Politics and Governance in Bangladesh: Uncertain Landscapes 2017011563, 9781138707610, 9781315201337

515 36 13MB

English Pages [213] Year 2018

Report DMCA / Copyright


Polecaj historie

Politics and Governance in Bangladesh: Uncertain Landscapes
 2017011563, 9781138707610, 9781315201337

Table of contents :
Title Page
Copyright Page
Table of Contents
List of figures
List of tables
Notes on contributors
Introduction: contesting political space: who governs in Bangladesh
Agrarian legacy: from feudal deference to mastaanisation
Institutions and organisations: the reality-rhetoric gap
Bengali or Bangladeshi? The foundations of present political contention
Deep structures: the social and cultural prison
Chapter outline
1 Party dysfunction and homeostasis in Bangladesh: the old disorder restored (or not)
Democracy’s trajectory in Bangladesh
Rules of the dysfunctional game
The caretaker interregnum and the elections of 2008
The old (dis)order restored
Breaking away from homeostasis
Misperceived homeostasis?
A third scenario
2 Where are the drivers of governance reform?
The challenge
A paradigm for state building
How does Bangladesh fit the paradigm?
The potential drivers of governance reform
Donor support for governance reform
The way forward
3 Citizen-centered governance: lessons from high-performing Asian economies for Bangladesh
Governance and development
Lessons of development from high-performing economies with special reference to Singapore
Democracy and development
Future directions: a blind alley or a rising sun?
4 Governance, rights and the demand for democracy: evidence from Bangladesh
Governance in Bangladesh between frustration and hope
Support for democracy: ambiguous democrats and ambiguous theocrats
Ambiguity around governance: a question of being pragmatic
5 Deconstructing the natural state? Is there room for de Tocqueville or only Gramsci in Bangladesh
Significance of the natural state for civil society
The concept of permeability
Apparent and hidden behaviour
‘Natural, limited access states’ applied to Bangladesh
Social origins of present deep structures
The surreal context for civil society
The demand side role of civil society
Holistic and marginal risk: sustaining the demand side
The PMUS case
6 When things go wrong in NGOs: what can be learned from cases of organisational breakdown and ‘failure’?
NGOs in Bangladesh
Narratives of NGO decline, disintegration and ‘failure’
Case study: Comilla Proshika
Making sense of ‘things going wrong’
What can we learn from all this?
Conclusion: ‘failure’ as process and progress
7 The significance of unruly politics in Bangladesh
Introduction: why political rule- breaking matters
Some ideas about unruly politics
Unruly politics in Bangladesh
Conclusions: what unruliness means for the governance agenda
8 Governance challenges in Bangladesh: old wine in not so new bottles?
Using governance to understand political developments in Bangladesh
Bangladesh: a brief governance update
Peeking into the future

Citation preview

Politics and Governance in Bangladesh

Since its Independence in 1971, Bangladesh has made remarkable progress in terms of reducing poverty levels, achieving high levels of economic growth over a sustained period of time, and meeting its Millennium Development Goals (MDG) targets set by the United Nations. With some justification, Bangladesh is considered an international development success story, and the country appears to be well on track to meet its policy target of becoming a middle-income country by 2021, the same year the country will celebrate 50 years of Independence. This book explores the central issue of Bangladeshi politics: the weakness of governance. The coexistence of a poor governance track record and a relatively strong socioeconomic performance makes Bangladesh an intriguing case which throws up exciting and relevant conceptual and policy challenges. Structured in four sections – Political Settlement, Elites and Deep Structures; Democracy, Citizenship and Values; Civil Society, Local Context and Political Change; Informality and Accountability – the book identifies and engages with these challenges. Chapters by experts in the field share a number of conceptual and epistemological principles and offer a combination of theoretical and empirical insights, and cover a good range of contemporary issues and debate. Employing a structurally determinist perspective, this book explains politics and society in Bangladesh from a novel perspective. Academics in the field of governance and politics in developing countries, with a focus on South Asia and Bangladesh, will welcome its publication. Ipshita Basu is Senior Lecturer in International Relations at the University of Westminster, UK. Joe Devine is Senior Lecturer in the Department of Social and Policy Sciences, University of Bath, UK. Geof Wood is Emeritus Professor of International Development and Visiting Professor at the Centre for Development Studies, University of Bath, UK.

Routledge Studies in South Asian Politics

1 Nepal and the Geo-Strategic Rivalry between China and India Sanjay Upadhya

7 Public Policy and Governance in Bangladesh Forty Years of Experience Nizam Ahmed

2 Security Community in South Asia Muhammad Shoaib Pervez

8 Separatist Violence in South Asia A Comparative Study Matthew J. Webb

3 Refugees and Borders in South Asia The Great Exodus of 1971 Antara Datta 4 India’s Human Security Lost Debates, Forgotten People, Intractable Challenges Edited by Jason Miklian and Ashild Kolas 5 Poverty and Governance in South Asia Syeda Parnini 6 US–Pakistan Relations Pakistan’s Strategic Choices in the 1990s Nasra Talat Farooq

9 Pakistan’s Democratic Transition Change and Persistence Edited by Ishtiaq Ahmad and Adnan Rafiq 10 Localizing Governance in India Bidyut Chakrabarty 11 Government and Politics in Sri Lanka Biopolitics and Security A.R. Sriskanda Rajah 12 Politics and Governance in Bangladesh Uncertain Landscapes Edited by Ipshita Basu, Joe Devine and Geof Wood

Politics and Governance in Bangladesh Uncertain Landscapes

Edited by Ipshita Basu, Joe Devine and Geof Wood

First published 2018 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2018 selection and editorial matter, Ipshita Basu, Joe Devine and Geof Wood; individual chapters, the contributors The right of the editors to be identified as the authors of the editorial matter, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Every effort has been made to contact copyright holders for their permission to reprint material in this book. The publishers would be grateful to hear from any copyright holder who is not here acknowledged and will undertake to rectify any errors or omissions in future editions of this book. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Names: Basu, Ipshita, editor. | Devine, Joe, 1963– editor. | Wood, Geoffrey D., 1945– editor. Title: Politics and governance in Bangladesh : uncertain landscapes / edited by Ipshita Basu, Joe Devine and Geof Wood. Description: Abingdon, Oxon ; New York, NY : Routledge, 2018. | Series: Routledge studies in South Asian politics ; 12 | Includes bibliographical references and index. Identifiers: LCCN 2017011563| ISBN 9781138707610 (hardback) | ISBN 9781315201337 (ebook) Subjects: LCSH: Bangladesh–Politics and government–1971– | Political culture–Bangladesh. Classification: LCC JQ635 .P67 2018 | DDC 320.95492–dc23 LC record available at https://lccn.loc.gov/2017011563 ISBN: 978-1-138-70761-0 (hbk) ISBN: 978-1-315-20133-7 (ebk) Typeset in Times New Roman by Wearset Ltd, Boldon, Tyne and Wear

In loving memory of Dr Jalal Alamgir. A generous, passionate and promising intellectual who was taken away much too soon.


List of figures List of tables Notes on contributors Acknowledgements Introduction: contesting political space: who governs in Bangladesh

x xi xii xiii



Agrarian legacy: from feudal deference to mastaanisation 5 Institutions and organisations: the reality-rhetoric gap 6 Bengali or Bangladeshi? The foundations of present political contention 7 Deep structures: the social and cultural prison 10 Chapter outline 13 1

Party dysfunction and homeostasis in Bangladesh: the old disorder restored (or not) HARRY BLAIR

Democracy’s trajectory in Bangladesh 18 Rules of the dysfunctional game 20 The caretaker interregnum and the elections of 2008 22 The old (dis)order restored 24 Breaking away from homeostasis 29 Misperceived homeostasis? 30 Prognosis 31 A third scenario 32 Conclusion 35


viii 2

Contents Where are the drivers of governance reform?



The challenge 39 A paradigm for state building 39 How does Bangladesh fit the paradigm?  42 The potential drivers of governance reform 47 Donor support for governance reform 54 The way forward 58 3

Citizen-centered governance: lessons from high-performing Asian economies for Bangladesh



Introduction 62 Governance and development 68 Lessons of development from high-performing economies with special reference to Singapore 74 Democracy and development 79 Future directions: a blind alley or a rising sun? 82 4

Governance, rights and the demand for democracy: evidence from Bangladesh



Introduction 86 Governance in Bangladesh between frustration and hope 89 Methods 93 Support for democracy: ambiguous democrats and ambiguous theocrats 93 Ambiguity around governance: a question of being pragmatic 98 Conclusion 103 5

Deconstructing the natural state? Is there room for de Tocqueville or only Gramsci in Bangladesh GEOF WOOD

Introduction 107 Significance of the natural state for civil society  108 The concept of permeability 110 Apparent and hidden behaviour 110 ‘Natural, limited access states’ applied to Bangladesh 111 Social origins of present deep structures 113 The surreal context for civil society 114




The demand side role of civil society 116 Holistic and marginal risk: sustaining the demand side 116 The PMUS case 117 Conclusion 121 6

When things go wrong in NGOs: what can be learned from cases of organisational breakdown and ‘failure’?



Introduction 125 NGOs in Bangladesh 127 Narratives of NGO decline, disintegration and ‘failure’ 130 Case study: Comilla Proshika 132 Making sense of ‘things going wrong’ 134 What can we learn from all this? 137 Conclusion: ‘failure’ as process and progress 140 7

The significance of unruly politics in Bangladesh



Introduction: why political rule-breaking matters 143 Motivations 144 Some ideas about unruly politics 150 Unruly politics in Bangladesh 158 Conclusions: what unruliness means for the governance agenda 164 8

Governance challenges in Bangladesh: old wine in not so new bottles?



Introduction 168 Using governance to understand political developments in Bangladesh 169 Bangladesh: a brief governance update 172 Peeking into the future 174 References Index

177 192


1.1 1.2 1.3 7.1

Bangladesh Freedom house scores, 1972–2010 Political stability and social peace in South Asia, 1996–2010 Bangladesh media ratings, 2002–2010 The number of hartals in democratic and military regimes, 1947–2002

18 33 34 159


1.1 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13 3.14 4.1 4.2 4.3 4.4 4.5 4.6 8.1

Vote and seat share, 1991–2008 hDI of South Asian and selected Asian countries, 2013 Growth rate of GDP (% of year) of South Asian and selected Asian countries Global competitiveness index, 2012–2013 (selected variables in selected Asian and BRIC countries) Democracy index of South Asian and selected countries World competitiveness index, basic requirements: selected Asian countries, 2010–2011 Voice and accountability Political stability and absence of violence/terrorism Government effectiveness Rule of law Most problematic factors for doing business Central government expenditure as percentage of GDP of South Asian and selected Asian countries quality of democracy (Freedom house) EIU classification Freedom index, 2011 Perspectives on an Islamic State Probit results for ‘ambiguous democrats’ and ‘ambiguous theocrats’ Factor loadings for indices Ambiguity and justifications for democracy: t-test statistics OLS results for pragmatism index and ethicality index Probit regressions: is democracy always preferable? Comparison of Bangladesh’s WGI performance

19 63 64 65 66 71 72 73 74 75 78 79 80 81 81 95 97 99 100 101 102 169


Ipshita Basu is Senior Lecturer (Associate Professor) in International Relations at the Department of Politics and International Relations at the University of Westminster. She is a political sociologist and her research specialises on the politics of identity and distributive justice in relation to development and governance, with a regional focus on South Asia. Her articles have appeared in Development and Change and Contemporary South Asia. Joe Devine Head of Department and Senior Lecturer in International Development at the Department of Social and Policy Sciences, University of Bath. His main research lies in well-being, poverty and inequality. Geof Wood is Emeritus Professor at the Department of Social and Policy Sciences, University of Bath. He is a sociologist, specialising in International Development with a regional focus on South Asia. Harry Blair is Visiting Fellow in Political Science at the University of Yale. His research focuses on democratisation in developing countries, especially on civil society and decentralisation. Graham K. Brown is based at the Department of Social Policy, University of Western Australia. He is a political scientist with interest in inequality and political mobilisation in South East Asia. Naomi Hossain is Research Fellow at the Institute of Development Studies, Sussex. She is a political sociologist and her work focuses on the politics of poverty and public services. Habibul Haque Khondker is Professor at Zayed University, Abu Dhabi. His research interests cover theories of globalisation and global modernity. Pierre Landell-Mills was Country Director of World Bank, Bangladesh and served in the World Bank for 26 years, until he retired in 1999. He has a Master’s Degree in Economics from Cambridge University. David Lewis is Professor of Social Policy and Development at London School of Economics. His research is concerned with international development policy, particularly the role of NGOs and civil society in development.


This book was made possible by the generous support of the Institute of Governance and Development (IGD; previously Institute of Governance Studies (IGS)), BRAC University, Dhaka Bangladesh. The papers were originally presented at the 40 Years of Governance in Bangladesh: Retrospect and Future Prospects in December 2011. We owe special thanks to Dr Rizwan Khair (former Director of IGS) and Barrister Manzoor hasan (Advisor and former Director of IGS) for their stewardship. The IGD has a long-standing reputation for producing the annual State of Governance in Bangladesh and State of Cities reports which are highly influential in policy and public debates. Part of the reason for producing this book is to build on the pioneering initiative that Manzoor hasan and Rizwan Khair began by consolidating critical research and policy uptake on the governance question for Bangladesh and beyond. We are also grateful to researchers and administration of IGD and BRAC University who pulled out all the stops to bring together established and emerging scholars from across the world for this book. Rubayet hamid, Nahid Waliul, Salina Aziz, Ekram hossain and Salahuddin Ahmed have made pivotal contributions to make the conference and subsequently this book possible. Finally, we are grateful to the long-standing friendships between the editors and the authors of this book as one knows that such efforts rely very much on voluntarism and the commitment to work beyond the rigours of our daily jobs and personal lives. Thanks for your patience, solidarity and continued support.

Introduction Contesting political space: who governs in Bangladesh Ipshita Basu, Joe Devine and Geof Wood

Bangladesh has been politically independent of Pakistan for 44 years now. Both the movements against British colonial rule and later against the Punjab-led elites of West Pakistan were motivated around ideas of and hopes for political freedom as the route to improved wellbeing and material livelihoods for the hard-pressed peasantry of East Bengal. In the scheme of things, perhaps 44 years is too short a period to deliver those initial high expectations. However, many were in a hurry, not least the donors who were eventually prepared to swamp the delta with aid across many government sectors, and then across NGOs when they despaired of the government’s ability to implement policy without largescale corruption and rent seeking among political classes. While aid support for successive governments has never been abandoned, it increasingly included the agendas of public administration reform and improving capacities for voice from civil society, while offsetting risks of ongoing corruption through enclaved projects with a fig leaf of government ownership, but actually run through relatively autonomous programme staff, involving many expatriates. This formula persists to date. When aid became more poverty focused for the major donors at the end of the 1980s, so their preoccupations with governance also grew. Thus emerged a normative set of goals derived from western mythology of liberal-democratic, pluralist societies holding governments and powerholders to account. This liberal bourgeois rhetoric misrepresented real political practice and class-based rent seeking in the industrial and post-industrial west, but was naively, or conspiratorially, applied as a benchmark to societies emerging from recent agrarian, quasi-feudal pasts. In other words it was part of the modernisation pantheon, the classic teleological discourse of development. And persistent failure of real behaviour to match the rhetoric has led to a wringing of hands among the aid community and their comprador bourgeois accomplices, heralding a search for drivers of change either from enlightened reformers within government or from a strengthening of the demand for improved governance and accountability from civil society actors, with a sprinkling of incentive-based ideas about political economy. Clearly, however, persistent failure to enact a western-derived normative agenda of good governance has to be explained in more realistic terms than just venality and changing incentives. Something deeper is going on.


I. Basu et al.

This book, derived from a meeting in Dhaka in late 2011 to reflect upon change and development in Bangladesh, as well as from subsequent exchanges of the meeting’s participants, offers this reality check to the normal governance discourse and wringing of hands. While referencing the Bangladesh experience, we suggest that our arguments have much wider application to an understanding of societies with recent pastoral and agrarian pasts yet jettisoned into the glare of globalisation. We all agree from different perspectives that too much analysis of ‘obstacles to development’ (alas still a prevailing paradigm after 50 years of development studies) has ridden roughshod over the facts of complex social cultures, ignoring history and the values and motivations of populations as they contest for survival and wellbeing in conditions of scarcity. This is not to say that we do not have values ourselves. We wish much for others, but mainly we wish their power to set their own agendas following the core principle of subsidiarity – that destiny should be devolved to the smallest possible unit of action and identity without threatening the interests of the wider collective. And threats are not differences, which can be tolerated. Of course this simple proposition is itself innocent, as we can never be sure of another’s real preferences, surrounded as they are by the powerful ideas, values and theologies of others. So we have to recognise that power, and the resistance of subsidiarity, is not just in the first dimension. Robert Dahl’s study of New Haven understood that, as did Stephen Lukes a decade later. Much governing occurs in the dark places away from the gaze, but that is more difficult in smaller universes where Naomi Hossain’s, one of the contributors to this book, unruly politics comes into play – the indignant wife challenging the dull patriarchy of a conforming husband in short loop accountability. Subsidiarity is about contested spaces, spaces nested within each other, reaching aggregate levels of significance for the big public goods issues and for the coalitions of interests setting the parameters within which others have to duck and weave. So subsidiarity is both a politically desired value for us as well as a concept, a framework for understanding the real political behaviour of contending actors in the society. We need to say this, because we would not wish the reader to think that any analytical relativism on our part is an abdication of judgement about the harm that men, and of course women, can do to each other, either through violence or the credible threat of it, or through other discriminations, material but also psychological. But we must accept that the circumstances of change and development of a society like Bangladesh is more likely to be a product, an outcome, of its historical mix of social institutions and cultural perfomances than an outcome of alien values which are not even realised in their alien birthplace. How strange it is to suppose that the Bengali should be anything other than Bengali, or that if there is a deluge of alien induced opportunities, that they will not be seen through Bengali eyes, and refracted by them to achieve compatibility with other, long-standing social practices and priorities – a mobile phone not for communicating with a chosen lover, but for tracking and controlling the whereabouts of a woman obliged to seek work outside the home and para. For mobile phone, substitute the idea of governance. Why should not that be refracted too? Can this alien language be deployed to legitimate forms of power,



derived not from ideological popularity and trust but from post-feudal forms of personalised control backed up by threats of violence perpetrated through gangs of loyal clients? Talk the talk with no intention of walking the walk, because to do so would be political suicide. So we need to stand back from subtractionist, teleological analysis remonstrating with a target society for not reproducing the governance mythology of the supposedly superior west, which is significantly further along the road of transparency, accountability and responsive government for all, and thus a model for other backward societies to follow. Our deliberate use of emotive terminology hopefully reveals the absurdity of such forms of analysis. No self respecting social scientist should be indulging in this fanciful discourse – analysing a society by contrasting it to what it does not have. Instead, we must ask the more descriptive question: who governs, and how; who contests, and how? Engaging with these questions will go some way to unravelling the way power is actually exercised in the society, the forms of contestation which arise, and whether such a dialectic reaffirms the structures of nested subsidiarity (i.e. stasis) or transforms them incrementally in ways which enable more autonomy for ordinary people over some more aspects of their destiny. Political freedoms, in other words. If we can educate the reader through asking these questions and using the case study of Bangladesh we will have contributed to a better understanding of the interface between political process and the quality of governance. It will tell us something about drivers for change, too – whether improvement occurs through struggle and threats to privilege, or through enlightened and clever elites perceiving those threats and preadjusting to them via reforms of some sort. What is our repertoire for this ‘education’? What is our curriculum? A decade ago, Wood (2000) proposed a metaphor of prisoners and escapees to characterise the core problem in terms of a contrast between those acting within history and those seeking to confront that history through a normative paradigm of good governance. For history, read ‘deep structures’. Their presence is the key reality check. Thus while many outsiders have seen NGOs and other civil society movements as representing the escape party from the all embracing prison of social structure and cultural performance, others ask how can such NGOs be expected to break free of surrounding and prevailing organisational culture? Why should they be different rather than reproducing within themselves the practices they purport to challenge? We have plenty of evidence of NGOs in Bangladesh failing to break free and being incorporated or reincorporated into business as usual. David Lewis draws our intention to this in Chapter 6 of this book. It is difficult, even dangerous, to stand apart when most others around you conform. So what is business as usual? What does the prison look like? Does the analogy hold up or should it be revisited? This introduction seeks to engage with this agenda as the thread linking the following chapters to each other. In particular it requires a link between the structural schema of North et al.’s limitedaccess state arguments and the more ethnographic accounts of mastaan and unruly politics in Bangladesh.


I. Basu et al.

The prisoners and escapees metaphor helps understand the entrapment of all classes and groups in Bangladesh in deep structures and institutional practices characterised by extensive patron-clientelism, kin and other loyalties, and sets of multi-period, non-transparent obligations and favours. There now seems to be a broad acceptance of that analysis, with a shared understanding of why ‘escape’ is so risky and difficult. That analysis was assisted by the use of the ‘institutional responsibility framework’. Since then, there have been three further conceptual innovations. First, the adaption of that institutional responsibility framework into a comparative regime model to show how rules of the game can vary as a function of institutional domain choices (Wood and Gough 2006). Second, Faustian bargain arguments have been developed to show how problematic institutions are reproduced through continuous resort to them for short-term security (Wood 2003). And, third, North et al. in Violence and Social Orders (2009) have advanced the thesis of limited-access states to show how exclusive elites have to enter steadily wider coalitions of interests to protect their original ones, thus extending access to state-related rents while continuing to limit the access of rent providers and thus falling short of the goal of an open-access state. All three of these developments help understand the prison better, enabling analysis to be more dynamic and fluid and thus sensitive to all kinds of potential game changers, as we have seen in various ‘Springs’ since Prague and more recently in North Africa. Through these additional lenses, we can ask: what are the forces in and around Bangladesh which might enable more voice, more threats of exit, reduce loyalty to the warders and a shift towards a more open-access state, organised through the principle of subsidiarity? However, in order to pursue our reality check on this normative pursuit of the open-access state, we have to outline the epistemological perspective which runs across many of the contributions in this volume. Political analysis always reveals an explanatory tension between contingent facts and longer-term structural conditions. Despite a particular, and indeed personalised, history in the formation of political behaviour in Bangladesh, it is necessary to retain a structurally determinist perspective. In other words, do the deep structures of Bangladesh, and countries like it, inevitably produce rent seeking political behaviour and political actors? And, therefore, if it was not the present alternating rent-seekers represented by either the Awami League and the Bangladesh Nationalist Party (BNP) with their various partners, it would be someone else acting in similar ways. This would be consistent with the idea that there is a political ‘warder’ class in Bangladesh occupying a structural position in the society performing functions which essentially characterise the state and its relationship with non-state society. This explanatory tension requires an analysis which interweaves the specificity of Bangladesh history with the generalised structural conditions of a society moving away from a recent agrarian past into ‘we know not what’. This epistemological perspective also entails an appreciation of processes of dynamic reproduction. Even from a structurally determinist perspective, we are not saying that there are fixed patterns of relationships and rules (institutions in a Northian sense) which exist independently of the actors who populate their nodal



points. The socialised behaviour of political and public actors is learned, inherited and morally legitimated within peer groups. That behaviour is a function of social incentives, maintained through reward and sanction. The repeated nature of such behaviour produces the rules and de facto codes of conduct and thus institutions are both reproduced and refined. This is a structuration perspective (Giddens 1984; Long 1992). The significance of this is that political actors in Bangladesh and elsewhere invest heavily in their behaviour and the de facto or informal codes of conduct which emanate. They are, as it were, highly committed to being inmates. That, of course, is partly a function of the absence of realistic choice to behave otherwise – life on the outside being even more uncertain and unreliable than the inside. It is in this sense that we are gloomy about the prospects of Bangladesh moving, in any near future, towards a more open-access state – because of active processes of reproduction not just given sets of conditions. In order to explore that gloomy reality check upon the prospects for a more open-access state, characterised by the normative principles of good governance (even if honoured in the breach by most societies, including donor ones) we do have to offer a stylised account of these deep historical structures in Bangladesh which shape present political behaviour. We do so, first, by considering the significance of the recent agrarian past of Bangladesh which was the framing conditions for the current second generation of a post-liberation political class in the country.

Agrarian legacy: from feudal deference to mastaanisation Although the liberation narrative for Bangladesh juxtaposed an incipient nation of small peasants against the landlord and bureaucratic-military elites of the Punjab, the East Bengal delta was more complex in its agrarian structure (Wood 1981) and by no means as homogenous as the narrative implied. Of course, that narrative was politically heuristic and thus a deliberate construction within a fission and fusion framework of political identity. The reality was that while there were key pockets of minifundist, small peasant farming as in the DhakaComilla belt, elsewhere there were more significant landlords, sharecropping tenants and effectively landless labour. And even within the minifundist region, the archetype of Awami League mythology, there were key indicators of inequality, dependency and patron-clientelism (Huq 1976). And of course, East Bengal/Pakistan in 1971 was very rural, unusually so even for South Asia. There was a tiny urban elite, mainly in Dhaka with landed property as the basis for bureaucratic, education, trading and manufacturing roles – an essentially comprador elite incorporated into the East Indian Company and British colonialism (see IGS 2011a). This admittedly over-simplified characterisation reveals to us that only a single generation ago, Bangladesh was agrarian and pre-capitalist in its mode of production and exchange. Power was exercised essentially through family and patron-client forms of exchange through concentric circles of moral proximity and intimacy, and through diffused, multi-dimensional, multi-period


I. Basu et al.

games. Clearly much theory and empirical observation is collapsed and condensed into this summary account of power. However, this conceptual formulation of agrarian forms of power in Bangladesh can be traced through to contemporary socio-economic structures and takes its form in the idea of ‘deep structures’. Pursuing this line of argument further, South Asian, ex-British and recently agrarian societies, are characterised by severe socio-cultural as well as economic inequalities. There is considerable deference to those above you on the social scale, however expressed – caste, bangsho1 and so on. Within this elite-mass, agrarian inheritance, most people thus have a deep rooted fatalism and de facto tolerance of non-transparent, non-accountable, undemocratic practices. They have expected their leaders to behave as zamindars2 with unchallenged authority. These feudal legacies have now evolved via the arrival of political parties into a mastaanisation of the countryside and the society at large (Devine 2007),3 an extension of urban mafia-like institutions, indicative of rapidly evolving opportunities,4 mediating between people’s immediate needs and imperfect markets and states in which these needy ‘clients’ have no power or influence. The tolerance of these institutionalised practices thus reflects socio-cultural history, but now reinforced by more superficial threats and inducements.

Institutions and organisations: the reality-rhetoric gap This context provides us with a key contradiction for Bangladesh, and for other societies like it in South Asia and beyond: namely the gap between the institutions which effectively shape the society and the formal organisations which apparently populate the landscape. Thus, for example, a political party may exist as a formal organisation competing in elections, providing governments and participation in the Jatiya Sangsad, but its institutional profile resembles a traditional dal5 comprising leaders and followers bound together by highly personalised, multiplex tiers of obligation, favour and dependence as well as interdependence. This contradiction, sometimes understood as the gap between rhetoric and reality, is well known in Bangladesh among its political actors and among its governance-oriented donors. It provides the setting for a complex charade of smoke, mirrors and shadows. Thus the political class among Bangladeshis and donors in Bangladesh live in a dual world of implicit institutional practices and explicit organisational pretension. It is interesting that the donors collude in this duality, based either upon aid imperatives6 or upon the idea that organisational reform might change underlying institutional principles – a kind of reverse causation. So we might expect political parties to be unable to separate themselves from prevailing institutional cultures since they are in the business of securing and exercising power by whatever means, even including violence as long as it is deniable. In his chapter in this volume, Wood considers the stronger test of this ‘reality/rhetoric gap’ proposition – the formally created non-governmental organisations, which potentially indicate a transition to open-access society.



They are supposed to be breaking away from business as usual, but they find it difficult to do so due to their essential embeddedness in the social institutions of the society. This argument is significant in providing a structural or socially grounded critique of the limits of the demand side to act as a driver for a more open-access state, in Northian terms.

Bengali or Bangladeshi? The foundations of present political contention Let us now apply our core epistemology to understand the terms of contestation of power in Bangladesh since liberation. It is worth very briefly outlining the sequence of regimes which have ruled in Bangladesh: 1971–75 1975–81 1982–90 1991–96 1996–2001




Awami League (evolving into one party BAKSAL). Military, following a series of coups (August–November 1975), headed by Major-General Ziaur Rahman. Military, following assassination of Ziaur Rahman, headed by Lieutenant-General M. Ershad. Democratically elected Bangladesh National Party (BNP), headed by Begum Zia, wife of deceased Ziaur Rahman. Democratically elected Awami League (AL), headed by Sheikh Hasina Rahman, daughter of deceased founding father (Bangabandhu) Sheikh Mujibur Rahman, assassinated in the coup of 1975. Democratically elected BNP, supported by Jamaat-e-Islami, a small fundamentalist party. During this period, there was increasing disturbance, including the assassination of AL leaders, with violence attributed to out-of-control radical Islamist groups. Official interim Caretaker Government, headed as per constitution by the immediately past Chief Justice, with technocratic/ appointed advisors. This interim administration attempted to convene elections in December 2006 on the basis of an out-ofdate election register, which would have favoured the recent incumbent party BNP. AL objected. The resulting stand-off produced: Military-backed Caretaker Government (CTG), operating outside formal constitutional provision (the ‘coup’ of 11 January 2007), but with a mission to clean up politics and pursue corruption. The period until elections on a new register (December 2008) was regarded by donors as a window of opportunity for improving governance. CTG was headed by Fakruddin Ahmed, an economist with past service in the World Bank, ex Governor, Bangladesh Bank and ex-Chairman of Palli Karma Sahayak Foundation (PKSF – a wholesale social fund supporting microfinance providers).


I. Basu et al.



Democratically elected AL government, with a massive landslide (270 seats, including nominated women in the Jatiya Sangsad, with BNP 35 and Ershad’s Jatiyo party 30), reflecting the anti-incumbent tradition of voting. BNP and main opposition parties boycott the election. Of the 300 parliamentary seats, 154 were uncontested. Polling day was marred by violence. AL takes office having secured just under 80 per cent of seats.

This sequence of regimes indicates that the military are never far away from seizing control if law and order breaks down or an incumbent political party seeks to steal an election, with law and order implications. In a curious, ironic way, this hovering presence of the military might indicate a mature natural, limited-access state rather than a basic or fragile one. The constitutional provision for an interim caretaker government (CTG) was designed to prevent incumbent theft. In early 2007, the official CTG was not strong enough to prevent that prospect, hence the military intervention again. When in opposition, parties have declined to sit in the Assembly in an attempt to deny legitimacy to the incumbent party, but as a result have not been performing their accountability functions. The 2009–13 AL government has used its landslide majority in the Jatiya Sangsad to legislate for the removal of the provision for a CTG prior to election, again raising the prospect, at least for the opposition, of incumbent theft. Behind this formal sequence of contingent facts lie, of course, deep sociopolitical processes inseparable from circumstances of liberation and the structural position of the new Bangladesh both internally and externally. The first challenge had to be the transformation of the Awami League from an inclusive populist movement, deploying a strong sense of Bengali nationalism in contrast to a Punjabi or combined Pakistani identity, into a responsible and competent governing party. Its liberation narrative constructed the delta as comprising a homogenous class of small peasant landowners with a history of exploitation from ‘external’ Hindu landlords (zamindars) through land rents and more recently a Punjabi dominated trading class, extracting rents on commodity production (especially jute) and thereby underdeveloping East Bengal at the expense of Punjabi ‘metropolitan’ accumulation. It was a dependista narrative underpinning the deployment of the interactive identity of Bengali-Muslim. Thus internal class differentiation between Muslims, whether between the intellectual Dhaka elites and the rural masses, or between Muslim jotedars (rich peasantry), raiyats (landholders), bargadars (sharecroppers) and bhumihin (landless), was ideologically and tactically underplayed. The transformational challenge was further compounded by the decimation of the urban educated elite who were due to populate the new state as bureaucrats and politicians. But instead of the ‘A team’, the incoming leadership had to rely upon loyal followers without relevant education and experience. The society and its infrastructure had also been stripped bare and plundered by a combination of Punjabi occupying forces and then Indian army ‘liberators’,



while new markets with new logistics had to be found for primary products as the basis for accumulation and rents.7 The surviving Awami League leadership first greeted the returning hero, Sheikh Mujibur Rahman, then promptly used his name to legitimate their predation upon the society, using the artificial monopolies arising from state management of markets as the basis for excessive rent seeking. There was increasing anarchy across the society as the ‘inheritance’ of the rent seeking Awami League was disputed by excluded factions among the army and from other liberation groups. This resulted in the short-lived attempt to create a stronger monopoly of control through BAKSAL—formally a coalition but so dominated by the Awami League as to amount to one party rule. This reaffirmation of the limited-access state under no-growth conditions sharpened the zero-sum game element of political competition, prompting the August 1975 violent coup by those elements in the army and bureaucracy whose own liberation claims were ignored by the exclusionist Awami League inner, and some would say family, circle. The final outcome of three months’ instability in November 1975 laid the foundation for the essential, and contemporary, dispute over the soul of Bangladesh in terms of liberation and national identity. The incoming military-bureaucratic government of Maj.-Gen. Ziaur Rahman (up to 1981) based its initial legitimacy on his leadership of liberation military forces in exile and then alongside the invading Indian army, and the mission to rescue the new Bangladesh from the anarchy and plunder of its first few years. Beneath this lay other missions: to restore rank and seniority in the army; to establish the domination of these interests over the faction of the army that had remained inside East Pakistan using guerilla tactics to confront the Pakistani crackdown; to moderate the secular Bengali identity akin to the politics of West Bengal in India, by emphasizing the Islamic heritage of being Bangladeshi; to re-align Bangladesh away from Indian dependency (backed by Soviet influence) and towards both western aid as well as middle-eastern petro-dollars as de facto guarantors of Bangladesh’s independence in the South Asian region; and thus to promote urban and small business development as a constituency of clients to compete with the Awami League’s rural peasantry. The route to sustainable legitimacy for this regime was the formation of the Bangladesh National Party, combining a moderate Islamic identity with this new constituency. This ‘opening’ story, still heavily summarised, has been recalled for its importance in establishing the basis of the major present competition for control over state resources between two major factions, each intent upon excluding the other while bent upon mutual delegitimation. The Ershad period between 1982 and 1990 itself reflected factional competition within the army, but in other respects continued Ziaur’s development legacy and tactics for legitimation in the formation of the more regional Jatiyo Party. His constituency building was two pronged: through decentralization and the promotion of local government; and through permitting a wider expansion of NGO activity, thus seeking support of their clients, created out of aid patronage. Ershad’s period has left some interesting, and potentially positive development legacies, but he and his party have been consigned to a junior coalition role, given that his party has been unable to


I. Basu et al.

reach out beyond its regional base. Thus in late December 1990, the stage was set for the basic structure of political contention as we know it today, two decades later. There has been a recent interlude of military-backed government arising out of the inability of the 2006 CTG to manage elections with an up-todate electoral role. This government (2006–08) started with some popularity for its narrative of clean up and anti-corruption, involving the house arrest of the two main protagonists and some of their associates. The subsequent landslide, anti-incumbent election has brought back ‘business as usual’, a reproduction of ‘prison’ conditions, perhaps slightly more open given the increasing prevalence of Hossain’s unruly politics, but also some evidence in support of Blair’s stasis thesis (Chapters 7 and 1 in this volume).

