choice model of politics.37 He suggests that members of the labouring poor will support the new agrarianism when they expect it to lead not just to higher food prices but also to
higher real wage rates. In these circumstances, it is possible that poorer
rural families will set aside differences with their local employers and
come to see an urban-biased state as the source of their poverty. Rational
political actions would then include demands for cheap water and electricity, higher farm output prices, and lower farm input prices, much as
Tikait has demanded. To the extent that poorer people in the countryside
can be provided with more heavily subsidized grains through the Food
Corporation of India/Fair Price Shops system, so much the better.
On the other hand, and this seems more likely to us, it is also possible
that poorer people have joined the new agrarian politics because they have
been coerced to do so by their employers, or because they have responded

Varshney (1995).


The poor and the state

to the symbolism and sense of social obligation of these social movements.

Jim Bentalls patient fieldwork in western Uttar Pradesh established very
clearly that Tikait was drawing support from Jat landless families far
more than he was doing from the numerically preponderant Scheduled
Castes (Bentall 1995). The BKU mobilized support within Jat khaps
(lineages), and by reinventing Jat concerns for the authenticity of soil
culture. But Tikait was also successful in making an appeal to political symbols and forms which generate and reinforce undifferentiated
notions of a community of commodity producers (Hasan 1998: 109).
The fact that this appeal edged at times into support for the Hindutva
project might not have discouraged poorer men and women, including
some Scheduled Castes, from taking part in a politics which so raucously
denounced the vampires who drained the blood of rural India: the merchants and petty traders, to be sure, but also the operators of electricity
sub-stations, and the urban-based officials who declared farm support
prices and who handed out permits to sell sugar cane to state-owned
Here, perhaps, is where the real strength of Mahendra Singh Tikait
resides. Further south, in Maharashtra, Sharad Joshi, an ex-UN official
and late arrival in the ranks of Indias farmers, has struggled to present a
more intellectualized account of the problems of Indian agriculture. To
be fair, Joshi has led effective and imaginative campaigns around particular crops: onions, for example, or tomatoes. And he is certainly able to
use evocative language. He told Bentall that urban India was run by the
black British, and that the countryside now laboured under the yoke of
a second colonialism.39 For the most part, however, Joshi has kept his
distance from campaigns for lower farm input prices, preferring to note
the damaging ecological consequences of cheap water and pesticides. He
has also kept his eye on the workings of the Commission on Agricultural
Costs and Prices (CACP) in New Delhi. Tikait, meanwhile, has sought
to fill the shoes of Charan Singh, the powerful Jat farming leader who
dominated UP politics in the 1960s, before he moved to the national
stage. More so than Singh or Joshi, however, Tikait has been prepared
to speak and act roughly. Much like Laloo Yadav, Tikait has been keen
to promote an ideology of avenging masculine power among his Kisan
Union activists. According to a Times of India article from 9 August 1989,
Despite his innocuous exhortations, the Chaudarys [Tikaits] followers
have repeatedly gone on the rampage; burnt down generators; refused

See Corbridge and Harriss (2000: chapter 5), Corbridge (1997). On the sugar permits,
see Attwood (1991) and Jeffrey (2001).
Interview on 26 March 1992, cited in Bentall and Corbridge (1996: 30).

Protesting the state


to pay official levies and dues; and attacked the police, often hindering
investigation into criminal offences (quoted in Gupta 1998: 355).40
The reporter goes on to suggest that, What matters [here] is the perception that all these actions are part of the helpless resistance to
the organized might of the state, something that he finds dangerous
[when] Mr Tikait succeed[s] in channelising this perception into organized protest (Gupta 1998: 355). But what he does not pick up on is the
pleasure of violence itself, or of vandalism, which is something we should
not underestimate. Thomas Blom Hansen writes of the importance of
social imaginaries, of desires of recognition, and the attraction of the
public spectacles of violence and assertion that Shiv Sena has employed
so successfully over the years (Hansen 2001: 7), and this insight can
usefully be generalized. Mahendra Singh Tikait and Laloo Yadav have
also understood the constitutive role of violence, and the sense of selfworth that many poorer people, and men especially, can gain from acts
of destruction or even self-destruction. Making demands of the CACP
will not mean a great deal to such people, but marching on Delhi has its
own attractions. Riding free on a train, bearding the conductors, claiming
the streets of the capital near the Boat Club Lawns, chipping away at a
statue of one of Indias great and good, all of these things can be liberating. They speak to an account of politics as transgression and also of
playfulness, even as they highlight what Scott has called the weapons of
the weak. They also speak to very direct sightings of the state, one of the
most concrete of abstractions as Marxists have sought to remind us.41
Reforming the state
The state is also seen by its employees and by those charged with reviewing their modes of conduct, including consultants for external agencies
or higher levels of government. We have made this point already, and
yet it bears saying here that there are important movements within the
state to push in the direction of pro-poor governance, or what might
be considered supportive of that goal. We have already reported the
contributions made by a more activist judiciary, and it is worth noting that one of the leaders of the MKSS, Aruna Roy, has a background
within the Indian Administrative Service. In her recent paper with Nikhil
Dey, she reports that attempts to link the MKSS to the Press Council


Not all farmers approved of such actions. One farmer with nineteen acres told Bentall
that, Tikait has upset the discipline of the farmers. Anyone can cut an irrigation channel,
anyone can burn a transformer. But I dont like that (cited in Bentall and Corbridge
1996: 39).
Marx referred to money as a concrete abstraction. For a discussion, see Harvey (1982).


The poor and the state

of India and the National Campaign for the Peoples Right to Information had been pushed forward in part because of some committed
bureaucrats and [because] the National Academy of Administration at
Mussourie . . . played a supportive and facilitating role in the process
(Roy and Dey 2001: 13). We await further work on the way in which
administrators at Mussourie are instructed on the human technologies of
rule which they are supposed to enforce in their future careers.42 It would
be surprising, though, if the rather traditional, and decidedly masculine,
forms of socialization that have made a virtue of academic excellence
and self-reliance were not being leavened now by courses that promote
a measure of reflection on the proper role of the state vis-`a-vis different
social groups.
Better training, of course, is unlikely to be turned into sustained action
when administrators are brought face to face with established elites in
rural political society, or where a training regime is unsupported by
incentive systems that reward pro-poor initiatives by individual officers
or teams. This might especially be the case in the state-level cadres of the
police and administrative services, where reform is perhaps most needed.
Government officials have to make a living, and even the best of them
will want to secure postings that allow them to educate their children
properly. That can be difficult in remote areas, or in the face of consistent pressures from local elites to transfer officers who cause them
But, again, smaller or larger spaces for empowerment can open up.
Arnold Harberger has argued that a handful of heroes can turn around
whole economies or societies, and it would be unwise to discount the
difference that someone as respected as N. C. Saxena can make in his
or her department (see above).43 And even if one leans more to a structural account of the spaces of social action, as we do, the story that we
related earlier about Polus B, the adivasi schoolteacher in Bihar (now
Jharkhand) who wanted to cut down and sell ten jackfruit trees, packs
a punchline that is worth recounting (see chapter 1). Polus B sold his
trees to a dalaal for just Rs. 20,000 even through the trees were worth
Rs. 80,000 net of logging and transportation costs. He did so because
he simply couldnt face the money, time and social (humiliation) costs of
dealing with various officials in the Forest, Revenue and Police Departments who had power over him. In 1993, however, a group of Forest

Potter (1996) remains the best account. See also Srivastavas (1998) interesting account
of the Doon School in forging the world-views of some among Indias elites.
Harberger (1993).

Protesting the state


Department officers in Bihar began a campaign to deregulate the trade

in those tree species that were commonly grown on private lands, and
which were not found in abundance in Protected Forests. The jack tree
was one of these species, and the officers made their case on the ground
that their time could be better spent regulating competitive timber markets. They also argued that the current system was stacked against the
interests of poorer people in Jharkhand, and on this basis, at least, they
managed to gain some support from sympathetic MLAs and from a new
Forest Minister who was keen to make his mark. The Transit Permit
system was relaxed on a trial basis in 1994, and after a short burst of
tree-felling the price of jack timber first stabilized and then increased in
real terms. Significantly, too, the rate of jack tree plantings also began to
increase at this time in response to new price signals. For the first time
in thirty years, a healthy number of new saplings were being put in place
alongside old trees that had begun to rot where they stood. Much as these
officers had supposed, a new regime of deregulation could have positive
social benefits for the poor while also leading to an overall increase in
homestead tree cover.44
We should not get too carried away with this success story, for the old
regime of nationalization was reimposed in 1996. A majority of officers
in the Forest Department had opposed the new rules, and they joined
together with a large number of dalaals and politicians to put pressure on
the government in Patna to rescind its orders of 1994. Powerful interests
are not easily dislodged, as we have repeatedly made clear in this book.
Nevertheless, there are fissures in large organizations like government line
departments, or between different branches of government, and these can
provide signposts to a form of politics that some within the state are keen
to exploit.
The same argument holds true for the World Bank or the UK governments Department for International Development (DFID). It would be
misleading to suggest that these organizations are not mainly responsible
to their shareholders or subscribers, and the geopolitical interests which
they are bound to uphold. The World Bank is not about to support Cuba
or Libya. That said, these are complex organizations, and it is possible
for agendas to be argued within them that might seek to empower the
poor in direct and possibly quite radical ways. The debate around the
production of the World Banks 2000/2001 report on attacking world
poverty brought some of these issues to the fore, and if the debate was
not finally settled to the satisfaction of the radicals it was successful in

Corbridge and Kumar (2002).


The poor and the state

bringing several matters into the open. Clare Shorts leadership of DFID
from its founding as a full ministry in 1997 to her resignation in 2003, following the USUK invasion of Iraq, also opened the way to new debates
on the direction and purpose of British lending policies. Her emphasis
upon the poor and the poorest, as well as on issues relating to gender
equality and sustainability, needs to be seen for what it was: an imperfect
but reasonably honest attempt to bring new voices to the table of development thinking and practice. These initiatives are now being followed
up by DFIDs Drivers of Change agenda, which promises to mainstream
considerations of politics and political economy into the departments
attempts to understand and effect pro-poor changes.45 In some small
way these voices are contributing to new sightings of the state in what
Escobar has called the economically less accomplished countries of the
Our major concern in this book has been to consider how poorer men and
women in India make sense of the state. By the state, of course, we have
generally meant the everyday state described by Fuller and Benei, or that
dull, routine, business-as-usual . . . state [that manages] literally millions
of transactions at the grassroots level (Oldenburg 2003: 28), and which
extends from the District Magistrate to the office boy of the . . . state
public sector Textile Corporation (Oldenburg 2003: 3) and on to the
shadow state described by Barbara Harriss-White. None of these sightings will be unmediated, for they always have regard for past experiences,
conversations with friends or relatives, newspapers read and so on, and
with some broader sense of the state idea. But the range of possible
transactions is immense, and it would be unwise to write off the possibilities for contesting actions around various statepoor encounters. We

In a background paper authored by Sue Unsworth on Understanding Pro-Poor Change

it is suggested that DFID and other donors find it easier to say what needs to be done
to reduce poverty than how to make it happen. In a draft paper for the Drivers of
Change initiative, the DFID team goes on to declare that Unsworths paper has enjoyed
remarkable resonance within DFID, and that there is now widespread recognition that
DFID and other donors need to make greater efforts to ground their programmes in
understanding of country contexts, including a good grasp of the likely relationships
between basic political, economic and social change over the long and medium term
(Government of United Kingdom, 2004, quoting Unsworth/DFID 2001). It might fairly
be argued that this recognition has come late in the day, and that DFID is still hedging its
bets on how to support pro-poor political organizations and agendas, but even a cursory
reading of this document will confirm that serious debates are being worked through
within the organization.