Deep structures: the social and cultural prison Behind the regime headlines listed above, there has been a continuity of problematic practices and behaviour within government among politicians and public officials and between them and the rest of the society, without effective regulation in governance terms either from apparatuses of government or from the civil society. These practices have been analysed by various writers over the last decade including Wood (2000), Landell-Mills (2002) and, more recently, BRAC’s Institute of Governance State of Governance Reports (IGS 2006, 2008a, 2009a, 2011a). The IGS analysis offers a deeper reflection on causes of poor practice, than the more factual, descriptive information from TIB and elsewhere. Political and organizational behaviour derives from deep structures (cultural and social institutions) which pervade the society and is thus continuous across regimes. In other words, the successive ruling parties cannot be distinguished from each other with respect to problematic underlying behaviour. They are more accurately to be understood as large factional agglomerations of patronclient favours and loyalties which extend from top to bottom of their organisations. Governance analysis typically examines formally constituted organisations such as: political parties, local government, civil service, police and the judiciary. However, they all behaviourally exhibit the deep structure phenomena. This comprises the subversion of formal organisational rules and expectations about conduct and delivery in order to meet personalised objectives for oneself, one’s clients or patrons (depending on direction of informal transaction and reward), for one’s immediate, near and wider kin, as well as others to whom past obligations are owed or to create loyalties for the future. This can be understood as concentric circles of attachment. To achieve these objectives there is widespread rent seeking and straightforward corruption,8 alongside tampering with formal systems of Human Resources (HR) management in order to ensure compliance and meaningfully continuous threats of punishment and sanction. In other words, there is widespread capture of formal (i.e. supposedly transparent and therefore accountable) systems (e.g. HR and financial management, as well as delivery of formal rights and entitlements) by informal practices which are therefore hidden, socially



exclusionary, entailing a distortion of public policy by redirecting resources and assets away from intended beneficiaries with supposed entitlements to private purposes where need has not been objectively established and authorised. Public goods and public resources are therefore plundered for the informal, personalised gain of specific individuals or social persona. Any ethical democratic, ‘poverty removal’ and ‘good governance’ objectives of both domestic and international ‘donor’ tax-paying citizens or ‘escapees’ are therefore betrayed. Within this general analysis of a cultural and social prison in which most of the society can – to follow the metaphor – be considered ‘inmates’, albeit for some reluctantly, the challenge, for those seeking both public goods objectives as well as targeted ones favouring the poor, is this socially and institutionally reproductive behaviour of leaders and therefore powerholders across the society. These powerholders are distorting even the idea of policy and incorporating dependent clients, thereby maintaining and reproducing these personalised institutional practices. And leaders and powerholders operate in a pyramid structure at all levels – a hierarchy of predation upon the weak, exploiting deep-rooted cultural traditions of social deference (Hossain 2010). The poor, however much they might wish and seek to escape from this entrapment, dare not risk escape as individuals, despite occasional rude outbursts. As enforced clients, they unwittingly reinforce the institutions which oppress them and deny them right and opportunities – a Faustian bargain.9 The behaviour of such leaders manifests itself inter alia through the practices of political parties and their active members, especially when they are incumbents of office and thus controllers of resources – financial and social as in connections and linkages. The 2008 State of Governance Report (IGS 2008) places a ‘driver’ emphasis upon political party behaviour interfering at all levels of government for private gain. Thus local government is penetrated and managed by local party goons or mastaan, and if local officials including the District and decentralised line ministry cadres do not comply with requests for engineering contract opportunities, for rent seeking on Vulnerable Group Feeding programmes, for allocation of jobs (sinecures) to informally nominated teachers and so on, the local goons will complain to their party seniors and such officials will be punished with poor reports, unattractive transfers, postponed promotions and increments. This process became further reinforced by the announcement in mid-March 2010 that elected MPs (not the nominated women members) will each receive a block allocation of Tk.15 crore10 to spend in their constituencies. Not only does this seriously distort any targeted policy to guide local government and local delivery of line Ministry services, but the use of these funds is not strongly and objectively audited. The same interference occurs in central government, with the Establishments Division subjected to pressure regarding transfers and promotions, as well as the Public Service Commission in its recruitment practices; and in the invoicing and allocative decisions across the line Ministries. The culture of fear among bureaucratic personnel is so strong that no level is prepared to make and take responsibility for formal, documented decisions, which are thus continually pushed


I. Basu et al.

upwards to the political level, where even Ministers might refer quite minor issues to the Prime Minister’s office (PMO) for political vetting, in case someone somewhere could be offended. However, despite this continuation and evolution of ‘business as usual’, there does seem to be a Rubicon when the judiciary and justice system as a whole comes into play. Even though most people experience any security they might enjoy via informal relationships (i.e. the Faustian bargain), there is some expectation that if the state has any value to the society, then it is via the judiciary as the ultimate guarantor of security and personal rights, even if individual access is expensive and socially exclusionary.11 (Retired) Chief Justices are supposed to be so above politics (i.e. representing the state rather than any incumbent government) that they were designated in the Constitution as leaders of Caretaker governments before the recent abolition of the CTG device. Yet the 2008 State of Governance Report (IGS 2008) has a substantial analysis of the increasing politicisation of the judiciary in terms of senior appointments, allocation of cases to be tried, and a pattern of outcomes in which the ruling party interest (i.e. its members, followers, funders and friends) is invariably protected, while the opposition interests are attacked. While the last CTG could be congratulated for separating magisterial and administrative functions in the District and subdistrict (upazila) levels, this separation of judiciary and executive belies a deeper more entrenched truth of the politicisation of the judicial process and thereby the justice system as a whole. Thus the last bastion of any semblance of open-access democracy, an independent judiciary, is threatened. The whole legal process is delegitimised, although a surreal game of court dispute continues, in which no decision is accepted up to the Supreme Court level, and enactment of a decision may anyway not occur if the police have also been bought off. There is a mockery of the superstructure. By drawing on the 2008 State of Governance Report, we are triangulating this ongoing analysis of deep structures (institutions) which explain the ‘way business is done’ within the emerging organizations of the state and governance in Bangladesh.12 There is a persistent cross-cutting theme that no ‘estate’ in Bangladesh has independence, leaving only a few critical beacons which require continuous and predictable support to ensure that independence. This has profound implications for the country’s governance profile, especially on the demand side of governance since any leverage through direct intervention on the supply side (favoured by donor aid programmes) is inherently ineffective in engaging with deep structures. Supply-side reform may be considered necessary, but it will never be a sufficient condition. Motivational, aid-supported, projects without dealing with incentives will never turn these socio-cultural embedded institutional practices around on their own. A good example of that proposition is police reform. Empowered citizens need well functioning supply side organisations of delivery to improve their wellbeing, but the ‘drivers’ logic indicates that, for instance, donor priority13 should remain with the demand side in order to rearrange the incentives (at least the non-financial ones) for good supply side behaviour.



Chapter outline The chapters in this book share a number of the conceptual and epistemological principles outlined above. Each chapter is a conscious attempt to depart from subtractionist and teleological accounts of how target societies like Bangladesh compare with an ideal-typical model of Western liberal democracies. Nevertheless, while there is a conscious attempt at avoiding approaching governance in Bangladesh with a false and pre-given lens, the chapters are underpinned by some form of normative framework with its own share of values and ideas on how democracy can be strengthened in a way that is more contextualised and relevant to Bangladesh. Also, while it is clear that it is unhelpful to approach any society through values and principles derived from far-removed histories and contexts, there is an inherent value and need for comparison. Perhaps more than ever before, governance analysis needs a comparative framework that captures the particularities and nuances of the society under investigation and enables a sensitive and contextualised analysis while situating the society against other, and possibly universal, norms of rights and entitlements. This is the reason the framework developed by North, Wallis and Weingast in their recent book Violence and Social Orders (2009) is so useful and empirically relevant. Ideas found in the book have been applied and developed further on a number of instances by the other contributors of this book (David Lewis, Naomi Hossain, Pierre LandellMills). Both North et al.’s (2009) characterisation of limited-access orders where entrenched elites use their power to acquire wealth and to remain in power by using patronage and coercion, and Wood’s metaphor for Bangladesh as a ‘prison’ where all classes are ‘trapped’ in deep structures and institutional practices that are shaped by extensive patron-clientelism prompt the evaluation of the two concurrent dynamics of constancy and change in the governance of Bangladesh. First, how are permanent structures reproduced through the repetitive social behaviour of public and political actors that shape, give form and maintain institutions? And, second, what are the drivers of change that could potentially generate doorstep conditions for transitioning in incremental, small but substantial ways from limited-access to open-access conditions? In many ways the chapters in this book are explorations of the resilient deep structures that give governance and politics in Bangladesh a coherent recognisable form, which have to be understood and are mediated by both internal and external actors. And concurrently they are also investigations on those evolving dynamics that can potentially give way to change and contestation in ‘unruly’ (see Chapter 7 by Naomi Hossain), ‘unexpected’ and unprecedented ways in Bangladesh. In Harry Blair’s Chapter 1 on ‘Party Dysfunction and Homeostasis in Bangladesh’ he uses the metaphor of homeostasis to describe recent politics in Bangladesh as having the unique quality of restoring itself to its dysfunctional state even after perturbations occur. The dysfunctional instability of politics in Bangladesh has in fact become a stable system with a resilience which is founded on


I. Basu et al.

‘Rules of the Game’ which have evolved since the 1990s. Crucially, these rules are understood and observed by the coalition of parties, political leaders, bureaucrats, the police and most of the general populace. As a result, he argues that even though the period of the Caretaker Government of 2007–08 marked an attempt to break political norms and interactions, it eventually failed and the old order was quickly restored once the Awami League government had been elected back to power. Thus even violent protests in the form of hartals are bounded within this system and are not insurrectionary or even aimed at seriously damaging public or private property and institutions as such. If Blair’s chapter focuses on the resilience of the Bangladesh political system, then Pierre Landell-Mill’s Chapter 2 explores the possible drivers for change. He applies the North-Wallis-Weingast schema of progression from a limited-access to an open-access state through the negotiation among different groups of elites. He finds that while there are stakeholders such as businessmen, the youth and the middle classes who have much to gain from a more accountable and transparent system, the high level of divisive politicisation in most arenas of public life blocks and weakens the demand for genuine reform. In his analysis LandellMills highlights the relevance of deep structures, wherein he explains that the state of poor governance in Bangladesh and the obstructions to reform have to be considered in light of deep social and cultural factors, especially the clientelist and hierarchical relations that prevail over all aspects of public life. Habibul Haque Khondker’s Chapter 3 draws attention to the nuances and complexities of the prescribed model of market-based economic growth and liberal democracy. Bangladesh has had consistently high economic growth despite poor governance but on the back of a flawed democratic model. He compares Bangladesh with other high-performing Asian economies, particularly Singapore. He contends that while there are useful lessons to be learnt from successful Asian economies, the Bangladeshi model of economic development should be founded on combining a strong state with democracy and not at the cost of sacrificing democracy for development. He describes the Bangladeshi model of democracy as an ‘authoritarian democracy’, wherein there is an overwhelming demand for electoral democracy but at the same time the state structure is paternalistic and dysfunctional. In his chapter he explores the ways in which democracy in Bangladesh can be deepened to a more citizen-led model. In line with Blair’s chapter Khondker recognises stasis in the political regime of governance, wherein the particular characteristics of the regime have remained more or less intact through the transition from civilian to military rule and from a presidential system to a party system. Chapter 4 on ‘Governance, rights and the demand for democracy: evidence from Bangladesh’ by Joe Devine, Ipshita Basu and Graham K. Brown explores the extent and nature of public support for democracy in Bangladesh by offering a detailed analysis of the Democracy Poll survey conducted by the Daily Star newspaper14 and Nielson in 2010. It is indeed curious that although the form of democracy that has evolved in Bangladesh has become a core part of the governance problem rather than a solution, political attitudes surveys find that there is



overwhelming public support for democracy. The aim of this chapter is to explore the level of support for democracy and whether it can accommodate different types of democratic regimes. In particular the chapter extends a comparison between pragmatic and ethical attitudes to democracy, i.e. whether people view democracy as a means to an end and therefore are willing to sacrifice some aspects of democracy if it fails to deliver institutional benefits or whether they see democracy as an end in itself. The chapter concludes that the pragmatic approach to governance is something that the Bangladeshi urban middle class are more likely to consider. This chapter therefore engages with some of the ideas outlined by Khondker. One of the main concerns of the remainder of the chapters in this book is the space for civil society activism in Bangladesh. The presence of an autonomous civil society is a crucial litmus test for an accountable democracy. But to what extent does a country like Bangladesh with a vibrant NGO community and a history of political mobilisation actually qualify as an open and accountable state? Geof Wood’s Chapter 5 ‘Deconstructing the Natural State? Is there room for De Tocqueville or only Gramsci in Bangladesh’ opens this discussion by asking whether civil society in Bangladesh can be free of the state and how much North et al’s doorstep condition of transition from limited-access to an open-access order with separable public institutions guided is possible? Through the case study of an NGO in Bangladesh, Wood demonstrates the concept of permeability for contexts such as Bangladesh, where formal domains of state, market, community and households are permeable and not separable with impersonal rules of the game. Hence for people in Bangladesh this entanglement of public and private, and state and non-state institutions and the ‘subversion of autonomy’ that it carries is not something that arouses shock but is accommodated as part of inhabiting the ‘prison’. The issue of permeability is picked up further in David Lewis’ Chapter 6 ‘When things go wrong in NGOs: what can be learned from cases of organisational breakdown and failure?’ Lewis takes up the issue of NGO failure and shows the significance of replacing a technical diagnostic approach of organisational failure with one that is more cognisant of the political environment in which NGOs function. Through the case study of Comilla Proshikha he illustrates the significance of the external environment in which NGOs operate where there are high levels of risk of predation by state, clientelist institutions, political parties and religious groups. NGOs in Bangladesh are, alongside government and business organisations, areas of commanding heights that are worthy of control by elites. Those who are leading NGOs have to have great political agility to negotiate the boundaries between the organization and formal politics. When leaders fail to manage these boundaries, organisations implode as there is a crisis of legitimacy. Finally, Naomi Hossain’s Chapter 7 on ‘The significance of unruly politics in Bangladesh’ asks why politics is so undisciplined – violent, lawless, irregular, defiant in Bangladesh? Through a political economy analysis involving an analysis of the garment industry in Dhaka, Hossain sheds light on the reasons for


I. Basu et al.

the unruly political culture and the core problems of elite politics. The chapter draws attention to activism of the poor and powerless groups and to the reality of political society, i.e. a world which is more volatile, vulnerable and sporadic than the controlled managerial world of donor-controlled civil society. The chapter is an apt conclusion to the narrative on governance in Bangladesh, where problematic structures have taken resilient statist roots, and therefore where room for protest and change from the powerless is negotiated in unruly, unpredictable and desperate ways.

Notes 1 Bangsho: lineage, which are understood hierarchically in the allocation of social status. 2 Zamindars: the old landlord class, used to describe anyone (men) behaving like traditional patrons. 3 Mastaan refers to urban underworld gang leaders behaving rather like the Mafia, but as a phenomenon, they are becoming the new patrons of the countryside. 4 The economy is, after all, growing through the expansion of the garments industry and other globalised economy activity, providing an expanded base for rent seeking. 5 A Bengali term best translated as ‘faction’. 6 That is: a collective exercise of myth making as the basis of continuing to be employed in the aid business and to enable donor governments to claim they are doing something worthwhile. 7 The attempt to find in Cuba a new market for jute exports quickly led to the exclusion of Bangladesh by the US from PL480 concessionary grain transfers. 8 When considered from a Weberian, or open-access society, standpoint. 9 Such analysis leads to the necessity for forms of collective action and resistance, perhaps organisationally protected by civil society organisations who can mobilise, protect and intermediate. 10 Crore = 10 million. 11 Political attacks and interference in the judiciary is a familiar cause of regime change in Pakistan, and certainly undermines regime legitimacy wherever it occurs. 12 And there are other business and media sectors which display the same vulnerability to ‘capture’ – especially the corporatisation and party ownership/control of the print and televisual media. 13 Donors are referred to here as the only possible uncontaminated, external, disinterested player. 14 The Daily Star is the country’s largest circulating daily English language newspaper and is known for its liberal and progressive position in society.


Party dysfunction and homeostasis in Bangladesh The old disorder restored (or not)1 Harry Blair

The politics of Bangladesh are often perceived as dysfunctional, disruptive, hovering just on (and sometimes over) the edge of serious instability, which is forestalled temporarily by caretaker governments, military rule or (most recently) a combination of the two. A different approach, however, would hold that over the past two decades this seeming dysfunctional instability has become in fact a kind of stable system with a structure and rules of its own, which its actors follow. This system exhibits a kind of homeostasis, in that it contains an inherent tendency to recover from perturbations and return to its former state, which it then maintains. The system also exhibits contradictions, however, in that it gives its principal actors incentives to break the rules. Thus far, rule-breaking efforts have been frustrated and the system has returned to norm, but this self-correcting tendency may not last indefinitely. Eventually all social systems give way to new dispensations. “Homeostasis” has long been a powerful concept in the biological sciences including medicine. It denotes the property of a system to react to external changes with self-correcting mechanisms. In the human organism, common examples are body temperature and blood glucose; external agents may cause either to go beyond the normal range, but protective mechanisms will generally return either measure to its norm. With most infections or diseases, the body likewise takes action to rid itself of the invading pathogens to restore a healthy stability. The concept can also be employed as an historical metaphor, for instance in considering the European upheavals of 1848, when democratic insurgencies threatened to overturn autocratic rule but national systems soon returned to their previous condition. The Paris Commune of 1870 provides another good example here. Eventually virtually all the Western and Central European countries moved to democratic systems, but authoritarian rule in its many variations was for centuries the homeostatic condition. And for most of capitalism’s history as well, economic panics and recessions have tended to dissipate as demand and supply gradually recovered. For Bangladesh over the last two decades, homeostasis has consisted of a volatile two-party system which has drawn a large portion of society and bureaucracy into its maelstrom, fueled massive corruption and disrupted the economy, but which has nonetheless persisted, resisting all attempts to change it.2 As with


H. Blair

any system that lasts over time, this one has had a certain structure with a set of “Rules of the Game” that facilitated its continuance. At times key actors have attempted to break the Rules and change the Game, with the latest attempt being the Caretaker Government (CTG) of 2007–2008, but so far the old system has re-established itself in essentially the same form each time. What is it that enables the polity to endure in this fashion, and what are the chances that it may evolve into something different? These two questions comprise the central focus of this chapter. This chapter offers a continuation and updating of an analysis I published in 2010,3 which endeavored to explain the dysfunctional political system of the 1991–2007 period and the caretaker regime of 2007–2008. The present offering begins with a brief reprise of that analysis, looks briefly at the December 2008 election, and then goes on to look at subsequent political events and their debilitating impact on the political system. The chapter concludes with some speculation on how developments in these areas collectively might affect the future of politics in Bangladesh.

Democracy’s trajectory in Bangladesh By coincidence, Freedom House started compiling its “Freedom in the World” index in 1972, just after Bangladesh had achieved its independence in December 1971, so it is possible to chart the country’s democratization trajectory virtually from its beginning. Figure 1.1 shows the Political Rights and Civil Liberties scores individually and then added together for the entire 1972–2010 period.4 New democratic era

Yearly ranking (1 = best; 7 = worst)





Democratic era


2 4 6 8 10 12 14 1972




Political rights


1992 1996 Year Civil liberties

Figure 1.1 Bangladesh Freedom House scores, 1972–2010.



Total score


Party dysfunction and homeostasis 19 The first two decades saw a series of wild gyrations with the successive democracy/dictatorship fluctuations during the regimes of the country’s first three leaders. Sheikh Mujibur Rahman began as the hugely popular father of his country (Bangabondhu or Friend of Bengal, the title bequeathed on him posthumously) but then turned authoritarian after a couple of years and was assassinated in August 1975. Ziaur Rahman opened his rule with a military coup in late 1975 but within a couple of years transformed himself into a genuinely popular elected leader, as is reflected in the Freedom House scores for the latter part of his era. He was assassinated in 1981, however, and within a year H.M. Ershad seized power in another military coup. He attempted to replicate Zia’s self-reinvention as a popular elected leader, but succeeded only partially, as can be seen in Figure 1.1. Over the course of the Ershad years, the two opposition parties recovered themselves to form a formidable opposition to his government. Both parties were legacies of previous leaders, the Awami League (AL) headed by Mujib’s daughter Sheikh Hasina Wajid, and the Bangladesh National Party (BNP), led by Zia’s widow Khalida Zia. By the end of the 1980s, the two parties had crafted a “politics of the street” consisting of mass rallies, marches and, most effectively, hartals (strikes) that could essentially bring economic and administrative activity in the major cities to a halt for days at a time – all tactics that set the tone for political life during the next two decades. Responding to the orchestrated protests with a mixture of guile and repression, governance under the Ershad regime began to deteriorate,5 reaching a low point at the end of 1990, when finally the army responded to a presidential command to impose martial law by instead directing Ershad to resign office immediately, which he did. A “free and fair” national election in early 1991 set into place what might be called a “two-party-plus” system, with the AL and the BNP as the major players, dividing between them the overwhelming majority of votes and seats, as shown in Table 1.1. Minor parties were drawn into electoral alliances but after the first two elections were not needed to form majorities in the Jatiyo Sangsad (parliament). Thus the main political drama has been played out between the two major parties, with the lesser groups being largely shoved to the sidelines over time. Table 1.1 Vote and seat share, 1991–2008 (percentages)

Awami League + Bangladesh Nationalist Party All others Totals

Votes Seats Votes Seats Votes Seats





61 76 39 24 100 100

71 87 29 13 100 100

83 85 17 15 100 100

81 86 19 14 100 100


H. Blair

After an initial burst of democratic enthusiasm at the beginning of the 1990s, the system settled into kind of routine momentum, which appears clearly in Figure 1.1. Civil Liberties continued from year to year at a steady score of 4 on the Freedom House scale, while Political Rights ranked somewhat better at a 2 during most of the 1990s and then gradually worsening to a 4 in the following decade. During most of this time, however, the country was consumed by what seemed to be a highly dysfunctional politics of instability verging on chaos, in which neither the party in power nor the opposition observed the basic conventions of a Westminster parliamentary system. The ruling party totally excluded the opposition from any meaningful role, routinely using state power – especially the police – to harass and intimidate it, while the opposition appeared to take every opportunity to disrupt normal life and cause sufficient turmoil that the army would find it necessary to intervene once again and eventually give it a chance to come to power through new elections. The polity suffered a near-death experience in 1996 when the ruling party engineered a blatantly rigged election, which the opposition boycotted, but a combination of public repugnance, international outrage and donor pressure forced an electoral rerun under a nonparty Caretaker Government (CTG), during which a non-partisan administration managed the country for a 90-day period running through the national election. The system had been subjected to the first serious attempt to break out of the Rules of the Game, but it righted itself and used the CTG system again successfully for the 2001 election. During what I have labeled the “Democratic Era” in Figure 1.1, Bangladesh did pass the Huntington “two turnover test” – that the ruling party be turned over through elections at least twice6 – but otherwise the dysfunctional polity went on, continually seeming to approach the edge of a total breakdown. Oppositionled hartals closed down the urban areas for days on end; gangs of mastaans (small-time thugs under the direction mafia-style patrons) disrupted normal life, often with police connivance; institutions like universities, professional associations and even NGOs were colonized by the parties and became divided into “panels” affiliated with them; the lower judiciary was used as an enforcer for the ruling party; and the list goes on.

Rules of the dysfunctional game But despite the appearance of breakdown and chaos, a quite well-defined set of Rules of the Game for public politics emerged during the 1990s, understood and observed by the parties, their leaders, the bureaucracy including the police, and most of the general populace. This set of Rules was never publicly articulated as such, but it was followed almost all the time. A seemingly dysfunctional system had become institutionalized. These were the essential elements of the Rules: •

Electoral democracy: “Free and fair” elections are held on time with a Caretaker Government in charge for a 90-day period.

Party dysfunction and homeostasis 21 • • • • • •

• • •

All power to election winners: The ruling party shuts out the opposition from any meaningful rule in parliament and takes over the state bureaucracy, operating a patronage-fueled, rent seeking regime. Local governance as patronage mechanism: Union parishads are subordinated to ruling party control, constitutional requirements for elected upazila and district councils are ignored. Opposition to the barricades: The past election is denounced as fraudulent, parliament is boycotted, hartals (general strikes aiming to shut down urban life) become the norm but violence is bounded, not insurrectionary. Gangsterism in public life: Party-based mastaan networks with police collusion intimidate opposition, operate extortion rackets, enjoy virtual impunity. Organized life commandeered: “Panels” allied to particular parties infest institutions everywhere, forcing students, professionals, even nongovernmental organizations to choose sides. Relative print media freedom: Marcusean “repressive tolerance” is permitted for elites,7 though some self-censorship is exercised, and journalists are subjected to significant harassment. Broadcast media are more closely monitored and pressured. Two-track judiciary: High Court and Supreme Court enjoy autonomy, while the lower court system is controlled by the Executive and used to support the ruling party. Electoral renewal: A new cycle begins with each election. A safety mechanism: Should a party in power try to break out of these Rules by fixing an election, some combination of national and international outrage, donor pressure and possible military intervention will ensure that the dysfunctional homeostasis continues in place.

The single most important word in the list just above is “but” within the fourth bullet point. The opposition party postures and fulminates, organizes huge processions (generally with truckloads of paid demonstrators brought into Dhaka for the purpose, buttressed by the opposition party’s student wings), compels businesses to shut down for the duration of its hartals, forces public transport off the roads (thus closing government operations), and in general threatens to create such chaos that organized life will break down altogether. But after 12 or 24 (or 36 or 48 or more) hours, the hartal winds down, demonstrators return to their everyday activities, work resumes and life comes back to normal. Some less well disciplined demonstrators may have provoked the police into a firing (or the police may have created an incident that fomented the provocation), and a small number of people have been injured or even killed. But there is never any actual threat of insurrection or even serious damage to public or private property. The real purpose of the hartal is not to overthrow the state or even to force the imposition of martial law that will lead to a new election, but rather to provide a strong reminder that the opposition is alive and well, and stands ready to take power after the next election, whenever it comes. In a true Westminster


H. Blair

system, a “Loyal Opposition” takes a full part in the cut and thrust of debate, subjects government proposals to strong scrutiny, acts as a watchdog to detect government malfeasance and abuse, offers alternatives to policies in place and in general endeavors to present itself in parliament through these methods as a viable alternative to the party in power. But by denying any meaningful role for the opposition, the ruling party in Bangladesh prevents the opposition from employing any of these avenues to making itself heard. Through its constant boycotts, of course, the opposition cuts itself off from the opportunity to become a Loyal Opposition, but it is I think fair to say that the basic hostility begins when the winning party assumes state power after an election and a new parliament begins business. This system survived several jolts, including assassinations of several leading AL politicians and numerous murders of lower-level party functionaries on both sides, a brief reign of terror in part of the country led by a fundamentalist Islamist group, a set of simultaneous bombings in 63 of the country’s 64 districts, and a short wave of suicide attacks on courthouses. But none of these shocks had any real lasting effect on public life. Things went on within the set of Rules I have outlined. In the run-up to elections scheduled for the beginning of 2007, however, the system began to unravel along several lines: •

• •

The ruling party (the BNP) was widely believed to be arranging the appointment of a party sympathizer to the post of Chief Advisor (i.e., administrator) to the Caretaker regime that would guide the country through the election period. The new Election Commissioner and his deputies were thought to be BNP partisans, a concern that intensified when it was discovered that millions of bogus names had been added to the electoral rolls. The opposition Awami League (AL) announced it would boycott not just parliamentary sessions, but the actual upcoming elections.

These efforts on the part of the BNP formed the second serious attempt to break out from the Rules of the Game, and they encountered much the same response from the opposition AL as had been the case in 1996.

The caretaker interregnum and the elections of 2008 Foreign donors reacted with public dismay, and the United Nations threatened to cease recruiting Bangladeshi troops as peacekeepers if the military supported a biased election.8 Almost immediately, the military took steps to replace the Chief Advisor, and the new CTG – under civilian management but clearly operating with military guidance – put elections on hold and shut down party activity for the indefinite future. The CTG extended its time in office to almost two years and undertook a number of serious reforms. The most important effort was to expel the two party leaders from the political scene and from the country as well: the “minus two” solution. As was

Party dysfunction and homeostasis 23 rumored at the time and corroborated just recently by newspaper reports quoting the Wikileaks cables from the American embassy in Dhaka,9 the military behind the CTG worked diligently during its first six months to promote the “minus two” plan, taking a page from the playbook Parvez Musharraf had used a few years earlier to press Pakistan’s two party leaders toward exile and trying to entice leaders from both major parties to form a “king’s party” to form a government of national unity. Initially the plan seemed to be succeeding, as the two begums appeared to accept exile and defectors from the BNP10 responded favorably to the unity government idea, bringing forth various proposals for internal party reform toward increased democratization. General Moeen Uddin Ahmed, the Army chief of staff and senior military leader behind the CTG talked of a “new brand of democracy” in April. Senior AL officials also proposed reforms, but apparently did not follow through by agreeing to support a unity government. Lacking sufficient traction, the unity strategy as well as movement toward party reform eroded and within a few months had stalled out altogether. A good part of the reason for the unity government idea’s collapse was that the “minus 2” strategy had unraveled by May 2007. Sheikh Hasina (who had already left the country) returned unexpectedly in May, which led Khaleda Zia to change her decision to depart. The CTG’s fallback strategy was to press criminal cases against both leaders. Hasina was arrested in July 2007 on corruption charges, and Khaleda a few months later on similar grounds. Trials were initiated but dragged on, and eventually were essentially abandoned as the CTG realized that the two leaders continued to hold sufficient grip on their respective parties that it would be impossible to hold the promised national election to restore democratic governance by the end of 2008 without their cooperation. Needless to say, the price of that cooperation included putting aside the trials. In sum, CTG efforts to reform the political system failed. The December 2008 national election did take place with the begum-led parties intact (minus the defectors). Thus the third effort to break the dysfunctional system – albeit an attempt to break it in a positive direction in contrast with the first two attempts of 1996 and 2006 – also fell short. But the CTG was not a complete failure. It did undertake a number of serious political reforms through promulgating ordinances (which as an interim governing authority it was allowed to do in lieu of legislation). Among the most important ones were: • • •

A computerized national voter ID system was put into effect, which greatly reduced the potential for fraud. The lower judiciary was finally separated from the executive branch after decades of promises from both parties and demands from the Supreme Court. An Upazila Parishad11 council was reinstituted with a directly elected chairperson after a lapse of two decades and more unfulfilled promises from both parties to restore it.


H. Blair

The Anti-Corruption Commission, which had languished during the democratic era, was strengthened and set to work looking into corrupt activities over previous years. A Right to Information ordinance in 2008 gave legal status to an idea that had been mooted several times over the years but never actually put into place.

National Elections were duly held in late December 2008 and were pronounced “free and fair” by various teams of outside observers.12 Although the AL was expected to win, following the pattern of ruling parties getting replaced in successive elections, everyone was completely surprised at the result,13 which gave the AL just over 48 percent of the valid vote – more than any party had won since Ershad’s faux victory in 1988 – and over three-quarters of the seats in the Jatiyo Sangsad. The BNP was reduced to less than 10 percent of the seats, an alltime low for the leading opposition party in the history of Bangladesh. In early 2009, then, the major questions facing the Bangladesh polity were: • •

Would the AL’s huge majority give it sufficient comfort that it could govern like a Westminster system party in power and grant the opposition a share in the structure of governance? After licking its electoral wounds, would the BNP conclude that the route back to power lay in becoming a loyal opposition that would participate in parliament, responsibly lay out an alternative agenda, work on rebuilding its constituency base, and bide its time until the next national election?

To put it a different way, could the BNP emulate the Canadian Tories, who in the 1993 election went from being the governing party to losing all save two seats, but in the ensuing years pulled themselves back together and returned to power in 2006, winning the next two national elections as well? Similar examples are offered by British Labour coming to power in 1997 after successive drubbings at Conservative hands, followed by a reverse sequence of three decisive Labour victories and then a Tory return to power in 2009. Or would things in Bangladesh return to the old dysfunctional pattern?

The old (dis)order restored The Ninth Parliament started off well enough in January 2009, as the muchreduced BNP delegation did attend the opening session. Among the first orders of business were to “ratify” (i.e., confirm) the ordinances that had been promulgated by the CTG. The new parliament did ratify a large number of these ordinances, including those dealing with re-establishing the upazila parishads, separating the judiciary from the executive, and strengthening the AntiCorruption Commission (ACC). In addition, it passed a new Right to Information law, thereby emulating India, where such a law had been passed in 2005 and had become a powerful instrument for improving state transparency.