Protesting the state


have made this point, in part, by reference to Laloo Yadav and the politics
of izzat, but we should also make it clear that the state is challenged every
day in the small acts of resistance that people deploy against government
officials or systems of rule. James Scott is right in this regard. The meeting not kept, the fine not paid, the form of address that is refused all of
these can help poorer people to maintain a sense of dignity and self-worth
in the face of aggressive and overbearing officialdom.
This does not mean, however, that we are inattentive to the vast
and enduring asymmetries of power that structure many or even most
exchanges between the state and poorer men and women. Richer people
and higher-ranking state officials also come to meetings with an agenda,
assuming they turn up at all. They also use words and gestures to establish the field of their social power, and they sometimes join with others
to support political projects, including a lopsided version of liberalization and Hindutva, that Corbridge and Harriss have described as elite
revolts. In the face of these actions, it would be a disservice to the poor
to suggest that sustained progress in the direction of empowerment will
come without a fight, or by individual actions alone. The major lesson
we have been taught by the MKSS in Rajasthan, or by the CPI-M in
Kerala and West Bengal, or even by Laloo Yadav and the RJD in Bihar,
is that change comes through concerted struggle and from highly organized attempts to pare back the power of established social and economic
To the extent that members of the development community play down
the need for conflict and contestation they will be led to misleading
and possibly false conclusions about the relevance of a good governance
agenda. In Bihar, a reasonably effective campaign to free the state from
the power of the Forward Castes has not been accompanied by a sustained attempt to improve the states infrastructure. Nor has it improved
the material conditions of existence of Bihars poorest communities. The
states coffers continue to be looted, and the Scheduled Castes continue
to find themselves in weak positions when they compete for jobs or make
claims on sarkar. The fact that they now encounter Yadavs in the Block
Development Office and Rajputs in the fields is arguably of little consolation. For many labouring families the options are simply to survive, to
engage in struggles over pay and conditions at work, and/or to seek the
exit option by migrating to Delhi, Haryana or Punjab. In this context,
a form of government that eschews corruption, or which shares information freely with its clients, or which is effectively decentralized to the
Blocks and panchayats, would undoubtedly be welcome. As Adrian Leftwich has pointed out, however, to suppose that good governance of this


The poor and the state

type will come easily to a state like Bihar is misleading. The incentive
systems that are generally in play in political society militate against
changes in this direction, and there are few signs of a challenge emerging
to the conditions under which governmental business in Bihar is presently
And yet, having made this point, it is not inconsistent to say that the
good governance agenda can make a difference in a state like Bihar (or,
indeed, in Malda District, West Bengal), so long as we extend our field of
assessment to include developments that happen slowly, episodically, and
perhaps in only a few localities at first. The depoliticization thesis tends
to think about politics in terms of large structures and big events. Quite
rightly, it points out that the empowerment of poor people is difficult
to effect when large numbers of them, and poor women especially, are
unable to read or write, and when most labouring families lack assets that
they can use to acquire loans and social status. It also drives home the
point that richer and more powerful individuals can be expected to defend
their sources of privilege. What this thesis tends not to pick upon, however, are two propositions that point in the direction of Scotts weapons
of the weak, and the possibility that progress can be achieved by political
actions that are small-scale, that sometimes are unintended, and which
often happen at a lag.
These propositions derive, first, from the more Foucauldian account
of politics to which Scott comes close on occasions. This is an account
which thinks of politics as a vast palimpsest of competing actions and
counter-actions that take shape around the multiple capillaries of power
that bring poorer people, in this case, into contact with the state and its
technologies of rule. In addition, they derive from the more pragmatic
traditions of political (or policy) analysis that can be found in the community of development experts. As we shall show in chapter 8, there
are good reasons for thinking that members of this community, including
some employees of the World Bank, know full well that the good governance agenda will work best where political systems are also pro-poor.
They also understand that opponents of good governance will have to be
marginalized, co-opted or even bought off, as Judith Tendler showed in
the case of health-care reform in Cerea, Brazil (see chapter 5). Politics,

It is possible, as we indicated in chapter 6, that the Scheduled Castes will benefit from
the increased competition which is apparent within the OBC community in Bihar. Their
leaders might yet strike deals with representatives of communities (such as the Yadavs
or Kurmis) that have mainly thought about voter mobilization in horizontal terms. It
needs to be said again, however, that capturing some part of the state, or a patronage
democracy, is no guarantee that the state will be made to run more efficiently in the
medium or long term. This is the point of entry for the new public administration, for
all its faults and occasional unworldliness.

Protesting the state


that is to say, is not absent from this form of analysis, but is conceived
more in regard to tactics than strategy. It is perhaps also recognized that
not shouting about politics can be part of an effective political agenda,
particularly when attention turns (as it is beginning to turn in the development agencies) to the possibility of empowering poorer people against
the state. This is something we shall consider further in the concluding

Post-colonialism, development studies and

spaces of empowerment

Post-colonialism and political society

The discipline of development studies does not have a good reputation among students of post-colonialism. Indeed, it is hard to think of
two intellectual and political traditions that are further removed.1 Postcolonial scholars are deeply suspicious of the Eurocentric and depoliticizing instincts of development studies. This is a common thread in the work
of Partha Chatterjee, Arturo Escobar and James Ferguson, however much
they are divided on the possibility of development itself. Chatterjee and
Ferguson do not fully share Escobars pessimism about the past fifty years:
the age of misdevelopment that supposedly brought about only famine,
debts and immiseration. But they do insist that the ambitious plans of
the development industry are repeatedly frustrated by structures of power
and politics that are opposed to easy talk of citizenship, good governance
and benign economic growth.
In the everyday worlds of popular politics, Chatterjee maintains, deals
are struck by poorer people with those who mediate for them in exchanges
with the state and governmental agencies. This is the dirty and sometimes
dangerous world of political society. For Ferguson, meanwhile, the necessary and repeated failure of development projects to secure their stated
aims is linked to the extension of state power over potentially rebellious
populations. The development business, and the counterpart discipline
of development studies, is neither ineffective nor especially insincere, but
its power effects are often profoundly disempowering for poorer people. Worse still, perhaps, the rhetoric of participatory development and
good governance is pushed down the throats of people who experience

Sylvester suggests that development studies does not tend to listen to subalterns and
postcolonial studies does not tend to concern itself with whether the subaltern is eating
(1999: 703). This is an exaggeration, of course, indeed a caricature (consider the work
of Chambers 1983, Narayan and Petesch 2001, and Parpart, 1993), and we shall need
to deconstruct it later in the chapter. Nevertheless, it does capture a prevailing sense of
stand-off: see also Briggs and Sharp (2004), Corbridge (1993a), Goss (1996), Mohan
and Stokke (2000).


Post-colonialism, development studies and empowerment


authority in entirely other and more troublesome ways. In place of the

hard work of politics and confrontation they are offered the candy-floss
of good intentions and the traumas of structural adjustment.
Our work in eastern India suggests there is a lot of truth in these observations. The Government of India has made play over the past fifteen
years of its agendas for a new form of public administration. These agendas have embraced the schemes we have discussed here (the Employment Assurance Scheme, Village Education Committees and Joint Forest
Management) as well as plans for the privatization of service provision
in areas like urban water, sanitation and electricity supply. Taken in the
round, they are meant to engineer a new set of relationships between
government(al) agencies and members of the public. The aim, increasingly, has been to create an informed citizenry that is at once able to make
demands of the state (including for employment and oversight of public
officials), and which is also required to stand on its own feet (in the sense
of taking responsibility for making payments to service providers).
Similar proposals for good governance have been in play in countries
across the developed and developing worlds, and have given rise to
a huge and often critical literature. Chatterjee might be right to suggest
that civil society is a western concept, but the construction of fully civil
societies is hardly complete in countries like the UK and the US. Poorer
people in London and Los Angeles make use of political society to access
some of the services they consider to be theirs as of right, if not always
by law.2 It is clear, nonetheless, that the good governance agenda faces
particular problems in Districts like Vaishali and Malda in eastern India.
As we showed in part II, rates of participation in the EAS in these Districts
are dismally low, both in terms of labour days gained by poorer men
and women and in terms of the workings of village open meetings. Such
meetings as are held are very often captured, as most students of eastern
India would expect, by members of socially or economically dominant
groups, as well as by political fixers. They are also run by and for men
rather than women. The intended beneficiaries of the EAS, for their part,
are not always unhappy that decisions are taken for them, or that resources
are diverted to the construction of pucca buildings or roads, as they were
in Bidupur.3
Much depends on where and how the monies are spent, and on
how people think about the opportunity costs of more direct forms

See the interesting discussion in Amitava Kumar (2002); also Davis (1991).
By the same token, the fact that people are sometimes unhappy with these arrangements,
and do complain, suggests that they are not always content with the forms of mediated
politics that dominate in political society. They also have a sense of their rights as citizens
or as members of a population.


The poor and the state

of participation. Precisely because poorer men and women in forestdependent villages expect the benefits of JFM to accrue mainly to richer
households, they refuse to participate fully in JFM executive committees or forest guarding activities. As we explained in chapter 3, poorer
people also have close regard for the need to maintain their non-state
social networks. They have well-informed expectations about the enduring nature of power asymmetries, and will think twice about making
demands that would offend an important employer or patron. Injunctions
in favour of participation will often fall on deaf ears in these contexts,
as will suggestions that government officers are being retrained to serve
poorer people as clients or citizens. As we have stressed throughout this
book, ordinary people form their sightings of the state at least in part
through their everyday encounters with lower-level government officers.
Villagers know that if they want to get a handpump built, or a house
constructed under the Indira Awas scheme, they might have to bribe
the VLW.4 And when it comes to dealing with more senior government
officers, they will expect meetings to be brokered by an intermediary.
Many of them will also expect to be treated as subjects rather than as
citizens or clients, even if they want more even exchanges. In Malda
District, indeed, the attitude of schoolteachers to parents (see chapter 5)
is most often one of grudging compliance; a sense, perhaps, that members of a frankly uncivilized crowd have on occasion to be humoured
or listened to, even where their complaints and suggestions are later
It is said that power corrupts, and it is certainly the case that the good
governance agenda is corrupted by existing power-holders in eastern
India, as well as by a lack of investment in state capacity. Ferguson is right
to ask questions about why such pallid and contradictory discourses are
produced time after time in a world riven by enormous power inequalities.
Chatterjee is also right to maintain that it is the identity and qualities of
the agency that mediates power in political society that is often key to the
livelihoods and sense of dignity of poorer people. Who claims to speak
for who, and on what terms?, he asks.
In Midnapore, as we have seen, it is the CPI-M which largely plays
this role and which dominates political society, sometimes to good effect.
Evidence from our Debra field sites suggests that Atul Kohlis arguments
from the 1980s are right in important respects: poorer people can be
empowered by a well-organized, left-of-centre political party that puts
pressure on (or directs) the local state to commit resources and power to

This knowledge also speaks to a well-informed and in some respects sophisticated account
of how different government agencies work in practice.

Post-colonialism, development studies and empowerment


the weaker sections of society.5 Kohli made this argument with reference to the land reform campaigns of the 1970s and early 1980s, and to
the Left Fronts commitment to the registration of sharecroppers (Operation Barga) and the politicization of local government (the red panchayats). Our work has complemented this and other previous studies
by highlighting the ways in which the CPI-M in Debra Block, Midnapore, was able to position itself as a major supplier of information, advice
and work (through the Employment Assurance Scheme) to members of
poorer households. Men and women in Debra Block were more than
three times as likely as villagers in the four remaining study sites to gain
work from the EAS (although a majority of them remained uninformed
about the schemes main provisions). The children of poorer households
in Debra Block were also more likely to attend functioning public schools
than their counterparts in Old Malda or in Bihar, and parents were more
able to exercise control over schoolteachers than was the case elsewhere.
This was true notwithstanding the close links between the CPI-M and
the teaching unions.
In some respects, therefore, poorer people in Debra Block came to see
the state in more positive terms than did poorer people in the remaining
field areas. Notwithstanding the cynicism about politics that Arild Ruud
reports for Burdwan District, West Bengal, and which we found often
enough in Midnapore, it remains the case that many poorer people in
these red fortresses (Ruud 2001: 122) come to see the state in terms of
the abstractions provided by the CPI-Ms actions and rhetorics, as well as
with reference to the Indianized accounts of politics that Ruud describes
(Ruud 2001: 1314). We would maintain, indeed, that the CPI-Ms insistence on making the state work for the poor amounts to something more
than politics as usual, or the polluting activities of politicians who trade
in deals rather than principles. Politics might be dirty in several respects,
and effective politicians are certainly those men and women who get
things done, but in Debra Block we detected something else besides.
The effectiveness of the CPI-Ms party workers was also judged by their
superiors, and with regard for a model of politics that was ideological,
disciplined and rule-oriented (see chapter 6).
This much was apparent in terms of the partys control over the
Employment Assurance Scheme in Debra Block. Efforts were made to
direct resources to people who could be defined as poor in income and
class terms, even as efforts were made, as well, to manage information

There are also problems with Kohlis analysis. Drawn primarily from interviews with
the upper echelons of the CPI-M, it portrays a party that is more coherent and more
ideologically committed than that which exists on the ground today. For an evaluation of
Kohlis work on West Bengal, see Williams (1999).


The poor and the state

flows in such a way that poorer households would not form unrealistic
expectations of the state. The CPI-M encouraged disadvantaged people to see the state (and thus the party) as an institution that could be
trusted to take the part of the poor, and to make goods and services available to them in a reasonably efficient and impartial manner.6 In important
respects it was old-fashioned, which is to say that it refused some of the
claims that are now advanced in favour of participatory development or
empowerment. The party insisted that class was more important than
gender, and that the delivery of development benefits counted for more
than primary education provision.7 It also refused Amartya Sens rather
easy suggestion that the maximization of freedoms along any one dimension will lead to an enhancement of freedom along all other dimensions.
The CPI-M proposed a step-wise model of politics which insisted on
tight party discipline and information control as a means to effecting
even minimally pro-poor governance. It proposed, in effect, and perhaps
quite reasonably, that poorer people had to be protected (or empowered)
not just against other classes, but against some of their own dispositions
and desires.
In Bihar, in contrast, desire would seem to be openly at the heart of
the politics prosecuted by Laloo Yadav and the Rashtriya Janata Dal. To
the extent that Laloos supporters form a favourable view of the state
it is often in terms that are personalized (letat cest moi) and by means
of a performative politics of aspiration. Vocabularies of acclamation are
phrased in terms of taking (back) control over the state, or of making the
state work for the private financial and emotional interests of groups like
the Yadavs or Kurmis. This renders them no less authentic, of course,
but, equally, there is no need to mistake these important signs of sectional
empowerment for a politics that is pro-poor in a wider sense, or in the
sense of being oriented to the Scheduled Castes.
At the same time, however, as we have also explained, the contrasts
between Bihar and West Bengal can be drawn too starkly or onesidedly. Poorer villagers in our Ranchi field site have benefited from the

This is not to say that poorer people always internalized this sighting of the party or the
state. Our work suggests that party workers are trusted to a degree, but poorer people
also understand that key decisions often remain beyond their reach.
During fieldwork in West Bengal in 1997, Williams was repeatedly told by CPI-M representatives that the party had solved the gender issue simply by the reservation of
panchayat seats for women. Of course, such empowerment from above often changes
precious little on the ground, and most grassroots activists saw the partys mohila samiti
(womens front) as an organization of limited importance; it was concerned with activities
that were limited to the domestic sphere. More generally, as we have stressed in chapter
6, the role that class struggle plays within the CPI-M is now more about preserving a
political order than it is with radically transforming it.

Post-colonialism, development studies and empowerment


efforts of the Mukhiya to put pressure on the local state to direct resources
to his panchayat, even if this was at the expense of other localities in the
Block (see chapter 5). In Bidupur Block, Vaishali District, meanwhile, we
have seen that a politics of desire and acclamation takes shape within
a political society that is more competitive than in Debra Block. For the
middling castes, especially, the state is brought into view, and made to
do considerable governmental work, on a basis that secures a relatively
even allocation of resources to panchayats and between non-SC villages.
Neither here nor in Bhojpur is the state dominated by a small number of
actors. State failure is apparent in both Districts, along with a measure
of looting, but it is no more apparent than it is in Old Malda Block, West
Bengal, where government agencies have been captured by members of
the locally dominant elites. Public schools function badly in Malda, and
health-care facilities are miserable. The physical landscape of Old Malda
Block tells a tale of neglect and exploitation that is worse even than in
Bidupur, where the local MLC did at least make sure that roads were
built. The shadow states that Barbara Harriss-White describes in Tamil
Nadu are out in the open here, and speak to a pervasive criminalization
of politics that does few favours for the rural poor.
Development studies and civil society
Post-colonial (and Marxist) critics would be right to point to the limitations of a good governance agenda in many Districts across eastern
India. Poorer people often have to protect themselves as best they can by
sidling up to mediators or protectors in political society. This is no easy
matter when that society is dominated by the very individuals or groups
who are opposed to their empowerment in a more structural sense. Most
of the participation that we observe in a Block like Bidupur (including
meetings to hand over small payments to VLWs, or meetings between
village leaders, contractors and BB) takes place in political society. We
should not conclude from this, however, that political society eclipses civil
society in eastern India, which seems to be Chatterjees assumption, or
that the development industry, and development studies, is committed
to a project of depoliticization, which is the thesis maintained in different
ways by James Scott and James Ferguson. We want to close the book (here
and in chapter 9) by saying why we find these arguments problematic. We
do so first with regard to theoretical and empirical arguments. We then
turn to questions of politics and public policy.
We can start with the distinction that Chatterjee maintains between
political society and civil society. This is an instructive distinction, and
Chatterjees focus on the governmental rights of illegal and paralegal


The poor and the state

populations is a necessary corrective to glowing homilies about the unending construction of modern citizens. As he puts it at the end of his first
Schoff lecture: it is morally illegitimate to uphold the universalist ideals
of nationalism [and citizenship] without simultaneously demanding that
the [group-based] politics of governmentality be recognized as an equally
legitimate part of the real time-space of the modern political life of the
nation (2004: 25).8
But like all binary distinctions, this one is also overdrawn. Our discussion of rent-seeking behaviour in north and central Bihar (chapter 5)
shows that government officers cannot always be classified as if they were
members of either elite or vernacular society. They face competing pressures from different groups in civil and political society, as well as from
within the departments or agencies they work for. These pressures caution them against excessively predatory and sectional forms of behaviour.
We also showed how the CPI-M in West Bengal has been expanding a
sense of citizenship for villagers in Midnapore, even as it directs many
of the demands they can make on the state as members of a population.
We would maintain, indeed, that the gram baithaks that we reported in
chapter 4 are one means by which a purposeful and well-organized mediator in political society can use its power to open up parallel spaces for
individuals or households in civil society. Perhaps most of all, however,
we would argue that at least some of the transactions that we observed in
Vaishali, including meetings between BB and the District Magistrate to
agree a new list of EAS schemes, are conducted in that uncertain space
where the workings of political society are informed by the precepts of
civil society.
It is not just a question, then, of who can mediate? (Chatterjee 2004:
64). What matters just as much is how mediation occurs, and to what
extent and how it promotes (and draws upon) ideas of citizenship. Politicians like Mayawati and Laloo Yadav enjoy well-deserved reputations for
linking governmental politics to sectional agendas. Farmers groups have
also been beckoned into existence by the exercise of governmental power

It will be helpful to know that Chatterjees target (or foil) in his first lecture is Benedict
Anderson. Anderson, Chatterjee suggests, is committed to the view that liberatory politics
will only be found in the unbound seriality of the everyday universals of modern social
thought: nations, citizens, revolutionaries, bureaucrats, workers, intellectuals, and so on
(Chatterjee 2004: 5 summarizing Anderson 1998). By contrast, the unrelieved nastiness
of ethnic politics (Chatterjee 2004: 6) is a product of the bound serialities that produce
governmental populations. Chatterjees point, in contradistinction, is that real people do
not live in the empty and homogenous spacetime of capitalism/modernity/the nation
that Anderson celebrates. Instead, The real space of modern life consists of heterotopia
(Chatterjee 2004: 7), and this is where we find the real and dirty worlds of everyday
politics, or the politics of the governed.

Post-colonialism, development studies and empowerment


over farm input and output prices, meant in part to serve the needs of
the urban poor and the BPL population. Mediators such as these have
worked to empower the Backward Castes with a sense of pride and honour, or to protect the interests of different populations of farmers and
It can still be argued, however, that the forms of mediation proposed
by the RJD and BKU are substantially different to those offered by the
MKSS in Rajasthan, or which are advanced by the KSSP and CPI-M
in Kerala. What we see in these cases, supported by a national right
to information movement and by rigorously decentralized structures of
governance, is more akin to the construction of citizenship and the direct
empowerment of individuals, households and communities. These acts
of construction, moreover, which are always tenuous, and which are open
to reversal, are supported in rural India, as they have been in many other
countries, by more general processes of education (the acquisition of literacy, the ability to make better use of information) and participation in
labour markets (particularly where contracts are more formal or impersonal). They are also supported by the right to vote periodically in elections, and by often fitful understandings that individual men and women
have rights as individual men and women.
There is a wider argument to make here, as well. Our work in eastern
India does not incline us to the view that civil society is more important in
the politics of the governed than is political society. Although we found
significant examples of engaged participation in Districts other than
Midnapore including autonomous village forest protection committees in Bero Block, Ranchi District, and a Village Education Committee
made up mainly of Yadav parents in Bidupur Block, Vaishali District we
found many more examples of participatory failure, at least in the terms
demanded by some development agencies. The embeddedness of many
government officers in local society also militates against quick and easy
mobilizations of a good governance agenda. At the same time, however,
we are troubled by the elements of closure that are built into a model of
political versus civil society. By declaring so firmly that civil society is a
western concept, and that the politics of civility are enjoyed in India only
by the closed circles of various elites, there is a danger that the governed
are denied any means of access to what Chatterjee tellingly describes as
the walled up . . . enclaves of civic freedom and rational law (2004: 4).
Where else is civil society made if not in political society?
What we detect in Chatterjees work is a premature form of closure that
is at odds with the multiple and hybrid sightings of the state that are made
by ordinary people in most of the world. When Chatterjee asks who
claims to speak for who, and on what terms? he takes care to distance


The poor and the state

himself from the suggestion that a poorer person, household or community stands to benefit in equal measure from the forms of mediation that
might be effected by an employer/moneylender, by the apparatchiks of
the Shiv Sena, or by cadres of the CPI-M. Each of these agencies can
secure governmental benefits for a named household or social group,
but they will do so in ways that demand different forms of reciprocity
from the beneficiaries. The normative, in other words, is not refused
by Chatterjee, even when political society comes close to being celebrated en masse. What Chatterjee tends to underestimate, however, is the
possibility that the slow-burning processes of democratization and secularization, along with some of the initiatives that are being put into play
by proponents of good governance (or the new public administration),
are creating new spaces for empowerment and citizenship within what
he calls political society. And there are parallels here, we will maintain,
with the discursive strategies employed by James Ferguson in his arresting account of the substitution of bureaucratic policy for popular politics
in the development of Lesotho.
Ferguson, we suggested in part 1, has written an important work of
social science, one that has concerned us throughout this book. He shows
with great clarity how development agencies in Lesotho combined with
the government of that country to depoliticize its public life. Development was apparently reduced to a series of technical interventions,
many of which failed in their own terms, but which worked collectively to
extend bureaucratic state power over previously marginal social groups
and areas. Ferguson also gives some thought to the question of what is
to be done? in the face of these developments, and he refuses to endorse
the simplistic nostrums of the anti-development school. Power has to
be engaged, it cannot simply be wished away.
At the same time, however, Ferguson imposes on his work a form of closure that puts us in mind not only of Chatterjee, but also of James Scott,
yet which in important respects is more severe. Scott, after all, maintains
in Seeing Like a State that the skills and social practices of poorer people
will be disregarded only when a high modernist regime is able to trample over civil society. Where this is not the case, we must presume that
the intentions or plans of the state will be disrupted by lower-level public
officials and by the weapons of the weak, much as we have seen in eastern
India. In Fergusons account, the closure is effected with regard to what
he calls the the development apparatus more generally. Readers of The
Anti-Politics Machine will recall that the empirical part of the book opens
with a one-paragraph citation from a World Bank country study of Lesotho that was published in 1975 and which purported to describe Lesotho
when it gained independence in October 1966 (Ferguson 1990: 25).