Party dysfunction and homeostasis 25 But even as these favorable events were occurring, other much less auspicious developments were also taking place, beginning to belie the hopes that followed upon the December 2008 election. Indeed, some of the critical CTG ordinances were severely undercut in the very process of getting ratified. In short, the promised new dispensation began to unravel almost at once. In what follows, I will endeavor to trace the most important threads that started to come undone. Parliamentary boycotts Almost immediately after the Ninth Parliament’s opening, the BNP and its allies began to boycott its sessions. By the end of the first year it had skipped 65, or 76 percent, of its 86 meetings – the highest rate of boycotting since electoral democracy was restored in 1991. In the succeeding 14 months, the party improved just a bit, missing 91 of 121 sittings for a boycott level of 75 percent.14 Actually absenteeism might well have been higher, but for the rule requiring MPs to show up at least once every 90 session days to retain their seats and receive their salaries and benefits. Civic disruption Large-scale opposition-led hartals did not re-emerge as quickly as the boycotts, but by summer 2010 the BNP was launching and enforcing 36-hour city-wide strikes in Dhaka and other urban centers. The pace picked up, and by a year later a 48-hour hartal shut down the major cities altogether, in the process almost completely dominating the news. Hartal-related stories and photographs took up seven of the nine news items on the front page of the Daily Star’s edition on July 10, for example. The hartal was largely non-violent, certainly by the standards set in the previous decade, but given the rhetoric employed by the BNP, it seemed more than likely that agitation would escalate to less peaceful tactics before long. Along these same lines, the university student fronts for the parties resumed their campus disruption that had hobbled the academic enterprise so severely ever since the Ershad days but had been brought under control during the CTG. And while I’m not sure that professional and civil society organizations have as yet come under severe pressure to form “panels” aligned with one party or another, it is probably a safe bet to assume that this process is well under way. Upazila parishads (UZPs) These bodies, enshrined in the 1972 Constitution as the middle (thana, as they were called at the time) level of elected local governance between the zila (district) and union (multi-village) councils were never actually created until the Ershad government put its plan into place in the 1980s. After first launching the scheme in 1982, the Ershad regime held direct elections for council chairs in


H. Blair

1985 on a nonpartisan basis, with the idea that those elected could be co-opted into supporting a government political party, a move which evidently proved largely successful. Elections were held again in 1990, but the Ershad government itself collapsed by the end of that year, and along with it the UZPs. Despite pre-election promises, neither the BNP nor the AL reinstituted this middle tier of governance, principally because of opposition from the Members of Parliament (MPs). Their hostility is not hard to grasp when one realizes that the country has 300 MP constituencies and around 485 upazilas, meaning that the average MP riding contains one or two UZPs, which would likely serve as bases for serious rivals to challenge sitting MPs and in any event would surely compete with MPs in controlling and dispensing government development funds. Thus while the AL government did reintroduce the UZP system in 1998, it appointed the sitting MP within whose constituency any given UZP lay as its advisor (and in any event the AL never implemented the scheme, most likely in order to make sure the MPs would have no rivals at local level). The CTG revived the UZPs in an ordinance, dropping the MP advisory role, and was ready to conduct a UZP election in 2008 before the parliamentary poll, but it acceded to demands by both parties to run the latter contest first. Faithful to its promise, the new AL government did conduct a UZP election in January 2009 and the UZP chairmen were sworn into office. But the new parliament then had to ratify the CTG’s ordinance on UZPs, which it did in March 2009. However, in the process it stipulated that each UZP would have to follow advice from the MP within whose territory it lay, that it would have to consult the local MP about any development work it wished to propose, and that it must send minutes of all meetings to the MP concerned within 14 days. The bill passed unanimously, accompanied by a “thumping of desks” in the legislative chamber, indicating the MPs’ enthusiastic approval of the measure.15 In this way, control over the UZP system was handed over to the MPs. UZP associations have tried to lobby against this development, even bringing a constitutional lawsuit at one point, but have had no success. It is also worth noting that the AL government has followed its predecessors by giving the party’s secretary general (its chief operating officer) the Ministry of Local Government, Rural Development and Cooperatives as his portfolio, thus making official the partisan control of local governance. To sum up, the ruling party has at its command a powerful mechanism to steer development spending at local levels in ways that will lubricate its own political networks. Constituency development funds Adding to what the new UZP scheme offered to the MPs, the AL government also created a constituency development fund that would grant Tk.15 crores to each sitting MP over the five-year parliamentary cycle, to be spent at the latter’s discretion on infrastructural projects within his constituency. Such funds are not

Party dysfunction and homeostasis 27 uncommon within the Commonwealth, as a number of countries have initiated them recently, while India has had one in place since the early 1990s. India’s Member of Parliament Local Area Development Scheme (MPLADS) has over the past decade allocated Rs.2 crores annually (about US$400,000) per MP, where the average constituency size is roughly two million people. In Bangladesh, Tk.3 crores per year comes to about the same thing for constituencies averaging about 500,000 inhabitants. This amounts to somewhat more than 80 US cents per head for the Bangladeshi MP to spend, against about 20 US cents per head for his Indian counterpart – a dramatic difference.16 The avenues for mischief will be similar, no doubt – where will the MP want to allocate the projects and, more importantly in many and perhaps most cases, who will get the contracts to undertake the construction with its inevitable scope for corruption? When one adds together the Bangladeshi MP’s control over the upazila parishads and the Tk.15 crore development funds, his ability to direct patronage to his own political networks will be hugely enhanced. And given that over 76 percent of the current Parliament consists of Awami Leaguers (with another 9 percent from the AL’s allied Jatiyo Party), the ruling party will be well placed indeed to lubricate the party machinery it will want to mobilize for the next election. Caretaker government On May 10, 2011, the Supreme Court overturned the country’s Thirteenth Constitutional Amendment that had created the Caretaker Government system, which had successfully steered the country through its last three national elections. Under the CTG, a small group of non-partisan advisors headed by a Chief Advisor had managed the country for a 90-day period (actually a two-year period for the 2007–2008 CTG) after the parliament had been dissolved and had been able to superintend an election that virtually all observers and critics (save for the losing political party) found to be “free and fair.” In May 2011, however, the Supreme Court declared the Thirteenth Amendment to be “prospectively void and ultra vires of the Constitution,” i.e., unconstitutional and henceforth invalid. At the same time, however, it said that the next two national elections “may be held under the provisions of the above-mentioned Thirteenth Amendment.”17 A few days later, the Chief Justice explained this added provision as being necessary for the safety of the people and the safety of the state.18 The AL government wasted no time reacting to the Supreme Court’s decision. On June 30 it utilized its more-than-two-thirds majority in the Parliament to pass the Fifteenth Amendment to the Constitution, calling for future national elections to take place within 90 days prior to the dissolution of Parliament (i.e., before the end of its five-year term), during which the cabinet would remain in power without restrictions to manage the state, but the Parliament itself would not meet.19 In short, the sitting cabinet would itself superintend the election of the next parliament – the very situation that precipitated the bogus election of February 1996, which in turn led to the creation of the caretaker system in the


H. Blair

first place. Needless to say, the BNP reacted by saying that it would not participate in any such election, and it has evidently made opposition to the caretaker abolition the centerpiece of its campaign to bolster its cause.20 The situation presented a nice symmetry in that this time it was the AL setting itself up to manipulate the next election, whereas in 1996 and 2006 it was the BNP that took the initiative to do so. The AL government’s action constituted the fourth attempt to break out of the Rules of the Game that had been in place since 1991. Anti-Corruption Commission While still in power, the BNP government passed an act in 2004 upgrading the ineffective Bureau of Anti-Corruption to an Anti-Corruption Commission (ACC), but the new body did not accomplish anything of note before the CTG took over in late 2006. The CTG in turn strengthened the ACC, which instituted cases against leaders in both parties including the two former prime ministers after the CTG’s move to exile them failed, as recounted above. The new AL government did ratify the CTG ordinances regarding the ACC, but then in February 2011 amended the 2004 Act by stipulating that the ACC must get government permission to begin any inquiry against a public servant, to include judicial officials and MPs as well as bureaucrats. In a word, the new move effectively hobbled the ACC from taking any effective action against corruption. Meanwhile, the AL government has dropped the cases earlier launched during the CTG regime against its own officials while pursuing those initiated against BNP leaders. After first filing charges in October 2009, finally in September 2011 began the trial of Tarique Rahman, Khaleda Zia’s son and alleged head of an extensive extortion and money-laundering enterprise under her prime ministership.21 Judicial separation Though the 1972 Constitution in Article 22 proclaimed a separation of powers between the executive and the judiciary,22 vesting control of the lower judiciary (i.e., all courts below the High Court and Supreme Court) in the Supreme Court, the executive took over this control in 1972 and with some small modifications has retained it ever since. The Supreme Court had ordered a separation on Constitutional grounds, but throughout the 1991–2006 democratic era it continued to allow the executive extensions in complying.23 The AL and BNP continued to include a promise to separate in their election manifestoes, but once in power each new government somehow forgot about implementation. It was too convenient to keep the lower judiciary within the law ministry so as to be able to control it from the prime minister’s office. Finally in November 2007, the CTG did implement a separation with an ordinance, and a new lower judiciary was set up with its own judges. The new AL government ratified the ordinance, but added a significant change: district magistrates (deputy commissioners) would retain the power “to take cognizance

Party dysfunction and homeostasis 29 of offences,” meaning they would have the prerogative to initiate criminal cases, though they would not try them. This power to launch criminal cases at the behest of the executive has been the source of considerable manipulation and mischief previously, and of course the danger is that the practice will continue. In addition, as of summer 2011, the judiciary still lacked control over postings, promotions and disciplining judges. Such powers were lodged in the judiciary according to the Constitution’s Articles 115 and 116 but were abrogated in 1975 and have never been restored. They remain with the executive. To sum up, a partial separation of the judiciary from the executive has been achieved, but a high degree of control remains under the control of the prime minister and the executive. As of today, the judiciary in Bangladesh can scarcely be said to be independent. Right to information One last development appears to have not only promise but also some actual hope of realization. In the spring of 2009, the parliament also ratified the Caretaker Government’s Right to Information (RTI) ordinance in the form of the Right to Information Act 2009, which mandated citizen access not only to government organizations but also NGOs operating with state-provided or foreign funds. The Act required government bodies to appoint Designated Offices to respond to RTI requests and also set up an Information Commission with some punitive powers to deal with complaints. Two years later, as of April 2011, it was reported that more than 1,000 government bodies had appointed Designated Officers, as well as 201 NGOs. So far, the media appear to have made relatively little use of the RTI (perhaps because of the 20-day period allowed for responses to RTI queries, a constraint making use of the act difficult for journalists with deadlines), but several NGOs have provided training programs to enable citizens to utilize the RTI provisions,24 while some others have taken advantage of it to demand information about state activities.25 The Information Commission has also been active in answering complaints.

Breaking away from homeostasis To sum up, since 2009 the Awami League has moved on a number of fronts to restore the dysfunctional polity of the 1991–2006 period, thus dismantling the CTG’s accomplishments during the 2007–2008 interregnum. •

In local governance, the AL ministry has turned over the renewed upazila parishad system to the mercies of the MPs by giving them a veto power over all parishad activities, and it added even more to their ability to dominate the local scene by awarding each MP a US$2 million discretionary fund for local expenditure. To deal with corruption, the ministry defanged the Anti-Corruption Commission by requiring it to take government permission before investigating any government official, including MPs.


H. Blair

The AL ministry has severely compromised the long-awaited separation of the judiciary by retaining control over initiating criminal cases and over posting, promoting, and disciplining judicial officers. In a move to undermine the dysfunctional polity altogether, for future elections the AL government has abolished the highly successful Caretaker Government system that had managed the last three elections and has put itself in sole charge of the upcoming 2013 poll.

Quite understandably, the opposition has responded by declaring the 2008 election a fraud (despite all evidence to the contrary), boycotting the parliament and taking to the streets with hartals closing down the major cities for days at a time. It might seem that the negative homeostasis of 1991–2006 has returned, but a closer look invites the very strong suspicion that we are instead looking at a fourth attempt to break out of the homeostatic pattern. The first two attempted breakouts aimed to move the system toward an even more negative trajectory. But the BNP effort to manipulate the first 1996 election and its later exertions to tilt the 2007 election in its favor were turned back, and the political system was righted. The third attempted breakout came with the CTG’s efforts to establish a more accountable and less corrupt polity in 2007–2008, but these efforts were also turned back for the most part by the incoming AL ministry in 2009. The question facing Bangladesh today is whether the AL’s maneuvers are serving only to restore the old dysfunctional but sustainable (at least in the short and perhaps middle term) homeostasis, or do they actually represent a new breakout effort, once again in a negative direction designed to ensure a ruling party’s return to power in a rigged election.

Misperceived homeostasis? It is certainly possible to perceive homeostasis reasserting itself when in fact an abrupt and sweeping change is about to occur. An excellent example for Bangladesh can be seen in the political crises of 1969 and 1971.26 In both cases a popular movement against a dictator paralyzed the state and the economy. As government authority came to a halt, a de facto parallel polity organized by the Awami League arose to keep vital services going throughout East Bengal. The military mobilized additional troops from the West and declared martial law. Political leaders called for more autonomy, while student networks and huge demonstrations demanded bolder steps. In the 1969 case, there were a number of skirmishes between army/police and unruly crowds, a few of which culminated in firings. At the macro level, the incumbent dictator was shown the door, and soon martial law excesses were terminated, life returned to normal and political prospects improved in the form of a national election that promised to give real power to elected representatives. Many activists wanted to press for outright independence, but AL leader Sheikh Mujibur Rahman’s goal continued to be a federation of the two wings, with the AL’s six points as the foundation. Two years later, it was not surprising that the

Party dysfunction and homeostasis 31 same regional leader who had rejected the insurrectionary impulse in the first crisis should prove reluctant to declare independence in the second one. In his speech of March 7, 1971, subsequently hailed as the definitive call to independence, Sheikh Mujibur Rahman did in fact say that “This struggle is for independence,” but then he also met with General Yahya Khan on 16 March and for several days afterward (Zulfikar Ali Bhutto came in from West Pakistan to join the talks on March 21) to discuss possible federal solutions that would keep Pakistan together, even as student groups and party leaders tried to organize for violent struggle. Today we still don’t have all the details of those macro-level discussions, but at that late date it appeared there was still some hope for a peaceful resolution. But by March 23, when the Pakistan Day holiday turned into People’s Independence Day, the die was effectively cast, and two days later the Pakistan Army unleashed its brutal repression on the populace. The outcome of the second crisis was the opposite of homeostasis, but at the time and on the ground, some very important actors thought that things would return to some kind of normality, as they had before. As we know, however, homeostasis was not in the cards in March 1971, and instead of a return to the status quo ante the armed struggle for independence began.

Prognosis Judging on the basis of the AL’s terms in office since 2009, two plausible scenarios would seem to emerge, which could be called less pessimistic and more pessimistic, with the Caretaker issue as the crucial factor. The less pessimistic one would see reactions to the AL’s attempt to break the Rules of the Game as bringing sufficient pressure to effect a homeostatic return to them. The BNP has made the caretaker issue the centerpiece of its agitational efforts, and international concern can be expected to grow as the anticipated 2014 election approaches. And of course giving MPs control over the upazila parishads, defanging the Anti-Corruption Commission, and undermining judicial separation all conform to the homeostatic model, simply restoring the structure that had been in place during the 1991–2006 era. The Tk.15 crore constituency development fund for each MP is something new, but certainly fits in with the other aspects of that period. If the AL can be persuaded to go back to the neutral caretaker system, as the Supreme Court has urged,27 the basic Rules can be put back in place, and a general dysfunctional politics can be resumed, at least until the next attempt to break out of the Rules occurs from either the AL or the BNP. The more pessimistic scenario contemplates the AL holding firm on its resolve to dismantle the caretaker system, ignoring all pleas domestic and foreign to restore it. The BNP would surely boycott the 2014 election, and the situation would be back to where it was in January 2007. Would the military intervene again? If the country appeared to be headed toward a genuine breakdown, it probably would. But would it do so with the intent of returning the country to civilian management after seeing virtually all its attempts to bring about political reform collapse in the previous round? Or would it conclude that


H. Blair

the two-party system’s dysfunctionality was just too endemic, that political morbidity was the inherent accompaniment to political democracy in Bangladesh, and that therefore it should maintain direct control of the country indefinitely?

A third scenario In contrast with the bleak and bleaker pictures offered just above, a more optimistic scenario can also be sketched out. There are several reasons to think that the political system can change course to a more positive trajectory which, while it won’t resolve all the dysfunctionalities at macro-level, could at least set some aspects of the polity in a better direction. The general economy For some years now, in spite of the dysfunctional political system (or perhaps in part because of it, in that the very intensity of the interparty political struggle so preoccupied the leadership of both parties that it was possible for economic policy decision makers to operate without much interference from above), the country’s economy has done quite well, especially more recently. According to the World Bank, gross domestic product per capita grew at an average of 2.7 percent during 1991–2001 (the first two democratic governments), then 4.1 percent during the last BNP regime, 5.2 percent during the CTG era and, finally, at 4.7 percent during the current AL ministry. To be sure, economic growth per se is no guarantee against social and political turmoil, but a healthy growth rate does allow both leaders and citizens to breathe a bit easier than when times are tight and the economy is declining. Political stability in the neighborhood Certainly political instability has been a serious problem for Bangladesh, a fact reflected in the World Bank’s ranking the country very low in its annual World Governance Indicators (WGI) report. The Bank uses its “Political Stability and Absence of Violence” indicator to measure “perceptions of the likelihood that a government will be destabilized or over-thrown by unconstitutional or violent means, including domestic violence and terrorism.” Since 2004, the WGI has put Bangladesh in the tenth percentile or lower each year, ranging down to the fifth percentile in 2005. Such a ranking is discouraging, sure enough, but it is not uncharacteristic of the South Asian region overall, as is evident from Figure 1.2, which shows the WGI rankings for all the principal countries of the area over the entire WGI coverage.28 The comparisons will be cold comfort for Bangladeshis at first glance, but when one reflects that democratic India has done scarcely better on the WGI index over the time period, and that other countries (Sri Lanka, Nepal) regarded as reasonable prospects for democratic governance have done much worse in many years, the picture does not look so bad. If India got through its

Party dysfunction and homeostasis 33 40 35

Percentile rank

30 25 20 15 10 5 0




2003 India

2005 Year Nepal

2007 Pakistan

2009 Sri Lanka

Figure 1.2 Political stability and social peace in South Asia, 1996–2010. Source: World Bank WGI 2011.

bad chapters and Sri Lanka shows promise of doing so today, then surely Bangladesh can as well. Civil society and the Right to Information Law It is well known that civil society organizations in general, and externally supported NGOs in particular, have been chary about advocacy at the national level ever since the late 1970s, when it became clear that policy advocacy was not likely to succeed and was very likely to lead to state repression. There were occasions when the NGO community did involve itself in macro policy issues, as when the government NGO Affairs Bureau showed signs of trying to control NGOs in the late 1980s, or toward the end of the Ershad regime in 1990 when virtually the entire society was turning against the isolated dictator, but these instances were rare indeed. Then in the early 2000s when several large NGOs did slip into political involvement, the NGO apex organization split, and the community felt itself getting drawn into the polarization then intensifying between the two major parties, it became even more clear that national advocacy on policy issues would be a mistake. But advocacy that can avoid challenging the political parties at national level has been quite a different story for a number of NGOs and CSOs. For some


H. Blair

years now, NGOs like Nijera Kori have fostered local-level advocacy, and others like the Bangladesh Environmental Lawyers Association and Bangladesh Legal Aid Society Trust have taken up public interest lawsuits. In the short time since the passage of the Right to Information Law in 2009 (it took effect in July of that year), quite a few groups have begun to empower individuals and local CSOs inter alia to gain access to public services, enforce environmental regulations, and make official land records available.29 As expected, the RTI Law has faced problems with uncooperative (as well as uninformed) officials, lack of follow-up by petitioners and so on, but it does appear to have made a good start.30 Critical to its success will be the media, for unless awareness of RTI’s potential becomes widespread and large numbers of citizens learn how to use the new law, it will make little difference in how the state conducts its business. Fortunately, media conditions have improved significantly in Bangladesh over the past several years, as illustrated in Figure 1.3. The Freedom House press freedom survey rated the country in the mid- to upper-60s (where 0 is the best score and 100 the worst) up through 2007, but then shows an improvement for the last three years, finishing at 54 in 2010, which is shown in the bottom line of Figure 1.3. The significance of this comes out when the media score’s components are separated, as in the top three lines of the figure. Here the Economic Environment 10


Media rating









Legal environment


2006 Year

Political environment

Figure 1.3 Bangladesh media ratings, 2002–2010. Source: Freedom House.



Economic environment


2010 Total

Party dysfunction and homeostasis 35 (ownership concentration, corruption impact on content) remains essentially the same over the period, while Legal Environment (state regulation) improves only slightly. The major change occurred in the Political Environment, which gauges access to information, editorial independence, censorship and self-censorship and harassment/intimidation of journalists. The rating here improved from 30 points in 2007 to 21 in 2010 and accounts for most of the shift in the Total measure at the bottom of the figure. In a parallel survey, Rapporteurs Sans Frontières, an international watchdog organization focusing more on individual journalists than media organizations, ranked Bangladesh at 151st out of 167 countries in the mid-2000s, but by 2009 upgraded it to 121st out of 175, a distinct improvement.31 Taken together, these two surveys offer strong evidence that the media have become significantly more able to find out about state activities at all levels and to publicize them. This progress scarcely indicates that Bangladesh has become a paragon of media freedom, for journalists continue to be harassed and arrested, but the general environment for journalism does appear considerably improved over a few years ago. What all this means is that it is more possible today for CSOs/NGOs or individuals to gain access to state information and for the media to publicize what they have uncovered – in short to exercise a kind of accountability over the state that was simply not possible in an earlier era.

Conclusion In the short run, it appears clear that Bangladesh’s systemically dysfunctional polity of the 1991–2006 era has come a long way toward restoration. The two major parties have resumed their duet of excluding the opposition from any meaningful role on the one side and combining parliamentary boycotts with civic disruption on the other side. Anti-corruption efforts and judicial separation remain largely dead letters. The promise of a revitalized upazila parishad structure has been effectively strangled in favor of the MPs, whose domination of local governance has been aggrandized with additional constituency developments funds placed at their discretion. And finally, the one saving quality of democracy that lies in the ability of the citizenry to recover from their mistakes by changing their rulers appears to be severely compromised by the abolition of caretaker governments superintending future elections. The overall outlook appears to offer a choice between a bad homeostasis and a decline from that condition. But Bangladesh does remain a country observing the Rule of Law at the upper end of its judicial system (High Court and Supreme Court), allowing a great deal of freedom to civil society (so long as it steers clear of party politics at the macro level), becoming significantly more transparent with the implementation of the Right to Information Law, and maintaining a reasonably free media. Collectively, these factors present a fair chance for the system to avoid a collapse of the present dysfunctional homeostasis and hopefully even to ameliorate it in the years to come.


H. Blair

Postscript More than four years after this chapter was written, the dysfunctional homeostasis model continues to hold though barely. Making a case for its validity is becoming ever more difficult, as one of the principal actors has continued to successfully flout its rules, perhaps just for the short term but quite possibly for a much longer duration. As shown in the paper, the Awami League began dismantling the model shortly after its sweeping electoral victory in December 2008, most notably by abolishing the Caretaker Government. In response, the BNP adopted CTG restoration as the main theme of its protest marches and hartals, but the AL held fast, offering instead an “all-party cabinet” to manage the 2014 elections. The BNP rejected the offer, continuing its protests and then, playing the AL card from previous elections, called for an election boycott, hoping that as before international outrage, donor pressure and the threat of military intervention would force the AL’s hand. In the event, however, it became clear that the BNP had misplayed its own hand, for the AL went ahead with elections, ignoring objections from abroad and whatever concerns donors expressed in Dhaka, while the military, likely discouraged by its previous efforts to right the system, stayed clear of involvement this time around. Thus the AL won 234 of the Jatiyo Sangsad’s 300 directly elected seats and the BNP was entirely shut out. Since the 2014 election, the AL has moved to consolidate its position on many fronts.32 A national broadcast policy announced in 2014 called for banning any content demeaning the military or police or “contrary to the public interest,” newspaper editors are facing criminal charges, and hundreds of citizens have been charged for online postings under in an Information and Communication Technology Act passed in 2013. Not surprisingly, in its annual Freedom of the Press report for 2016, Freedom House downgraded the country from “partly free” to “not free,” and by 2016 Rapporteurs sans Frontières had lowered its press freedom ranking from 121st out of 175 in 2009 to 144th out of 180 in 2016. State use of the constabulary to harass and repress opposition advocates – always a problem under dysfunctional homeostasis in Bangladesh – has increased, especially with the Rapid Action Battalion (RAB) created in 2004 as an anti-terrorist unit with special powers and strengthened under the AL, leading to more arrests, charges, and reports of torture, extrajudicial killings and disappearances. The Anti-Corruption Commission remains toothless, unable to bring cases against officials without government permission. Meanwhile, the BNP has continued to demand a restitution of the CTG and fresh polls, following its strategy of protest and hartal just as in the past, but with no effect. By way of backdrop, economic growth continues to provide a buffer against popular discontent, with an average 4.8 percent GDP per capita growth rate during 2011–2015 according to the World Bank (and predicted by The Economist Intelligence Unit in June 2016 to continue growing at a similar rate over

Party dysfunction and homeostasis 37 the next several years). Accordingly, economic distress appears to be unlikely to fuel opposition to the incumbent government. As of mid-2016, if the present trajectory continues, it may well show that the dysfunctional homeostasis model has run its course. The CTG safety mechanism is unlikely to be restored for the election due in 2019, and the AL government will have sufficiently solidified its dominance of the constabulary and bureaucracy, as well as the judiciary and media in all probability, so as to ensure a favorable election result, whatever the BNP is able to muster in the way of protest and complaint. A new model would then become necessary to explain the more authoritarian polity that Bangladesh will have moved into.

Notes 1 This chapter reflects my experiences in and ruminations on Bangladesh over the four decades and more than 20 visits to the country since my first trip there in 1973 to the Bangladesh Academy for Rural Development at Comilla. 2 My use of homeostasis as a metaphor is imperfect, in that in medicine it denotes a self-healing quality, whereas I am employing it to analyze a political system that has managed to restore itself to a condition of poor health and in addition has generally needed some exogenous factor to promote the process. Even so, the concept of selfrestoration (if not exactly self-healing) provides a great deal of mileage in explaining the recent politics of Bangladesh. 3 See Blair 2010. Parts of the present paper appeared in a presentation (Blair 2011) at a conference in Seattle honoring the work of Paul Brass, on September 4, 2011, and an earlier version formed part of Blair et al. (2004). 4 The Political Rights measure focuses mainly on political participation, including parties, elections, civil society, and accountability of state officials to the citizenry, while Civil Liberties centers on freedom of speech, religion, assembly, rule of law and human rights. Both indices are scored 1 (best) to 7 (worst), with the combined score ranging between 2 and 14. For a full explanation of the Freedom House methodology, see www. freedomhouse.org/template.cfm?page=351&ana_page=374&year=2011. 5 Freedom House did not publish its annual survey for 1989, which accounts for the gap in the graphs in Figure 6. Had the scores been tallied, they would have shown a decline from 1988. 6 See Huntington 1991: 266ff. 7 See Marcuse 1969. 8 Bangladesh had been a major supplier of such troops, who for several years amounted to around 12 percent of active duty army strength. The UN payroll for the peacekeepers formed a major component of army income. 9 It should be noted that, to the extent that Wikileaks accurately reproduced these cables, their contents reflect only what various people in the American Embassy wrote, which is not necessarily what actually happened in any given case. September 2011 saw a spate of newspaper articles based on the leaked cables (all of which had been made available through Wikileaks on August 30, 2011). This and the following paragraph are based largely on several of these articles (especially Liton and Ashraf 2011a; Star Reports 2011a, 2011b, 2011c, 2011d, all in the Daily Star; and Staff Correspondent 2011a, 2011b, 2011c, all in the New Age). 10 Most prominent among them was the BNP’s Secretary General, Abdul Mannan Bhuiyan. 11 Upazilas were essentially the erstwhile thanas renamed, analogous to taluks and tehsils in India. Thus the upazila parishad is similar to the panchayat samiti in India.

38 12 13 14

15 16

17 18 19 20 21 22

23 24 25 26

27 28 29 30 31 32

H. Blair Despite constitutional requirements, Bangladesh has never set up an elected zila (district) council system. See for example, Eicher et al. 2010; EU 2009; NDI 2009. “Everyone” included me as a member of the National Democratic Institute’s observer team (cf. NDI 2009). Data from Liton (2010, 2011). In March 2011, it was proposed in the parliament to change the rule for ouster from 90 sitting days to 90 calendar days, which would cut down somewhat on the boycotting, as the body generally meets for no more than 90 sittings in a calendar year (Liton 2011). But the proposal appears to have languished, as it did not make the list of amendments to the Constitution passed in June 2011 (Staff Correspondent 2011d). Suman 2009; also Staff Correspondent 2009. For a study of the Indian program, see Blair 2011. It should be noted that in 2011 the Indian program expanded to Tk.5 crores per MP, which will mean roughly US$0.50 per inhabitant, now about five-eighths of what will be available to each Bangladeshi MP. I am indebted to Prof. Nizam Ahmed of the University of Chittagong for details on the Tk.15 crore fund. See Sarkar 2011a, 2011b. See Sarkar and Liton 2011. The May 10 verdict was little more than a brief statement of the court’s decision. The full version of the Court’s decision had still not appeared at the end of December 2011, despite earlier promises (Sarkar 2011b). See Liton and Hasan 2011. Such a plan would put Bangladesh at odds with the general Commonwealth practice, under which elections are held after the dissolution of the sitting parliament (Latif Mondal 2011). In October 2011, the BNP used the caretaker issue to launch a series of large demonstrations around the country. See Suman 2011 and Juberee 2011 for examples from Sylhet and Bogra. See Liton and Ashraf 2011b; also Court Correspondent 2011. The demand for such a separation had been around for a long time. The Awami League had promised a separation ever since its 1949 draft constitution, and in 1957 the East Pakistan provincial legislature passed an act to that effect, but it was never implemented. In October 2005, for example, the Supreme Court granted the twenty-first such extension. An example here is Nijera Kori 2011. For a collection of short case studies, see World Bank Institute 2011. See Halim 2011; Khan 2011; Sobhan 2011. This and the next two paragraphs are based on one early and premature assessment of the two crises (Blair 1971) and one much more recent review (Ludden 2011). Both arrive at a similar conclusion. There are many other analyses of the 1971 events, of course. Two that affirm the account given here are Blood (2002) and most recently Bose (2011). The release of the Court’s full decision may well accelerate pressure to restore the Caretaker system for the 2014 election. The WGI began its coverage for the year 1996, then published every two years until 2002, after which it has appeared for every year. World Bank Institute (2011) provides a good sampling of case studies involving use of the RTI. India faced many difficulties in implementing its RTI Act passed in 2005 (see, for example, Kulkarni 2008), but over time it became increasingly effective (see Roberts 2010). Accessed on October 24, 2011 respectively at www.freedomhouse.org/template. cfm?page=16 and http://en.rsf.org/spip.php?page=classement&id_rubrique=1034. For two excellent accounts events and trends after the 2014 election, see ICG 2015 and 2016.


Where are the drivers of governance reform?1 Pierre Landell-Mills

The challenge Since its independence in 1971, Bangladesh has repeatedly confounded the pessimists. Kissinger’s much repeated caricature of Bangladesh as a basket case was far off the mark. Not only has Bangladesh been able to feed its expanding population, but it has also dramatically improved its social indicators and become a significant exporter of manufactures. But it has not nearly reached its full potential. Its long-term economic growth has been limited to 4–5 per cent, far short of the 8 per cent or more that might easily have been achieved (Kapoor et al. 2000; Lateef 2006). The reason for falling so far short has been poor governance with its origins in the elite’s inability to reach a robust political settlement establishing widely accepted rules on how the country’s resources should be managed and how the benefits should be allocated. The purpose of this chapter is to explore how the political economy constraints might be overcome and what contribution the donors might make, if any, to promoting a solution. The chapter has four main sections. The first section briefly sets out a conceptual framework that aims to help explain how states put in place the institutions that create an enabling environment to allow accelerated social and economic development. The second section examines how Bangladesh fits into this framework. The third section discusses who might be the drivers of governance reform, what constraints they face and how these might be overcome. The final section examines the contribution donors might make to support them.

A paradigm for state building Political economy studies aimed at explaining why certain countries have progressed faster than others have multiplied in recent years.2 An emerging paradigm that provides the conceptual framework for this chapter has the following main elements. Using the terminology of Douglas North (North et al. 2009), countries at different levels of development are seen as lying along a continuum from having ‘limited access order’ that is fragile (least developed countries where competing warlords prey on each other)3 to basic (most developing countries) to mature (with well established institutions that govern relationships


P. Landell-Mills

among the elite) to countries with an ‘open access order’ which have the most advanced economies and institutions. There is no inevitable or natural progression along the continuum, but generally the more developed the country, the closer it is to having ‘an open access order’. Within these broad categories there is considerable political and institutional variation. Limited access orders each represent a solution to critical need for some form of political order to limit indiscriminate violence and chaos. If external factors (e.g. external donors or peacekeepers) intervene, it may unsettle the fragile balance. The degree of ‘access’ is a measure of the state’s control over non-state organizations and the extent to which citizens are equal before the law. It also reflects the extent to which the rulers are accountable to all their citizens. Limited access orders are characterised by entrenched elites that have as their main agenda the use of power principally for the acquisition of wealth which in turn is used for patronage and to pay for the means of coercion that enables them to remain in power. As a consequence low priority is given to pro-poor development. The regimes are largely authoritarian (neo-patrimonial) and generally seek to manipulate elections, if held at all, in favour of those in power. They negotiate from a position of power, given their control of the state’s security forces (described by North as ‘the means of violence’)4 with other groups within the elite who may also have control over other means of violence (e.g. gangs of thugs, strikes and the like) to maintain themselves in power. The spontaneous mobilisation of citizens willing to face down the security forces, made more possible by social networks, is always a possibility if those in power grossly abuse their position. However, while such movements may bring down the regime in power, in the absence of charismatic and visionary new leaders a positive outcome is by no means assured. The extent to which limited access orders may be stable depends on how successful those in power are by incorporating factions who are capable of challenging them partly by negotiation and partly by superior force. This ultimately depends on reaching an agreement to share access to economic benefits euphemistically and somewhat misleadingly referred to as ‘rents’. Thus different factions with access to the means of violence are constantly contesting those in power and the degree of political stability depends on the extent to which there is a robust consensus on the rules of contestation. Put another way, the key factor determining the degree of maturity of the limited access order is the strength and permanence of institutions and respect for the law governing their functioning (e.g. property rights, trading rights and the method of selection or election of the rulers). Let me now apply this framework to the challenge of social and economic development. In developing countries governance reform (in the sense of some movement towards an ‘open access order’) has generally made slow progress, though with some significant exceptions.5 Periodically there are uprisings or military coups that seem initially to be harbingers of reform, but often result over time in little more than one self-serving faction replacing another. While the number of countries with elected governments has increased in recent years,

Where are the drivers of governance reform? 41 elections in themselves have often not brought noticeably more accountable government. We may thus well question whether there is anything inevitable about the emergence of open access orders (i.e., accountable governments); some countries may remain for decades, or even centuries, stuck with authoritarian regimes run by predatory elites unable or unwilling to allow the political order to evolve in step with economic development towards more accountable governments. It cannot be disputed that there is a crude long-term correlation between governance reform and economic progress, driven by wealth creating elements within the elite who see advantage in putting in place more accountable institutions that respect the rule of law and produce predictable outcomes that encourage and facilitate productive investment. Better governance is seen as emerging through a protracted process of negotiation among competing factions within the elite and the need for rulers to respond to the growing expectations of a better educated and informed citizenry. This postulated process of institutional development in no way assumes moving towards a Western democratic model, but does require states to establish not only a favourable environment for investment but also mechanisms for governments to become more accountable to their citizens, giving them genuine voice in how resources are allocated and services delivered. If they do not, then there is a high probability that citizens will eventually take to the streets. Outside elections and setting aside mass movements, which can achieve swift and radical change in exceptional circumstances, the only meaningful option available to citizens who are not part of the elite power structure is to pursue constructive collective action through organised structures. Allowing space for such actions is a measure of the maturity of the political system. Ever present are the security forces which rulers need to keep on their side. Their role depends on the circumstances of each country. Sometimes they may have a tradition of intervening, while elsewhere for historical reasons they may be reluctant to take sides. If they are united and enter the fray, their intervention is likely be decisive in the short term, but most often will resolve little in the long term. In the end there must be either a negotiated political settlement among the groups contesting power or descent into the anarchy of a failed state. The outcome will depend in part on the quality of political leadership. This chapter will now explore how this paradigm applies to Bangladesh and, in the light of this analysis, how better governance might emerge. Specifically how might internal pressure for reform arise, and can it be promoted? More particularly, the chapter discusses the potential roles of the various Bangladeshi non-state actors in demanding better governance and, related to that, what might be the role and impact of donors? Behind this discussion lie some critical questions: •

Why do businessmen, especially those engaged in highly competitive export markets who are seriously handicapped by poorly managed infrastructure, insecurity and the unpredictable and uneven application of the law, not press harder for a more favourable enabling environment for business activities?


P. Landell-Mills

Why do citizens, and more particularly the rank and file of the main political parties, allow themselves to be effectively marginalised? Why do the young who have few employment opportunities not challenge the corrupt behaviour of the older generation that dominates politics and seemingly deprives them of a better future? Why does the growing middle class appear to have such a minimal impact in demanding better governance? If the political process is essentially a matter of bargaining among the elite, why are the outcomes so sub-optimal?

• • •

To a significant extent these questions overlap. The chapter looks for answers by exploring the constraints faced by businessmen and organised civil society in Bangladesh. But first, we need to set out the context before seeking answers to these questions.

How does Bangladesh fit the paradigm? We need first to briefly recall the key relevant political events since Bangladesh achieved independence in 1971. During the last 40 years, Bangladesh witnessed erratic governance marked by some successes and a multitude of failed reforms. The first Awami League government faced extreme difficulties in trying to recover from the disruption and devastation of the war and failed to build a consensus among the contesting factions. An attempt to resolve these difficulties by imposing a one-party state ended with the assassination of Prime Minister Mujibur Rahman and members of his family in 1975. Over the next 15 years Bangladesh endured a series of military regimes which ended with the collapse of the Ershad government in 1990 and free elections overseen by a caretaker government headed by the recently retired Chief Justice. For a period it seemed that democratic institutions might become rooted, but this hope quickly faded as patronage politics took a firm grip, with those in power unwilling to allow the main opposition parties any governance role. The two main parties alternated in government over the next 13 years with no obvious sustained improving trend in governance. Then, a sharp deterioration in governance in the period 2004–06 resulted in the growing perceived illegitimacy of the regime, culminating in cancelled elections and a military takeover in January 2007. The army installed and oversaw a Caretaker Government (CTG) managed largely by technocrats. Some important reforms were initiated by the CTG (e.g. the separation of the Executive and the Judiciary, the enactment of a Right to Information Act, the reinvigoration of an independent Anti-Corruption Commission), but soon the CTG came to be viewed as ineptly managed and eventually stalled, leading to an irresistible demand for a return to elected government. The CTG lost popularity also partly as a result of their failure to stem rising food prices. At the end of 2008 to its credit the CTG successfully held new elections with an updated electoral roll acceptable to all the main political parties.6 The electoral process had

Where are the drivers of governance reform? 43 strong donor support and was widely regarded as fair. The Awami League won an overwhelming victory. The refusal of the Bangladesh Nationalist Party, the main opposition, to recognise the elections as fair is symptomatic of deep distrust within Bangladesh politics. When in opposition, political parties have periodically declined to sit in the National Assembly in an attempt to deny legitimacy to the incumbent party and, as a result, have not performed their accountability functions. Instead, they have questioned the legitimacy of the government and have engaged in repeated street protests (hartals) that have dislocated economic activities. Generally, politics has been characterised by violence and, occasionally, by assassinations, in place of orderly democratic processes. Street power has been integral to the struggle for political dominance. Bangladesh’s recent political history is greatly illuminated by the North– Wallis–Weingast paradigm as explained in some detail in a recent paper by Mushtaq Khan (Khan 2010a). Khan relates the political struggles for power to the elite capture of whatever rents were available as the economy shifted from being predominantly agrarian prior to Partition to primitive industrialisation in the period 1950 to 1980 and thereafter in the context of a rapid expansion of labour intensive exports. What remains to be fully explained is the logic for the formation of the different competing political groupings which seem to lack any deep-seated political ideology. Political parties seem only loosely related to distinct identifiable political interests, but each has an entrenched core of loyal members held together through patronage and consequently have impressive staying power. In this sense, the parties resemble political ‘tribes’ or mafia gangs where loyalty is a deliberate strategy for survival or at least for self-advancement. In the absence of distinct well articulated political platforms, the underlying struggle to capture ‘rents’ has been the driving force. In the immediate post-independence period, the political leadership was confronted by the virtual impossibility of satisfying the many claimants to a share of the limited available ‘rents’ – mainly those derived from the expropriation of West Pakistani assets and those of Hindus and Biharis. Asset stripping and the exploitation of nationalised businesses resulted in economic decline and hence fewer ‘rents’ to be captured. The attempt to control the growing violence by introducing an authoritarian one–party state resulted in fierce opposition from those excluded from the spoils, leading to the overthrow of the government and 15 years of unstable military dictatorship. Khan (2010a) explains how the MultiFibre Agreement which allowed Bangladesh duty-free access to the US and European markets fortuitously provided Bangladeshis with new ‘rents’ derived from the rapidly expanding highly profitable garment industry from 1980 onwards. At the same time growing external aid became another important source of rents. The military regimes’ failure to reach an accommodation with the elite factions they had side-lined or to supress those unwilling to be co-opted eventually led to their overthrow. President Ershad was eventually forced to stand down by a united front of the opposing political parties. To resolve the struggle for power,


P. Landell-Mills

they agreed to a caretaker government organising free and fair elections. Unfortunately the basis for moving towards an open access order – a robust consensus on the rules for a participatory democracy, the equitable sharing of ‘rents’ and an acceptance of the need to establish the conditions for growing the ‘rents’ (i.e. investment leading to accelerated economic development) – was still not in place. Instead the party that narrowly won the election adopted a winner-takesall approach and the opposition became increasingly convinced that the next election due in 1996 would be manipulated by those in power to ensure their reelection. And, indeed, the February 1996 election was boycotted by the main opposition parties, giving the BNP all the seats. Only 21 per cent of the electorate voted. This led to a series of violent hartals that eventually forced the ruling party to accept an imaginative constitutional innovation for appointing an independent caretaker government headed by the most recently retired Chief Justice whose principal task was to oversee new elections. These took place in June 1996, with the Awami League becoming the ruling party. The caretaker arrangement for holding elections worked tolerably well both in 1996 and 2001, but broke down in 2006 as a result of the government’s attempt to ensure that a Chief Justice favourable to the party in power was appointed and would retire at the right time to take charge of the elections as the head of the next caretaker government.7 In other words, the contesting political factions had yet to reach a consensus on the ‘rules of the game’ and hence were unable to take Bangladesh forwards towards an open access order. The military intervened to restore confidence in the elections, but the decision of the present Awami league government to abolish the constitutional provision for a caretaker government could easily lead to a new crisis when the time comes for the next election, due in 2013. Following the North–Wallis–Weingast paradigm, we would also expect the role of the military to have been critical. And, as we have seen, that was indeed the case though without being captured by either of the two main political parties. At critical moments in Bangladesh’s recent history, the army has intervened to bring about regime change. But, as was noted earlier, its experience of ruling (1975–90) was troubled and unpopular. Consequently, since the establishment of democratic governance in 1991, the military have preferred to remain out of politics provided its privileged status was respected and it continued to receive generous budget allocations. The Bangladesh military have not become much involved in business activities either, preferring to make money from its participation in UN peace-keeping assignments. To access these benefits the army needed to maintain a good international reputation. It has not therefore wished to be used for the suppression of internal disturbances. This role has been left to the police who, having no credible independent power base, have over the years been obliged to support those in power. Nonetheless, when the political process seemed dangerously blocked in 2006, the army did intervene and, despite their unpopularity in this role, if the main political parties cannot agree on how to hold the next election, there is always a possibility that the army will again intervene, though always with an overwhelming public

Where are the drivers of governance reform? 45 expectation that new elections will be held quickly to restore ‘legitimate’ democratic government.8 In brief, Bangladesh may be judged to lie somewhere between a basic limited access order and a mature access order and it is far from clear at present whether it is progressing towards a mature order. Willingness to undermine the independence of the Supreme Court is evidence rather of regression, as are the actions that limit the independence of the Anti-Corruption Commission. While applying the North–Wallis–Weingast paradigm is helpful in understanding Bangladesh politics, it is formulaic and incomplete. For a more complete picture of the political economy, it is important to incorporate into the analysis an understanding of the deep-rooted socio-cultural determinants of political and bureaucratic behaviour. Behind the regime changes recounted above, and the complex contestation relating to access to ‘rents’, there have been continuous problematic practices and behaviours within government, involving both politicians and public officials. These practices have been extensively reported by various writers over the last two decades9 (see IGS 2006, 2007, 2008a, 2009a; Landell-Mills 2002; Wood 2000). The proximate cause of Bangladesh’s governance problems is to be found in the nature and functioning of its main political parties. These are large, factional, patron-client agglomerations from top to bottom. Local government, civil service, police and the judiciary are all drawn into the patron-client nexus which subverts formal rules in order to meet personalised objectives for oneself, one’s kin and one’s clients or patrons, as well as others to whom past obligations are owed or to create loyalties for the future. These have been pictured by Wood as concentric circles of attachment. The result has been extensive corruption and nepotism to satisfy social and political obligations. There is widespread capture of the formal (i.e. supposedly transparent and therefore accountable) governmental systems of financial and personnel management. Formal rights and entitlements are overridden by informal practices which are socially exclusionary and distort public policy by capturing resources and assets from intended beneficiaries. This patronage system, which Wood (2000) has aptly described as ‘imprisoning’ Bangladesh society, challenges reformers keen to promote a ‘good governance’ agenda. Those with political power operate in a pyramid structure – a hierarchy of predation upon the weak, exploiting deep-rooted cultural traditions of social deference (Hossain 2010). The poor, however much they might seek to escape from this entrapment, dare not because of their dependence on patrons, thus reinforcing the institutions that oppress them. This analysis suggests that change can only come from determined collective action, promoted by civil society and community-based organisations which may be able to mobilise, protect and intermediate. At the same time, it poses a question: to what extent is it possible to be a reformer (i.e. in Wood’s parlance ‘an escapee’) when one is so caught up in the reproduction of the system. This gloomy picture has one bright patch. Since 1991 Bangladesh has seen its incumbent government thrown out whenever elections have been held – in 1996,


P. Landell-Mills

in 2001 and, after a hiatus when the army intervened, in 2008. This demonstrates that there is a significant swing vote that is not ‘captured’ by one of the main political parties. The problem is that elections turn out to be a crude form of accountability which has not led to a significant reform of the political culture. IGS’s 2008 State of Governance (SOG) report highlighted political party behaviour that interferes at all levels of government for private gain. It describes how local government is penetrated and managed by local party ‘goons’ or ‘mastaans’,10 who terrorise officials to allocate contracts to favoured persons, divert resources, for example, from Vulnerable Group Feeding programmes, and allocate jobs to their nominees. Officials who do not comply are punished with poor reports, unattractive transfers and postponed promotions and salary increments. This process of local politicisation has become further institutionalised by the decision in early 2010 to each give elected MP a block allocation of Tk.150 million to spend in their constituencies. The same political interference, according to the SOG reports, occurs in the central government, with the Establishments Division pressured to make staff transfers and promotions based on political allegiance not merit, the Public Service Commission to adopt discriminatory recruitment practices and the Ministry of Finance is subject to similar improper pressures in allocating resources among the line ministries. Indeed, the culture of fear among bureaucrats is said to be so strong that few officials are prepared to make formal, documented decisions, which are therefore continuously pushed upwards to the political level, where even ministers might refer quite minor issues to the Prime Minister’s Office for political vetting, in case someone somewhere would be offended.11 Few if any significant organisations in Bangladesh are truly independent of those in power. Even Grameen Bank was brought back under government control and Nobel Prize winner Muhammad Yunus, its founder, was ousted from his position as managing-director following his attempt to start a new rival political party. Taking into account these deep-rooted socio-political processes, let us return to the North–Wallis–Weingast paradigm of different groups within the elite contesting for power. In Bangladesh’s case there are two main political groups locked in what seems like an unending struggle, with no side giving ground and little sign of a permanent resolution, even though at times for a while one side or another has clearly gained the upper hand. This appearance of stasis is misleading; a number of key features in the political economy landscape are changing. A new dynamic is at work with accelerating globalisation. The information revolution that has accompanied the emergence of electronic digital technology weakens the capacity of those in power to control and manipulate citizens in the street and contain their frustration. Education and economic growth is producing a better informed and more assertive populace that wants to see an end to corruption and unaccountable government. The expansion of export manufacturing – there are now over 3000 garment factories – is severely hamstrung by the consequences of poor governance limiting its future growth and frustrating a powerful vested interest.

Where are the drivers of governance reform? 47

The potential drivers of governance reform If one accepts that those in power have no strong incentive to reform, but instead an obvious interest in resisting reform, then we must look for other drivers of collective action to press for governance reform.12 There are two distinct categories – one motivated by the search for profits and the other by a sense of civic duty. The first group is made up of businesses and businessmen identified above, who form part of the elite and are mostly linked by kinship to the politicians. Families of the elite, wishing to hedge their bets, have a foot in both or, if they can, in all camps. Second, there are civil society organisations (CSOs), broadly defined to include virtually all the other non-state actors – more specifically: professional associations, independent policy research and advocacy centres, the independent media, NGOs and the women’s movement.13 Many of the leaders of these organizations also have kinship links with the elite. Recognising their civic motivation in no way assumes that CSOs are faultless or have escaped the divisive politicisation and patronage that plagues public institutions. It is widely recognised that CSOs need to be encouraged and supported to become more transparent and accountable to their members and those they seek to help by strengthening their own internal governance mechanisms and adopting appropriate codes of conduct (IGS 2006). An increasing number of civil society organisations are affiliated in various ways to political parties, ranging across the spectrum from associations of professionals to NGOs. Although the memberships of these groups are large, the clientelistic dependency of their leaderships limits their capacity to exact accountability from the state and in some cases to function as effective pressure groups for reform. Given the extensive divisive capture of most of the key nonstate organizations by the political parties, it would seem that the prospects for reform are not good. However, this is too pessimistic; it is worthwhile examining more carefully the potential contribution of each of these groups to promoting reform. For a start, let us consider in more depth business groups and prominent businessmen as drivers of governance reform. Historically, European kings wishing to consolidate their realms, and keen to plunder their neighbours, needed money to pay for their armies and their military adventures. Those who produced wealth – the traders, guildsmen and farmers – sought to negotiate favourable arrangements with their sovereigns who awarded towns special privileges and rights in exchange for the right to impose taxes and ceded to property owners’ rights of tenure. In order to gain access to the resources they needed for their political adventures, rulers were forced to share power (Bates 2001). Following a similar logic, today’s businessmen in Bangladesh could equally be powerful drivers of change if they so chose, negotiating with their political leaders to put in place the right conditions to allow their businesses to prosper – efficient and reliable infrastructure, a predictable and honest judiciary and regulatory framework, honest and well behaved police, and put a stop to the harassment and corruption of businesses by public officials.


P. Landell-Mills

Based on many conversations with prominent businessmen over a decade or more, I have found no obvious signs that they are forcefully and persistently demanding better governance or that their advocacy has contributed much to governance reform. They do, of course, complain frequently though fatalistically about poor governance. But there is little evidence that they are prepared to exercise their financial muscle to seek change, even though they do not lack resources if they choose to become more militant. The reason for their acquiescence is not hard to find. As they readily admit, they have calculated that it is easier, cheaper and quicker to bribe away the obstacles to their operations, rather than try to change the political system of extortion and risk making enemies of those in power.14 They are undoubtedly right, but only if one takes a short-term view. There are, in addition, more subtle and complex reasons why the business community find it hard to achieve collective action, despite the strong incentives in the highly competitive world of labour intensive exports to reduce business costs.15 First, given Bangladesh’s extraordinarily divisive political system, almost everyone is politically aligned or is seen to be aligned and politicalbusiness family loyalties run deep. Again using Geof Wood’s terminology, they are as ‘incorporated’ into the political patronage system, though not as adversely as are the poor. Moreover, trust is inevitably in short supply and since the political system feeds on corruption, it is hard for businessmen to ‘escape’. Second, because the political scene is so volatile, businesses apply a very high discount rate to longer-term benefits, focusing instead on achieving short-term profits. In any event, investment in garment manufacture, which dominates the industrial sector, is essentially one of rapid cost recovery; an investment in a new factory can be paid off in less than five years. And perhaps, as some have speculated, there is carved into the psyche of Bangladesh’s delta people, exposed to centuries of devastating floods and shifting land, a profound sense of instability and uncertainty that favours a short-term perspective. In calculating whether to make a stand and fight poor governance or take the easy route of payoffs, businessmen and those they bribe must make a difficult calculation. How much will the market bear? If the bribe-takers and extortionists demand too high a price, they risk driving businessmen into bankruptcy or militant advocacy. At the margin some will refuse to pay. Out of these transactions arises a ‘market rate’ in bribery and extortion, which Bangladesh business leaders claim amounts to well under 5 per cent of business costs and consequently is affordable. But this does not, of course, include the huge costs resulting from poor infrastructure, inefficient ports and the like that are attributable to poor governance and bureaucratic incompetence. It is hard to assess the cost of these failures accurately. In 2007 the president of the Bangladesh Enterprise Institute, Farooq Soban, stated that all types of corruption adds as much as 25 per cent to the cost of doing business,16 though it is hard to verify this estimate. Whatever the correct figure, none would dispute that corruption is a huge drag on the economy. Over the years the number of businessmen in Parliament has risen steadily and now stands at over 60 per cent. One may assume that this also reflects the

Where are the drivers of governance reform? 49 growing presence of businessmen in the ranks of both the main political parties. So far, it is hard to detect signs of much overt pressure for governance reform from these businessmen turned politicians. Indeed, it is likely that the main motivation of the businessmen entering politics is to secure special privileges for their businesses, thereby tending to perpetuate rather than reform the existing clientalist system. What then might alter the business community’s calculus that makes for relative passivity in the face of relentless political pressure to extract ‘rents’? First, if their businesses evolve in a way that requires investments with longer payback periods, then they will want greater security and predictability. Moreover, relentless competition in global markets will add to the pressure to cut costs and raise productivity. Second, the profile of business families is changing as second and third generations start to take over. These children and grandchildren have often been educated abroad – many are graduates of US, UK and Australian business schools – have spent time working abroad and have been acclimatised to, and understand, the advantages of a very different, more modern, business environment that is associated with an open access order. They are much better educated and more sophisticated than their parents and have a much broader vision of Bangladesh’s potential under better governance. Consequently, we may hypothesise that they are more likely to press for reform. Turning now to civil society, where might one find drivers of change? First, let us consider professional associations. These might be expected, in the collective interest of their members, to campaign for governance reforms in matters that adversely affect their members. Surely, for example, one might hope that the bar council would wish to foster integrity and professional good practice in the courts, that the association of chartered accountants would wish to impose a strict code of conduct on their members to protect their reputations, and so on across the professions. But to date this does not seem to be the case, or at least to no significant extent. This may be explained partly by a lack of tradition, partly by narrow shortterm self-interest, but by far the most important reason is that these institutions have become heavily politicised. Consequently, we are forced to conclude they are not at present significant forces for improved governance and in some cases actively obstruct reform. For example, the Press Club has been unhelpful in getting the Right to Information Act passed and implemented. However, over time and with growing global pressures and a more educated public, greater professionalism may emerge, perhaps aided by external assistance, and these organisations may gradually acquire a more effective voice in advocating reform. Similar considerations may apply to Bangladesh’s trade unions, although the weaknesses and constraints in their case are more acute. There are as yet no research studies that trace in depth the links that bind the unions’ leadership to the political parties, but it is often alleged that the considerable rents extracted by organised labour are shared up the political chain. It is hard at present to see much positive contribution to governance reform coming from this quarter, indeed rather the opposite. In any event, union membership has fallen dramatically in recent years as a result of privatisation.


P. Landell-Mills

In the case of Bangladesh’s independent policy research centres, the story is more promising. These have grown rapidly in number, with many engaging constructively with the state on governance issues, while resisting political cooption. For example, the Institute of Governance Studies at BRAC University has produced valuable research on the state of governance in Bangladesh which is impressively frank and outspoken about the problems of governance and the need for reform, and has trained senior civil servants on these matters. Notable, too, for their independence and professionalism are the Centre for Policy Dialogue and the relatively new Policy Research Institute; both have engaged with the government on a range of issues related to governance and the economy. And there are many other examples. While the immediate impact of these initiatives on public policy is hard to detect, they have unquestionably contributed to raising difficult issues, pushed for reform and contributed to building public awareness of the systemic weaknesses in governance that increases the pressures on the political elite to consider reforms. Chambers of trade and industry and various business-sponsored policy research institutes lie between the categories of CSOs discussed above, being neither professional associations nor independent centres for policy and research, since they represent or at least reflect business interests. For the most part, they are not engaged in actively campaigning for better governance. The Bangladesh Enterprise Institute is an interesting case in point; it has published a long list of reports, but surprisingly few have a direct bearing on the governance discourse. Its 2009 annual report, the latest posted on its website, is curiously silent about governance problems, though in 2007 BEI did organise a round table discussion on how to address corruption. An active independent media, with professional investigative reporting, is generally regarded as a prerequisite for voicing public demand for better governance and for scrutiny of public office. Bangladesh has a vibrant and diverse print media, and a fast-growing electronic and broadcast media. However, their independence is vulnerable both to corporate interests and to capture by political parties. Despite these constraints, and the obvious need to improve the quality of investigative reporting, there is a wealth of material in the media exposing all manner of misdeeds by public officials. To what extent the specific allegations are solidly based is hard to say and undoubtedly there is a considerable scope for malicious reporting. Nonetheless, the media do provide an important platform for public discussion of governance failures, for example, through televised debates and ‘question time’ sessions in which politicians and other public representatives are publically ‘held to account’ (IGS 2006). In short, it cannot be denied that the media raise awareness of governance failures and exert pressure on the government to reform. Unfortunately, the government’s reaction to this pressure often seems unconstructive, more often with threats to curtail press freedom rather than prompt actions to address the issues raised. In Bangladesh, the development NGOs have played an exceptionally prominent and valuable role in supplementing public agencies in the delivery of services, especially in relief work, health, education and, above all, micro-credit,

Where are the drivers of governance reform? 51 whether as independent service providers or as subcontractors to government. Indeed, development NGOs have often pioneered major innovations in such varied areas as oral rehydration therapy to reduce deaths from diarrhoea, family planning, and stipends and simple village schools to encourage girls to remain in education. There are over 2000 development NGOs. A few, like BRAC, have become major national organisations, with substantial self-generated funding. Many are known for their effective leadership, efficiency and constructive impact on service delivery. But NGOs have been much more cautious in entering the sensitive debate on governance reform fearing, not without reason, that such an initiative might encourage the government to curb their activities. Thus, many have preferred to remain relatively quiet on governance issues, such as corruption that damage their development programmes, rather than put at risk their pro-poor development programmes. Although NGO action on governance remains relatively limited, in recent years programmes to build citizen demand for good governance have become more prominent. Some of the large NGOs (BRAC, Proshika, etc.) have established advocacy departments and networks, yet there has been no attempt at mass mobilisation. Evidence-based advocacy tools (e.g. citizen report cards, surveys, opinion polls) are still not widely used and few NGOs are experienced in applying such techniques, though this also now beginning to change. A recent study quantifying the impact of social mobilisation in rural Bangladesh (Kabeer 2012) concluded that donors’ encouragement of development NGOs to take up service delivery had shifted their core activities away from social mobilisation. However, the study, which investigated the impact of Nijera Kori on its members’ ‘democratic knowledge, practice and engagement’, found that Nijera Kori’s focus on education, information sharing and social mobilisation raised its members’ ‘political consciousness that qualifies trust in public institutions with implications for enhanced democratic accountability and an alternative civil society approach to improving democratic citizenship’. Whether the NGO movement more widely will increasingly engage in demanding better governance, with the risk of a political backlash, remains to be seen. By and large, NGOs have sought to protect themselves from being caught up in Bangladesh’s divisive politics by remaining neutral, although they have not entirely avoided politicisation. After the 2001 election, some NGOs were increasingly perceived to be party-affiliated,17 and this damaged the non-partisan reputation of the NGO sector. Challenged by the government, NGOs generally drew back from the political front line but this has not discouraged the present government from threatening to revise the NGO law in a way that would severely curtail their scope for advocacy work. Although Transparency International Bangladesh (TIB) lies somewhere between an advocacy NGO and an independent research and policy centre, undoubtedly it is the most prominent and successful of all Bangladesh CSOs in addressing governance issues. Established in 1996 as a Trust and registered in 1998 as an NGO, TIB has ‘undertaken various activities, research, information, dissemination, campaign and advocacy to bring about greater awareness amongst


P. Landell-Mills

the general public of the nature, extent and implications of corruption in Bangladesh across all sectors public, private and non-governmental’. TIB is committed to ‘values of democracy, justice, rule of law, transparency, accountability, integrity and impartiality’. Its mission is stated to be ‘to catalyze and strengthen a participatory social movement to promote and develop institutions, laws and practices for combating corruption in Bangladesh and establishing an efficient and transparent system of governance, politics and business’.18 TIB’s activities and influence have grown steadily. It has not only undertaken surveys and investigations into specific areas of corruption, but has also initiated a form of mass mobilisation with the creation of Committees of Concerned Citizens in some 45 districts to act as watchdog bodies to deter coruption, drawing support from ‘constituencies of citizens’. While there has been no in-depth assessment of the impact of these Committees, they certainly have the potential to create strong local pressures on public agencies to be more accountable. A recent UK Aid evaluation of TIB concluded that TIB had helped increase transparency and accountability in education, health and local government institutions through national and local campaigns and engagement. At the national level, it has contributed to replacing the executive boards of the Anti-Corruption Commission, Election Commission and the Public Service Commission through advocacy and lobbying during the period of the last Caretaker Government. Other examples of research-based lobbying led to improvements at Chittagong Port and the introduction of the Public Procurement Rules 2007. In short, TIB is a striking demonstration of the constructive role a civil society organisation can play in promoting good governance. At the same time, its efficacy is directly a function of the receptivity of the government of the day, somewhat weakened by its dependence on external funding. It has been a good deal less influential over the past two years compared to the period of the last CGT due to the present government’s weak response to the challenge of corruption. Another example of a CSO active in the field of governance is the Manusher Jonno Foundation (MJF ) founded in 2002 to promote ‘human rights’ and ‘good governance’ in Bangladesh. It is committed to promoting voices of people whose rights are being denied and violated, and challenging the vested interests and established hierarchies in society that perpetuate poverty. It supports groups of citizens interested in creating an enabling environment where human rights and good governance are upheld. MJF has maintained its independent character and has built a reputation for integrity and quality in its field of activities. It has grown steadily, disbursing over US$10 million annually in recent years. There are a large number of very active women’s organisations in Bangladesh. These have played a significant role in promoting women’s rights, human security and gender-sensitive governance (Nazneen and Sultan 2009). They include Bangladesh Mahila Parishad, the oldest and largest of the women’s organisations, with more than 130,000 members nationally and around 500 local units, and the radical organisation Nari Pokkho. Their energetic lobbying and advocacy work has yielded considerable successes in campaigns at national and

Where are the drivers of governance reform? 53 local levels on issues such as violence against women, women’s political representation, constitutional and legal reforms, and the role of religion in public life and national policy. These questions have absorbed their energies; they seem so far not to have been drawn into tackling issues of corruption or the central governance weaknesses. Although these organisations have largely avoided politicisation, they, as so many NGOs, are seen to have often become heavily dependent on donors and hence donor agendas, though it is not clear how these might differ from the agendas they would have chosen on their own. Advocacy CSOs currently active in the area of governance take risks. For example, the present TIB Executive Director, Iftekhar Zaman, has received a number of death threats and TIB’s offices have been searched by government officials on more than one occasion. It is clear that its work in exposing corruption has upset a number of highly placed individuals. But, at the same time, the informal system through which real power is achieved and exercised is far from monolithic. Ties of loyalty and kinship that criss-cross the members of the elite enable activists to obtain informal protection from the worst abuses. There is a mix of ‘saints’ and ‘sinners’ within the elite families, just as there are informal links and networks across politics, business, academia and public service, providing a context for intra-elite negotiations where eventually modest steps towards reform and modernisation may be decided. It is irrefutable that reforms could be devised that would result in long-term win-win outcomes for both the elite and ordinary citizens; the challenge in the case of each reform is to overcome the resistance of the few within the elite that benefit disproportionately from the status quo. Reviewing the wide disparities in economic performance of countries over the long term, researchers had for long sought to explain why some countries have performed markedly better than others. Many explanations have been offered for this – different levels of resource endowment; cultural, ethnic, climatic and geographical differences, and so on – but in the end it is hard to escape the conclusion that the quality of political leadership is a major part of the explanation. At each stage in the political history of Bangladesh the skills, vision and motivations of those in power have been critical. There is little doubt that, with all the other factors unchanged, but with more visionary leadership, Bangladesh would undoubtedly have performed far better and have progressed much further towards an open access order – and could do so now if the leadership so chose. But this is the subject of a separate study. To summarise, the North, Wallis and Weingast analytical framework, when applied to Bangladesh, is helpful in exposing the true nature of the political struggle set apart from the formal rhetoric of political leaders. In truth, despite having the trappings of a ‘modern’ state, Bangladesh’s modern institutions (political, executive and legal systems) are all deeply compromised by the underlying logic of a limited access state. This has provided a fragile stability – enough to allow moderate economic and social progress, but with a high degree of drag from poorly performing governmental organisations and little genuine commitment to promoting inclusive development.


P. Landell-Mills

The remaining conundrum is how, or under what circumstances, ‘prisoners’ are likely to become ‘escapees’. I can only make some tentative suggestions on this at this stage, as the question merits more research. One factor, surely, is overseas education. More and more young Bangladeshis are going abroad for master’s and doctoral degrees; we may reasonably assume that this has a profound impact in ‘modernising’ attitudes and drawing those concerned away from the traditional forces that have bound people to the traditional social structures. This will even be true for those gaining higher degrees within the country. A second factor is financial independence from salaried positions in the private sector and new wealth derived from successful businesses.

Donor support for governance reform Across the developing world donors have for years struggled to address problems of governance. In the 1970s and 1980s the problem of poor performing countries was mainly attributed either to inappropriate economic policies or weak capacity and poor public administration. This gradually morphed into discussions of how to strengthen ‘public management’ (World Bank 1983). The term ‘governance’ first appeared in the development literature in 1989 World Bank’s Africa Report (World Bank 1989). More precisely, it was a group of prominent Africans19 on the World Bank’s Council of African Advisers who insisted that the problem of corruption and weak governance should be placed centre stage. Then the pace quickened. By the mid-1990s multilateral and bilateral aid agencies were openly discussing corruption as an overriding constraint to development. Nonetheless, there was still little formal recognition of the underlying political nature of poor governance and the links to winning and holding onto political power. If there had been, the confusion and disappointment over aid conditionality might not have arisen. Again and again the donor community has promoted technical ‘supply side’ actions to address corrupt procurement, weak financial management and flawed personnel management. The formal diagnostic was that developing countries lacked the knowledge and management capacity – problems which could be solved by technical assistance. The implicit underlying assumption was that political leaders and the donors shared the same agenda: to achieve open, efficient and accountable government. Project failure was attributed to lack of ‘political will’ implying, rather curiously, that the political leaders lacked the will-power to do what they wanted to do. This of course was not the problem; they were doing exactly what they wanted to do. But the reality was that the agendas differed markedly – political leaders were primarily concerned to remain in power in order to enjoy the patronage and the personal financial and social advantages that were derived from it. Governance failures were not seen by politicians as political failures, but rather as the source of resources to achieve their personal political, social and financial ends. Given the endless ruthless competition for power, it is hardly surprising that political leaders have very short time horizons. To confuse the picture, the rhetoric of government officials

Where are the drivers of governance reform? 55 was hypocritically aligned with the concerns and aspirations of the donors. They knew how to please with fine words to secure a steady flow of aid money. But in practice their actions followed the completely different logic reality of political opportunism and self-advantage, leading to disappointed donor expectations year after year. This has been as true in Bangladesh as elsewhere. The formal incorporation of political analysis into donor thinking started in the early 2000s, with a number of ‘political economy’ country background papers.20 They exposed the awkward political realities surrounding the contestation for power and highlighted the associated corruption. These studies presented the managers of aid agencies with a serious dilemma. The issues surrounding poor governance were revealed to be intensely political and intractable; it was very hard to see how a donor could constructively respond with the tools and approaches they had used up until then. The issues were immensely sensitive, threatening to torpedo the very basis of the aid partnership. In most cases, the response was to continue with ‘business as usual’. The situation in Bangladesh was no different. For the donor community, weak governance has for long been and remains a critical and intractable challenge. Most donor programmes are profoundly undermined by the prevailing weakness described earlier. Consequently, the impact of aid falls far short of its potential and donors over the years have expressed their frustration and disappointment that the funds they make available from their own taxpayers so often seem wasted. In truth it is hard to see a way out. Reform must come from within Bangladesh, yet in practice there is hardly any progress. Endless donor hectoring of government officials for failing to ‘deliver’ reforms sours the relationship and achieves little. A governance assessment commissioned by a donor in 2008 marks a watershed in that donor’s thinking about governance reform in Bangladesh. There were a number of similar studies at around the same time.21 It offers a thorough analysis, building on an earlier ‘drivers of change’ study prepared in 2002. It took on board the socio-political economy factors presented above, recognised the failure of donor support for the ‘supply side’ governance reforms and noted the presence of a vibrant civil society, but then focused attention mostly on the role of NGOs: A positive aspect of accountability mechanisms is the country’s vibrant civil society, comprising business groups, NGOs, think tanks, electronic and print media. Business groups stand out particularly for their ability to influence (macro-economic) policy. NGOs have played an active role in enhancing poor people’s ‘voice’. They put pressure on the state to fulfil its obligations and to protect the civil, political, cultural and economic rights of the poor, and vulnerable and excluded people. This has led to achievements in terms of ‘islands’ of effective local government and service delivery. The challenge remains for NGOs to institutionalise reforms and have a national impact. That is, of course, a significant challenge, given generally antagonistic relationships between NGOs and government, hindered further by weaknesses in NGO governance.22


P. Landell-Mills

The report’s apparent expectation that development NGOs may have a larger role relative to other non-state actors in civil society in demanding better governance is not well supported by their performance to date (Kabeer et al. 2009). While NGOs have made a major contribution to Bangladesh’s development, they have mostly had a minor part in promoting governance reform.23 By and large NGOs have not been very active in pressuring government to become less corrupt, improve its financial accountability, reform its procurement processes, seriously tackle corruption by sanctioning public officials who take bribes, nurture free media or respect the rule of law. Equally, it is risky for development NGOs to advocate the reform of political parties, especially as the political parties are often quite hostile to them and are critical of NGOs’ own (i.e. NGO) governance mechanisms. More powerful pressures are needed, and these must come from other non-state actors which have greater clout and stronger links with the ruling elite, especially the business community and the independent media. Thus, in the formulation of their assistance strategies donors’ official strategic thinking has so far failed to fully internalise the critical insight that governance reform has to come from more broadly based internal pressures. The need is for donors to radically rethink their strategies to recognize that supply side programmes alone are inherently ineffective in a context where the ruling elite have a different agenda. Administrative and system reforms are necessary, but on their own they will never result in sustainable reform. For such reforms to succeed, they must backed by strong political leadership. Thus, the donors need to take account of the incentives faced by each of the key stakeholders relevant to the reforms. The question to be asked in each case is whether the stakeholder concerned is genuinely ‘incentivised’ to support the reform or, at least, not to oppose it. In practice, for each case of corruption there are usually a relative small number of beneficiaries, in contrast to the much larger numbers who are adversely affected. Understanding how the costs and benefits arise and are allocated is critical to devising a successful reform assistance strategy, Supply side projects reforming the core administrative functions unaccompanied by measures to change the contextual incentives are unlikely to succeed on their own.24 The key unmet challenge is to find ways to start to turn the socio-culturally embedded institutional malpractices around. A good example of this approach can be found in the recent UK Aid-supported police reform programme.25 The police service in Bangladesh has hitherto been regarded as a profound and pervasive source of widespread rent seeking and corruption. The Police Reform Project aimed at a transformation from ‘colonial’ to ‘community’ policing, and involved setting up 11 model thanas since 2007 as ‘pro-people, service-oriented’ police stations with service delivery centres. In addition, victim support centres have been established alongside more supply side reforms. The programme has sought to bring supply and service into a relationship with the demand side pressures for improved standards of policing and evidence. Ask most ordinary Bangladeshis what they want most from their governments and ‘personal law and order’ will most often be

Where are the drivers of governance reform? 57 the answer. So this programme linking supply with demand is a strong example of moving in the right direction.26 The ‘drivers of change’ studies strongly indicate that donor priorities should remain with the demand side in order to seek to rearrange the incentives for good supply side behaviour. However, it also seems that a significant proportion of donor governance assistance relies upon finding or developing a critical mass of change agents who can be co-opted as the project directors of supply side projects. But this will only happen through a deliberate strategy of continuous follow up, contact and team building reinforced by external, demand side pressures not to revert to ‘business as usual’. Even if the social and political economy analysis summarised above were internalised by the donors, it would still be difficult for the donors to translate the resultant logic into effective programmes. None seems to have much room for manoeuvre to act on this logic within the parameters of diplomacy and a strong preference for risk avoidance. Consequently, donor approaches remain inherently conservative and elitist in their conception of how fundamental change occurs, i.e. via enlightened leaders rather than frustrated followers. It is the ‘enlightened leadership’ assumption which, however false, still implicitly underpins the justification of supply side interventions even though the strategy has self-evidently failed. The short-term incentives dominating the thinking of the political elite make it highly unlikely that there will be a strong push for reform from the top even when a leader is so inclined. As argued above, poor governance is inherently a political failure and not therefore receptive to technical solutions alone. Without question, supply side capacity needs to be strengthened, but supply side projects will not deliver improved governance until and unless the political culture becomes more receptive to reform. Two key outcomes sought by ordinary citizens – reduced corruption and greater security – are particularly elusive. In Bangladesh reform in these areas seems to make little headway. Indeed, more generally governance reforms appear stalled or have actually gone into reverse in their implementation.27 Corruption remains pervasive,28 while maastanisation especially in the countryside appears to be a growing threat to personal security.29 In the face of these trends, donors need to radically reorient their strategies to give priority to supporting the whole range of non-state actors in demanding better governance. And they should seek to increase their impact through strengthened co-ordination of their governance programs, informed by a shared analysis and an agreed new governance assistance strategy. Critical to this approach is for donors working in Bangladesh to understand better the nature, role and potential of civil society (broadly defined to include all non-state actors). This is consistent with donors’ recognition that to be effective in supporting the strengthening of civil society to participate in governance reform and development more generally, more analysis is needed of the nature, composition, governance and strengths and weaknesses of civil society organisations (Griffin and Judge 2010). For a start it would be useful for the donors jointly to commission the mapping of the civil society ‘sector’ and then


P. Landell-Mills

to share in its assessment. Out of this should come a shared appreciation of the potential of civil society to campaign for better governance and some agreement on how the likely drivers of reform might be supported and encouraged. This is not to conceal the concern I have that donors are not well equipped to make the fine political judgements needed to navigate these treacherous waters. Although, up to this point, we can argue that such an initiative fits well within the Paris Accord for donor harmonisation, it is hard to reconcile such activism with the ‘alignment’ of donor policies with those of the host government, whose priorities are likely to be quite different. In Bangladesh’s case, the local mechanism for donor coordination – the Local Consultative Group – calls for a joint donor-government chairing of the various sector groups. However, significantly, in the case of the governance group, the government was reluctant to appoint a co-chair and only did so after a long delay. Lastly, we must always be vigilant that donors do not simply overwhelm civil society organisations in a new enthusiasm for demand led governance reform, creating a ‘santised political space’ which fails to promote any real political reform (Hossain 2012a).