Post-colonialism, development studies and empowerment


Ferguson has no trouble in showing how the major assumptions of that

report, at least as indicated by its opening paragraph, constitute Lesotho
as a blank page upon which people living in a subsistence economy can
be developed with the help of foreign aid and technical assistance. The
report does this by blithely disregarding questions of landownership, the
scale of the cash economy, and the past histories of forced labour flows
within the colonial economies of southern Africa hence the charge that
it depoliticizes public life.
Now, this might well be a correct reading of how the World Bank saw
Lesotho in 1966/75, but we need to recognize how this one paragraph
is made to do much more work in Fergusons book, and how this work
supports a particular, and in our view controversial, reading of the relationships that prevail between the development industry and directed
social change. Ferguson asks his readers to follow him in making three
assumptions, each of which needs to be rendered problematic. First, we
are asked to assume that the World Banks view of Lesotho, even in 1966,
is accurately and completely represented by the one paragraph that Ferguson cites and reproduces. Second, we are asked to assume that what
might have been an uncontested World Bank account of development in
Lesotho in 1966/75 continues to structure, indeed dominate, the Banks
lending policies and decision-making into the early 1980s, when Ferguson was conducting fieldwork in Lesotho, and the late 1980s when he
was writing his book. The assumption constitutes the Bank as a unitary
institution which displays a singular incapacity to learn. Third, Ferguson
invites us to believe that aid workers in Lesotho throughout this period,
including those men and women who worked for the Canadian International Development Agency (CIDA) on livestock and range management
projects in Thaba-Tseka, either fully internalized the assumptions made
in a World Bank country report of 1975, and/or were unable to reshape
or contest these assumptions in their working lives. To put it another way,
while Ferguson encourages us to believe that the intended beneficiaries
of CIDAs livestock interventions are able to see through and undermine
the so-called bovine mystique (he does this quite brilliantly and with
lacerating wit: Ferguson 1994: chapter 7), he fails to extend the same
courtesy to the aid workers he spent time with. Oddly, for an anthropologist, these men and women are not allowed to speak for themselves. As
a result, they come across as guileless at best and possibly as dupes or
even knaves.
The approach to development agencies and discourses that we have
taken in this book is rather different. We accept that most of them have
blind spots, and are not about to support left-wing politics in the periphery. Many institutions will draw the line at supporting left-of-centre


The poor and the state

political parties that work within a democratic framework, as in Kerala

or West Bengal, albeit in part for legal reasons. We also accept the force
of John Harrisss suggestion that an agency like the World Bank will periodically embrace ideas that serve to depoliticize development, and not
just ideas about the economy (Harriss 2001). This much was evident in
the Banks attempts to turn Robert Putnams ideas about social capital
into the missing link of development, notwithstanding the logical and
political flaws in an analysis which fails to see the mutual dependence
of social capital and other forces of social change. It was also apparent
in the literatures that surrounded the Banks first attempts to promote
a strong agenda of participatory development. As we saw in chapter 4,
some of the very real arguments that need to be made in favour of specific forms of political and economic participation get lost amid a more
general rhetoric about the virtues of participation tout court. Much like
the agenda of good governance (chapter 5), the mainstream perspective
on participatory development presumes what it must demonstrate: that
strong individuals flourish in strong civil societies, and that distinct centres of bureaucratic power (or governance) can be brought under their
control. Politics gets lost in this mix, along with accounts of the structuring of local political societies. At its worst, politics is simply factored
out as a disturbance, and the skills of key actors in political society are
bluntly disregarded.
At the same time, however, we want to insist that the broader development community is less homogeneous than it is made out to be by
post-colonial critics. It is not only implicated in the politics of the governed, but is by turns a prominent source of critique of particular forms
of governmentality. It is important to note that governmentality is not
something that any of us can escape from, not, at any rate, if we want to
enjoy the benefits of welfare, citizenship or extended divisions of labour.
What matters are the forms in which the politics of governmentality is
played out. How, if at all, is power shared in political society? How are
individuals and agencies made accountable, and to whom? How is the
economy managed, and with what effects on incomes and livelihoods?
Questions such as these are the very stuff of development studies and
related policy-oriented disciplines, and necessarily so. But they are also
fiercely contested within these broad communities. The major development institutions are committed to fiscal rectitude and the promotion
of market-access economies (for both of which strategies strong arguments can be made).9 But they are also pushed by activist groups, staff

Albeit these arguments are sometimes negative in form and focused on the medium to
long run (fiscal ill-discipline hurts the poor in the longer term).

Post-colonialism, development studies and empowerment


members and contracted academics to reflect more seriously on development discourses that are political and which do challenge the status
quo, including support for accountability campaigns.10 The World Banks
World Development Report, 2004, devotes a Box (10.7) to the question
of Managing the thorny politics of pro-poor service delivery reforms,
and includes in its discussion the possibility of Marginalizing opponents before, during and after implementation, particularly those with
veto power (World Bank 2003: 198). It also deals explicitly with the
question of corruption, which is described as a regressive tax, penalizing poor people more than others (World Bank 2003: 196). Partly in
response to this, the Bank commends the use of public expenditure tracking surveys, shorter routes to accountability that link service providers to
clients directly, greater spending on health-care and education (so long as
monies are directed to efficiency gains and the removal of bottlenecks),
and the use of public-sector solutions (including the contracting in or out
of services) where market failures are likely.11
Our point, of course, is not that the World Bank doesnt also speak
loudly in favour of macro-economic policies that are broadly neoliberal,
or that we think it has brought sufficient pressures to bear to fund interventions that could reasonably be described as pro-poor. The Social
Development Unit of the Bank hardly competes on a level playing field
with divisions staffed by neoclassical economists and political scientists,
or which take their lead from their paymasters in Washington DC. Nor
do we think that shorter routes to accountability will be easy to construct in political societies of the sort we have encountered in Bihar and
West Bengal. The point we are trying to make has to do with politics
in a broader sense, and with what might be called the methodological implications of false closure. Fergusons work is instructive in this
regard, because for all its undoubted brilliance it buys into a conventional
left-cum-post-developmental view that both homogenizes development
theory, and, curiously, fails to recognize the full effectivities of what we
would prefer to see as an evolving and contested body of work. Governmental practices built around participation, accountability, human rights
and other technologies of rule will often fail to meet the needs of poorer
people who have to secure most of their needs in political society. But this
doesnt mean that the unintended consequences of such practices work


Amartya Sen, Mick Moore, Robert Chambers, Barbara Harriss-White and Robert Wade,
for example, have all carried out work in recent years for the World Bank, UNDP, and/or
DFID; it makes no sense to press them into service alongside the late Walt Rostow, say,
or even Anne Krueger, as exponents of some hopelessly teleological and managerial
account of the political economy of uneven development.
See also Devarajan and Shah (2004).


The poor and the state

exclusively to secure the greater power of the state. (In any event, the
extension of state powers should not always be linked negatively to the
senses of well-being of ordinary people. Much will depend on how that
power is used and contested.) Our work suggests there are reasons for
believing, as well, that governmental practices such as these can, slowly
and unevenly, be instrumental in providing poorer people with a greater
sense of self-worth, dignity and, more rarely, a degree of power over those
who would govern them. They can be made to work, that is, to change
the contours and effects of political society.
Spaces of empowerment
All of this matters, finally, because it impacts on the way that we might
think about the politics of what the development community likes to call
pro-poor governance. James Scott is careful in Seeing Like a State not to
dismiss development as such, as opposed to its high modernist variants.
And James Ferguson likewise takes care to conclude his book with some
reflections on where and how western academics, and others, might seek
to engage the asymmetries of power and knowledge that construct certain
peoples as objects for development (see chapter 9). In a very real sense,
however, the political trajectories that are pointed to by Ferguson are only
thinly fleshed out. The same might be said of Partha Chatterjees work,
for all that we admire it.
It is important that development doctrines are understood as forms
of governmentality and are subjected to rigorous critique. It is equally
important to insist on the vitality, and indeed validity, of the unbound
serialities that Chatterjee is moved to celebrate in political society. But
none of this means that we shouldnt also attend closely to the careful
accounting of political and developmental possibilities that is served up
in a report like the World Development Report, 2004. The authors of that
report sharply oppose the sort of one size fits all approach to policymaking that some critics, rightly on occasions, have associated with the
World Bank. They rather contend that public bureaucracies are likely to
play a positive role in the organization of modern life only where political
systems are pro-poor; where politics are clientelistic the task of policymakers should be to strengthen the voices of the poor and to provide nonstate options for livelihood enhancement. More generally still, the Bank
recognizes that, Despite the urgent needs of the worlds poor people,
and the many ways services have failed them, quick results will be hard
to come by. Many of the changes involve fundamental shifts in power
something that cannot happen overnight (World Bank 2003: 18).

Post-colonialism, development studies and empowerment


Sadly, this is the conclusion we are driven to as well. In Midnapore,

a committed left-of-centre political party has helped to shift power and
resources to the poor in some measure, but this does not mean it will
happen in a state like Bihar, or that it could be effected in short order.
Building political capacity is time-consuming at the best of times, and the
incentive systems which currently prevail in Bihar are hardly conducive to
the investments that would be required. Poorer people are not generally
empowered by an optimism of the will that lapses into wishful thinking
or simple romanticism. Not should the costs of radical (Naxalite) actions
be ignored in any accounting of the merits of political society in a District
like Bhojpur.
At the same time, however, an extreme pessimism of the will can be
politically corrosive, and is not called for. In this book we have begun
to indicate how the Left Front government in West Bengal is beginning to extend its economic (mainly land and labour-related) initiatives
into the fields of education provision and democratic accountability. In
the postscript we will report further on how our work with the states
Institute for Panchayats and Rural Development took up some questions
relating to the training of local government workers. And in Bihar, too,
there are real possibilities for the empowerment of poorer people, both
by the government and in the face of government. Once we accept that
poorer men and women make multiple sightings of the state, and that
they weave together diverse and sometimes contradictory accounts of the
states obligations to them, we can begin to imagine a terrain for politics which is correspondingly decentred and plastic. This terrain might
involve the manufacturing of better exit options as well as localized struggles for accountability around public spending. It might also involve the
production of education and health services outside the direct control
of the state. But whatever it involves, it will be bound up with forms of
engagement with the world of the second-best (at best).
James Scott is right to suggest that poorer people are rarely empowered
in the long run by governments which claim to see all on their behalf.
We can add to this that poorer people are rarely empowered by political
movements that claim to see the future in utopian terms. Poorer people
instead need to be supported in their diverse and discontinuous efforts
to bring agencies of the state more firmly under their gaze and control.
These efforts will often taken shape in political society, with the help of
local mediators. But if some initiatives are commended by members of the
development community this is no reason for discounting them. The poor
stand to benefit most from a politics that is honest rather than glamorous,
and from public policy interventions that recognize existing realities or


The poor and the state

incentive systems. The shape of those public policy interventions, moreover, is not set in aspic. The development community, for all its faults,
and for all that it must respect its paymasters, is a learning community.
As we have argued repeatedly in this book, it is one of the communities
that helps to produce the poor as an object for social policy, and it is a
community that helps some members of the poor to see themselves and
the state in different ways. No matter how much Ferguson (or Escobar)
suggests otherwise, it is a community that helps to weave together the
complex and contested tapestry of statepoor encounters that we have
described in this book, and which we attend to further, and with a closer
eye on the politics of the research community, in the postscript which

Postscript: development ethics and

the ethics of critique

James Ferguson ends his account of The Anti-Politics Machine with a useful
and very honest Epilogue which addresses the question What is to be
done? In the course of his discussion he makes a number of points that we
find helpful. In particular, he warns against a form of romanticism that
would turn the fieldworker into a hero or social activist. The truth is that
most of us should not expect to make a difference. Social change is most
often made slowly and in a non-linear fashion by the men and women
who become the subjects of social science. Ferguson quotes Foucault to
bolster this argument. As Foucault remarked of the prisons, when the
system is transformed, it wont be because a plan of reform has found
its way into the heads of the social workers; it will be when those who
have to do with that . . . reality, all those people, have come into collision
with each other and with themselves, run into dead-ends, problems and
impossibilities, been through conflicts and confrontations; when critique
has been played out in the real, not when reformers have realized their
ideas (Foucault 1981: 13, quoted in Ferguson 1990: 281). This view is
largely consistent with the arguments we have developed here, although
we would want to signal more clearly than Foucault the role that directed
change can make, particularly when it is being pressed by the wealthy
and the powerful. We would also insist that the critiques that are played
out in the real are informed by the views of social workers (in this case)
or other experts; their plan of reform, that is to say, is already part of
the sightings of the prison system that are made by inmates and warders.
Ferguson also makes the argument that western academics who want
to take the part of the global poor should expect to be active politically within their own countries, and here too we find ourselves in strong
agreement. Given the power of the major development institutions, and
given our view that these are contested institutions, it would be inconsistent not to advocate getting engaged with these centres of power. This
isnt easy, of course, for most academics will not get a chance to work
with the World Bank, say, or UNDP or DFID, and far less to make policies in these institutions. As ever, a measure of realism and modesty is


The poor and the state

required, and perhaps also a measure of determination if one is committed

to working there, or, indeed, in activist groups or campaigning organizations.1 Robert Chambers and Norman Uphoff are two academics who
have sought to make a difference in this way, and we have signalled our
admiration for their efforts. They have taken steps to make social science
matter, to take the title from a recent and much-quoted book by Bent
But still the question remains of the relationship between the fieldworker and the field. This has been the subject of a great deal of writing of late, much of it concerned with questions of positionality and the
dangers of extractive research.3 We have learned a great deal from this
work, and we recognize that our initial academic research project was at
times very demanding of our respondents. If we were doing the project
again we wouldnt use a questionnaire survey than runs to seventy-two
main questions and many sub-questions. We simply havent been able
to analyse this data as effectively as we had hoped, and it could be said
that the excess of data we collected amounts to a tax on the people we
worked with. We also recognize that our presence in the villages gave rise
to expectations of what we might do for certain households or the community at large that we could not hope to fulfil. Like most fieldworkers,
we did our best to dampen down these expectations, and if we could help
villagers with a small task we did so.
Much like Ferguson, however, we found ourselves asking the question,
is there something (else) that we could or should be doing in a more
positive sense? In our view, this is a perfectly reasonable question to ask
and we suspect that most fieldworkers ask it at one time or another. By the
time that fieldwork for the ESRC-funded research project was drawing
to a close, early in 2000, we had forged strong links with activist groups
and NGOs in Bihar, as well as with the State Institute for Panchayats and
Rural Development (SIPRD) in West Bengal and DFIDs West Bengal

Working for the UNDP neednt prevent activism in a campaign like Jubilee 2000 (now
Jubilee Research), for example. We also know of colleagues who have worked for the
Save the Children Fund before moving on to a spell at the World Bank. But let us also
be clear that we do not believe that this is the only way that academics can engage what
non-academics so often like to call the real world. Academics engage the worlds of
which they are necessarily a part in all sorts of ways, and most often with a clear purpose
and self-critically. Teaching, writing and training fellow researchers will remain the major
engagements of most academics, and very properly so. These engagements are not without
effect, nor are they undertaken lightly or without regard for the ethical issues that frame
this postscript.
Flyvbjerg (2001). It should be obvious, but well say it anyway: activists of the New Right
also want to make social science matter, and they too will be pushing for jobs and power
within organizations like the IMF and World Bank.
Good starting points are Caplan (2003) and Scheyvens and Storey (2003).

Development ethics and the ethics of critique


project team in New Delhi. We began to exchange ideas and results with
some of these agencies in the way that researchers generally do. In spring
2000 we decided to approach DFID (India) to see if it would fund an
action research project that would return some of our research findings
to villagers and other stakeholders in Malda and Bhojpur Districts. We
chose Malda because the good governance agenda was less established
there than in Midnapore. We chose Bhojpur because of its long history
of pro-poor struggles: we thought this would make it a good site for the
consciousness-raising work we had in mind. It would be dishonest to
say that academics who work for universities in the UK or Canada arent
under pressure to get research grants, but this was not our main objective.
We wanted to take a year off the main project the one that would give
rise to refereed books and papers in order to get more involved on the
fuzzy border between development studies and development practice.
Happily, DFID agreed to fund the research, although at first it only
wanted to finance the work in West Bengal. (We shall come back to this
shortly.) The project went by the name of Enhancing Pro-Poor Governance in West Bengal and Bihar (EPPG), and work began in West Bengal
in September 2000. The research team worked in tandem with members of SIPRD, and began by convening village meetings in both Old
Malda and Debra Blocks. These were attended by about 200 and 150 villagers respectively, including a large number of people from marginalized
households or communities. We provided information on peoples rights,
the system of local self-government, and development programmes, and
we tried to give villagers a sense of how these varied between Districts.
Follow-up activities in the Malda panchayat included participatory learning exercises, the distribution of handbills, and the creation of a village
library. We also tried to get a sense of whether villagers found the meetings useful. We discovered that rates of information retention varied from
42 per cent to 88 per cent over the course of a month, and retention rates
were highest when the information was practically relevant to a persons
livelihood strategy. (The mode of dissemination mattered much less: see
also table 4.6.)
Three intermediary stakeholder meetings were then held in Malda with
137 local councillors, government officers, and members of NGOs and
political parties with whom we had worked at the village, Block and District levels. We asked participants to reflect on some of the problems of
local governance that we had identified. For the most part, though, we
invited them to push the agenda in a direction they felt comfortable with,
and which might point to concrete suggestions for public policy reform.
These recommendations were developed further in an inter-stakeholder
(general) meeting that was held in Malda, in December 2000, and in a


The poor and the state

state-level workshop that was held in Calcutta in January 2001. Thirtyseven recommendations for changing local government policy and practice were discussed at the workshop, and some of these were reshaped for
further analysis by DFID (India) and the government of West Bengal.
Many of the recommendations that came from stakeholders in Malda
referred to the non-implementation of existing government acts or directives, and the need for enforcement actions. The later meetings also
allowed participants to reflect more generally on the governance roles of
local councils. Discussion extended to the possibility of bringing health
and education under effective pro-poor controls, and the means by which
local councils might mobilize their own resources more effectively and/or
untie some of the funds made over to them.
Do we think these workshops made a significant difference to the political landscapes of Malda District or West Bengal more generally? Of course
not. But we do believe the project met the test of not doing harm and that
at least some villagers became more aware of their rights vis-`a-vis the state
(or party), whether or not they could act on them. SIPRD also seemed
to think the project was worthwhile. It drew on the EPPG project when
it redesigned elements of its training programmes for participatory rural
development. DFID was right to push us to work with a government
agency in a state that showed strong signs of pro-poor governance.
The research team then turned to Bihar, where DFID was reluctant to
fund work of any kind. In 2000, indeed, it was committed to not funding
work there, apparently on the basis: (a) that monies very often were captured (this is in regard to development projects, of course); and (b) that
the political culture of the state was not pro-poor (the prevailing incentive
system allowed politicians to make their careers without attending to the
needs of poorer men and women).4 Realizing that DFID was unlikely to
be shaken in its negative view of Bihar, we argued nonetheless: (a) that
there were spaces of empowerment for the poor in Districts like Bhojpur
and Vaishali; (b) that some of these had been forged by officers working to close down the space for rent-seeking, and others who promoted
Janata Durbars; (c) that other spaces had been forged through continual engagements with and against the state by campaigning groups and
NGOs, and most notably by the struggles led by the JP movement and

We need to be careful here. This is what we were given to understand from conversations
with DFID officials, although this might not have been official DFID policy. By the same
token, we were not provided with coherent arguments for this apparent discrimination
against Bihar. Did DFID suppose that the political cultures of Andhra Pradesh or Madhya
Pradesh, or even parts of West Bengal, were so much better than those in Bihar, and if
so, why? As we said in the main body of the text, there is a danger that the word Bihar
functions as a stereotype of corruption and unreason when the ground realities are both
more complicated and less depressing.

Development ethics and the ethics of critique


the CPI-ML, as well as those waged in support of the dignity and honour
of the Backward Classes by the Janata Dal; and (d) that if DFID was
serious about working on behalf of the worlds poorest it could hardly
avoid Bihar and Uttar Pradesh.
It is likely that persistence as much as strong arguments persuaded
DFID to let us go ahead with the workshops we planned for Bhojpur. In
any case, preparation for the workshops began in February 2001, and a
lot of effort went into the task of engaging villagers ahead of the workshop
planned for our Sahar Block field site in June. A discussion paper was prepared and circulated in Bhojpuri, and villagers were asked to let us have
their thoughts on a possible agenda for discussion/action. We drew on
these responses to help write a folk drama that was performed as part of
the village workshop. Dugdugi drew more than 600 people to watch it, and
used the figure of an idiot savant (the bhaku) to draw attention to the functioning of local government institutions in a more active panchayat than
villagers were used to one where ward commissioners and alert voters
put pressure on government officers to run their programmes properly
and with an eye to meeting collective needs.
Workshops were then held in June and July with Block-level NGOs
and activists from across Bihar, including literacy activists and BEP
motivators; with Block- and lower-level civil servants from Bhojpur
District; with Block-level political party functionaries from Bhojpur
District; and with newly elected panchayat representatives and members
of VECs and Mahila Samoohs (groups to promote womens empowerment
and female school-going) from Bhojpur, Rohtas and Buxar Districts. Participants in the Block-level workshops were invited to consider a range of
questions relating to peoples empowerment, government reform, political culture and social change, and a refined version of this agenda was
later used to structure the discussions of a District-level workshop that
combined members from the separate Block discussions. The project
was concluded by a state-level workshop that was at first confined to
civil servants, activists, NGOs, academics and people from the media
or concerned foundations, and which on its third day was given over to
state-level politicians.
Again, did we expect these workshops to galvanize political society
in Bhojpur or in Bihar more generally? Not exactly, although we had
hopes that they would provoke further reflection and some action. Several
politicians spoke openly at the state-level workshop about the incentive
systems that were in play in Bihars political cultures, and about the difficulties they faced in making electoral capital out of health and education
issues (something that was true also in West Bengal). Others confirmed
that political parties were organizing exclusively on caste lines, and that