The way forward Across the world citizens are ever more impatient and frustrated by corruption and other abuses of state power. Mass movements such as the Orange Revolution and the Arab Spring are challenging self-serving authoritarian regimes. In China, the number of strikes and demonstrations has risen dramatically over the past five years.30 Less dramatically, but probably in the long term more significantly, through a myriad of new initiatives citizens are demonstrating ways for CSOs to engage public authorities constructively to negotiate for better governance. New tools and approaches are being invented and honed to force public authorities step by step each year to become a little more transparent and accountable. For example, India’s poorest villagers are using the Right to Information Act to find out their entitlements and demand an end to public officials embezzling their benefits. In the Philippines a new procurement law enables citizens to monitor public contracting and purchases. In Cambodia, Nepal, Peru and Tanzania people are using social accountability tools such as citizen report cards, participatory expenditure tracking and community score cards to demand an end to the theft of public money and the poor delivery of public services. And citizens in even the most repressive countries have learnt to use the media, Facebook and Twitter to expose the malfeasance of public servants. All these activities are being gradually protected by legislation and new institutions – anti-corruption laws, whistleblower protection, codes of ethics, citizens’ charters and anti-corruption commissions are beginning to appear everywhere. The tide of reform is strong and universal. The most effective strategies stress gradualism and constructive engagement wherever allowed. The aim is to achieve ‘negotiated accountability’ that builds on incremental gains and has a long-term perspective. Those now in power in Bangladesh would be wise not to resist these pressures. But if they do,

Where are the drivers of governance reform? 59 the public is likely to become more militant, following the example of citizens elsewhere that have taken to the streets. Poor peasants have often in the past resorted to ‘rude’ accountability (Hossain 2010) in response to local abuses. So far, violent protest in the urban areas in the form of hartals has been instigated and firmly controlled by the opposition political parties. But the urban middle classes may easily lose patience and engage in a new form of street action that the present elites will not control. It would seem wiser for the elite to learn from experience elsewhere and engage in genuine well thought-out reform rather than procrastinate and risk the disruption of growing popular protests which may spin out of control. This chapter opened by asking why the main stakeholders in Bangladesh’s development – businessmen, the youth, the middle classes and the families of the elite – who have all so much to benefit from good governance seem so ineffective in demanding accountable and transparent government. In response, at the first level of analysis, it was argued that the high degree of divisive politicisation of almost all relevant organisations, whether public or private or not-forprofit, have effectively blocked the demand for reform. However, if we go more deeply into why this is the case, one must recognise that there are deep social and cultural factors, linked to Bangladesh’s history and environment, which give rise to a hierarchical and clientalist society, that largely explain the nature Bangladesh politics and hence poor governance. We must equally recognise that none of these factors is static. Many elements are changing. Despite poor governance, per capita incomes continue to rise, the health of the population is improving, education levels are rising and, with them, people’s aspirations. Bangladesh is being drawn increasingly into the global economy and its sons and daughters are more and more exposed to global trends and practices. Ideas spread ever faster and more extensively with the rapid development of new information technology. All this is progressively enlarging the space for civil society activism. In recognition of these trends, donors are reappraising their assistance strategies and are seeking more effective ways to help strengthen civil society organisations to enable citizens to demand governance reform. Change is inevitable; with imagination and wise leadership it could benefit everyone. For the whole array of non-state actors that wish to see governance reform, the challenge is to find ways for each to contribute constructively to promoting reform. There are a myriad opportunities and a wide range of proven tools to be used. For donors, the challenge is to rethink their methods and modalities to find new and better ways to assisting these non-state actors, working together rather than competing. This should be made the central thrust of their country assistance strategies. Finally, it is important for donors to recognise that more research is needed to build a better understanding of the character and capabilities of the different drivers of governance reform and the context in which they operate. In particular, it would be useful to know far more about the internal arrangements, practices and dynamics of the main political parties, both the formal and the


P. Landell-Mills

informal realities. The same is true of the main business associations. And it is imperative to build up a much more comprehensive database on the other nonstate actors. Lastly, there needs to be an ongoing and sustained assessment of governance reform initiatives as they unfold.

Notes 1 This chapter draws on two recent studies undertaken by different teams that included the author, which were prepared for donors and have not been published. With regard to these, the author wishes to acknowledge in particular the contributions of Alex Duncan, Geof Wood, Naomi Hossain and Mirza Hassan. 2 For example, Acemoglu and Robinson 2006; Barrington Moore 1966; North et al. 2009. 3 Including failed states. 4 North et al. (2009) builds on Max Weber’s proposition that ‘a state is a human community that (successfully) claims the monopoly of the legitimate use of physical force within a given territory’ (see Weber 1919). The issue is one of the legitimacy of the elite using violence to stay in power. 5 For example, Botswana, Singapore, South Korea and, more controversially, India and Malaysia. 6 Holding a neutral and fair election was the original justification for the CTG; this was a promise it kept, including preparing a ‘clean’ voters’ list and voter identity cards with photos. 7 This was achieved by raising the Chief Justice’s retirement age. 8 For an interesting discussion of the relatively autonomous role of the military in Pakistan and Bangladesh, see Alavi 1972. 9 Also see the research publications of Transparency International Bangladesh (TIB). 10 Mastaan used to refer to urban underworld gang leaders behaving rather like the Mafia but, as a phenomenon, they are becoming the new patrons of the countryside. 11 From interviews with senior civil servants. 12 The concept of ‘drivers of change’ first appeared in 2003 in a report on Bangladesh (Duncan 2003). 13 Not covered for lack of research, but certainly no less important, are faith-based groups. 14 Interviews with the author in 2010. 15 It is worth noting that the Bangladesh Garment Manufacturers’ and Exporters’ Association has a reputation for acting more independently of the political parties than most other business associations, reflecting its greater vulnerability to increased costs. 16 At a roundtable discussion on Anticorruption Drive and the Way Forward organised by the Bangladesh Enterprise Institute on 5 September 2007 (www.bei-bd.org). 17 Most notably Proshika. 18 www.ti-bangladesh.org. 19 Ex (and future) Presidents, Prime Ministers and others who had held very senior government positions. 20 For example, with a series of studies of ‘drivers of change’. 21 The main governance analyses undertaken by the aid agencies in Bangladesh are: DFID’s 2002 ‘Drivers of Change’ study and the 2008 ‘Country Governance Analysis’; the World Bank’s 2002 ‘Taming Leviathan’ and its ‘Bangladesh Country Assistance Strategy 2009’; the ‘Power and Change Analysis’, commissioned by the government of the Netherlands in 2008 as part of its Strategic Governance and Corruption Assessment; the EU’s Mid-Term ‘Review of the Country Strategy Paper for Bangladesh’ (2007–13); USAID’s ‘Democratic Governance’ analysis of 2009; and

Where are the drivers of governance reform? 61

22 23 24 25 26 27 28 29 30

UNDP’s ‘Mid-Term Evaluation of Progress towards Outcomes in Governance in Bangladesh’, 2009. This list only includes studies and reports issued between 2002 and 2009. DFID’s Country Governance Analysis 2008, section 12, page 4. Available www. albacharia.ma/xmlui/bitstream/handle/123456789/31775/1620Bangladesh_Country GovernanceAnalysis_20088.pdf?sequence=1 (accessed 18 May 2017). There are some notable exceptions, e.g. Transparency International Bangladesh, Manusher Jono and BRAC’s IGS. As demonstrated by continuing problems of unaccountability and corruption after 25 years of supply side support through training and public sector reform projects from across the donor community. To develop a safer and more secure environment based on respect for human rights and equitable access to justice through policing, which is more responsive to the needs of poor and vulnerable people (2005–08). This assessment is based on field work in 2008 and needs to be revalidated in the light of the performance of the present government to check whether the reforms have been continued. For example, the operations of the Anti-Corruption Bureau and the implementation of the Right to Information Act. The Corruption Perception Index for Bangladesh is 2.4, placing it 134 out of 178 countries (i.e. in the lowest quartile). Recent studies of four villages in rural Bangladesh under the ‘Wellbeing in Developing Countries’ research programme at the University of Bath provide ongoing evidence and analysis of these processes, confirming the IGS SOG (2008a) analysis. See, for example, www.bbc.co.uk/news/10434079 and globalspin.blogs.time.com/2011/ 02/28/in-chinas-jasmine-crackdown.


Citizen-centered governance Lessons from high-performing Asian economies for Bangladesh Habibul Haque Khondker

Introduction This chapter attempts to assess the political and economic developments in Bangladesh and outlines the possibilities of creating an inclusive democratic society. Once derided as “international basket case,” Bangladesh’s economic achievements, as the country turns 42 (in 2013), are not to be sniffed at. The Economist magazine in its November 3, 2012 issue has praised Bangladesh’s economic achievements (Economist, 2012), and the Guardian quoted economic pundits’ prediction that following the BRICs, Bangladesh is likely to be one of the 11 “newwave” economies (Elliott, 2012). The changed perceptions in the international media, that has been traditionally skeptical about Bangladesh’s economic growth prospect, followed the favorable assessments made by multilateral development institutions as is evident from Tables 3.1 and 3.2. Bangladesh’s growth is part of the economic growth acceleration in South Asia. In terms of Human Development, Bangladesh is one of the frontrunners in South Asia, excelling India, its giant neighbor. However, Bangladesh gets low rating in its democracy and governance indicators despite the existence of democracy for 26 years (of 42 years) as indicated in Tables 3.3 and 3.4. The puzzle that Bangladesh presents is: development as measured in the improvement of social indicators can take place in the context of moderate economic growth; and economic growth can take place against the backdrop of a factious and limping democracy and a weak, sometimes dysfunctional, government. The continued economic growth (5 percent per annum on the average since 1990) may be sustained and furthered provided improvements are made in the quality of governance. Compared to Southeast Asia, Bangladesh (for that matter, South Asia as a region) is still lagging behind in Human Development Indices (HDI). Hence, Bangladesh needs to strengthen its governance to catch up – even surpass – Southeast Asian development outcomes. In this regard Bangladesh may benefit from the development experiences of the high-performing economies of Asia. Such lessons combined with innovations of new strategies will help improve the quality of governance in Bangladesh. Much of the development discourse in Southeast Asia at the time of its rapid economic growth was characterized by the debate over democracy and development, often presented as a stark choice. The debate was premised on what

0.566 0.500 0.568 0.714 0.524 0.526 0.768 0.662 0.714 0.725 0.725 0.674 0.790 0.851 0.940 0.975

68.9 67.2 65.4 76.8 68.8 65.4 74.9 75.2 74.1 68.7 73.5 69.4 74.2 81.1 83.4 81.1

Life expectancy at birth

Source: UNDP, Human Development Report 2013.

Bangladesh Bhutan India Maldives Nepal Pakistan Sri Lanka Vietnam Thailand Philippines China Indonesia Malaysia Singapore Japan Norway

HDI Value 4.8 2.3 4.4 5.8 3.2 4.9 8.2 5.5 6.6 8.9 7.5 5.8 9.5 8.8 11.6 12.6

Mean years of schooling

Table 3.1 HDI of South Asian and selected Asian countries, 2013

1,529 5,293 3,468 5,276 1,160 2,550 4,943 2,805 7,694 3,478 7,476 3,716 13,685 52,569 32,295 47,557

GNI per capita in $ppp 146 141 134 109 157 145 97 128 103 112 101 124 61 26 12 1


Low Low Medium Medium Low Low High Medium Medium Medium High High High Very high Very high Very high

Status of Human Development


H. H. Khondker

Table 3.2 Growth rate of GDP (% of year) of South Asian and selected Asian countries

South Asia Afghanistan Bangladesh Bhutan India Maldives Nepal Pakistan Sri Lanka China Indonesia Malaysia Singapore







9 8.2 6.6 7.8 9.7 18.0 3.7 5.8 7.7 12.7 5.5 5.8 8.6

8.8 14.2 6.4 12.6 9.2 7.2 2.8 6.8 6.8 14.2 6.3 6.5 8.8

6.3 3.4 6.2 10.8 6.7 6.2 5.8 3.7 6.0 9.6 6.0 4.7 1.5

7.1 20.4 5.7 5.7 8.0 –2.3 3.8 1.2 3.5 9.2 4.6 –1.7 –0.8

7.9 8.2 5.8 5.8 8.6 4.8 4.0 4.1 7.6 10.3 6.1 7.2 14.5

7.5 8.0 6.3 6.3 8.2 5.0 3.8 2.5 8.0 9.6 6.4 5.3 5.5

Source: Adapted from Table A1, Asian Development Bank, 2011, p. 249.

Amartya Sen calls the “Lee hypothesis” named after Singapore’s founding Prime Minister Lee Kuan Yew, who, according to Sen (1997), “has defended authoritarian arrangements on the ground of their alleged effectiveness in promoting economic success.” One of the leading exponents of the “Lee hypothesis” stated: “the crucial variable in determining whether a Third World society will progress is not whether its government is democratic but whether, to put it simply, it has “good government’ ” (Mahbubani, 1992: 8). In the ensuing reformulations, “governance” as a catch-phrase came to dominate the development discourse, privileging the role of the strong state as a pivot for development. It was hence argued that good governance is more important than democracy and, therefore, improvement in the quality of governance will bear fruitful developmental outcomes. Such an argument implied postponement of democracy in favor of good governance undergirded by a strong state. This chapter challenges the argument that posits a choice between a strong developmental state and democracy and embracing the former as a secured path to achieve economic development. A strong state governing economic development model may not work in Bangladesh given her historical specificities. This is not an argument in favor of the status quo. Bangladesh’s social stability and economic growth may be threatened by a factious democracy. The privileging of the present “electoral democracy” to the neglect of improvement in governance may undermine economic growth and social stability in the long run. A citizen-centered governance is the ultimate guarantor of both sustaining democracy and development in Bangladesh. The chapter argues that a citizen-centered governance which is democratic in substance has been a key feature in high-performing Asian economies. The premise of our argument is a thesis developed by German political sociologist Claus Offe, who argued that democracy is parasitical on a strong state (Offe, 2011). Conversely, a weak state is a bad host and a poor driver for democracy. Following this line of argument, and modifying it slightly, we will state

92.2 99.4 91.4 71.2 66.4 95.0 90.1 94.0 91.7 95.3 94.1 96.8 94.2 91.8

51.4 78.2 60.0 43.5 33.1 87.0 77.0 66.9 82.5 79.5 68.7 103.0 100.0 84.8

Secondary school enrolment 5.0 38.3 10.1 9.0 9.0 15.0 23.7 35.1 29.0 18.0 61.0 75.0 45.0 49.0

Internet users

Source: World Competitiveness Report, 2012–2013,World Economic Forum, Davos.

Bangladesh China India Nepal Pakistan Sri Lanka Thailand Vietnam Philippines Indonesia Malaysia Singapore Brazil Russia

Primary school enrolment 56.5 73.2 72.0 43.8 61.6 87.0 113.2 143.4 92.0 97.7 127.0 149.5 123.2 179.3

Mobile telephone subscription per 100 ppl 26.7 51.0 31.6 33.6 13.6 20.3 30.0 29.3 24.6 33.0 33.7 44.4 18.4 28.6

Gross national savings

Table 3.3 Global competitiveness index, 2012–2013 (selected variables in selected Asian and BRIC countries)

42.9 25.8 68.1 34.1 60.1 79.0 41.7 38.0 40.5 25.0 52.6 100.8 66.2 9.6

General government debt as % of GDP

Source: Economic Intelligence Unit, 2010 Total of 167 countries.

0.00 2.50

Authoritarian regimes China 136 (3.14) Afghanistan 150 (2.48)

9.58 9.58 7.00 6.92 6.50 8.33

10 10 10

4.33 7.42 5.17 1.83

31 (7.77) 40 (7.28) 55 (6.64) 60 (6.53) 71 (6.19) 74 (6.12)

1 (9.80) 2 (9.65) 3 (9.52)

Electoral process and pluralism

82 (5.89) 83 (5.87) 104 (4.55) 108 (4.24)

Hybrid regimes Singapore Bangladesh Pakistan Nepal

Flawed democracies France India Sri Lanka Indonesia Malaysia Philippines

Fully democratic Norway Iceland Denmark

Rank and overall score

5.00 0.79

7.50 5.43 5.71 4.29

7.14 6.07 6.07 7.50 6.79 5.00

9.64 9.64 9.64

Functioning of government

Table 3.4 Democracy index of South Asian and selected countries

3.89 2.78

2.78 4.44 2.22 3.89

6.11 5.00 5.00 5.56 5.56 5.00

10 8.89 8.89

Political participation

5.36 2.50

7.50 5.00 4.38 5.63

7.50 6.25 6.88 5.63 6.25 3.13

9.38 10 9.38

Political culture

1.18 3.82

7.35 7.06 5.29 5.59

8.53 8.82 8.24 7.06 5.88 9.12

10 9.71 9.71

Civil liberties

Citizen-centred governance 67 that a strong state is a prerequisite for deepening democracy. The two processes, strengthening the state and deepening democracy, must go hand in hand. And a citizen-focused governance can achieve meaningful developmental outcomes and aid in deepening the process of democracy. After considering the arguments in favor or against the so-called participatory governance and aligning it to the debates over deliberative democracy, we argue that the link between the nature of the state and the quality of democracy needs to be reconsidered. We follow Guillermo O’Donnell’s distinction between electoral democracy and citizenship democracy which he defines as “a way of organizing society with the object of assuring and expanding the rights of its people” (O’Donnell, 2004: 54). This chapter is organized in five sections. Following this introductory section, the next section discusses the relationship between governance and development outcomes. In the third section we consider the issue of lessons of development from high-performing Asian countries. We consider both models of development, highlighting the challenges of copying or learning from others and the importance of innovation. In this regard, we focus on Singapore as an exemplar to show how lessons can be learned and contextualized. In the fourth section, we raise some issues with regard to the nature of democracy, especially the role of citizen-centered governance in deepening democracy. This leads us to the issue of the political context of governance. In the concluding section, we outline a political framework for Bangladesh and postulate some strategies towards deepening democracy and institutionalization of citizencentered governance. Tables 3.1–3.4 provide the context of discussion, indicating that the Asian economy is the most dynamic of the world regions and Bangladesh has the potential to be a high-performing economy in the not-too-distant future provided certain aspects of governance framework are strengthened. Bangladesh has shown resilience in economic growth even at a time of global economic slowdown, with remarkable achievements in some of the social indicators, especially those pertaining to Millennium Development Goals (i.e., reduction of maternal mortality and infant mortality and gains in primary education). Bangladesh has achieved economic gains in the context of a democratic polity that has been characterized as “hybrid regime” by The Economist Intelligence Unit (EIU). The democracy in Bangladesh is top-down and contentious to a fault, threatening political stability. It is interesting to note that both Singapore and Bangladesh are classified as a “hybrid regime” by the EIU, yet in terms of both quality of governance and economic achievements, the two countries are studies in contrast. Table 3.3 indicates that in some of the basic indicators that increase competitiveness Bangladesh has shown reasonable promise. Primary and secondary school enrolment is one such indicator. The main assumptions of this chapter are as follows. A citizen-centered governance is a prerequisite for the wellbeing of all the people by creating a socially stable and culturally vibrant society – in a word, a thriving society. The potential for a citizen-centered governance which is a prerequisite for the sustainable development of Bangladesh remains unrealized largely due to the deficiencies in


H. H. Khondker

governance. In order to retool the structure of governance towards citizenfocused more work needs to be done to develop an appropriate political framework that would ensure a citizen-centered governance by granting more rights to the people, paving the way for an inclusive democratic society. We consider the idea of citizen-centered governance as a means as well as an end of the development process. A meaningful development outcome is also likely to make the democracy stronger and more robust. We argue that lessons must be learnt selectively from the experiences of the high-performing Asian economies that achieved rapid economic growth or are showing promises of development overcoming historical and structural disadvantages. It would be useful to see what lessons can be learnt and what cannot in view of the cultural (or even civilizational) differences and historical specificities.

Governance and development The discourse of governance has dominated discussions of economic development since the 1990s thanks to the changed pragmatic focus of the international development agencies, notably, the World Bank. The discussion began in earnest in the wake of the rise of the “Asian Tigers” in the 1980s, the collapse of the socialist states in East Europe, culminating in the collapse of the Soviet Union in the late 1980s and early 1990s, with a renewed interest in the role of the government or the state in economic development. The World Bank’s flagship publication, the World Development Report of 1997, was called The State in a Changing World. Learning from East Asian development was a major theme in countless discussions on development. The World Bank’s annual development report as well as the new literature of the time used the phrase “good government,” proclaiming “Good government is not a luxury – it is a vital necessity for development” (World Bank, 1997). Whatever the terminology, it deals with matters of state involvement in economic management and development. The role of the state was emphasized in the late development in the nineteenth century (Gerschenkron, 1962). The state played a crucial role in post-Meiji Japan leading to a miracle economy in the 1970s (Johnson, 1982). Gunnar Myrdal’s (1968) notion of “soft state” presaged today’s “dysfunctional” or “failed (or failing) state.” The remarkable socio-economic development in East Asia became a model in the 1980s and 1990s, with the World Bank devoting an annual report to the role of the state/government in East Asian economic development. The notion of governance is biased for its state-centric approach. The concerns of the citizens are not always adequately reflected. This lacuna was overcome with the introduction of the discourse on participatory governance. Even in the World Bank agenda of good government, the idea of involvement of civil society and public-private partnership received due attention. The discussion of participatory governance, though opening up the possibility of more openness, transparency and accountability, remains somewhat inadequate for its failure to address the deeper issues of democratic participation and going into the sources of the enabling environment for participation. Discussions of participation of public or stake-holders

Citizen-centred governance 69 often remain vague. A critical school of participatory governance pointed out various inequities of the participants is the setting of agenda, etc. The view that suggests that people or masses are clients takes a corporatist approach, whereas the idea of citizen-focused governance is based on the recognition of the rights of citizens, which presupposes deepening and opening up of more space for democracy. In order to ensure that governance is not just state- or elite-centric, participation was encouraged. However, participation entails certain problems. First, who should participate? Can the poor participate, or do they delegate their participatory role to their representatives? How does one go to grassroots level? Many of these became part of the slogans of participatory development. Some writers became so skeptical of participatory development and participatory governance that they coined such phrases as “tyranny of participation” (Cooke and Kothari, 2001). In order for participation to be meaningful it is important to recognize, respect and institutionalize the rights of the citizens as citizens. The idea of participation of stakeholders is seemingly a forward step towards participation, but several problems remain. The issue of governance, especially participatory governance, must take into account the idea of citizenship and the success of governance is dependent on the recognition of citizenship rights of the populace. Who decides who the stake-holders are? The idea of stake-holder is corporatist and thus not founded on true democratic principles. It is aimed at forming a consensus but does not capture all the constituencies who are affected. The principle of equal participation is at the heart of meaningful democracy. A number of questions concerning participation need to be analyzed if we want to address the issue meaningfully, otherwise it would remain nothing more than a normative judgment at best and a wish at worst. In the Declaration of the Right to Development adopted by the United Nations in 1986 the “right to development” is presented as an “inalienable right”: “The right to or entitlement to participate in and contribute to and enjoy economic, social, cultural and political development, in which all human rights and fundamental freedoms can be fully realized.” This declaration, like many other UN declarations, is not legally binding on anyone and thus can be treated at best as a normative judgment. What is needed is to develop a rationale and modalities for participation. Citizenship is a precondition of participatory governance, i.e., good governance, and also of democracy. The right to participation becomes meaningful only in the context of citizenship rights. Citizenship rights are the cornerstones of substantive democracy. However, there are instances where democracies ignore the importance of citizenship rights; and conversely, non-democracies may take into account citizenship rights for instrumental reasons. It is a truism to say that good governance produces good developmental outcomes. However, once we disaggregate the governance indicators the relationship is not as straightforward as it appears to be. Governance can be defined as the process of decision-making and the implementation of those decisions using established institutions. Governance can be used in several contexts, such as corporate governance, international governance, national governance and local


H. H. Khondker

governance. Since governance is the process of institutionalized decision-making and the process by which decisions are implemented, it entails an analysis of the institutions as well as the traditions of decision-making and the process of implementation of the decisions for a desired societal outcome. In this definition, government is one of the key actors in governance. Other actors involved in governance vary depending on the historical, economic and political contexts. In some contexts ethnic or religious divisions may play a role. In rural Bangladesh, local power structures, landed elites, trade unions such as farmers’ or fishermens’ associations, professional associations, NGOs, religious leaders and political parties are constituents of the ensemble of governance. The situation in urban areas is equally, or may be, more complex. The above definition may be too broad to be analytically useful. However, it hints at the non-triviality of the concept and alerts us of the thin line that separates and sometimes conflates governance with the environment of governance. According to Worldwide Governance Indicators (WGI) project: Governance consists of the traditions and institutions by which authority in a country is exercised. This includes the process by which governments are selected, monitored and replaced; the capacity of the government to effectively formulate and implement sound policies, and the respect of citizens and the state for the institutions that govern economic and social institutions among them. (http://info.worldbank.org/governance/wgi/#home) WGI uses six criteria in its conceptualization of governance: (1) voice and accountability; (2) political stability and lack of violence/terrorism; (3) government effectiveness; (4) regulatory quality; (5) rule of law; and (6) control of corruption. Based on research evidence, DFID indicates that there is a strong relationship between effective governance and development. According to the Department for International Development (DFID, 2011): Better governance is positively associated with increasing primary education and adult literacy, and reducing infant mortality. Effective political governance is critical to economic growth including through improved investment and productivity. Improved governance is essential for achieving the MDGs. Democratic governance tends to provide developmental benefits through responsive governance over the long term, but other forms of government have also proved to be high effective at reducing poverty. Political instability and policy uncertainty have significant negative effects. A high score in Voice and Accountability show either the presence of democracy or a certain openness in the polity where information is allowed to percolate for pragmatic reasons. South Korea (69.19), South Africa (65.40), Brazil (63.50), India (59.24), Indonesia (48.34) and the Philippines (46.91) are democracies which guarantee the free flow of information; hence they all share high

1 (6.12) 114 (3.71) 81 (4.30) 125 (3.52) 132 (3.39) 73 (4.42) 30 (5.27) 99 (4.02) 60 (4.62) 33 (5.19) 3 (6.05)

8 (5.73) 115 (3.24) 58 (4.03) 130 (3.03) 112 (3.34) 55 (4.06) 49 (4.37) 125 (3.14) 61 (3.98) 42 (4.62) 1 (6.13)


Source: World Competitiveness Report, Economic Forum, Davos.

Hong Kong Bangladesh India Nepal Pakistan Sri Lanka China Philippines Indonesia Malaysia Singapore

Rank and score 1 (6.77) 133 (2.25) 86 (3.43) 139 (1.81) 110 (2.75) 70 (3.82) 50 (4.44) 104 (2.92) 82 (3.56) 30 (4.97) 5 (6.22)


Table 3.5 World competitiveness index, basic requirements: selected Asian countries, 2010–2011

10 (5.67) 80 (4.49) 73 (4.53) 86 (4.41) 133 (3.19) 124 (3.60) 4 (6.11) 68 (4.60) 35 (5.15) 41 (5.01) 33 (5.20)

Macroeconomic environment

28 (6.29) 106 (4.96) 104 (5.16) 109 (4.81) 123 (4.27) 35 (6.18) 37 (6.16) 90 (5.42) 62 (5.78) 34 (6.18) 3 (6.67)

Health and primary education


H. H. Khondker

Table 3.6 Voice and accountability

Bangladesh Brazil China India Indonesia Iran South Korea Malaysia Nepal Pakistan Philippines Russia Saudi Arabia Singapore South Africa Sri Lanka Thailand UAE Vietnam




33.17 62.09 5.76 59.61 45.67 8.17 67.78 29.32 30.28 25.48 44.71 23.55 4.80 34.61 63.94 32.69 31.25 22.59 7.69

37.91 60.66 5.21 59.24 45.97 7.10 68.24 32.22 31.75 24.64 44.39 20.85 3.79 36.49 62.55 30.80 33.17 26.06 8.05

38.38 63.50 5.21 59.24 48.34 6.63 69.19 31.27 30.80 27.48 46.91 20.85 3.79 37.44 65.40 31.75 30.33 24.17 8.53

Source: Worldwide Governance Indicators, 2011.

scores. The score of Bangladesh (38.38), a rambunctious democracy, and that of Singapore (37.44), a country not known for her democratic vigor, are comparable. In the case of Singapore, the pragmatic structure of governance leaves enough room for feedback – sometimes through letters to newspapers or online discussions, or even direct feedback to various government departments. The passage of a Right to Information Act in 2009 and the establishment of an Information Commissioner are significant steps in Bangladesh towards enabling the free flow of information. From Table 3.7, we can see that political stability is an important component of governance. Singapore is one of the world leaders in this regard. The UAE also has a high score as it provides a safe, stable social setting which is a public good in itself, but also makes a contribution in encouraging foreign investment and tourism. Former Malaysian Prime Minister Mohamed Mahathir, whose role was crucial in the development of Malaysia, commented in a speech in the Philippines: When people think only of the freedoms of democracy and know nothing of the implied responsibilities, democracy will not bring the goodness it promises. Instead, it will result only in instability, and instability will not permit development to take place and the people to enjoy the benefits of freedom and rights that democracy promises. (Mahathir 2012)

Citizen-centred governance 73 Table 3.7 Political stability and absence of violence/terrorism

Bangladesh Brazil China India Indonesia Iran South Korea Malaysia Nepal Pakistan Philippines Russia Saudi Arabia Singapore South Africa Sri Lanka Thailand UAE Vietnam





24.51 34.13 38.94 17.78 13.46 32.21 62.50 60.57 35.09 11.53 30.76 11.05 37.50 85.57 27.40 5.76 57.21 74.03 57.69

23.55 50.48 31.73 18.26 2.88 22.59 54.32 46.63 12.01 14.90 9.13 13.94 56.73 79.80 36.05 4.80 60.09 77.40 56.73

9.13 33.17 31.73 14.42 4.32 21.63 61.05 55.76 3.36 6.73 5.76 7.69 23.07 84.61 40.38 15.86 23.55 73.07 49.51

9.90 48.11 24.05 10.84 18.86 6.13 50.00 43.39 5.66 0.47 6.60 18.39 35.84 97.65 44.33 21.22 12.73 75.94 51.41

Source: Worldwide Governance Indicators, 2011.

South Asia fares poorly on political stability. Part of the volatility can be explained by unequal distribution of resources, i.e., social inequality, but it also exposes the weakness of the government in maintaining social order and peace. The culture of confrontational democracy is another contributing factor. Overall violence is a problem that affects the quality of life as it also degrades the competitiveness of a country. Bangladesh is surely more violent than China, Indonesia, Vietnam and Nepal, and less so than Thailand, Sri Lanka, India and Pakistan (see Table 3.7). Bangladesh has to work hard at both the institutional as well as behavioral level to reduce the level of violence. Violence has a great social cost and is linked to poor enforcement of the rule of law. China, Indonesia and Vietnam have better enforcement of laws than that of Bangladesh. Containing violence will enhance confidence of the business establishments and lead to higher investments. Clearly, Singapore is a world leader in terms of government effectiveness (see Table 3.8). For any Asian country it would take a long time to achieve such a high standard. Besides, Singapore’s small population (5.3 million), which is about one-third of the population of Dhaka, is an advantage. Singapore in addition possesses one of the world’s most talented and dedicated civil services, which also helps. Singapore’s leadership is known for its high degree of pragmatism, part of which is translated into greater involvement of the public in aiding and fine-tuning the government’s various plans and policies.


H. H. Khondker

Table 3.8 Government effectiveness

Bangladesh China India Indonesia Malaysia Pakistan Philippines Singapore Sri Lanka Thailand Vietnam





24.87 45.36 53.65 40.00 76.58 30.73 50.24 100 46.82 62.92 34.14

32.19 52.68 52.19 43.90 82.43 31.21 51.70 91.57 43.41 60.97 39.51

19.51 55.60 51.21 44.39 84.87 40.97 49.26 96.09 42.92 66.82 38.53

21.53 59.80 55.02 47.84 82.29 25.83 51.67 100 49.28 58.37 45.93

Source: Worldwide Governance Indicators, 2011.

Lessons of development from high-performing economies with special reference to Singapore In recent years, in the wake of the development of the “Asian Tigers,” a number of second-generation countries, such as Malaysia, Thailand, Indonesia, the Philippines and Vietnam, have made significant economic development. The lessons of these countries also have great relevance for Bangladesh. One has to study deeply how an appropriate model of development can be framed drawing upon the lessons of successful countries in the region. The evidence from the highperforming Asian economies in the 1980s and early 1990s revealed that governmental efficiency often compensated for the absence of a vibrant democracy in enabling meaningful and inclusive development. Some writers called it “softauthoritarianism”; it played a strong role in propelling desired economic development. In the last two decades most of the “soft authoritarian” regimes have transited towards the inexorable path of democratization. Singapore is often touted as a model for good governance. Can lessons from Singapore be utilized for improving the quality of governance in Bangladesh? One might suggest that the idea of importing lessons from Singapore simply ignores the political realities of Bangladesh. There is some truth in that. At the same time, the differences in political culture and historical experience cannot be taken as excuses for ignoring some of the lessons that can be drawn from Singapore or elsewhere. For example, Malaysia seriously studied the developmental lessons of Korea and Taiwan in forging its development strategies. Cultural differences do not always neutralize the positive effects of learning from the best practices. Singapore was ranked no. 2 in the Global Competitiveness Report of 2012–2013 (Schwab et al., 2013). Of 144 countries, Singapore was ranked no. 1 in government regulation and in transparency in government policy making, and no. 2 in government services for improved business performances. Bangladesh

Citizen-centred governance 75 was ranked no. 85, 101 and 120 respectively in the above categories. In the regulation of securities and exchanges, Singapore was ranked no. 3 and Bangladesh no. 130. In the reliability of police services, compare Singapore’s rank of 3 with Bangladesh’s 126. The most striking, however, is the ranking in “public trust in politicians.” Singapore ranked no. 1 and Bangladesh 124. The GDP per capita of Singapore according to the UN Human Development Report 2012 stood at US$52,569. Singapore ranked 26 in the Human development Index. While Bangladesh was ranked 144 of 174 countries according to the Corruption Perception Index of Transparency International in 2012, Singapore was ranked 5 following Denmark, Finland, New Zealand and Sweden. Recognizing the role of government in economic development it can be stated that, where citizens are taken into confidence and where the public policies address the needs of the public rather than a segment of the public or private interests, meaningful developmental outcomes are more likely. Meeting the needs of the ordinary citizens not only makes development meaningful, it also empowers citizens to play a part in the development process. A stable democracy is more likely to take root in a society where citizens are empowered. The history of democracy reveals that democracy evolved largely because of the growing demands from an empowered citizenry, hence a true democracy is citizenship democracy. It is not just the elite and the middle class who wanted democracy, the call for democracy often came from the bottom for inclusion (Markoff, 1996). Singapore stands out for its strong commitment to the rule of law, a precondition for economic success (see Table 3.9). While Singapore remains the leader in the rule of law, Thailand, Malaysia and India also receive high marks. India has had a long tradition of weak state institutions. Yet the tradition of rule of law in India is deeply entrenched. The independence of the judiciary in India is a vital contributing factor for Indian democracy. If Bangladesh could follow the standard of rule of law in India, it would make a significant improvement in governance. The only country in Table Table 3.9 Rule of law

Bangladesh China India Indonesia Malaysia Pakistan Philippines Singapore Sri Lanka Thailand Vietnam





16.74 39.23 61.72 44.97 68.42 30.14 52.63 94.28 59.80 67.46 39.71

22.00 35.40 60.28 27.75 61.72 24.88 37.32 87.55 57.41 67.46 42.10

16.26 31.73 53.11 26.79 67.46 22.48 32.05 94.73 56.93 55.50 42.10

26.54 40.66 54.50 31.27 65.40 25.59 34.59 93.36 52.60 49.76 38.86

Source: Worldwide Governance Indicators, 2011.