The poor and the state

training camps for party workers had more or less stopped in the 1980s.
Many MLAs and party workers were unaware of the governments development programmes. Nevertheless, a number of issues were raised in a
spirit of open and constructive dialogue, and members of every political
party were encouraged to discuss their concerns in public with other participants in Bihars political cultures.5 The workshops also gave rise to a
large number of practical recommendations and suggestions for taking
them forwards. Several of these were in support of the idea that NGOs
should play a bigger role in structuring political society in Bihar, and for
linking between the state and poorer households. Members of the project
team worked with existing NGOs to provide more secure documentation
of the range and scope of NGO activities in Bihar (and Jharkhand). Copies
of the resulting documentation have been widely circulated. The research
team was also encouraged by the decision of the Secretary, Department
of Panchayats, to work closely with the project. He attended the village
meeting in Sahar and expressed interest in developing the folk drama into
a television programme that could be used in the training of panchayat
representatives. The time was right for such an enterprise. The Secretary
found himself in charge of a department that was required to form a more
robust set of rules and regulations for the panchayati raj institutions that
were being revived across Bihar. We might add that we are still in the process of trying to set up a small organization that would work with poorer
villagers in two or three panchayats around their rights to information and
benefits from local government agencies. We find inspiration here in the
work of the MKSS in Rajasthan (see chapter 7).
But let us now come to our punchlines. We have set out these activities at some length not simply as a matter of report, or because we have

One minister said that politics was only about power, and that politicians were interested
in the glamorization of poverty, but not in its alleviation. It should be noted, however,
that the education minister (Ram Chander Purbe) disputed the view that education was
not a big issue for politicians. He argued that the trickle down of democracy was creating
a hunger among people for education, and that MLAs were having (or would have) to
respond to this. He further suggested that a combination of VECs with functioning panchayati raj institutions would give parents and VEC committee members greater statutory
powers over teachers, including the power to withhold wages in some circumstances. It
is possible this will happen. We hope to test this and other suppositions/predictions by
means of resurvey work in 2009. Another minister said that government schools would
improve only if someone in the country had the courage to ban private schools. Finally,
still another minister took the research team to task for supposing that it was middlemen and political fixers who pushed for pucca construction works, when this is what the
people at the grassroots were demanding. As we said, a range of views were clearly and
forcefully expressed, and if there was general support for our research findings this was by
no means complete or across the board. A representative of the CPI(M-L) complained
that our work, and the workshop, were not focused enough on the disempowering effects
of the governmentgoonda (muscleman)contractorbroker nexus.

Development ethics and the ethics of critique


drawn on the EPPG project in chapters 36. Nor are we trying to suggest
that we have done something unusual, or that ethical objections cant be
raised against us getting involved in the first place. Ferguson maintains
that outsiders should get involved only where it is possible to identify
interests, organizations, and groupings that clearly represent movements
of empowerment, and when a demand exists on the side of those working
for their own empowerment for the specific skills and expertise that the
specialist possesses (Ferguson 1990: 286). We largely share these sentiments while noting that there are important issues to be decided about
the insider/outsider distinction in fieldwork, and that Manoj Srivastava
is positioned very differently than Corbridge, Veron and Williams and
we acknowledge that if we have met the first of these tests it is less certain
that we have met the second one (although we would argue that we have
done). In any case, the point that we want to develop here is rather different, for it concerns the ethics of critique more generally, and what Max
Weber described as the duties of a public intellectual in the service of
moral forces (see Introduction). The EPPG project gave us fresh insight
into the sorts of questions that development professionals are bound to
ask, and which are not going to go away any time soon. These questions
made us wonder if too many academics claim a monopoly on virtue and
fail to render problematic the ethics of the forms of critique to which
they themselves subscribe.6 In our view, there are at least four legitimate
forms of critique that academics might engage in, all of which deserve
to be treated with a mixture of caution and respect, and each of which
speaks to different aspects of the politics of empowerment and statepoor
First, there is the tradition of refusing judgement. This tends to be associated with positive economics, and with a defence of the status quo,
but it is also common among anthropologists. Refusing to make overt
judgements is often very difficult and can be commendable. Imagine a
liberal anthropologist from the UK trying to make sense of active supporters of the National Rifle Association in the United States, or groups
trying to promote the teaching of Creationism in schools. Straightforward
descriptions of these groups (insofar as descriptions are ever straightforward) can be an effective way of representing difference and of allowing
moral judgements to be made by the respondents themselves.

Let us be clear that this goes both ways (see also note 1). Development professionals
sometimes underestimate the contributions that academics make. In addition, the report
format that is often imposed by development agencies, with its emphasis on discrete
(numbered) points, an executive summary, and even bullet points, imposes a cost in the
same way that any form of writing imposes a cost (it stops the writing of other stories).
The urge to simplify, or to get to the point, is understandable but not always attractive
or even helpful.


The poor and the state

At the other end of the spectrum there is the post-structuralist strategy

of subjecting everything to critique. Foucault seemed at times to subscribe to this viewpoint; indeed, one might argue that it is consistent with
his expansive conception of power and its effects. The aim is not necessarily, or even, to suggest ways forward in the sense of concrete policy
alternatives. It is rather to add to the foment of debate and to put into
play new ways of thinking which might provide resources for some individuals or groups in the constant jockeying for power and position that
promotes the (re)structuring of everyday life. Critique thus becomes an
act of permanent revolution, or perhaps even a playful decentring of ideas
and practices that are taken for granted. Some criticisms of the very idea
(or discourse) of development fall into this category, if category it is.
In between these strategies are the more conventional strategies of the
political Left and Right. These forms of critique imagine a situation where
the poor are empowered a world where land and power are better distributed, for example, or where men and women have equal opportunities
to access the market and dismiss strategies that fall short of a proposal
for reaching this world as reformist, or missing the real issues. Critiques
of this sort are important because they push policy-makers to think more
critically about the nature of the constraints they face. They also direct
attention to popular struggles that might push back the envelope, and to
the dangers of promoting policy agendas without specifiying the political
forces that would bring them into being.
Lastly, however, there is the form of critique to which so many postdevelopmentalists and some leftists (not to mention neoliberals) object
what we might call the reformist or pragmatist tradition which begins
with a specific question and works outwards from it. This tradition also
has much to commend it, and we feel this more strongly now that we have
worked inside the belly of the beast. One complaint that reformists might
direct to members of the post-developmental community, and indeed to
James Ferguson in this specific respect, is that they falsely homogenize a
range of development initiatives under the heading of the development
discourse. Another complaint that would apply to individuals like Esteva
and Prakash, and perhaps also to Escobar, is that they refuse to specify the
costs of their proposals.7 There is an important ethical issue here. How
legitimate is it to commend strategies of delinking or spatial closure, say,
or of returning to a culture of the soil, if the opportunity costs of these
actions are not made clear to those who are expected to heed the call? By
the same token, it can fairly be argued that the utopias of the Left or the
Right communism or free-market capitalism carry less moral weight

Esteva and Prakash (1997); Escobar (1995).

Development ethics and the ethics of critique


to the extent that their proponents refuse to consider the likely costs of
these regimes, and of the social upheavals that would be required to get
there (assuming the horizon is not ever receding).8
We can also consider a more practical argument that can be made in
favour of this form of critique (as one of many). If we assume: (a) that
the world is not perfect or perfectible, (b) that what is called development comes in many versions, and (c) that pro-poor political coalitions
are not easily built; and if we further assume that an actor wants to take
the part of the poor in some way (that is, he or she refuses the first form
of critique), then it is not clear that this form of engagement (or critique)
is uncalled for. More positively, while we agree with Harriss-White that
Development policy needs rethinking as that set of political and institutional forces required to prevail against the obstacles to a democratically
determined accountability (2003: 247), it is not obvious: (a) that this
takes us very far in generating specific policy initiatives that would address
the problem of corruption in West Bengal or Bihar, or of forcing governments to share information with poor people in such a way that their
citizenship rights are genuinely deepened; or (b) that the formulation of
policies that would address these issues would look radically different to
some parts of the good governance agenda that has been put into play,
and periodically reviewed and developed, by the Government of India or
leading agencies from within the NGO and international development
communities. (This is why we wrote favourably of one part of the World
Banks World Development Report, 2004 in chapter 8.)
What we are calling for is a greater recognition that all these forms of
critique must be in play at different times, and that they all raise ethical
questions. Working on the EPPG project has made us more aware of the
need to ask specific, policy-related questions, and of the need to refuse
the romanticism that many left-leaning academics bring to development
issues. People who work for the development community, including the
NGO community, like to ask academics, well, what would you do then?
Academics, for their part, like to brush the question away. Thats not our
job, is a common reply, and it is one that Ferguson uses in his Epilogue.9
We respect this answer, and we recognize that Ferguson develops two
other forms of critique to quite superb effect. We would simply say, in

It is a logical error to suppose that a form of critique that shows that capitalism is associated with negative outcomes a, b and c (say unemployment, pollution and inequality)
is in itself an argument for socialism or something that is not capitalism. That argument
would need to show that the alternative is not also associated with a, b or c, and/or is not
damaged by negative outcomes d, e, f and g.
The first response to this sort of objection must be that the book never intended or
presumed to prescribe, and that this is not what the book is all about (Ferguson 1990:


The poor and the state

addition, that what the EPPG brought home to us is that destructive

forms of critique are ultimately weakened (made less relevant and perhaps
more dangerous (by refusing to specify costs)) to the extent that they are
not also informed by the constructive forms of critique that public policymaking demands.
We think this is what Max Weber had in mind when he spoke about the
duties of a person who stands in the service of moral forces. As Mitchell
Dean explains, Weber wanted to prosecute an analytics of government
[that encourages] us to accept a sense of responsibility for the consequences and effects of thinking and acting in certain ways (Dean 1999:
36). That sense of responsibility would extend to raising inconvenient
facts (Dean 1999: 36, citing Weber 1972: 147), and thus to critiquing
the conventional techniques, practices and rationalities of government
and self-government (Dean 1999: 37), but it would also extend to commending specific forms of government or self-government that would
seek the empowerment of individuals and groups against specific states
of domination (Dean 1999: 37). It is this specificity, perhaps, that public policy-making ultimately teaches, and which calls for a less dramatic
conception of politics than some academics feel comfortable with. It also
confirms that development studies, in this case, is not exterior to the
world it describes, but is constitutive of that world.

Appendix 1 Major national programmes

and policies related to poverty
alleviation, 1999

Special employment and poverty alleviation programmes


Rural self-employment programmes




Integrated Rural Development Programme

50% centrally sponsored scheme with national coverage
since 1980 (197680 pilot scheme in selected Blocks). Aims
at providing self-employment through acquisition of productive assets and skills through provision of subsidy and
bank credit. Targeted at rural BPL population, largely small
and marginal farmers, agricultural labourers and rural artisans. Special safeguards for SC/STs, women, physically
handicapped; priority to assignees of ceiling surplus land,
Green Card holders under Family Welfare Programme and
freed bonded labourers. Performance during Eighth Plan:
total allocation (Centre and State) = Rs. 5,048 crores;
108 lakh families covered.
Training of Rural Youth for Self-Employment
50% centrally sponsored facilitating component of IRDP
since 1979. Aims at providing basic technical and managerial skills through training. Targeted at rural BPL population
between 18 and 35 years. Special safeguards for SC/STs
and others, like IRDP. Performance during Eighth Plan:
total allocation (Centre and State) = Rs. 370 crores; 15 lakh
youth trained.
Development of Women and Children in Rural Areas
Sub-scheme of IRDP started in 19823 on pilot basis, later
extended to all Districts. Aims at improving living conditions
of women and, thereby, of children by promoting womens
income-generation activities through self-help groups and
providing access to basic social services. Targeted at groups
of 10 to 15 women among BPL families. 50% to SC/STs.