H. H. Khondker

3.9 comparable to Bangladesh is Pakistan, where various extenuating factors have weakened the institution. Based on historical evidence drawn from both the East and the West, Francis Fukuyama (2011) argues that a successful modern liberal democracy combines all three following sets of institutions in a stable balance: the state, the rule of law and accountable government. The state has a tendency to concentrate and use power. The rule of law and accountable government, on the other hand, limit the state’s power. There are tensions in this relationship. According to Fukuyama, Afghanistan has had elections since 2004 but the state has no capacity to implement laws in much of its territory. Russia has a strong state and holds democratic elections, but its rulers do not feel bound by the rule of law. Singapore has both a strong state and the rule of law but only an attenuated form of democratic accountability (Fukuyama, 2011). Starting from the strong developmental state model, Singapore is evolving to a more vibrant democratic polity. Does Singapore provide a model or blueprint for national development elsewhere? If the answer is in the affirmative, what are those lessons and how can such lessons be learnt or implemented in other contexts? We explore this issue with a simple question: are there models of development? Can nation-states learn from each other? In the world of corporations we hear about the “best practices”, about ideas and practices that enhance productivity, help the company grow, make it more efficient and so on. Other companies often try to emulate these “best practices” and even try to improve on them. Admittedly, nation-states are not corporations, hence lessons to be learnt in the format of “best practices” must be considered with caution. The precondition for learning is a willingness to learn. This presupposes a change in attitude. Leaders in some countries do see themselves as heads or CEOs of corporations and would seek out ideas and policies that are effective. Taking cultural differences and cultural sensitivities into account, they adopt pragmatic policies that help solve pressing problems. A dedicated corps of public officials is a precondition for implementing national development. Public officials in Singapore work with dedication. Singapore has been able to recruit and train a cadre of civil servants who would be able to deal with their counterparts anywhere in the world. Public participation is often ensured via various advisory committees and consultative bodies. Even in sensitive areas such as censorship boards there are advisory committees. Different advisory committees are formed to get input from cross-sections of the population. The Feedback Unit of the Ministry of Community Development holds regular discussions with invited members of the public for consultation on various public policy issues (Chua, 1995: 23). Even in such matters as the design of the mosaic at Changi International Airport, anthropologists of the National University of Singapore were consulted who vetted the suitability of the motifs. For example, the establishment of casinos in Singapore was preceded by months of public debates. Research money was allocated to study the effects of gambling on Singapore society. University departments would often be approached by the government encouraging research in certain social issues. There is a close relationship between the public universities and the state. Since the universities

Citizen-centred governance 77 are run on taxpayer’s money, they have an obligation to conduct socially useful research. Every time a senior official was to go on a foreign mission, enquiries would be made to the various departments of the university asking if any member has done any research on that country, to give the official a good grasp of the place, society, etc. before he or she left Singapore. Singapore provides a trove of public policies ranging from public housing, high-quality education, crime control, social discipline, traffic safety, health care, fighting commercial corruption, to management of public parks. The following lessons from Singapore’s development stand out. In the first phase of development, creating jobs was given the utmost importance. Productivity increase comes from an educated labor force. Hence education was given priority at the same time. Labor discipline is very important, and was ensured. Control of violence and ensuring social stability was ensured. Neutralizing forces that stand in the way of development is important. These developments made Singapore an attractive destination for multinational companies looking for suitable locations. In the subsequent phase of development, as Singapore sought to move away from a labor-intensive to a knowledge-intensive economy, adequate attention was given to skill development and research and development; part of that goal was achieved through creating world-class research universities. Important lessons can also be learnt from the experiences of such BRICS countries as China and India. China lifted more than 600 million people out of poverty between 1981 and 2004. Rapid growth and urbanization have been central to China’s poverty reduction in the past 25 years, as have a number of reforms, including the opening of the economy to global trade and investment. Even as the overall level of poverty has dropped, inequality has increased, and the remaining poverty has become concentrated in rural and minority areas. China has been able to craft a strong state, a legacy of socialist state making. China’s ability to deliver goods and services provides the state with a great deal of legitimacy. Plans are drawn and implemented effectively. Indiscipline and dishonesty including bribery are dealt with harshly, often with the death sentence. The recent developments in India show the vital role of democracy and citizen participation. The active NGO movements, a vibrant private sector that was able to internationalize itself, are positive signs. Although India’s track record in mass education is weak, its gains in higher education are laudatory for their high quality. Taking the example of migration governance, Bangladesh lags behind both the Philippines and India. Bangladesh can gain simply by cut and paste of the migrant labor policies from India. Both the Philippines and Indonesia in their post-authoritarian phase provide interesting models of economic growth. For decades, both Indonesia and the Philippines were known for high-level corruption and crony capitalism. Since the fall of the Suharto regime in 1998 in the wake of the Asian financial crisis, Indonesia has had two successful elections, ensuring a peaceful transfer of power. Indonesia’s successful democratic transition has aided its solid economic growth. Philippines has posted a 6.3 percent economic growth rate in 2012. The


H. H. Khondker

Table 3.10 Most problematic factors for doing business

Bangladesh Brazil China India Indonesia Iran South Korea Malaysia Nepal Pakistan Philippines Russia Saudi Arabia Singapore South Africa Sri Lanka Thailand UAE Vietnam

Inadequate infrastructure


Inefficient government bureaucracy

Policy inconsistency

21.8 15.1 7.8 17.0 9.5 9.9 4.8 6.4 8.6 8.7 16.5 4.9 8.8 6.1 8.0 8.1 5.3 3.7 7.8

18.5 6.4 8.5 16.7 15.4 4.2 5.6 9.6 10.0 11.6 24.4 22.8 2.5 0.3 11.5 9.1 14.5 1.9 5.7

17.3 10.5 10.9 13.5 14.3 10.6 16.8 12.9 11.8 8.3 18.3 13.3 10.2 2.4 19.6 9.9 11.7 7.4 4.2

6.7 – 9.9 6.1 7.4 13.7 16.7 8.3 10.4 11.2 7.9 1.5 0.1 0.7 3.2 9.3 12.9 4.7 11.8

Source: World Competitiveness Report, 2011 Economic Forum, Davos.

main challenge for the Philippines seems to be a growing population which fuels local unemployment. But thanks to the quality of education, Philippines has been able to derive dividends from the migration of professionals and skilled workers. Overseas Filipino workers (OFW) remitted US$24 billion in 2012 (World Bank, 2012). There are several lessons that can be learnt from the experiences of development in Southeast Asia. First, the value of regional peace and a positive frame of mind in dealing with inter-state relationships. ASEAN provides a good model. There are problems but there are also avenues to resolve conflicts peacefully. The government in Asian economic development has remained a major player, despite the predominance of the rhetoric of neo-liberalism and alleged shifts towards a free-market economy (see Table 3.11). In terms of the role of government in economic development the lessons are rather straightforward. The government should be a catalyst, an enabler. In the globalized world, the government has to be an “adaptive system” ready to change and reinvent itself. In the twenty-first century, with the advent of a knowledge-based society, governments alone are ill-equipped to deal with all the problems of development in a society. Participation of citizens via various channels dedicated for this purpose in fact helps any administration. In improving its governance, Bangladesh can draw upon the best practices from various high-performing countries. There is no common set of best practices; they are related to the specific institutions of society. How and what

Citizen-centred governance 79 Table 3.11 Central government expenditure as percentage of GDP of South Asian and selected Asian countries

Afghanistan Bangladesh Bhutan India Maldives Nepal Pakistan Sri Lanka China Indonesia Malaysia Singapore






19.6 13.9 35.9 25.8 59.3 14.5 18.4 24.3 18.7 20.0 24.8 14.5

19.7 13.4 35.1 26.3 60.8 15.9 20.8 23.5 18.7 19.2 25.0 13.1

19.3 15.8 34.4 28.4 63.1 17.4 22.2 22.6 19.9 19.9 26.4 16.9

19.1 14.3 38.6 29.0 64.9 19.9 19.9 24.9 23.0 16.7 30.3 17.8

21.6 16.0 39.5 25.3 51.2 20.2 20.5 22.9 23.0 16.4 26.8 17.6

Source: Adapted from Table A21, Asian Development Outlook, 2011, p. 269.

policies should be learnt and what should be avoided are the moot issues for the policy makers and planners. It is not suggested here that Bangladesh can or should follow only the model of Singapore. No country can blindly follow or imitate the model of economic development or governance of another country. One of the major problems of imitating the strategies followed in another country is historical time. What was feasible in Singapore in the late 1970s or 1980s in the context of the world economy at that point in time may not be feasible or practical in the context of 2013. Yet important lessons can be distilled from the experience of successful cases. Certain key principles, if not the model, are transportable. Countries, like corporations, can follow some best practices which will produce desirable results. There are many cases of such borrowings in history. Historically, Japan borrowed from various European countries (Westney, 1987), especially Germany, a whole range of techniques and policies and, now the best practices of Japanese corporations have become objects of emulation in the West. Some countries, however, are too proud to learn from others. An open mind and adopting a pragmatic mindset are preconditions for learning the best practices from around the world.

Democracy and development The examples of the Asian high-performing economies also indicate that in some cases democracy may be a facilitator for development; in other instances a confrontational democracy may retard the process of development. Government effectiveness, an important facilitator for development, may be hampered by larger political culture. The presence of democracy in Bangladesh is both a liability and an asset. It is a liability in the short run since in many instances it generates a paralysis; endless debates, intolerant and conflicted politics impede development process. It is an asset because in the end development as a process


H. H. Khondker

is a means to attaining an open society where citizens can live freely with dignity and without fear, in a word, a democratic society. The fact that the polity of Bangladesh is more of a democracy than an autocracy is an important achievement, but it must not be a source of complacency. Democracy is a necessary but not a sufficient condition for the attainment of developmental goals. Much work needs to be done to deepen the process of democratization so as to transform a procedural and flawed democracy to a substantive and meaningful democratic system which would, in turn, create an enabling environment for citizen-centered development, often referred to as inclusive development. In the end the goal of development in Bangladesh as elsewhere is to ensure the general well-being of the people. Well-being refers to positive and sustainable characteristics which enable individuals and organizations to thrive and flourish in an environment of freedom and dignity, hence democratic society. Democracy is a work in progress. Even in advanced democratic countries, serious problems impede democratization in the sense of deepening the practice of democracy so as to make it more inclusive. In view of growing income inequality, and recent concentration of wealth in the hands of a small minority, Paul Krugman (November 3, 2011) has characterized the US as an “oligarchy, American style”; he argues that an extreme concentration of wealth in a few hands is incompatible with “real democracy.” Weakening of democracy is a concern of both the advanced developed as well as the less developed countries in the world. The discussion here needs to be prefaced with a word on the quality of democracy. While Bangladesh has reasons to be proud of the fact that it is a practicing, albeit imperfect, democracy, the quality of democracy shows little sign of improvement, making people weary of democracy. In fact, the state of democracy in the world is not as bright as it seemed to be. There are two views. According to Freedom House, of the 194 countries in the world, 87 are free, 60 partly free and 47 not free. Bangladesh has been placed in the partly free category (see Tables 3.12 and 3.14). The Economist Intelligence Unit (EIU) uses a tighter set of criteria, to the extent that the US and India were rated as “flawed democracies.” According to EIU, of the 167 countries. Bangladesh was placed in the category of “hybrid regimes” (see Table 3.13). Bangladesh does reasonably well for its political rights and civil liberties. According to EIU, free and fair elections and civil liberties are necessary conditions for democracy, but they are unlikely to be sufficient for a full and consolidated democracy if unaccompanied by transparent and at least minimally Table 3.12 Quality of democracy (Freedom House)

Free Partly free Not free


Population (%)

87 (45%) 60 (31%) 47 (24%)

43 22 35

Source: Adapted from “Freedom in the World, 2011”.

Citizen-centred governance 81 Table 3.13 EIU classification

Full democracies Flawed democracies Hybrid regimes Authoritarian regimes


% of countries

% of world population

26 53 33 55

15.6 31.7 19.8 32.9

12.3 37.2 14.0 36.5

Source: Adapted from Economist Intelligence Unit.

Table 3.14 Freedom index, 2011

Afghanistan Bangladesh Bhutan India Maldives Nepal Pakistan Sri Lanka China Indonesia Malaysia Singapore

Freedom status

Political rights

Civil liberties

Not Free Partially free (PF) PF Free PF PF PF PF Not Free Free PF PF

6 3 4 2 3 4 4 5 7 2 4 5

6 4 5 3 4 4 5 4 6 3 4 4

Source: Adapted from “Freedom in the World, 2011”. Freedom House.

efficient government, sufficient political participation and a supportive democratic political culture. It is not easy to build a sturdy democracy. Even in longestablished ones, if not nurtured and protected, democracy can corrode. Although one-half of the world’s population now live under democracy of some sort, the quality of democracy needs monitoring and fine-tuning. EIU classified Bangladesh as hybrid regime, not even a flawed democracy. Now this could be due to the use of dated data. But human rights violations, violence and social disorder may have contributed to this image. Bangladesh started as a democracy, but the post-independence circumstances defeated the goals of that democracy, and the transition from parliamentary to presidential form took place in less than four years of independence. The military coup eliminated all traces of legitimate government. Between 1975 and 1990, Bangladesh had elections without democracy, staring with the “yes/no” vote of General Zia in 1977. Several elections were held during this period, when Bangladesh was ruled by military strong men; first, General Zia until 1981 when he was killed in an abortive coup attempt, followed by the rule of General Ershad from 1982 to 1990. When an elected non-military government came to power in 1991, it could not break from the legacy of the past. Bangladesh made a transition from what Leslie Bethel in another context


H. H. Khondker

describes as “elections without democracy to democracy without citizenship” (Bethel, 2000). The glaring defect of Bangladesh’s democracy is the nearabsence of the rule of law. The rise of a neo-patrimonial state in Bangladesh has been a major restraint for the transition to a citizen democracy. The absence of a rational bureaucracy and the rule of law has led to a lack of trust by the citizens in government. With such a low trust, citizens do not see government as a benefactor. This creates a vicious circle. The donor agencies have been active in promoting democracies with little effect. Strengthening the civil society organizations (CSO) is one strategy. However, donor-driven democratization is a weak substitute for indigenous state building and deepening of democracy.

Future directions: a blind alley or a rising sun? The economic growth trends in Asia are positive and Bangladesh is part of the Asian growth trajectory. Drawing on the lessons of high-performing economies in Asia, sometimes referred to as “dragon economies,” and the examples of BRIC (Brazil, Russia, India and China), this chapter argues that Bangladesh has great potential for rapid economic growth, but that in order to realize the potential, key aspects of governance need to be improved. Often the idea of drawing lessons from high-performing economies are mentioned. However, a cut and paste approach to improvement in governance is more likely to fail. Bangladesh needs to reinvent an appropriate model of governance in light of her own geohistorical and cultural specificities and contingencies. A major rethinking is necessary with regard to the nature of the governance. Bangladesh has also been very successful in her NGO movement and other countries have learnt lessons from the experiences of BRAC and Grameen. Bangladesh has been able to share its knowledge of diarrhea treatment and prevention with other countries of the world. The Bangladesh government, in improving the quality of governance, can look to the successful NGOs in Bangladesh for innovative ideas of participatory development. Other than sporadic examples of collaboration between some of the NGOs and the government, the full potential of partnership remains untapped. Here the mindset of the bureaucrats and myopic vision of the political leaders complicate the structural impediments of the political system. Political leaders are often suspicious, even jealous, of the success of the NGOs and fail to give due credit to their collective contribution to the improvement of the social development of Bangladesh as evident in the improvement of the human development indicators. What can be done to change the political framework of Bangladesh? Here, I venture some observations only. First, there has to be a consensus that democracy has to be strengthened; this would lead to the development of specific strategies towards deepening democracy and institutionalization of citizenfocused governance. Southeast Asia has seen many reforms and much political engineering. Bangladesh remains solidly grounded on the age-old Westminsterstyle parliamentary democracy. As Bangladesh society moves into the twentyfirst century, the state remains solidly entrenched in the twentieth century, in

Citizen-centred governance 83 fact, early twentieth-century politics. Much work needs to be done to initiate political reforms. The two contending parties seem to be bogged down on the modality of elections. It is not the election that makes democracy work; it is the quality of democracy, the appropriate political framework, that the politicians should be concerned with. Transparency about the source of party funding and election funding may be a good start. All assets of the political parties and the sources of their income must be publicly disclosed. Bangladesh has been a practicing democracy, although various writers have expressed doubts about the quality of democracy. The arguments presented in the characterization of Bangladesh’s political regime as “authoritarian democracy” (Khondker, 2007) remain unchanged in 2013. The weakness of democracy in Bangladesh is systemic. The political culture in Bangladesh is not favorable to a transition to citizen democracy and sustaining an electoral democracy. The real problem, however, can be attributed to the weak state structure. The state structure remains paternalistic and dysfunctional. This is why a transition from military to civilian rule or a transition from a presidential form to a parliamentary form did not change the political system or regime type. Bangladesh remains dominated by the rule of a single person at the helm of affairs. It is clearly a neo-paternalist system with political power concentrated in the hands of the Prime Minister. The weakness of the state structure hinders or delays the institutionalization of democracy. The strengthening of the state is contingent on a number of factors, not least the political will and the understanding of the political leaders that the strength of the state, building of institutions to consolidate state or building of a developmental state is a requisite for national development. The bureaucratic inefficiency, absence of professionalism, lack of competence, corruption and tardiness, absence of clear cut rules, poor implementation of rules, and the list goes on – all these problems of government symbolize state failures. Or more appropriately failings of the government. State functionaries, needless to say, readily pass the blame onto the politicians for not allowing them to function. There may be some truth in that, and is symptomatic of paternalistic politics or traditional leadership as viewed by Max Weber in the late nineteenth century. Weber differentiated three forms of power: traditional, charismatic and rational. Bangladesh remains a queer combination of patrimonial political system and democratic government. This hybridized form is deleterious both for the state as well as the government. The democratic system will be consolidated by creating a rational, neutral state system based on the rule of law. This is not to say that the creation of a rational state will solve all the problems automatically. There are inherent contradictions between democracy and the rule of law. This is one of the paradoxes of democracy. It can be minimized by the practice of deliberative democracy that would recognize and guarantee rights of the citizens (Benhabib, 2001; Gutman and Thompson, 2002; Mouffe, 2005). A good deal of rethinking is needed first to nurture the three institutions and then to find a balance between them. Active citizenship will help overcome this contradiction to some extent.


H. H. Khondker

Decentralization is a precondition for deliberative democracy which will promote active citizenship. Citizens’ participation and activism has added a new dimension to democratization (Hardt, 2003). In South Africa, active citizenship has been effective to democratize at the level of local governance (Murray et al., 2010). Strengthening local government is a must for active citizenship. Active citizenship will foster proactive participation of the citizens. A democratic government must be the facilitator for such participation. The enabling role of the government is crucial. Citizens must be recognized as members of the nation state who belong and are entitled to rights. By empowering citizens, government can be truly democratic. As citizens are active in participation in various policy issues, this will be an education of their citizenship. Membership in a state is not just an one-off thing; citizens must renew their citizenship everyday by asserting their rights. Education and consciousness building will come out of praxis. If citizens remain passive receivers of state-help, dole, they will have no incentive for asserting their rights. Their participation must be not only reactive but also proactive. Some of the common problems can be noticed in the growing disillusionment of the citizens with regard to what they can receive from the government, growing distrust of government. In order to narrow the trust gap the government must play its part by being honest, by meaning what it says and saying what it means. The main thesis of this chapter is that for Bangladesh, the appropriate model of governance must take into account the voice and interest of the people, in other words, the deepening of democracy has to be a precondition for participatory governance in Bangladesh. There have been instances in Bangladesh in recent history, when an assertive or a reactive public forced the government to change their badly thoughtout policies. During the BNP government in September 2005, a badly thought-out policy of not supplying electricity to the peasants in the months when they most needed the electricity supply for irrigation led to a massive protest of the rural people near Rajshahi, who stood up against the non-supply of electricity. Of course, political leaders, such as Dr. Kamal Hossian and others supported the movement, but the movement was deeply rooted and autonomous which forced the government to reverse its decision. This was an act of civil disobedience, an uprising without a political leader. Mr. Golam Rabbani who was catapulted to the leadership position was unknown prior to this peasant movement. The second example comes in the protest against the proposed Bangabandhu International airport in Arial Beel, near Munshiganj. In the face of spontaneous protests, the government was forced to reverse its decision. The decision to go ahead with the project to construct an airport by acquiring 25,000 acres of waste land was not based on a feasibility study and utterly failed to consider the impact on the local community as well as the ecology. The protests and demonstration of the local villagers led to violence, even the death of a police officer, and scores of people were injured. Fortunately, good sense prevailed and the Prime Minister called off the proposed construction of the international airport. These are significant episodes in civic engagement, of citizen’s involvement or intervention.

Citizen-centred governance 85 In conclusion, we suggest that both democracy and development should be pursued in tandem to make development sustainable and meaningful. And these two goals can be pursued simultaneously without sacrificing one for the other. Citizen-centered governance entails a transition from electoral democracy to citizen democracy and a citizen-focused governance. And as citizen-centered governance portends to expand the space of democratic process. Alongside inclusive development, we need inclusive democracy. And the two goals are not unrelated.


Governance, rights and the demand for democracy Evidence from Bangladesh Ipshita Basu, Graham K. Brown and Joe Devine

Introduction Over the last 20 years, interest in good governance has progressively become more intense and focused. In part this reflects a conviction shared by academics, policy makers and practitioners that good governance can positively influence a range of key issues including poverty reduction, economic growth, the efficiency of service provision, the impact and effectiveness of development programmes and the building of more inclusive societies (Kaufmann et al. 1999, 2002; McGillivray et al. 2005). Conversely, poor quality governance is considered a barrier or hindrance to growth and wellbeing, and an incubator for corruption, violation of rights, discrimination, violence and disorder. At least this is how the theory goes. However – and not for the first time in its history – the experience of Bangladesh appears as something of a paradox when looked at from a less normative perspective. Of late therefore it has become almost a truism to note and then question the co-existence of two informed observations about modern Bangladesh. On the one hand, the country has made significant and consistent progress in socio-economic terms. It has thus enjoyed steady and prolonged macro-economic growth rates, fuelled by manufacture and remittance growth; made significant progress in relation to many of the MDG targets; and reduced the proportion of the population in poverty from 40 per cent in 2005 to 31.5 per cent in 2010 (BBS 2011; World Bank 2013). On the other hand, Bangladesh’s performance in governance terms has been poor, epitomised in its classification by Transparency International for five successive years (2001–2005) as the world’s most corrupt country. The co-existence of poor governance with successful growth and poverty reduction raises many questions and puts some core development assumptions to the test. Where some might see paradox in all of this, others may see collision. At a minimum, the post-1990 history of Bangladesh tells us that the two observations are not mutually exclusive. Governance discussions in Bangladesh, as elsewhere in the world, were originally steeped in a commitment to an alliance, uncontested since the end of the Cold War, of liberal democracy and market-based economic growth. This global commitment dovetailed nicely with political developments in Bangladesh, in particular the reintroduction of formal electoral democracy in 1990. The onset

Governance, rights and demand for democracy 87 of electoral democracy witnessed a radical shift in the country’s political landscape with both the military and the bureaucracy visibly withdrawing from frontline politics to allow for the emergence of a new political class (Blair 2010). With the exception of the 2007–2008 hiatus when the military re-emerged as the de facto backers of the caretaker government, electoral politics has helped institutionalise a form of democratic governance in Bangladesh which has remained intact and relatively stable albeit, as the contributions to this volume attest, deeply problematic. Since the introduction of formal democratic politics, Bangladesh has gone on to almost effortlessly pass Huntington’s famous two-turnover test (Huntington 1991). At the same time, however, electoral politics has evolved into a de facto two-party system in which both parties ‘compete’ for the right to monopolise the political system and dominate both public and private institutions. The competition is zero-sum orientated, and profoundly impacts the nature and dynamics of party-state relations as well as state-society relations. The result of this democratic process is an aggressive and unforgiving polarised political landscape, in which the interests of the state and those of the party in power, are almost completely aligned. Commitment to democracy is not just concerned with formal political processes and actors but also citizenship identity and practices. Bangladesh remains a relatively young country and democracy was one of the four founding principles of its first Constitution. Many citizens therefore have first-hand experience of the struggle for independence and democracy, and this memory is kept alive in the imagination of younger generations. However, while there seems to be strong public support for democracy (Duncan and Williams 2012), we know far less about the ability of citizens to meaningfully contest and hold the state to account. Some have argued that the room for manoeuvre to promote and protect citizenship rights in Bangladesh is highly restricted, and Wood’s (2000) analogy of Bangladesh society as a prison encapsulates this perspective nicely. Others, however, have argued that social relations are more fluid and have pointed to evidence of stronger citizenship identity and practices in cases where these relations are supported in the right way (Kabeer 2011). Even in this relatively more optimistic scenario, however, Kabeer admits that attempts to change existing power structures are ‘constantly undermined by various forms of unruly practices on the part of the more powerful sections of society’ (ibid.: 352). In any case, it appears that the reintroduction of formal democracy has not automatically resulted in a wider establishment of substantive citizenship rights. A significant part of the explanation for this takes us back to the power of the political parties. In other words, the nature and dynamics of citizenship is fundamentally shaped by the political hegemony resulting from the competition between the two parties vying for power. If therefore we consider citizenship-based or formal electoral practices, we arrive at a similar conclusion: that the form of democracy which has evolved in Bangladesh has become a core part of the ‘governance problem’ rather than a solution. This context leads us directly to the key question which underpins this chapter. Through an analysis of new empirical data, we want to explore the


I. Basu et al.

extent and nature of public support and commitment to democracy in Bangladesh. There are a number of converging reasons why this is question is hugely significant. First, we have relatively little information on public perceptions, expectations and demands for democracy in Bangladesh. The 2008 State of Governance Report, published by the Institute of Governance Studies, is one of the very few attempts to look at public attitudes about political legitimacy and performance in Bangladesh (IGS 2008a). Second, governance analysis and reforms have tended to focus almost exclusively on the supply measures of governance (for example, strengthening state institutions, capabilities and accountability) and given scant attention to demand measures (Wood and Landell Mills 2010). In part, as we have argued in our previous point, this is because we know so little about the demands for democracy, and know even less about how then to support these. The danger, however, with an exclusive focus on supply side measures, as Wood and Landell Mills rightly argue, is that it assumes that political elites are willing to engage with progressive change. This is clearly not a robust assumption in the case of Bangladesh. By looking at the demand side of governance, we therefore hope to better balance our understanding of governance dynamics. Third, by examining attitudes about political life and practices, we are able to look also at the extent to which particular governance arrangements are valued and likely to elicit public support and commitment. This is important because the so called ‘democratic project’ can translate into a myriad of institutional forms and arrangements, and different political settlements. Gauging the aspirations and preferences of citizens is a first and important step to understand which settlements are likely to gain more or less support; and by implication, acquire greater or weaker legitimacy in Bangladesh. Recently, our understanding of public opinions towards democracy has taken strides forward with the emergence of different ‘democracy barometers’. Work on the barometers aspires to rebalance political analysis by combining concerns about ‘what people think and aspire to’ with the more traditional research focus on the actions and behaviour of political elites and organisations. The barometers have also opened up a number of new substantive areas of research, including trying to understand what values or aspirations shape people’s support for democratic regimes. Bratton and Mattes 2001 (see also Mattes and Bratton 2007), for example, analysed data from the Afrobaramoter to examine the extent to which citizen demand for democracy was a principled affair (i.e. reflecting intrinsic support) or relied on the performance of governments (i.e. reflecting instrumental support). Their analysis led to two conclusions. First, that citizens’ support for democracy did not depend on a regime’s economic performance, suggesting that democracy was valued for what it is (i.e. as an end in its own right) rather than what it can do (i.e. as a means). Second, this intrinsic commitment to democracy, however, did not mean that governments could do what they wanted or act in a wholly unaccountable manner. Regime performance therefore remained important in terms of citizen support and legitimacy building. However and perhaps counter-intuitively, it was the ability of governments to provide ‘political goods’ such as the rule of law and the protection of freedom as opposed to

Governance, rights and demand for democracy 89 economic ones which influenced citizens’ assessment of performance. In another article, Bratton and colleagues take this idea further by testing it against a Globalbarometer which includes the Latinobarometer, the Afrobarometer, the Asian barometers and the Arab barometer (Chu et al. 2008). Not surprisingly, they found that people across the globe support democracy for quite different reasons. However across three of the regions, they found that popular approval for democracy depended more on the delivery of political goods than it did on economic ones. Latin America was the exception where national economic conditions were prioritised more than political considerations, but only marginally so. This kind of analysis is relevant to the context of Bangladesh and is central to the argument of this chapter. While there is evidence of general public support for democracy in Bangladesh (de Souza et al. 2008; IGS 2008a), we do not know if that support has any depth, or if that support could be equally extended to different types of democratic governance arrangements. Reflecting on the distinction between instrumental and intrinsic support for democracy therefore gives us an analytical lens to examine the nature of the political process in Bangladesh. Furthermore, while it is safe to assume that the survival of all democratic regimes depends to some extent on government performance, the priority given to ‘political goods’ as highlighted by the work of Bratton and colleagues, is significant. These political goods typically include the rule of law, the curtailing of corruption, the protection of rights and freedom, the ability of citizens to change government by lawful means and so forth – a list which mirrors many of the priorities included in good governance programmes. While successive Bangladesh governments have managed to deliver relatively strong economic goods, it is precisely their performance in delivering political goods which has cast a long and murky shadow over the current governance landscape.

Governance in Bangladesh between frustration and hope Since the reintroduction of electoral democracy in 1991, Bangladesh’s democratic journey has faced a number of challenges. Two of these challenges are particularly relevant to our discussion. The first challenge, alluded to in our introduction, relates to the dominance of the country’s two main political parties: the Awami League and the Bangladesh Nationalist Party; and in particular to the confrontational relations between them. The democratic experience in Bangladesh is one in which the party in power adopts a winner-takes-all approach and monopolises the state apparatus, while the losing party normally ends up boycotting parliament and pursuing a strategy of protest and disruption outside of parliament through hartals and street violence. The second challenge is the rising prominence of the religious right in mainstream politics and the place of political Islam in relation to Bangladesh’s commitments to secularism and religious pluralism. Below we look at each of these in more detail. Although many parties contest national elections in Bangladesh, only four (Awami League (AL), Bangladesh Nationalist Party (BNP), Jatiyo Party and the Jamaat Islami party) secure parliamentary seats; and only two (Awami League


I. Basu et al.

and the Bangladesh Nationalist Party) ever win the elections. Since the reintroduction of democracy in 1991 therefore, the two leading parties have moved from a position of prominence to one of dominance, creating effectively a twoparty system. This party system is remarkably stable and robust and yet, at the same time, is built on relations which are vitriolic, violent and devoid of trust. Commentators have explained the endemic violence and lack of accountability in Bangladesh’s democratic system as the result of a ‘politics of vengeance’ (Riaz 2004) and the ‘politics of distrust’ (Quadir 2010), where intense political rivalry and the opposition’s party’s de facto non-acceptance of the legitimacy of the party in power poses a constant threat to the democratic system itself. Political competition between the two parties is organised through patronclient relations that stretch across public institutions and the private sector and are also deeply embedded in Bangladeshi society, reaching right down to urban slums and rural areas (see Khan 2005; Kochanek 1993). Patronage is mobilised both vertically – through a chain of neta-karmi (leader-worker) relations based on kinship, localised and personalised ties; and horizontally – where the party in power uses the state machinery to distribute resources to maintain support groups in every sphere of public life (Islam 2002; Osman 2010). In governance terms, this results in the conflation of the interests of the state with the party that comes into power. Public institutions such as the civil service and the judiciary exhibit high degrees of politicisation with recruitment and promotions being determined on partisan terms (see IGS 2006, 2008a). Bangladesh is also one of the most centralised countries in the world (World Bank 2010), which means that the ruling party has a firm control over the operations of local government institutions, the dynamics of local politics and the disbursement of sizeable development funds. In line with a winner-takes-all form of politics, major development projects tend to go to areas controlled by the ruling party, while constituencies loyal to opposition parties are often excluded or at best under-resourced. Consequently, the party in opposition plays less of a constructive role in the democratic system and concerns itself mainly with contesting the legitimacy of the party in power and toppling its government (Quadir 2010). The culture of parliamentary boycotts illustrates this point well (Rahaman 2007). For example, the AL boycotted 156 out of 382 days of parliament working days during the BNP’s government of 1996–2001, whereas the BNP and its allies boycotted 222 of the 373 parliament working days over the tenure of the AL government of 2001–2006 (Rahaman 2007: 110). Over time, the politics of boycotting has been ramped up. The current opposition party (BNP) has boycotted the House for 316 days out of 370 days of sittings in the current Parliament – setting a new record of boycott in Bangladesh’s parliamentary history. The immediate result of boycotting is that political opposition is removed from Parliament and played out mostly in the streets through the medium of party-led hartals (strikes). Historically hartals played a crucial role in mobilising dissent and protest against colonial powers in Bangladesh and other parts of South Asia. Today however they are more associated with intimidation, violence and coercion; and inflict huge

Governance, rights and demand for democracy 91 costs – financial and otherwise – on the nation as a whole. In 2003, the UNDP undertook a nationwide opinion survey to assess public perceptions on the use of hartals. While 73 per cent of the 3,058 respondents believed that hartals were legitimate modes of protest as citizens have the right to dissent, 95 per cent believed they had a negative impact on the economy, 92 per cent claimed that they were not effective in pressurising the government to institute change, 51 per cent identified parliamentary debate as an alternative to hartals and 50 per cent believed politicians gained financially from hartals (UNDP 2005). Political violence routinely comes to a head in the run-up to national elections, and it was in the face of this challenge that the Thirteenth Constitutional Amendment in 1996 introduced the provision for a neutral caretaker government to oversee national elections in a fair and impartial manner, and the subsequent introduction of a new government. The system, often heralded as a unique feature of Bangladesh’s democracy, was in place for the parliamentary elections of 1996, 2001 and 2008. In the run-up to the ninth parliamentary election, fixed for 22 January 2007, the leading political parties refused to agree on a candidate to lead the caretaker government. This opened up space for the military to move in and allow the President to become the de facto leader of the caretaker government. Quickly, opposition parties objected and this led to increased violence and protest on the streets. In response, the President imposed a state of emergency on 11 January 2007. The very next day a military-backed caretaker government, with donor support, was appointed to run the country. This government remained in power for two years and set out to clean up politics and deal with corruption in order to create a level playing field for the next national election. These two years put Bangladesh’s democratic commitments to the test (see Ahmed 2011). Although initially there was some citizen support for the new caretaker government, over time this faded as a result of what appeared to be an authoritarian approach, which included arrests without charge and the heavy-handed influence of the army. This led to a popular demand for the restoration of democracy through national elections (Alamgir 2009). The Awami league government which came into power after a landslide victory in 2008 introduced a Fifteenth Amendment in June 2011, which abolished the system of caretaker government along with strict provisions against future military take-over of state power, stating that such acts would be treated as treachery and subversion. Paradoxically it was the Awami League which led the campaign to have the caretaker government introduced and, in line with the pendulum politics of Bangladesh, it is the Bangladesh Nationalist Party which has been leading the call for its reintroduction over the past few years! One of the general theses about Bangladesh – and which this chapter will directly engage with – is that although the quality of democracy may be wanting, people are steadfast in their commitment to democracy. This argument is often used, especially in the context of the events following 9/11 and the global discourse on the ‘war on terror’, to showcase Bangladesh as a Muslim majority country with a high regard for political freedom and secularism. Like democracy, secularism is one of the founding pillars of the first Bangladesh constitution.