Appendix 1

Performance during Eighth Plan: total allocation (Centre

and State) = Rs. 190 crores; 1.9 lakh groups formed; 30 lakh
Supply of Improved Toolkits to Rural Artisans
Sub-scheme of IRDP introduced in 1992, national coverage from 19956. Aims at enhancing incomes and product quality through 90%-subsidized provision of toolkits of
Rs. 2,000. Targeted at rural artisans (BPL), except weavers,
tailors, needleworkers and bidi (country cigarette) workers.
Special safeguard for SC/STs, again like IRDP. Performance
during Eighth Plan: total allocation (Centre) = Rs. 116
crores; 6.1 lakh toolkits distributed.
Ganga Kalyan Yojana
80% centrally sponsored scheme started in 1997, covers
all Districts. Aims at improving agricultural productivity
through exploitation of groundwater (borewells and tubewells, which are not provided under MWS). Targeted at
small and marginal farmers (BPL) (individuals and groups
of 515) who have not been assisted by another government
minor irrigation programme.
Rural wage employment programmes
Jawahar Rozgar Yojana
80% centrally sponsored scheme since 1989, after merger
of NREP and RLEPG. Aims at generating additional
gainful wage employment for unemployed and underemployed men and women in rural areas. Secondary objectives include creation of assets for the community and for
SC/STs, and positive influence on wage levels. Provides
employment opportunities at minimum wages. Targeted
at rural BPL population with preference for SC/STs and
freed bonded labour and with 30% reservation for women.
Second stream added in 19934 targeted at 120 identified
backward Districts in 12 states; aims to provide 90100
days of employment per person (merged into EAS). Third
stream introduced in 19934 (later called JRY Special and
Innovative Projects) for projects aiming at preventing migration, enhancing womens employment, etc. Performance
during Eighth Plan: total allocation (Centre and State) =
Rs. 17,473 crores; 40,362 lakh person-days of employment
created. The Draft Ninth Plan made the EAS the major

Programmes and policies for poverty alleviation






employment programme; JRY to be confined to the creation of rural infrastructure at the village panchayat level, in
consonance with the felt needs of the community.
Employment Assurance Scheme
80% centrally sponsored scheme introduced in 1993 as
pilot scheme in 1,775 backward Blocks, extended in 19978
to cover all rural Blocks in country. Aims at providing
100 days of assured casual manual employment at statutory minimum wage during lean agricultural season. Secondary objective is the creation of economic infrastructure
and community assets for sustained employment and development. Linked to environmental programmes (watershed
development, agro-horticulture, water and soil conservation, etc.) as per the felt needs of the District. Demand
driven, targeted at all persons in the age group 1860
who are in need of work. Performance during Eighth Plan:
total expenditure (Centre and State) = Rs. 5,278 crores;
10,686 lakh person-days of employment created; 259 lakh
persons registered.
Million Wells Scheme
80% centrally sponsored scheme since 1996, started as subscheme of JRY/NREP in 19889. Aims to improve agricultural productivity through fully subsidized provision of
open irrigation wells on private land. Targeted at small
and marginal farmers below BPL, two-thirds of which are
SC/STs and free bonded labourers (before 19934 exclusively to SC/ST). Performance during Eighth Plan: total
expenditure (Centre and State) = Rs. 3,727 crores; 7.4 lakh
wells constructed.
Indira Awas Yojana
80% centrally sponsored scheme since 1996 (before subcomponent of JRY). Provides funds for house construction.
Targeted at SC/STs.
National Social Assistance Programme
Programme launched in 1995 with three components:
National Old Age Pension Scheme, National Family Benefit
Scheme, National Maternity Benefit Scheme. Supplements
efforts of states in order to ensure minimum national standards of well-being. Provides social assistance benefits in
the case of old age (pension of Rs. 75 per month); death of
the primary breadwinner (lump-sum of Rs. 5,00010,000);
and maternity (Rs. 300 per pregnancy up to the first two


Appendix 1

live births). Performance during 19967: total expenditure

(Centre) = Rs. 384 crores; 58 lakh beneficiaries.





Urban poverty programmes

Swarna Jayanti Shahari Rozgar Yojana
75% centrally sponsored scheme launched in December
1997, replacing NRY (Nehru Rozgar Yojana) and other
schemes. Seeks to provide gainful employment through support of self-employment ventures and provision of wage
employment. Rests on foundations of community empowerment. Targeted at urban BPL population in all towns, particularly in identified pockets of urban poverty. Performance
(of NRY) during Eighth Plan: total expenditure (Centre and
State) = Rs. 498 crores. In 19934, 124 lakh person-days
employment created; 1.5 lakh beneficiaries for setting up
Urban Self-Employment Programme
Main component of SJSRY. Seeks to upgrade informalsector activities by encouraging the setting up of microenterprises with subsidized loans, skills training, and infrastructural and marketing support. Targeted at individual
urban poor and, particularly, at neighbourhood groups of
urban BPL women.
Urban Wage Employment Programme
Component of SJSRY. Aims at providing wage employment to unemployed and underemployed persons. Secondary objective is creation of socially and economically
useful public assets (coordination with NSDP). Targeted
at urban BPL population.

Special Area programmes

Drought Prone Areas Programme (Watershed Development)
Started in 19734 as integrated area development programme; 50% centrally sponsored: revamped in 19956
into approach based on Watershed Development. Aims
to promote economic development of village communities through optimum utilization of natural resources that
will mitigate effects of droughts and encourage ecological

Programmes and policies for poverty alleviation


balance. Seeks to improve economic condition of resource

poor and disadvantaged sections through creation and equitable distribution of stable resource base and increased
employment opportunities. Covers 947 Blocks in 155 Districts in 13 states. Expenditure between 19956 and 1996
7: Rs. 2,035 crores; almost 5,500 Watershed Projects to be
implemented in this four-year period.

Social services


Basic services programmes





Minimum Needs Programme

Launched in 19745. Seeks to establish network of facilities and social services in all areas up to national norms
in order to raise living standards and reduce regional disparities. Provides investment in elementary and adult education; supplementary nutrition; rural health, water supply, roads, electrification, housing; and environmental urban
slum improvement. Most important programme for delivery
of basic services.
Basic Minimum Services
Started in 1996. Seeks universal coverage by 2000 in regard
to primary health-care, primary education and safe drinking
National Policy on Education
Universalization of elementary education has been policy
goal since Independence. Focus on physical infrastructure
and teacher training. Revised Policy since 1986. Includes
18 centrally sponsored schemes, including: Mid-Day Meal
Scheme, Operation Blackboard; Non-Formal Education
Programme; and Post Literacy and Continuing Education.
In 19978, central plan allocation to elementary education
was Rs. 2,265 crores.
Total Literacy Programme
Principal instrument of National Literacy Mission launched
in 1988. Aims to eradicate (child and adult) illiteracy by
2005 through campaigns in specific areas. Performance
198897: covered 215 Districts; 666 lakh new literates.


Appendix 1



District Primary Education Programme

World Bank-assisted project launched in 1994. Aims to
achieve universal elementary education through specific
planning and target setting at district level. Operates in 149
Districts of 14 states. In 19978, central Plan outlay for elementary education was Rs. 561 crores (bulk of expenditure
borne by states).
Health and family welfare
Health for All Policy
Policy announced in 1983. Continued expansion of health
infrastructure, including primary health centres, but with
more attention to underprivileged and vulnerable sections.
Includes National Illness Assistance Fund that seeks to
ensure financial assistance for specialized medical treatment
from life-threatening diseases. Targeted at BPL patients.
Includes various disease-eradication programmes. In 1997
8, central Plan allocation to primary health care was Rs. 918
Family Welfare Programme
100% centrally sponsored scheme started in 1952. Seeks
to promote small family norm and reproductive and child
health through free and voluntary choice. Includes Reproduction and Child Health scheme, Pulse Polio Immunization scheme, etc. In 19967, central Plan outlay to family
welfare amounted to Rs. 1,535 crores.



Housing and sanitation


National Housing Policy

Revised in 1994. Recognizes importance of housing for overall development of rural people and urban poor. Includes
central assistance to national network of building centres;
housing schemes for Economically Weaker Sections and
Lower Income Groups; and IAY (see above).
Rural Water Supply Programme
Central assistance matching provision by state under MNP.
Rural Sanitation Scheme
Supplements MNP and other programmes. Targeted at SC/
STs. During 19967, central expenditure was Rs. 60 crores;
2.3 lakh latrines built.

Programmes and policies for poverty alleviation







Welfare of weaker sections

Special Component Plan
Started in 1979. Central assistance to states to make special
provisions to SCs under various schemes.
Tribal Sub Plan
Started in 1974. Central assistance to states to make special
provisions to STs under various schemes.
Development of women and children
Empowerment of Women Policy
National policy announced in Ninth Plan. Main scheme:
ICDS (see below). Various smaller schemes: STEP (Support to Training and Employment Programme); Employment and Income Generating Training Programme that
trains women belonging to weaker sections in nontraditional occupations (co-funded by Norway); Rashtriya
Mahila Kosh that extends credit to poor women of informal
sector; Mahila Samridhi Yojana that promotes thrift among
poor women, etc. Continuation of womens component in
poverty-alleviation programmes. Total central outlay for various schemes for women and children 19978: Rs. 1,026
Integrated Child Development Services Scheme
Started in 19756. Seeks to provide integrated package of
services, including supplementary nutrition, immunization,
health check-up and referral services, pre-school non-formal
education and health to children below six years. Covers
201 lakh children and 38 lakh mothers.

Various crop, livestock, fisheries schemes; schemes for irrigation, agricultural credit, supplies, technology, processing, marketing, etc. Some
schemes targeted at marginal and small farmers. In 19978, central plan
allocation to agriculture and allied sectors was Rs. 2,969 crores.

Public Distribution System
National food security and general subsidy scheme until
1997. Aimed at providing national and individual food



Appendix 1

security through distribution of subsidized food and maintenance of buffer stocks (since 1966); promoting foodgrain
production through ensured procurement and minimum
support prices for farmers; checking inflationary pressure
through subsidized food prices. Untargeted, but originally
biased to urban areas.
Targeted Public Distribution System
Replaced untargeted PDS in June 1997. Aims at ensuring availability of essential commodities at affordable prices
especially for the poor through provision of subsidized rice
and wheat (10 kg per month), sugar, kerosene, etc. Other
objectives like older PDS. Targeted at rural and urban BPL
families; reduced food subsidy for the non-poor. In 19978,
total subsidy will be Rs. 6,167 crores (Rs. 3,718 crores for
BPL families); 1,645 lakh families covered (587 lakh BPL

Appendix 2
in Hajipur

The 1999 general election

What follows is a report on the 1999 parliamentary election in Hajipur,

Vaishali, that was written for the research team by our co-worker,
Vishwaranjan Raju. Many of the themes that Vishwaranjan teases out
here (gender issues, the relationship of local and national political figures, the temporary stopping of the development state) are not specific
to Hajipur. Several of them can be observed in elections in West Bengal, for example. Nor should we assume that the everyday qualities of
political life in a District are captured in extremis by the events that occur
at election time. This report should rather be read as a complementary
piece to the accounts of political society in Vaishali that we built up in
part II of the book.
For the record, the parliamentary seat of Hajipur was won in 1999 by
Ram Vilas Paswan for the Janata Dal (United). The Janata Dal (United)
was part of the BJP-led National Democratic Alliance which swept to
power in New Delhi. Ram Vilas Paswan had won the seat in 1998 for the
Janata Dal. At this point he was an ally of Laloo Yadav. The runner-up in
the 1999 elections was Ramai Ram for the Rashtriya Janata Dal. Ramai
Ram would have been required to resign his position as an MLA and State
Minister in the government of Bihar had he won the parliamentary seat
of Hajipur. Ram Sunder Das was placed third (in 1998 he took second
place for the Samajwadi Janata Party (Rashtriya)). Ram Sunder Das had
twice previously represented the Hajipur constituency, and was an ally
of Laloo Yadav in the early-1990s. He had also been the Chief Minister
of Bihar. BB is UNR, the local MLC. BB had for some years been the
Janata Dal MLA for Raghopur. A long-time ally of Laloo Yadav, BB had
resigned his seat in favour of Laloo Yadav before the latter became an
MP. In return for this favour, Laloo Yadav rewarded BB with a seat in the
Legislative Council. The present MLA for Raghopur, Rajgir Chaudhury,
is something of a dummy candidate for BB, and is rarely mentioned by
villagers. In D village, in 1999, BB was able to mobilize the votes of
those (few) Rajput families in the village who had taken advantage of