I. Basu et al.

However, even the first government of Sheikh Mujibur Rahman quickly began to court Islamic political opponents who had been defeated by his political agenda based on ‘Secular Bengali Nationalism’. Shifts in ideological associations, religious or otherwise, have come to characterise successive political leaders in Bangladesh. This ideological oscillation has allowed for the emergence and growth of religious parties, most notably the Jamaat-e-Islami Bangladesh (JIB); and the establishment of Islam as the state religion in 1998. While these changes are often credited to the military regimes of Ziaur Rahman (1975–1981) and Mohammad Ershad (1982–1990) who allegedly courted JIB in order to earn legitimacy for military rule and win the approval of oil-rich Middle Eastern Islamic countries which were a major source of foreign aid (Hasan 2011), political leaders in the post-1990 democratic regime have also been busy making tactical use of religious ideologies in order to mobilise support and outflank their competitors. On 1 August 2013, the high court declared that the registration of the country’s most prolific Islamic party (Jamaat-e-Islami) was illegal and as a consequence banned it from participating in the next parliamentary elections. This news was greeted by some as a victory of Awami League over Jamaat-e-Islami, of secularism over religious-based parties. However, following the ruling, the Awami League publically distanced itself from any move to ban Jamaat-e-Islami, partly for fear of a potential backlash accusing them of being against religion. Religious parties in Bangladesh have always been electoral fringe players (Khondker 2010), but religion continues to influence government decisions and its social policy strategies (Asadullah and Chaudhury 2006), and to make itself more visible in social and public life (Devine and White 2012). Both religious and secular ideologies therefore are important for the main political parties, and leaders are content to tactically associate with both ideologies and distance themselves from any ‘bad press’ associated with either. More recently the question of militant Islam has also risen to prominence, with Bangladesh being seen by some as a home to more radical and fundamentalist elements (Riaz 2004; Seabrook 2002; Uddin 2006). There is no doubt that in Bangladesh, Islamic militancy is indeed a ‘complex web’ (Riaz 2008), capable of the kind of mobilisation that was implemented in 2005 to carry out the synchronised explosion of 500 bombs in 63 of the country’s 64 districts. It is however far less clear whether the existence of these groups constitutes a meaningful step towards religious nationalism as such; whether these groups have any real influence amongst the political elite; and whether they have significant citizen support. These remain important and, in some respects, open questions. The Fifteenth Amendment retained Islam as the state religion but also restored secularism and the freedom of religion in the Constitution. Political Islam in Bangladesh therefore seems to represent an ideological resource that taps into notions about Bangladeshi identity and which can be drawn upon tactically by the political elite rather than a measured approach to institute Islamic law or a public acceptance for religious fundamentalism in mainstream politics.

Governance, rights and demand for democracy 93

Methods The chapter draws on original data derived from a 2010 national household survey commissioned by the Daily Star, the largest circulating English daily newspaper in Bangladesh, and Nielson Bangladesh, a specialised information and measurement company. The 2010 survey followed up a Governance Barometer survey carried out by the Institute of Governance Studies in 2009 (IGS 2008a) and adopted a similar methodological approach. The overall aim of both surveys was to explore public perceptions on political life in Bangladesh. A total of 2,520 respondents took part in the 2010 survey from 44 districts covering all seven administrative divisions of Bangladesh. Of the respondents 30 per cent were urban residents and 70 per cent were rural. All respondents were adults (i.e. over 18 years of age), and the sample had an equal number of men and women. The survey asked a range of questions about attitudes towards democracy and governance more generally, mostly in the form of statements with which respondents could express agreement or disagreement, in some cases on a typical Likert scale (from ‘strongly disagree’ to ‘strongly agree’), in other cases a binary yes/no option. Our methodological approach was as follows. First, we examined descriptive statistics and cross-tabulations to identify different categories of respondents. As we discuss in the next section, crosstabulating questions relating to the position of Islam with respect to democracy produces some naively counterintuitive results, and we use these results as the basis of our categorisation. Using these categories, we proceed to examine the socio-economic determinants of support for democracy. In particular, as articulated above, we are interested in different justifications for democracy and how far these can be explained. We exploit a range of questions that elicit responses to a range of ethical and practical questions about democracy to construct indices using factor analysis. Factor analysis is based on the presumption that responses to such categorical questions are reflective of an underlying ‘latent’ continuous variable and uses the cross-correlation between different answers to these categorical questions to construct a factor loading to estimate the latent variable for each individual. We discuss the specific construction of the indices we use below. Regression analysis was performed using the Stata software with robust standard errors calculated using the survey sample frame.

Support for democracy: ambiguous democrats and ambiguous theocrats The starting point for our analysis is to assess the level of support citizens have for democracy in Bangladesh. We have evidence from previous research that levels of support have been traditionally high. Thus the 2001 World Values Survey found that as many as 98 per cent of respondents in Bangladesh approved of the democratic system, and this overwhelming endorsement was also evident in a more recent survey carried out under the State of Democracy in South Asia


I. Basu et al.

project (de Souza et al. 2008). It is important however to revisit the core question about support for democracy because, since the two referenced surveys, Bangladesh’s experience of democracy has had to face two years of militarybacked caretaker government (2007–2009) which was interpreted by some as a reversal of the democratic project (Diamond 2011). Furthermore following the 2007–2009 period, there have been continuous media reports and commentaries suggesting that the association of the democratic machinery – political parties in particular – with patronage, violence and intimidation has increased. The survey asked general questions about support for democracy in three different ways: • • •

How suitable is democracy for Bangladesh (on a five-point scale from ‘not at all suitable’ to ‘very suitable’)? What is the best form of government for Bangladesh (democracy, Islamic state, military government, or other)? Do you agree that democracy is preferable to any other kind of government (yes/no)?

In each case, support for democracy was overwhelming: 87 per cent agreed that democracy is preferable to any other kind of government, 83 per cent that democracy was the best form of government for Bangladesh, and 75 per cent that democracy was ‘suitable’ or ‘very suitable’ for Bangladesh. Despite therefore the apparent reversals or setbacks to the democratic project, levels of support for democracy remain remarkably high in Bangladesh. This is not a trivial finding given the recent state of politics in Bangladesh. Indeed, one might be tempted to ask the question: why in the face of so many political setbacks has the level of support for democracy remained so high in Bangladesh? The most obvious response to this is that Bangladesh citizens prefer democracy to other alternatives. We will examine this below. Here, however, it is important to acknowledge an equally plausible explanation which is rarely highlighted in the literature, namely that democracy in Bangladesh has become an established part of the polity and actually possesses quite resilient roots. This interpretation would certainly be consistent with a number of observations about Bangladesh, including high levels of political participation and a strong tradition of citizen activism expressed through various civil society initiatives as well as through forms of unruly politics (see Chapter 7 in this volume). Having established high levels of support for democracy, we need to explore further the aggregate statistics because they could camouflage the kinds of ambiguity we seek to explain in this chapter. Thus, for example, while the data clearly shows that people preferred democracy to any other kind of government and also felt that it was suitable for Bangladesh, there is an important difference in response rates. In fact, over 20 per cent of respondents felt that democracy was preferable to any other kind of government, but did not view it as suitable for Bangladesh. Rather than view these results as

Governance, rights and demand for democracy 95 evidence of inconsistent or unreliable responses, we see this as evidence of the kinds of ambiguous or pluralist attitudes towards democracy we want to explore in this chapter. A similar initially counter-intuitive finding emerges when we compare attitudes towards democracy and attitudes towards a secular/Islamic state. Bangladesh is the third largest Muslim country in the world and, as argued above, Islam is a prominent if not hyper-visible but contested feature of the current political landscape. One of the curious observations about the rise of ‘political Islam’ in Bangladesh is that it seems to have occurred without resulting in a concomitant erosion of the secular character of the polity. Hence the present government has gone out of its way to present the country as a secular democracy, and yet it retained Islam as the state religion in the recent constitutional amendment. At the same time as it proclaims its Islamic credentials, it has supported, if not engineered, some highly visible attacks on the most prominent of Islamic parties. The relationships therefore between secularism, Islam and democratic commitments in Bangladesh are complex, reflecting a fundamental and as yet unresolved tension around national identity. The question therefore of whether Bangladesh is a country of secular Bengalis or Muslim Bangladeshis may be old (Uddin 2006), but it continues to agitate and divide the country today. Simple cross-tabulations from the survey illustrate this point. Respondents were asked two questions that related directly to Bangladesh’s potential status as an Islamic state. The first asked simply how far respondents agreed with the statement that Bangladesh should be an Islamic State; the second asked respondents which form of government was most suitable for Bangladesh, with an Islamic State listed as one of five options which also included democracy and military rule. The point here is that the second question has ‘exclusive responses’ in that it forces respondents to select between an ‘Islamic State’ and ‘democracy’. The first question instead is non-exclusive. Table 4.1 cross-tabulates the proportion of respondents who agreed (or ‘strongly agreed’) that Bangladesh should be an Islamic State (i.e. the non-exclusive question) with the proportion who thought that an Islamic State was the most suitable form of government for Bangladesh (the exclusive option). For both questions, we have excluded nonMuslim responses. When presented with the non-exclusionary question, therefore, just over three-quarters of Muslim respondents agreed that Bangladesh Table 4.1 Perspectives on an Islamic State Per cent of total Muslim respondents

‘Bangladesh should be an Islamic state’

‘Islamic state is the best form of government for Bangladesh’

Disagree Agree Total




23.1 65.1 88.1

0.5 11.4 11.9

23.6 76.4 100.0


I. Basu et al.

should be an Islamic state (of these, over 50 per cent ‘strongly agreed’). However when asked to select between an Islamic State, democracy and other institutional options, almost nine in ten respondents selected against an Islamic State and the overwhelming majority selected ‘democracy’ instead. This leads to a number of important reflections. First, it is clear that respondents do not typically see a contradiction between an Islamic State and democracy (but neither do they see them as co-terminous). In part, this reflects the fact that citizens in Bangladesh are quite accustomed to the intermingling of religion and politics: the formal commitment to democracy has gone hand in hand with the identification of Islam as the state religion; some religious leaders have gone on to forge relatively successful careers in electoral politics; many religious parties have worked within the parameters of electoral democracy as opposed to calling for its rejection. For the many therefore, the rise of ‘political Islam’ does not pose the kind of direct threat to democracy that is often associated with other Islamicists’ movements (Nasr 2005). The analysis of Table 4.1 allows for a different interpretation which has also been suggested in work by de Souza et al. (2008). Their research looked at the experience of democracy across South Asia (Bangladesh, India, Nepal, Pakistan and Sri Lanka). Forty per cent of their respondents agreed that religious leaders rather than politicians should make the important decisions in the country. Response patterns across countries of course varied, with the majority support in Bangladesh and Pakistan. Rather than see this as an indication of citizens’ desire for a religious as opposed to democratic regime, the authors suggest that support for religion may be ‘an expression of unease with the received model of democracy and a desire to combine the existing model with other virtues’ (de Souza et al. 2008: 95). This is not only consistent with the analysis presented in Table 4.1, but resonates very strongly with perceptions that politics in Bangladesh is dysfunctional and dominated by money, corruption, goons and people with little background and training in formal politics (Khan 2003). Nonetheless, we have now two areas of political ambiguity: ambiguity over how preferable democracy is per se and, among Muslim respondents, ambiguity over its relationship to this issue of an Islamic State. By describing attitudes as ambiguous, we are not making normative judgments. The global ideal of democracy can be translated into many forms and our analysis so far suggests that the democratic process in Bangladesh, in both fragile and consolidated moments, offers new expressions of democracy permeated with distinctive and interesting meanings. Before examining why people might hold ambiguous views, it is worth examining who holds such views. In order to do so, we define two new analytical categories. We define ‘ambiguous democrats’ as those who expressed a preference for democracy over all other forms of government, but who also did not see it as suitable for Bangladesh. This group accounts for 20.4 per cent of responses. We define ‘ambiguous theocrats’ as those Muslims who thought that Bangladesh should be an Islamic state but who also expressed a preference for democracy over all other forms of government, including an Islamic state. This group

Governance, rights and demand for democracy 97 accounts for 7.8 per cent of Muslim responses (of whom around 70 per cent were also ‘ambiguous democrats’). We run simple probit regressions on these new variables with a range of standard social and demographic predictor variables: age, gender, level of education, income and urban residence. For ambiguous democrats, we also include religion. The results are presented in Table 4.2 and show clear and divergent trends. Basic demographic characteristics – age and gender – are not significant in explaining ambiguous attitudes, neither is education below the tertiary level. However, ambiguous democrats are significantly more likely to have tertiary education, tend to live in urban areas and tend to have higher incomes. What is striking about this finding is that other studies (de Souza et al. 2008; Shastri and Palshikar 2010) have found that people with these same characteristics (i.e. higher levels of education, urban residence and higher incomes) are far more likely to support democracy than those without. So what might drive support for democracy also influences people’s ‘ambiguity’ towards it – ambiguity again understood not in normative terms. Our analysis also suggests that ambiguous democrats are significantly less common among non-Muslims. This, we would argue, is less the effect of religion or some cultural trait and more to do with wider political economy considerations. With the rise of Islam in public life, the minority-majority dynamic has become more salient in Bangladesh, forcing many minority groups into significantly more vulnerable situations. Minority groups are therefore more likely to support democracy because of its association – at least in principle – with tolerance and respect. This is consistent with other findings in South Asia in which minorities value the ‘freedom qualities’ of democracy significantly more than their ‘majority’ co-nationals (Shastri and Palshikar 2010). If we then look at the data on ambiguous theocrats, we see that they are less likely to have tertiary education (although this result is not statistically significant), tend to live in rural areas and tend to have lower incomes. Put in more Table 4.2 Probit results for ‘ambiguous democrats’ and ‘ambiguous theocrats’ (1) Ambiguous democrats Age Male Primary completed Secondary completed Tertiary completed Urban residence Family income (000s) Non-Muslim Constant Observations

0.000481 (0.00241) 0.00711 (0.0655) −0.0530 (0.0785) 0.0271 (0.0945) 0.237** (0.113) 0.239*** (0.0896) 0.00831*** (0.00222) −0.282*** (0.104) −1.034*** (0.138) 2,949

(2) Ambiguous theocrats −0.00111 (0.00269) 0.164 (0.110) 0.0515 (0.0998) −0.0640 (0.119) −0.0405 (0.158) −0.191** (0.0954) −0.00797** (0.00388) −1.329*** (0.156) 2,672

Notes Standard errors in parentheses; asterisks indicate significance at level ***p < 0.01, **p < 0.05, *p < 0.1; equation (2) restricted to Muslim respondents only.


I. Basu et al.

straightforward terms, democratic ambiguity is primarily an urban middle-class phenomenon; theocratic ambiguity is more a phenomenon of the rural poor, although the strength of this association is weaker.

Ambiguity around governance: a question of being pragmatic In our introductory section, we discussed the different ways in which people might support democratic institutions, distinguishing between those with an intrinsic commitment to democracy from those with a more instrumental commitment (Bratton and Mattes 2001; Inglehart and Welzel 2003). Inspired by the fact that this distinction has yielded insightful results in a range of national and regional contexts, here we want to explore further the ambiguities we have found above and ask how far they can be explained by the ‘differences in support for democracy distinction’. This we contend is an important question for it helps move the debate beyond the question about whether Bangladesh citizens support democracy to ask a deeper question, with stronger diagnostic power, about the quality of support for democracy. The survey asked respondents to identify reasons they considered essential for democracy. Multiple responses were allowed. We divided the different responses into reasons which relate to the ‘pragmatic’ benefits of democracy and those that relate to the ‘ethical’ imperatives for democracy, the former relating to the more instrumental or overt support for democracy and the latter reflecting a more intrinsic or principled position. These were then coded to generate a ‘pragmatism’ and ‘ethicality’ index, and factor analysis was used to create a single index for each dimension. The factor loadings, along with the relevant statements, are reported in Table 4.3. Since factor analysis generates measures with a mean of zero and standard deviation of 1, it is not possible to ask whether people are more pragmatic than ethical in their support of democracy (or vice versa). We can, however, examine how far an individual’s commitment to democracy is lined to pragmatic or ethical reasons. This, as we will see below, gives surprising and strong results. The first question we can ask is descriptive: do our categories of ambiguous democrats and ambiguous theocrats differ significantly from the rest of the population on these two indices? In order to answer this, we run four t-tests to explore differences between each of our two groups and the rest of the population. As with the regressions reported above, we compare ambiguous theocrats only with other Muslims. The results of the t-tests are presented in Table 4.4. They show that ambiguous democrats are indeed significantly more ‘pragmatic’ in their attitude towards democracy than others, but do not differ significantly from the rest of the population in terms of ‘ethicality’. In contrast, ambiguous theocrats are significantly more ‘ethical’ in their attitude towards democracy than other Muslims, but do not differ significantly in terms of ‘pragmatics’. These differences are particularly noteworthy given the reasonably high level of correlation between the two indices, and can be taken as confirmation that the indices are indeed picking up separate attitudinal dimensions.

Elected representatives sit in parliament to make laws and hold the government to account

0.332 0.200

The courts are independent from political influence

The government is responding to meet the basic needs of people


Civil society groups can work freely without pressure or harassment.

Everyone has equal rights

0.287 0.445


We don’t have to bribe officials

Police and security services don’t harass people

We are free to speak our mind


Our basic needs are being fulfilled (education, health care, housing, clean water, sanitation)

Minorities are protected from attacks or harassment

Everyone is free to practise their own religion


We can complete our education without disruption

Media can report freely without fear or pressure

We are free to engage in political activities

We are free to criticise those in power

We can carry out business/economic activities without fear 0.423 of extortion

Government institutions and officials (court/police/ bureaucracy) serve the people

Ethicality index

Pragmatism index

Table 4.3 Factor loadings for indices










I. Basu et al.

Table 4.4 Ambiguity and justifications for democracy: t-test statistics

Ambiguous democrats Others t-stat. Ambiguous theocrats Other Muslims t-stat.

Pragmatic index

Ethical index

0.049 −0.011 1.800** 0.050 −0.005 1.066

0.013 0.001 0.360 0.162 −0.016 3.307***

Note Asterisks indicate significance at level ***p < 0.01, **p < 0.05, *p < 0.1.

What have we established so far, then? We have established that while overall levels of support for democracy in Bangladesh are apparently high, a sizeable proportion of Bangladeshis hold ‘ambiguous attitudes’ toward democracy – either because they report thinking that democracy is the best form of government but may not be suitable for Bangladesh (our ‘ambiguous democrats’), or because of ambivalence towards the relationship between democracy and the prospect of an Islamic State (our ‘ambiguous theocrats’). Further, we have established that these groups differ markedly from the rest of the population in terms of the reasons they hold to value democracy, but in different ways: ambiguous democrats have a significantly more pragmatic view of democracy, while ambiguous theocrats have a significantly more ethical view of democracy. Given this, the next logical question to ask is whether the same set of socioeconomic variables that predicted the different dimensions of ambiguity are also effective at explaining the variation in conceptualisation of democracy. In order to do so, we can replicate the regression analysis in Table 4.2, using the two indices of Table 4.3 as the dependent variables instead of the two categories of ambiguity. Because the indices are continuous rather than binary variables, we use an OLS regression instead of a probit. The results of this analysis are reported in Table 4.5. The significant predictor variables for the pragmatism index overlap somewhat with those for democratic ambiguity, although not completely. Education is therefore highly significant at all levels, with a larger effect at each higher stage of education, and urban residents have a significantly higher level of pragmatism in their conceptualisation of democracy than rural residents. In contrast to the democratic ambiguity measure, we do not find a significant impact of income on pragmatic attitudes, while we do find a strong, significant gender effect – men have significantly more pragmatic conceptualisations of democracy then women. In broad terms, therefore, our analysis points to the same conclusion that both ‘pragmatic’ and ‘ambiguous’ attitudes toward democracy are largely the domain of the urban, educated classes. In contrast, there is very little similarity between the predictor variables for ambiguous theocrats and the predictor variables for the ethicality index: ambiguous theocrats were found to predominate among poorer, rural respondents, while the predictor variables for the ethicality index are, instead, similar to those for the pragmatism index,

Governance, rights and demand for democracy 101 Table 4.5 OLS results for pragmatism index and ethicality index Variables

(3) Pragmatism index

(4) Ethicality index

Age Male Primary completed Secondary completed Tertiary completed Urban residence Family income (000s) Non-Muslim Constant Observations

0.00111 (0.00104) 0.0830*** (0.0274) 0.137*** (0.0364) 0.283*** (0.0443) 0.305*** (0.0569) 0.0769** (0.0308) 0.00132 (0.00116) 0.0390 (0.0456) –0.288*** (0.0538) 3,000

−6.88e–05 (0.00107) 0.0917*** (0.0280) 0.0314 (0.0372) 0.250*** (0.0452) 0.276*** (0.0581) −0.0339 (0.0315) 0.000770 (0.00118) 0.0757 (0.0466) −0.151*** (0.0549) 3,000

Notes Standard errors in parentheses; asterisks denote significance at levels ***p < 0.01, **p < 0.05, *p < 0.1.

finding higher levels among men and among the more educated (although the impact of education is less strong, and insignificant at the primary level). While the coefficient on urban residence is negative, as for the ambiguous theocrats, it is insignificant. We have established, then, clear evidence of a link between ambiguous attitudes towards democracy and a pragmatic conceptualisation of democracy, and that these appear to be more common characteristics of the urban, educated middle classes. While we observed a significantly higher level of ethical commitment to democracy among ambiguous theocrats, however, we have not been able to establish as strong a link between these phenomena in socio-economic terms. The implication of this is that we are missing some intervening variable(s) that mediates between conceptualisation of democracy and actual policy preferences. One plausible candidate here is attitudes towards secularism. In order to examine this, we construct another index using the same technique as above to measure respondents’ attitudes to secularism using questions that relate to religious institutions but not directly to democratic institutions or the nature of democracy.1 The correlation between this new index and our two previous indices is very low (0.019 for the pragmatism index; 0.013 for the ethicality index), confirming that attitudes towards secularism are indeed orthogonal to conceptualisations of democracy. In other words, both extreme secularists and extreme non-secularists can and do hold pragmatic and/or ethical conceptualisations of democracy. Armed with these three indices, we can now return to the basic question with which we began: is democracy always preferable to other forms of government? We run three probit regressions on this model (see Table 4.6): the first model only includes our socio-economic basket of variables, the second model includes the pragmatism and ethicality indices, and the third model includes the secularism index. The socio-economic control variables are fairly consistent across all the models. Education is positive and significant, with a larger coefficient at each

0.00356 (0.00237) 0.0886 (0.0610) 0.307*** (0.0771) 0.409*** (0.0971) 0.733*** (0.140) −0.191*** (0.0678) −0.00238 (0.00249) 0.210* (0.110) −0.145*** (0.0486) 0.0983** (0.0482) 0.560*** (0.132) 3,000

0.572*** (0.132) 3,000


0.00337 (0.00236) 0.0851 (0.0607) 0.290*** (0.0767) 0.391*** (0.0961) 0.711*** (0.138) −0.204*** (0.0675) −0.00254 (0.00247) 0.207* (0.109)

Notes Standard errors in parentheses ***p < 0.01, **p < 0.05, *p < 0.1.

Age Male Primary completed Secondary completed Tertiary completed Urban residence Family income (000s) Non-Muslim Pragmatism index Ethicality index Secularism index Constant Observations


Table 4.6 Probit regressions: is democracy always preferable?

0.00537** (0.00258) 0.0391 (0.0663) 0.354*** (0.0843) 0.352*** (0.104) 0.518*** (0.147) −0.278*** (0.0737) −0.00395 (0.00281) −0.134 (0.130) −0.146*** (0.0528) 0.0958* (0.0520) 0.571*** (0.0557) 0.632*** (0.144) 2,803


Governance, rights and demand for democracy 103 stage of education: higher levels of education correspond to higher commitment to democracy. Urban residence is consistently negative and significant. When we include the pragmatic and ethicality indices, we find them both significant but pointing in opposite directions – a finding all the more remarkable given the moderately high correlation between the indices. On the one hand, a pragmatic conceptualisation of democracy is associated with a lower commitment to democracy as always preferable to other forms of government. On the other hand, an ethical conceptualisation of democracy is associated with a higher commitment, although at a weaker level of significance. In the final model, however, when we include the secularism index, this is returned as very strongly and significantly correlated with a higher degree of commitment to democracy, and the ethicality index in this model is even weaker in significance, at the 10 per cent threshold. In one sense these findings are not that surprising, at least within the theoretical framework laid out in our first section. They confirm that the ‘demand’ for democracy can indeed be premised on different conceptualisations of democracy within the population, and they also suggest that the nature of the democracy debate in Bangladesh is not so much about the ‘rights’ and ‘freedoms’ encapsulated in the ethical conceptualisation of democracy (see Shastri and Palshikar 2010 for a similar observation on Bangladesh), but more about negotiating the relationship between the pragmatic benefits of democracy on the one hand and the position of Islam in relation to the state on the other hand. Unpacking support for democracy in this way reinforces the need to think more seriously about policy interventions to deepen commitments to democracy. If there is no onesize-fits-all model of democracy, it follows that there can be no one-size-fits-all policy intervention.

Conclusion In this chapter we have established that levels of support for democracy in Bangladesh remain exceptionally high. This is consistent with previous findings and reports from over the past 20 years. The fact that support remains high despite the recent turbulent political history of Bangladesh, is encouraging and bodes very well for the future consolidation of democracy in the country. Like other scholars, however, we need to be cautious when interpreting attitudinal responses to politics, especially those indicating high levels of support for democracy. Bangladeshis may overwhelmingly support democratic values and practices but this does not tell us much about the more important question of whether or not that support has any depth. De Souza et al.’s (2008) study into democracy in South Asia reported equally high levels of support for democracy in Bangladesh but then went on to argue very persuasively that the balance of democratic and non-democratic forces in Bangladesh was very precarious and lower than the South Asian average. In this chapter, therefore, we have tried to advance the debate on support for democracy. We accept that the questions which drive our analysis are timeless, and indeed responses will change over time. Although our analysis and findings are exploratory, we still contend that


I. Basu et al.

they are important because they resonate with recent historical experiences of democracy, generate important insights and point to new avenues of future and much needed research. Our first round of findings highlighted two interesting observations. First, although support for democracy is high, there is a significant number of ‘committed democrats’ who feel that democracy may not be appropriate for Bangladesh at the moment. Second, there is strong support for the idea of an Islamic State although this is significantly tempered when respondents are asked to choose between an Islamic state and a democratic form of government, with support for the latter being far more dominant. Do these two observations shed any light on arguments around the balance of democratic and non-democratic forces in Bangladesh? Our response to this question builds on observations outlined in the paragraph above. Democracy in Bangladesh, we argue, has strong and tested roots and therefore talk of its demise or rejection is perhaps exaggerated. However, we accept that the success of democracy is by no means a foregone conclusion and based on this affirmation, we propose an alternative interpretation of our observations. We would like to argue that citizens’ endorsement of an Islamic State (but, note, not of a theocracy) and their assessment that democracy may not be suitable for Bangladesh right now both indicate a desire not for a non-democratic regime but for a better democratic regime or an improved democratic model. In other words, both observations reflect a dissatisfaction with the democratic model as it stands but not necessarily with democracy per se. Furthermore, we would argue that the outcome of this critical reflection and questioning of specific democratic regimes which will give shape to the formation of future support for democracy in Bangladesh. We built our analysis into the quality or depth of support for democracy around a distinction, found in the literature, between intrinsic and instrumental support. The former represents a commitment to democracy as an end in itself and is presented in our analysis as an ethical disposition to democracy. The latter instead sees democracy more as a means to an end, and is presented here as a pragmatic disposition. The crucial difference between the two is that the latter is conditional; and therefore in an important respect, can be withdrawn or increased on a temporary or permanent basis. The distinction is also useful because it taps into governance and political reform dynamics. Broadly speaking the intrinsic or ethical position reflects a normative commitment to democracy and therefore resonates with the language of rights, equality, voice and representation. The pragmatic approach to democracy instead has a strong institutional focus and highlights key areas of regime performance such as independence of the judiciary, welfare provision, levels of corruption and so forth. Democracy in this context is a means to achieve positive outcomes in these different areas. The good governance agenda of course covers both rights and institutional reform but, as we argued in our introduction, in Bangladesh the emphasis to date has been on supporting institutional reform (i.e. the supply side of governance reforms). The broad pattern which emerges from our findings is that the pragmatic disposition to democracy tends to find resonance among urban, educated and

Governance, rights and demand for democracy 105 middle classes. Perhaps not surprisingly, this perspective is associated with a commitment to democracy which is lower than other respondents. The ethical disposition is less clearly associated with particular socio-economic groups, but significantly more associated with a commitment to democracy. This appears to reflect religious differences and, possibly, preferences regarding an Islamic State. At one level, these findings can be seen as consistent with our observations above about the questioning of specific models of democracy. For some the litmus test may be about regime performance (pragmatists), while for others it could be a question of values or principles (the ethical disposition). Two further reflections, however, are important. First of all, our findings confirm that higher levels of education translate into a higher commitment to democracy and the effect is greater at each higher stage of education. We found this to be true in the case for both pragmatic and ethical dispositions. The significance of education is consistent with findings from across South Asia where graduates were found to be seven times more likely to support democracy than non-literate people (de Souza et al. 2008). It is therefore reasonable to suggest that access to education, which increases with successive generations, will continue to have a strong impact on the way democracy develops in Bangladesh, at least in the short and medium terms. Second, the high levels of association between urban middle classes and both the pragmatic disposition to democracy and the position of ambiguous democrat (i.e. committed democrat who prefers democracy but does not see it as suitable for Bangladesh) is an important finding. Bangladesh has a rapidly growing middle class made up of a relatively young demographic with high disposable incomes. It is also a constituency about which we know very little, in particular its political aspirations and demands. A strong and plausible hypothesis to emerge from our analysis is that the urban middle classes tend to focus on the institutional benefits which democracy might deliver (the pragmatist disposition); and in order to secure these, certain aspects or qualities of the western liberal model of democracy may be sacrificed or relaxed (the ambiguous democrat). Put in starker terms: if democracy is unable to deliver and respond to the demands and interests of the urban middle classes, support may well be withheld or withdrawn. It remains a moot point what these interests may be. By looking at pragmatic and ethical dispositions to democracy we are able to conclude that the ‘demand’ for democracy can indeed be premised on different conceptualisations of democracy within the population. This opens up opportunities and risks in terms of governance interventions, as well as inevitable tradeoffs. Our analysis suggests that institution-based approaches to governance reform are more likely to win the support of the educated urban middle classes. This however may come at the cost of a lower or at least more relaxed commitment to democracy. Furthermore while we accept the need for middle-class buy-in for future social change in Bangladesh (Devine and Wood 2009), there is a real risk that this strategy can put severe constraints on more radical reform in terms of wider redistributive effects. On the other hand, while there is some evidence that a rights-based approach to democracy has a wide-support base and


I. Basu et al.

is linked to a higher commitment to democracy, equally important questions arise as to whether this approach can attract longer-term middle-class support and whether it opens up more space for religion to play an even more prominent role in politics. The idea of democracy has taken firm root in Bangladesh and the analysis presented here indicates that Bangladesh has also infused the idea of democracy with distinctive and transformative meanings and experiences. This iteration will continue in the future making it difficult to predict what form of democracy will evolve. It is therefore crucial that we pay more attention to the dynamics and relevance of support for democracy in Bangladesh. Pragmatic and ethical dispositions to democracy do not exist in a vacuum, but are both constitutive and reflective of wider constellations of values which can be found in the polity. Identifying and understanding the workings of these constellations and their interactions is, we would argue, the single most important governance challenge facing policy makers, practitioners and researchers alike.

Note 1 The questions used are: All major decisions should be taken by religious leaders (agree/ disagree); Do you consider yourself a religious person? (yes/no); Are you involved in a mosque/temple/church committee etc.? (yes/no); Are religious parties acceptable? (yes/ no). The scale was inverted to make more secular responses positive and less secular negative.


Deconstructing the natural state? Is there room for de Tocqueville or only Gramsci in Bangladesh Geof Wood

Introduction The discussion about civil society in countries like Bangladesh is a little surreal. Are we complicit in seeking to analyse institutions in a society through false lenses, through constructs which have been developed in the context of other societies, at other times, far away? Is the concept of civil society the appropriate device for observing civil society in a society where the concept is not embedded in its values and political culture? One response to such questions is that we need to develop analytic concepts which are relative to specific conditions and that, therefore, we need to be relativist. Thus we might refer to the notion of the Ummah as blurring a distinction between state and civil society, enabling theocratic rather than democratic government and governance. However, an opposing response is that such relativism removes all prospect of judgement about what is good for citizens, so that if the Taliban are the true representatives of Afghan values and culture, then it does not matter that they oppress women. That is not a good position or place to be in, analytically. Thus we need some comparative framework which both captures the intrinsic or ontological essence of institutional practice, while enabling us to situate a society against some universal principles of human rights and entitlements. Our suggestion is that the recent book Violence and Social Orders by North, Wallis and Weingast (2009) is seminal in providing this comparative framework which facilitates sensitive contextual analysis alongside the possibility of judgement. Although regarded by us as seminal, Violence and Social Orders is of course in a tradition of discourse about states subverting and incorporating organisations which appear as if they are part of civil society. Indeed that appearance is part of political illusion – a landscape of organisations whose legitimacy requires at least the appearance of independence even if they are in continuous and dependent negotiation and compromise with regime power-holders. In the UK, for example, during the 1960s’ Wilson Labour years and the ‘social contract’, apparently independent trades unions had a beer and sandwiches intimacy with Downing Street. Union leaders were in highly ambivalent positions: back-room deals potentially undermining their legitimacy to their own members, while needing the occasional strike or meaningful threat of one in order to demonstrate


G. Wood

to government that they retained a power to disrupt. These relationships indicated the idea of the corporate state. But we have earlier examples of the social contract in practice: the pre-World War II fascist regimes in Italy and later Germany comprised populist parties penetrating the entire organisational landscape of the society: professional bodies; youth movements; workers’ associations; the press; charities. And of course the pressures of that war itself suspended politics in the UK as key aspects of resource allocation were traded between collaborating leaders of the respective ‘estates’.1 And regimes apparently as far apart as the USSR and Japan both pervaded the putative organisations of civil society before, during and, for the USSR, after that war. Indeed there seems to be a persistent theme that periods of crisis collapse any meaningful distinction between state and society, as external ‘others’ are confronted, requiring unity at home. Post-war reconstruction in the USSR and Eastern Europe sustained strongly that ideology of unity above all else, embodied in deliberate reworkings of nationalism. Communist regimes in China and later South-East Asia reproduced these forms of excessive state penetration over society. And during the same period, authoritarian, military-bureaucratic regimes across Latin America (O’Donnell 1986) displayed strong features of corporatism and syndicalism, perhaps most famously under Perón in Argentina. Thus one might conclude from all of this that Gramsci offered the more accurate account of civil society as a supporter of the state, performing a range of politically aggregating functions under the guise of separation while being thoroughly implicated in the state’s project. By contrast, the tradition of de Tocqueville arises from a more specific experience associated with the more libertarian pursuits of post-Revolutionary France, its democratic lessons for the newly independent United States of America and the history of hostility and suspicion towards monarchs and their lackeys in the nations of the United Kingdom. And yet, it is the liberal-democratic-pluralist model derived from these rather more specific historical experiences which have become the universal benchmark not only by which societies are judged but also by how they are understood and analysed. In other words, societies are being analysed for what they are not instead of for what they are. It has been difficult to disentangle the normative from the descriptive as a result, which is why Violence and Social Orders is seen by us as so important. It offers some prospect of analytical connection between Gramsci and de Tocqeville via the notion of doorstep conditions between limited and open access states. We will come to their schema presently.