Appendix 2

his considerable links to the Block and District office, but not of those
younger Rajputs who turned to Ram Vilas Paswan.
Although most of Bidupur Block falls into the Yadav-dominated
Raghopur Assembly Constituency, the Hajipur parliamentary constituency is less obviously dominated by Yadavs. The 1999 election in
North Bihar was determined by the Laloo question: whether to be for
or against the husband of the Chief Minister of Bihar. Mindful that his
party was losing support to the Samata-JD (United)-BJP combine, Laloo
Yadav gave word that wherever an RJD candidate for Parliament lost out
in an RJD-Assembly seat, the sitting MLA would not have his or her ticket
renewed for the next Assembly elections. BB, then, although not the MLA
for Raghopur, was under pressure to ensure a lead for Ramai Ram in his
area (which he did, by about 17,000 votes). The MLA for the Hajipur
Assembly constituency, Rajendra Rai, faced similar pressures, but in a
constituency less dominated by Yadavs he was unable to prevent Ramai
Ram from trailing there by rather more votes.
The election looms
Poor rural Indians are intelligent enough to know that seeing colourful
helicopters in the sky means that election season has come again. They
can easily make the assumption that white khadi-dhari politicians with
huge crowds will again become the talk of villagers.1 A long procession of
motor cars, new models of cars, will again move around these villagers. At
the same time they become afraid, remembering what happened in the last
election. There were no commercial vehicles for the last three days of the
election. If anyone had an emergency the only option was to go to Hajipur
by bicycle. There was no doctor in hospital as somebody told them that he
was on election duty. School-going children were free to move in fields.
Education is the first and foremost part of government which has to
give greater sacrifice in name of election: teachers being polling officers,
schools being polling booths, and colleges being counting centres. The
whole system of education is stopped until a new government is formed.
This election is not different from others. When Vajpayee had come
to Patna, a rickshaw puller told me that the traffic police had changed
the route and just for one kilometre distance he had to move through
three kilometres. But the passenger paid him only at the past rate. A
hawker selling samosas and chaats [snacks] told me that at a rally for
Laloo many people came and ate up all his items and paid nothing in
return. While approaching D village [the research site] I was at Ganga
bridge. I saw many passengers having no other option but to reach

Wearing traditional or neo-Gandhian dress.

The 1999 general election in Hajipur


Hajipur by walking the whole distance of ten kilometres, putting their

luggage on their heads. At Paswan Chawk many flood-affected people
were seen waiting for governmental assistance but nobody was available from the government side to console them as all were on election
Big men (and women) and local politics
H Singh told me that in this parliamentary constituency 25,000 votes
are bogus and about thirty booths are be chiragee as nobody resides
here. According to him, thirty years back many villages were swept away
by a current of the Ganga. The residents of those villages were at the bank
of the river and a few of them migrated to Goraul or even Monghyr District. However their name on the voter list still remains. This is why Laloo
Prasad preferred to contest the Vidhan Sabha [Legislative Assembly]
election from here, considering 25,000 votes as solid votes, for which
the ruling party easily manages in its own favour. When I asked why the
parties in opposition do not raise this issue I was told that when we will
be in power we will encash this whole 25,000.
Recalling the earlier system when the post of minister was given to a
political leader, C Singh [a prominent villager] said that people had to
take time to meet with them as they would always be busy in meetings.
For the approval of any officials letter there used to be a meeting, and
with the consent of all the participants letters were either approved or
rejected. But nowadays a criminal also says that tomorrow morning I
will meet Mantri-ji [minister] and take approval for a particular contract
work. Now the residence of a political leader has become a shop where
for each piece of work for example, a signature there is a fixed rate.
The authority to examine these matters is generally delegated to a PA or
a very close relative.
One villager, V Singh, said that BBs visit to this village was directly
related to the need to assemble a big crowd at the time of nomination.
Earlier there used to be a victory rally, but now a leader has to show the
media that a wave is in his favour. For this reason BB asked all vehicle
owners [preferably four-wheelers] to participate in his nomination. K
Rai, a contractor whom I met at a tea stall, told me that when BB had
visited the Yadav tola he specifically asked M Singh to come at the time
of his nomination at Hajipur. M agreed to go on condition that a new
vehicle would take him there. K Rai, sitting in front of BB, accepted the
responsibility for arranging a new vehicle. For fare and petrol the total
cost was about 4,000 rupees. After the nomination was filed all respected
people were invited to the residence of BB, where KD Rai and others
had to contribute to the provision of beaten rice and curd. There he


Appendix 2

contributed two big drums of curd. This is one of the means by which
politicians who give out [government] contracts for work take a return
from their [the contractors] earnings.
It was rumoured within the Rajput tola that Ram Sunder Das had taken
Rs.10 lakhs to stand and contest from Hajipur constituency. However,
elderly people were not happy as Ram Sunder Das had shifted from the
RJD to the Samajwadi Janata Party as Laloo did not gave him a ticket.
One evening when I was at Deba Chowk I saw that Rajendra Rai, a
minister of Laloos government, visited and stayed for a long time. He was
at the time discussing with local volunteers and villagers how he can get
a victory over Ram Vilas. Actually he was in trouble as his local support
was almost nil. All of his contract work on the Employment Assurance
Scheme was actually done as per rule; that is, with the appointment of
an Abhikarta [Agent] (the Block Development Officer also told me that
in one of the meetings for the selection of an Abhikarta he was physically
present). The work was done by the Junior Engineer, who did not appoint
or involve local weighty personalities in his work. No benefits from the
work were diverted to these people. The people agreed that the work
done by Rajendra Rai is of comparatively better quality, but the setback
he faced in a democracy was that the votes of the masses were still
decided by a few weighty personalities in the village.
On my way to Patna I stopped at Onari Chowk to review the motivational level of people. This is a place where Dusadhs and Yadavs are
in equal number. A person selling fish at Onari chowk was telling his
people not to worry and that our leader [indicating Ram Vilas] has made
arrangements for the election. He was telling the other Dusadh people
that this time nobody will stop you voting. I joined the group and tried
to win his confidence. He said finally that R Singh, who is the stronghand of Ram Vilas-ji, has arranged for an AK 47 [rifle] for the period of
the election. This has come from Nepal. My observation was that at this
Chowk, if some special effort is not made by the administration, there is
a fair chance of some casualities.
The next day, in the village, P Singh, who last year was the Vidhan
Sabha [Assembly] candidate of the Janata Dal, and who was defeated and
had only 48,000 votes, again moved to the villages to collect his castes
[Rajputs] votes for Ram Vilas and Vajpayee. In D village he came and
stayed at C Singhs dalan. This place is a milestone for all leaders, as he
is the most reputed and respected person of his caste and of other castes
also. Every leader wants to please him, and if he is pleased politicians
assume that other villagers respecting the view of C Singh will not go
against him. I found here for the first time that a political leader was
facing questions from the public. One question was why Paswan himself

The 1999 general election in Hajipur


had not come here? If he has no time during the election then what is
the guarantee that in future he will come when he will become minister?
How will he recognize us when he is MP since 1977 and has never visited
this village or others nearby?
Campaigns, rhetoric, transport
Next afternoon Ram Vilas made his speech at Deba Chowk. An
announcement was made at a regular interval that due to some technical
fault in the helicopter he is coming by car. When Ram Vilas spoke he
used the word helicopter five times. Then I got into the issue of helicopters and talked with several people standing there and reached the
conclusion that the word helicopter added much to the reputation of
that leader. For rural illiterate people the helicopter is a symbol of the
gradation of the leader. For people for whom the concept of the country
is not clear where somebody tells that the country means adding up
Hajipur, Patna and Delhi it is not easy to convince that so-and-so is a
Cabinet minister or has held a higher designation. The only easy means
to convince them is that they have come by helicopter. The poorer people
can easily assume that his influence is much higher than that of leaders
who are moving in cars. Even then, moving by Tata Sumo [as opposed
to] moving by other [smaller] cars marks a difference in status.
Apart from four-wheelers, Rajdoot motor cycle rallies have also made
the election more colourful. At least fifty youths making a group, with
their party flags on their bikes, were common and were instructed to move
inside the village. The question arises from where they were getting funds
for petrol. For Ramai Ram they were mostly contractors who were always
seen at party offices at every Chowk. Common people used to speak of
them as mobile cash counters. When an RJD minister in a particular
zone requires some money he will make contact with a nearby contractor
who has been given work in the Laloo ministry. The story about Ram
Vilas is not very different. I hear also there are cash counters, and even
that you may not have to pay money: just by listing a few names you may
get a full tank of petrol, or some time you may have to show a coupon
issued by the party office. The question then is what this petrol pump
owner will get in return for free petrol distribution for the last seven days.
D.K.C is the owner of two petrol pumps at Hajipur. He simply wants a
ticket in the Vidhan Sabha election. Somebody says that his petrol pumps
had been allotted by MP quota by Ram Vilas and that he has already been
obliged. However, when I asked about distribution of petrol free or by
coupon a volunteer said that petrol is given only to those candidates who
may climb on stages and give speeches.


Appendix 2

The next morning supporters of Ramai Ram came to the village asking
for votes. They firstly approached the Dusadhs and Chamars. But the
villagers were in a mood to face them down. Only when the vote comes
do you come to us; where is your leader? If the leader has no spare time
for us at this time of the election, will he count money when he becomes
minister or will he come to us?
One of [Dass] supporters pulled out a diary stating that: I will note
down your problem and when Neta-ji will come to power he will implement it. Then one of the Dusadhs standing nearby reacted and said:
Sir, please dont make a fool of me. I am a ragpicker and I have found
many such diaries thrown at the garbage; this diary will also be thrown
at the same place.
In Yadav tola, I found that many people were fans of Mulayam Singh,
and said that it was only due to him that Sonia could not become Prime
Minister. Most of them, being Yadavs, were supporters of Laloo, but in
their opinion the alliance of Laloo with Congress was not correct. We
know the Laloo who was a follower of Jai Prakash Narayan, Lohia-ji and
who now claims that I am the only leader who is protecting for social
justice.2 But he does just the reverse of it. It is not others but our own
people who are teasing us.
There were comments also on the speech of Rabri Devi at Raja Pakar
Block [dominated by the Yadav caste]. She had said that whenever Ram
Vilas will come to you: You have to throw stones at him and this was
matter of a criticism. The message to her was that you may do unfair
measures while in party office but in public meetings you must follow
some accepted norms.
R T Paswan [of Gurmiswa] was saying at Deba Chowk that: I
have studied somewhere in the newspaper that our President is also a
Dalit. But he never takes steps to protect our rights. Then what is the use
of making a Dalit President? The President should instruct the election
commission either to make separate arrangements for Scheduled Castes
or resign from the post of Presidentship. Another person sitting behind
him said that is not a good sign for the Dalits as every leader had talked
about Dalits for many years and nobody in reality did anything for them.
Many leaders divi