Significance of the natural state for civil society Why do we pose the question in the title of this chapter? The term ‘natural state’ is a reference to North et al., characterising limited access states as natural states, subdivided between fragile, basic and mature. We think it is rather easy to assign Bangladesh to the natural, limited access category, though whether it is fragile, basic or mature is harder to determine, reminding us that typologies are heuristic and not necessarily to be applied too literally, although North et al. try to do so

Deconstructing the natural state? 109 for their example countries. A key premise of the natural state is that organisations are not free of the state, whether represented by key overbearing individual rulers/personalities, or by a more generalised social persona or political class – or more likely a mix of both. Not to be free of the state seriously undermines the idea of civil society in its de Tocquevillian sense, removing a whole range of ‘between elections’ democratic functions associated with accountability as well as policy refinement. It is our contention that in Bangladesh most of those organisations to which can be applied the collective term ‘civil society’ are not free of the state. However, the explanations for the actual relationships between civil and state organisations are multi-layered and not just attributable to a syndicalist conspiracy by ruling control freaks. Thus Bangladesh is not alone. The structural weakness of civil society, evidenced by its unfreedom, is shared with other societies at similar points of transition from relatively recent agrarian pasts with complex forms of indigenous social capital and institutional practices. The organisations of putative civil society do not grow in a vacuum. By posing the question in this way with an expectation of a complex answer, we are placing the aspirations for civil society in Bangladesh within a context of governance, rights and representation – in other words a normative agenda of citizenship. Although we have used the prison metaphor previously (Wood 2000) to describe the entrapment of most people in Bangladesh within systems of non-transparent, personalised transactions, there is nevertheless an active debate and deliberate activity by escapees and their international sponsors (i.e. the aid community) to advance the presence of a meaningful civil society in a variety of forms. So we are not observing a hegemonic, unchallenged suppression of rights, but rather a struggle albeit a rather uneven one. This is why the central question is worth asking. There is a dynamic interactive process, a dialectic, of incorporation as well as attempts to break free, indicating not a fixed, locked-in system but an ongoing contestation. We certainly agree with North et al. that the freedom of non-state organisations from the state is a key doorstep condition for an open access society. However, our argument cannot be a simple mechanistic one. The apparently straightforward idea of a ‘freedom indicator’2 oversimplifies the relative autonomy of the estates in any society. The ‘freedom’ formulation artificially disentangles formal roles from more complex underlying interests, motivations and relationships. An anthropologist would refer to these as cross-cutting ties, meaning that social players in apparently separate organisations with formally separate, contesting missions are also bound together by family and class identities in other equally important associations. As a result behaviour in both domains is moderated. Indeed some agendas never surface as they are informally pre-negotiated off the table, akin to Lukes’s second dimension of power (1974). Another way to capture this is to follow North’s earlier distinction between institution and organisation (1990) and repeated in Violence and Social Orders (North et al. 2009, p. 15). Thus institutions are ‘rules of the game’ (North 1990) and the ‘patterns of interaction that govern and constrain the relationships of individuals’. ‘Institutions include formal rules, written laws, formal social conventions, informal norms of behaviour, and shared beliefs about the world.’


G. Wood

Organisations are more purposive counterparts of institutions, comprising specific groups of individuals deliberately associating with each other to pursue common as well as individual goals. There is some parallel to the Tonnies distinction between Gemeinschaft and Gesellschaft,3 except that we do not want to overemphasise the contrast between small scale and large scale embedded in the Tonnies’ contrast. Perhaps there is also a ‘base’ ‘superstructure’ parallel from Marx, though he was more intent on capturing a causal relationship between underlying economic structure or infrastructure (i.e. base) and its management and representation through politics and culture.

The concept of permeability Let us return to cross-cutting ties and Bangladesh. In previous writing (LandellMills 2002; Wood 2000) governance issues in Bangladesh have been analysed in terms of ‘deep structures’ and institutional responsibility domains. In particular we have pursued the proposition that the formal domains of state, market, community (read: organised civil society and inherent social institutions) and household are permeable, rather than representing separate rules of the game. This permeability proposition is absolutely central to our overall argument. The liberal bourgeois myth of separated domains enabling the compensation and regulation of behaviour between them is fundamentally undermined when permeability exists. This is the conceptual challenge of the Ummah (Wood 2009). Under conditions such as for Bangladesh, this permeability is treated from the De Tocqueville perspective as negative in the sense of mutual contamination between domains, thus denying a rights-based existence for all the population. It is this proposition which led to the ‘prison’ analogy. The proposition has direct significance for understanding welfare regimes and the limitations of the Polanyian compensation schema.4

Apparent and hidden behaviour By deploying a distinction between institutions and organisations, and the language of deep structures and cross-cutting ties, we can begin to understand a contrast between apparent and hidden behaviour.5 Apparent behaviour occurs more completely in the organisational domain. Socially constructed actors interact with each other through their formally acknowledged roles. In this sense they are definitely ‘acting’. It is a structured drama characterised by role-specific language and costume. Indeed it is a world of cases rather than stories.6 Hidden behaviour refers to the institutional domain where the basic rules of the game apply. This is a world of diffuse,7 multiplex ties being activated simultaneously, a world of favours and obligations entailing multi-period games. This is an oral underworld, relying upon memory and triangulation. It is a world which demands the tacit understanding of its players and a sense of continuous membership to manage compliance and indeed complicity. Failing to honour expected obligations is risky and incurs sanctions and/or ostracism. Exits are hard to find. It is all embracing. Hidden behaviour is the real purpose of apparent behaviour.

Deconstructing the natural state? 111 It is important, at this point, to remind ourselves that the lived combination of apparent and hidden behaviour is a universal social phenomenon. It is the essential epistemology of good sociology. And, in that sense, the distinction is neither new nor in itself particularly interesting. Rather it is the relationship between the two domains in different limited and open access state conditions which matters. Thus we might say that in open access societies the hidden realms of our behaviour are more suppressed, or even repressed. This suppression may be reinforced by the lower interconnected density of primordial relationships, so that actors are freer to act their formal roles with more anonymity and be primarily identified with them. In other words there is less permeability. The domains of apparent and hidden are kept more easily apart. And because they are so, any permeable transgression is widely perceived as inappropriately self-serving and corrupt. Case law is paramount, the room for discretion tiny. But in limited access societies, the hidden behaviour bubbles more obviously to the surface, having a strong influence upon the acting out of formal, apparent roles. The expectations arising from the informal or ‘natural’ rules of the game gain currency in the formal drama. The space between front and back collapses. Yet the maintenance of a respectable veneer remains an important device in the capture of rents, providing a thin layer of legitimacy to essentially personal motives. The transactions between the hidden and apparent domains have other dimensions also. The competition for the formal roles can be intense precisely due to the veneer of legitimacy for private rent seeking and the consequent distribution of these rents among kin and friends. But prices have to be paid to reach the ‘audition’ and subsequently win the formal role. Those prices are paid in the hidden domain with an expectation of payback.

‘Natural, limited access states’ applied to Bangladesh Before considering the case of Bangladesh more specifically, the core argument in Violence and Social Orders should be outlined since it offers a highly recognisable framework for a deep structures approach to governance. For North et al. (2009, p. 72): The progression of natural states involves increasingly more complex societies, requiring increasingly complex institutions that support more complex organisations. In all natural states, economics is politics by other means: economic and political systems are closely enmeshed, along with religious, military and educational systems. This is their core epistemology reflecting their overall perspective. It has resonance with my arguments about permeability between the domain elements of the ‘institutional responsibility’ square (state, market, community and household). Adapting my previous arguments to theirs, negative permeability (the dysfunctional blurring of institutional boundaries) is most strong under fragile natural state conditions, but also significant in any movement from basic to mature


G. Wood

natural states. In this progression, although organisations become more codified, ‘the process initially occurs simultaneously in the public and private sector; indeed it is a primary reason that governments in most limited access societies appear so corrupt to observers from open access societies’ (p. 73). This is particularly a function of the personal identity of elites: ‘organisations in fragile natural states are usually tied to powerful individuals’ (p 73). But ‘as societies move towards basic natural states, these identities become less associated with specific individuals and more with social personas that become associated with powerful organisations’ (my emphasis, p. 73). Thus my earlier permeability argument8 is supported by their statement that: ‘Most important basic natural state organisations are closely associated with the (private) [sic] individual identities of the elites who inhabit them. These organisations span the boundary of public and private, personal and social’ (p. 73). Applying this insight to Bangladesh helps to establish the transitional phenomena which is familiar to observers, namely the melange of: powerful individuals; the activities of an exclusive political class (social persona); and the intimate connections between this class and elite strata in other, private and nongovernment sectors. C. Wright Mills talked of ‘circulating elites’, and with antiincumbent voting in the ‘winner takes all’ politics of Bangladesh, there is a revolving door between public and private roles. Ethnographically, this circulation occurs among members of large, sprawling kinship groups who at any point in time straddle the public and private, even if the actual personages rotate between the domains. Such kin, even several times removed, recognise a loose structure of mutual obligations and favours between each other, played as repeated, multi-period games. Such underpinning keeps the political class intact, even when the more extreme ‘tribal’ identities are seeking to de-legitimise the history and political claims of each other in absolutist terms. While such crosscutting ties might be a sign of maturity, there are, however, no guarantees of progression. Thus: No teleology pushes states through the progression from fragile to basic to mature states. The dynamics of natural states are the dynamics of the dominant coalition, frequently renegotiating and shifting in response to changing conditions. If adjustments lead to more power and rents based on personal identity, institutions become simpler and organisations less sophisticated, and the society moves to the fragile end of the progression of natural states. If adjustments lead to more power based on durable agreements, institutions become more complex and organisations more sophisticated, and societies move toward the mature end of the progression. No compelling logic moves states in either direction. (p. 73) This quotation contains several issues for us involving the tension between personality and structure. The progression towards mature natural states with limited access is partly a problem of ‘constraining personality’ (p. 74) and partly

Deconstructing the natural state? 113 the causally related issue of bringing about more specialist organisations which are both independent of the whim of the ruler while organically interdependent.9 In other words, are the interests of the coalitions of elites, the political class, to be achieved through the patronage of rulers as in fragile or even basic natural states or through a codified allocation of rents and privileges which are sufficiently inclusive of those voted out of office as to avoid disorder, coups and civil wars?10 And given that we are observing an extended patronage system down to grassroots, micro-level access, the issue of personal identity politics is a reference to a system not just to leaders at the top. And if we face the presence of personality from top to bottom of this system, then the issue of succession has to be resolved at all levels too before a natural state even reaches the ‘basic’ mode. While North et al. are substantially aiding the analysis of Bangladesh with this framework, there appears to be a further complication not really considered by them. Not all institutional rules and organisational forms can be classified in one single type (i.e. fragile, basic or mature natural, limited access states – let alone open access ones) as they do not move perfectly in step with each other for an excolonial society like Bangladesh. This is where the Alavi thesis (1972) also assists our thinking. The colonial process of establishing specialist organisations, relatively independent of any ruling coalition of elites, such as the legal system, educational entities, taxation and revenue authorities and so on, sets up an overdeveloped bureaucracy, with the appearance of Weberian principles, while democratic politics11 is left trailing as competing propertied classes (agrarian, commercial urban comprador and metropolitan) vie with each other for sectoral advantage while also seeking their collective interest in the protection of the principle of rents and privileges. Thus the veneer of formal sophistication may lie in some institutional arenas (i.e. Public Service Commission or the Establishments Division) due to colonial inheritance, while the rules of political competition remain crude, immature and highly fragile. In other words, there is co-existence of form.

Social origins of present deep structures From the inception of liberated Bangladesh, there has been a continuity of problematic practices and behaviour within government among politicians and public officials and between them and the rest of the society, without effective regulation in governance terms either from apparatuses of government or from the civil society. These practices have been analysed by various writers over the last decade especially, including myself (Wood 2000) and Landell-Mills (2002; from a review of Public Institutional Performance in Bangladesh in 1999). More recently BRAC’s Institute of Governance Studies (IGS) has provided similar and in-depth analysis in a sequence of State of Governance Reports (IGS 2006, 2007 and 2008a). Aspects of these reports have been informed by the research of Transparency International Bangladesh (TIB). The IGS analysis offers a deeper reflection on causes of poor practice, than the more factual, descriptive information from TIB and elsewhere. SOG 2008 actually appeared in 2009 and thus embraces behaviour of the present government actors as well as previous ones.


G. Wood

South Asian, ex-British and recently agrarian societies, are characterised by severe socio-cultural as well as economic inequalities. There is considerable deference to those above you on the social scale, however expressed – caste, bangsho12 and so on. Within this elite-mass, agrarian inheritance, most people thus have a deep rooted fatalism and de facto tolerance of non-transparent, nonaccountable, undemocratic practices. They have expected their leaders to behave as zamindars13 with unchallenged authority. These feudal legacies have now evolved via the arrival of political parties into a ‘mastaanisation’14 of the countryside and the society at large, an extension of urban mafia-like institutions, indicative of rapidly evolving opportunities,15 mediating between people’s immediate needs and imperfect markets and states in which these needy ‘clients’ have no power or influence. The tolerance of these institutionalised practices thus reflects socio-cultural history, but now reinforced by more superficial threats and inducements. Although the liberation narrative for Bangladesh juxtaposed an incipient nation of small peasants against the landlord and bureaucratic-military elites of the Punjab, the East Bengal delta was more complex in its agrarian structure (Wood 1981) and by no means as homogenous as the narrative implied. Of course that narrative was politically heuristic and thus a deliberate construction within a fission and fusion framework of political identity. The reality was that while there were key pockets of minifundist, small peasant farming as in the Dhaka–Comilla belt, elsewhere there were more significant landlords, sharecropping tenants and effectively landless labour. And even within the minifundist region, the archetype of Awami League mythology, there were key indicators of inequality, dependency and patron-clientelism (Huq 1976). And of course, East Bengal/Pakistan in 1971 was very rural, unusually so even for South Asia. There was a tiny urban elite, mainly in Dhaka with landed property as the basis for bureaucratic, education, trading and manufacturing roles – an essentially comprador elite incorporated into the East Indian Company and British colonialism (see IGS forthcoming). This admittedly over-simplified characterisation reveals to us that only a single generation ago, Bangladesh was agrarian and precapitalist in its mode of production and exchange. Power was exercised essentially through family and patron-client forms of exchange through concentric circles of moral proximity and intimacy, and through diffused, multidimensional, multi-period games. Clearly much theory and empirical observation is collapsed and condensed into this summary account of power. But this conceptual formulation of agrarian forms of power in Bangladesh can be traced through to contemporary socio-economic structures – what I have previously referred to as ‘deep structures’.

The surreal context for civil society This context provides us with a key contradiction for Bangladesh, and for other societies like it in South Asia and beyond: namely the gap between the institutions which effectively shape the society and the formal organisations which

Deconstructing the natural state? 115 apparently populate the landscape. Thus, for example, a political party may exist as a formal organisation competing in elections, providing governments and the occupation of the Jatiya Sangsad, but its institutional profile resembles a traditional dal comprising leaders and followers bound together by highly personalised, multiplex tiers of obligation, favour and dependence as well as interdependence. This contradiction, sometimes understood as the gap between rhetoric and reality, is well known in Bangladesh among its political actors and among its governance-oriented donors. It provides the setting for a complex charade of smoke, mirrors and shadows. Thus the political class among Bangladeshis and donors in Bangladesh live in a dual world of implicit institutional practices and explicit organisational pretension. It is interesting that the donors collude in this duality, based either upon naivety or upon the false idea that organisational reform might change underlying institutional principles – a kind of reverse causation. So we might expect political parties to be unable to separate themselves from prevailing institutional cultures since they are in the business of securing and exercising power by whatever means, even including violence as long as it is deniable. But let us consider a stronger test of a ‘reality/rhetoric gap’ proposition – the formally created non-governmental organisations, which potentially indicate a transition to open access society. They are supposed to be breaking away from business as usual. Their existence is supposed to be, in itself, a critique of non-transparent, hidden forms of informal, personalised forms of power. They are purposive, deliberate associations – a standard bearer for Gesellschaft over Gemeinschaft. They are supposed to represent ‘escape’. But for them too institutional norms overcome organisational objectives. Leaderships are highly personalised, recruitment takes loyalty strongly into consideration, policy is centrally directed despite the existence of formal committees, and there is exaggerated respect and deference to key, founder directors (i.e. specific individuals, not just social persona). In moments of dispute over succession, there is even the mobilisation of violence and the bribing of third-party officials. None of this should surprise us. Prevailing societal institutions set the tone for organisational cultures so that associations apparently created for deliberate missions quickly resemble patron-client communities like the rest of the society. The internal organisation of the major business houses is similar (see Fukuyama’s Trust (1995), and the analysis of the chaebol). Thus even the NGO ‘test’ indicates basic or even possibly fragile natural state conditions. The interesting conclusion is not so much the institutional influence over the organisational, but that this further layer of interaction exists: i.e. dealing with organisations as if they were not influenced by their institutional settings. Thus political parties are treated as real things which can be improved towards democracy once informed about it, as if lack of knowledge about ideal type political party functioning was the problem. And NGOs can be supported as if they are mission-based, objectives-oriented, rationally constructed organisations rather than communities of personally attached individuals. And the ‘game’ is not only played by donors but internally too, as a kind of sham or veneer.


G. Wood

The demand side role of civil society From this ‘institutions and deep structures’ perspective, it is clear that any escape route to engineer some kind of transformation from a limited access state to an open access one, and at the same time hold the design and delivery of essential rights, entitlements and services to account requires a demand side stance through social mobilisation and advocacy However, the key issue for a demand side driver for good governance is risk and its corollary: the security of agency. While there are different kinds of risk, the main focus here is upon the distribution of risk between political actors from within and outside the state: i.e. from those benefiting in some way from limited access conditions and those, hitherto politically excluded classes, seeking to open up access. There is a complex principal-agent problem to unravel. The ‘principal’ is this discussion are either actual local escapees (I ignore donors as ultimately insignificant in this discourse) or a more diffuse or latent sentiment that state performance should be better and characterised by open access conditions if only the free-rider problem could be overcome. Thus there is a mass prisoner’s dilemma dimension to extending the escape party! And then there are the agents: organisations and individuals. There have been some spectacular failures of organisational agents in recent years because they have either crossed a boundary into politics and/or because leaders have been financially corrupt (including the inappropriate diversion of donor funds into direct political mobilisation and electoral ambitions). The boundary between social mobilisation for advocacy and such organisations either entering or being captured by the political domain is a difficult one for ‘principals’ to police. Both contending political parties, or factions, have been assiduous in co-opting other movements (especially youth and student ones, but also unions) into exclusive agreements as front organisations.16 This resembles, as noted earlier, the Gramscian critique of civil society as providing support for authoritative regimes, i.e. as an element of fascist political theory, rather than the de Tocqueville version of free standing independence.17 And other large organisations outside the state (NGOs) have chosen to minimise the risk of boundary crossing or political attack for not accepting incorporation by reducing their advocacy roles either to below the parapet or to technical rather than rights arenas. In this way, they become ineffectual in a governance sense of either access transformation or accountability. The press and other media organisations tread this fine line between co-option and the risks attendant upon being independent.

Holistic and marginal risk: sustaining the demand side Of course the governance agenda is about moving people from the condition of dependent security to autonomous, rights-based security, from limited to open access conditions. But from the preceding analysis, we can see that the demand side of governance incurs holistic not just marginal risk for most people, especially for the poor. They are not in a social situation where they

Deconstructing the natural state? 117 can compartmentalise their lives – struggling in one bit of it, while kowtowing through necessity in another bit. This is the meaning of Naomi Hossain’s ‘rude accountability’ dilemma (Hossain 2010): outbursts of protests with rough language capable of being dismissed as affective rather than affectively neutral behaviour18 with no significance beyond the immediate moment. Thus agents, as organisations or individuals, need to be continuously supported by their principals – that is to have security of agency as the incentive to confront holistic risk, and to move from rude to sustained accountability either short or long loop.19 But what can that continuous and predictable support mean in a closed limited access society (prison) with interlocked, covariant relationships inside the perimeter fence? It is too easy to resort to international donors as the ultimate principals in a principal-agent cascade. Donors and international human and political rights organisations can and should contribute to predictability of support as long as they are prepared to shift from projectised time horizons to longer-term ones. That can be important signalling. But the necessity of long-term predictability of support by an NGO for people being locally mobilised into risky struggle and confrontation to achieve accountability means that the NGOs themselves need predictable support in some way over long periods. They need to be independent and free of the state and its key actors. Sustainability, via independence and freedom, is thus a major issue in the mitigation of some aspects of risk for both groups of poor people and the NGO activists who support them (since they are at risk too). NGO workers are vulnerable to attack, along with their families. That sustainability requires a broader ownership of ‘principal’ status from among the emerging middle classes of the country who have some resources and, perhaps via the spread of direct tax nets, an interest in more active citizenship. India, for example, has moved way beyond donor dependency (if it ever had it) in this respect.

The PMUS20 case To support the theoretical argument and the contextual analysis offered above for Bangladesh, we offer here a case study which in our experience is quite typical of the wider picture. The case of PMUS refers to a large NGO created in the mid-1970s and which had three decades of expansion before entering a period of extended crisis. PMUS had a wide, national, grassroots presence; at one time in the early 2000s this exceeded even the reach of BRAC,21 without having a similar HQ establishment or profile. It pursued a strategy of group mobilisation and supported income generation alongside conscientisation and empowerment objectives. It was innovative technologically especially in smallholder agriculture and horticulture and developed a large-scale microcredit, revolving loan fund, to support group entrepreneurialism. Increasingly it federated these groups and entered local-level politics through its members competing in local government elections as well as via various forms of protest. Nationally it reinforced this process by being a prominent leader of the Association of Development


G. Wood

Agencies in Bangladesh (ADAB); and by creating a think-tank in the early 1990s to promote broad civil society objectives as well as specific critiques of the annual budget from a pro-poor perspective. In this role, it participated in national and international seminars and conferences, and attracted large-scale funding from external donors. It appeared to be very successful. It developed a large HQ, an attractive out-of-Dhaka ‘ashram’ for its training sessions, conferences and some action-research. It also constructed a network of pucca area development offices. In other words it acquired significant property. Up to the mid-1990s, it also appeared to have independence from the state, partly because democratic politics only returned to Bangladesh in the early 1990s, and it managed to appear separate from and critical of the military regimes, while operating and expanding under their patronage. This sense of independence was also reinforced by having large and apparently secure flows of external donor funding. However the formation of the NGO Affairs Bureau in 1989, towards the end of the Ershad regime, was resented by PMUS as a controlling, regulatory intrusion.22 It became a warning of problems to come. However beneath this formal picture of a separated and independent ‘estate’, the leadership of PMUS could not divorce themselves and their organisation from the social interests in the society around them. As part of extended kinship groups, they had relatives across the other ‘estates’ and sectors in the society: the bureaucracy, the military, the police, universities, judiciary, the media, banking, the manufacturing economy, property developers and, of course, political leaders of the suppressed and then emerging political parties.23 These and other links entailed two way interaction. Jobs in PMUS were found for sons and daughters in return for access, introductions and protection. Property deals were sealed with commissions. Mastaan24 were paid off to facilitate construction and quiescent labour. Senior bureaucrats and later politicians were given favours to release land for development, and not to question too closely the use of donor funds. Such linkages have their ethnographic roots as well as serving key organisational objectives. It is a familiar tale of necessary co-option in an uncertain environment, a tale made famous through Selznick’s classic study of TVA and the Grassroots (1953). Thus in a generic sense, the leaders of PMUS would not be doing their job if they were not engaged in these various co-option practices. However, the issue under the limited access state conditions of Bangladesh is the nature of this institutional permeability. Do the practices of co-option undermine the autonomy of CSOs and the boundary between the ‘estates’? These practices have intensified during the recent period of PMUS when it has been enduring an extended crisis. The origins of this crisis derive from the explicit mission of PMUS that ‘politics matter’ in the removal of poverty. It is of course laudable that PMUS understood that the reproduction of poverty was a function of class relations with political elites both protecting propertied classes and themselves rent seeking across the economy. It was therefore broadly comfortable with the proposition that to be effective on behalf of the poor, the political arena had to be entered in some way. Thus group mobilisation at the grassroots was increasingly seen as the base of a pyramid, a large informal

Deconstructing the natural state? 119 constituency to support the advocacy and subsequently candidature in national elections of the principal leader.25 The first obvious foray occurred in the mid-1990s with a large rally in Dhaka, intervening essentially on the side of the then opposition party, the Awami League (AL), to overturn the bogus elections held by the BNP (Bangladesh Nationalist Party) in February 1996. This effort was successful in that subsequent more valid elections were held later in the year, bringing the AL to power for the 1996–2001 period. PMUS was now firmly tagged to the AL in the popular imagination, and indeed participated in various ways in policy development, especially in relation to land access for the poor. PMUS and its leader appeared to be strong as a result of this association, developing higher ambitions of expansion and influence with the prospect of ongoing donor support and an expanding revolving loan fund supported by the Palli Karma-Shayak Foundation (PKSF; a microcredit wholesaler). But, as friends advised at the time, this strength was also a weakness. With anti-incumbent voting, the BNP returned to power, primus inter pares in a coalition including Jamaat-e-Islami, in 2001, with scores to settle, especially with PMUS. A period of political harassment followed, entailing corruption charges against the PMUS leadership, and especially its principal leader. In these charges there were elements of possible truth, but also of ambiguity, and of invention – but enough potential truth to scare donors away and to suspend funds. Investigations and legal processes, indeed entanglement, followed.26 The previous patronage of the AL was now a millstone, so protection could only be sought through mobilising bureaucrats, other NGOs, think-tanks and journalists, and seeking supportive links to lawyers, judges and military personnel. Pleas to donors increasingly fell on deaf ears. PMUS limped through to the coup in early 2006, only to be included in a wider list of people and organisations to be cleaned up by the interim government. The elections in 2008 produced a landslide for the Awami League, but alas for the PMUS leader he lost his deposit in all of the three seats he contested, having failed to get an AL ticket. This humiliation broke the leadership of PMUS into two opposing factions: led by the leader, on the one side, and by his erstwhile deputy, on the other. This split and subsequent intense competition for control of PMUS prompted a nervous Board finally getting the courage to vote off its Chairman, namely the PMUS leader. The ‘charge’ was essentially bringing PMUS into disrepute by using it to support his increasingly personalised ambitions to enter politics and the Jatiyo Sangsad and, further, pursuing this agenda over many years with donor funding. Thus the Board had finally overcome its loyalty and shared the formal, GOB critique, actually now shared by both leading parties, that the de facto use of external donor funds to build up a constituency of electoral support was fraudulent and anti-nationalist. Of course the ousted leader hotly contested this critique, referring to his right as a citizen to enter politics. This story, as related so far, might imply that representatives of the state and the political class wanted separation between the state and such organisations in


G. Wood

the civil society domain. However, this is not the main lesson to be drawn, as subsequent events further reveal. The factional struggle between the contenders for control over PMUS over the last five years has required these contenders to recruit support from various elements of the state and political class. This process of cooption was initiated by the ex-leader, deploying his erstwhile links, albeit informal ones, with mastaan, goons27 and local police to attempt a physical reoccupation of the HQ building in Dhaka.28 With this knowledge, judges in the lower courts were influenced along with the police to turn a blind eye to this violence. To compete, the erstwhile deputy had to imitate this behaviour and open up matching contacts with the political class, and be prepared to mobilise funds for the legal costs of repetitive cases in successively higher courts. In return, places on the Board had to be reallocated to political nominees to ensure political management of PMUS in the future. Thus this process of factional struggle has ensnared both contenders and their respective supporters in an ever-increasing dependency upon both the governing party, and also functionaries in the state. By being unable to separate itself from the patronage of the ruling party, any long-term survival of PMUS requires it now to be deeply entangled with the state. What is interesting, and key to our argument, is that this entanglement and thereby subversion of autonomy does not arouse more perceptions of shock and horror among observers in Bangladesh. The explanation for this lies in the ‘prison’ metaphor (Wood 2000). Some other large NGOs have had uncomfortably similar stories and thus cannot credibly throw stones from their glasshouses. Others have more subtly and quietly nurtured their support among the different factions and branches of the political class and senior functionaries of the state through Board memberships, through actual senior appointments in their organisations, especially for recently retired senior officers, as observed by Lewis (2011). Young relatives, whose family connections have ensured them a good education, have had starts on their career ladder with NGOs, which are now highly professionalised, offering decent salaries and prospects. Thus both crisis, and the management of crisis avoidance, demands this entanglement. What are the wider structural implications of this entanglement? NGOs, the hundreds of them in Bangladesh, are both a threat and an opportunity for political classes managing a limited access state. In threat terms, they can potentially mobilise an alternative constituency, especially if supplied with non-state funds, thus undermining the state’s monopoly of the distribution of resources and positional goods, and its collective/coalition monopoly of rent seeking. NGO constituencies have demonstrated their weakness in open electoral terms, so far, and anyone who has attempted to launch a meaningful third-force political party has been rapidly and explicitly suppressed, even if led by a Nobel Prize winner!29 This is a very obvious process of limiting access to prospects of political office, and maintaining a strict control of that access. Indeed the two main contending parties, or more accurately political factions, are strongly focused upon delegitimising the claims of their respective opponents to democratic political status. The spoils are not to be shared, if at all possible. However, the prospect of being out of office does temper the behaviour while actually in office!

Deconstructing the natural state? 121

Conclusion We have proceeded from an assumption that an autonomously functioning civil society is essential for the governance and accountability dimensions of democracy, thereby addressing the generic problem of democratic deficit in most societies of the world. We have explored this widespread problem through the case of Bangladesh, famous on the one hand for the vibrancy of its development NGO sector while still emerging from an agrarian past, characterised by patronclient hierarchies, social deference and unaccountable elite domination. Although NGOs are not synonomous with civil society, we have referred to the case of a significant development NGO in order to demonstrate the intricacies of their relationship to the natural, limited access state and its actors. We have argued that these relationships circumscribe their room for manoeuvre, even when they are apparently challenging the state. So the key question to pose is: can we envisage a change in these conditions so that civil society does become free of the state? North et al. (2009, p. 148 and passim) argue for what they term ‘doorstep’ conditions enabling a transition from limited to open access societies. The crux of these conditions is a shift towards impersonality and the rule of law for elites. This is combined with ‘perpetually lived organisations in the public and private spheres’ (i.e. organisations that can exist independently of personalised elite patronage and incorporation) and ‘consolidated control of the military’. They also cite the work of Acemoglu and Robinson in the Economic Origins of Dictatorship and Democracy (2006)30 who paradoxically argue: ‘Because the elite loses under democracy it naturally has an incentive to oppose or subvert it; yet, most democracies arise when they are created by the elite.’ The easiest way to present the apparent paradox of Acemoglu and Robinson’s observation is through elite perceptions of the need to make concessions to other classes seeking access rather than risk losing everything through clinging to elitist, highly exclusionary power. In other words, for elites a controlled transition to democratic, open access conditions in which they still significantly share power is preferable to revolutionary upheaval which might sweep away their rents and privileges. So how might this work in Bangladesh? The first observation is that nouveau riche, urban middle classes are growing rapidly so that longer standing elites have to make room for them, thus shifting the society from fragile to somewhere between basic and mature natural state conditions. Second, the threat from a growing aspirant but substantially excluded, urbanised lower middle/ working class is palpable. Thus mastaan management of such emergent classes on behalf of elites is unstable and unpredictable. Third, a neo-Marxian mobilisation of poor peasant classes via NGO or other movement intermediaries is now highly unlikely, despite the past four decades of radical orthodoxy, since they have been successfully contained by a combination of mastaanised forms of cascading patronage (as a substitute for previous semi-feudal mechanisms of incorporated control) and migratory safety valves (either in overseas labour


G. Wood

markets or internal urbanisation). And fourth, crucially for the key doorstep condition, there will be a shift over the next decade along the personal identity scale from specific individuals (and their families and intimate associates) to social persona. In other words, not only is Bangladesh at a demographic conjuncture regarding threats to elites, but also facing a succession crisis in both main political factions. This time fact alone heralds the prospect of at least partially fulfilling the key doorstep condition of some shift towards impersonality31 and, with that shift, a stronger rule of law for elites. But do these four observations represent a sufficient combined ‘doorstep’ condition? Are they enough for ruling elites to accommodate other classes into a wider political space, managed through codified norms and rules offering autonomy for organisations of civil society rather than through continued use of mastaanised processes? However, with a few notable and noble exceptions of relative independence and autonomy,32 most organisations33 and corporations34 in Bangladesh are a long way from freeing themselves from current forms of political incorporation. The military is a long way from consolidated, civilian control. The judiciary is highly contaminated. Millenarian options associated with Islamic fundamentalism are close at hand for populist management of the excluded, volatile, urban déclassé. Educational provision, and thus ideas about citizenship under open access conditions, remains pitiful and reproduces exclusion or, at best, adverse incorporation. Market transactions remain highly imperfect through non-transparent social and cultural variables. Labour remains pre-commoditised (Wood and Gough 2006). The spheres of economic and political life thus remain interlocked. It is highly possible, then, that Bangladesh will continue with its mastaan forms of capitalism,35 that a shift to impersonality will be incomplete for at least another generation,36 and that rising populist threats to elite (albeit enlarged and accomodating) rent seeking will be managed through a combination of fascist millenarianism,37 interspersed with military coups. Donors and the few local, open access oriented escapees,38 will continue to wring their hands from the sidelines. Though, to return to an earlier point, we should not see this set of conditions and consequent outcomes as hegemonically locked in. The structural contradictions are present to enable ongoing struggle and contestation. Although the ‘springs’ in North Africa and the Middle East are not running a course into a pleasant summer, they are indicative of the proposition that limited access, natural states are ultimately unstable in a modern world of migration and global communications where elites steadily lose their monopoly of control over information. The widening knowledge of open access, democratic models elsewhere can act as a driver of internal movements for positive change, movements especially populated by emerging middle classes with a growing stake in how the society should be run. But, as North et al. have observed, these changes are not lazily linear, and can be reversed, and for Bangladesh that remains a threat of diversion into more obvious millenarian forms as a function of aspirant classes experiencing ceilings to their progress by rent seeking elites.

Deconstructing the natural state? 123

Notes 1 This reference to ‘estates’ derives from the medieval distinction between clergy, nobility and the commoners, akin to varna categories in Hinduism. This context of three estates, or orders, gave rise to the idea of the ‘fourth estate’. Thus Edmund Burke referred to the fourth estate in the gallery of the House of Commons, namely the Press. Thomas Carlyle also deployed the same idea in his History of the French Revolution. Today, the term ‘fourth estate’ refers to societal or political forces whose influence is not always recognised by powerholders. This is close to the contemporary idea of civil society. 2 For example, as favoured by the World Bank ‘social development indicators’ exercise. 3 Always difficult to translate into English, though ‘community’ and ‘association’ are usually used. 4 This is a reference to the state protection of labour rights in otherwise open markets (Polanyi 1944). See also Esping-Andersen (1990 and 1999), as well as Wood and Gough (2006) for the use of ‘de-commodification’, derived from the same separation/ compensation principle. 5 Another way of understanding this artificial layer of interaction is to recall Goffman’s Presentation of Self in Everyday Life (1971): we interact through constructed ‘fronts’ (formal organisations), taking care to disguise our ‘backs’ (deep structures or institutions). It is thus by reference to this ‘game’ that the contradiction exists. And there is widespread collusion in this game. It is as if we have ‘pretend’ societies, seeking normatively to act out the tenets of the liberal democratic pluralist state for an audience, while behaving pragmatically by another much more acceptable and culturally familiar set of rules derived from the intimate personalised structures of agrarian relations: patron-clientelism enriched by kinship, clan and regional identities, as well as the notion of the ‘batch’. 6 See Wood 1985 and 2007 on labelling which asserts that stories are the real ‘us’, while cases are the abstracted elements of our story specific to compartmentalised transactions to get formal business done. 7 The contrast between specific and diffuse is derived from Talcott Parsons’ ‘pattern variables’, in this instance a reference to role performance. 8 From Prisoners and Escapees (Wood 2000), but also a term used in a similar way by Oliver de Sardan (2008) writing about Africa. 9 Organically interdependent in a Durkheimian sense. This point is very akin to the structural-functionalist tradition in anthropology and political science. 10 See Mushtaq Khan’s paper for the World Bank (2010c) which deploys an earlier essay of North on Limited Access Order (LAO) to periodise phases of rent seeking by political classes in the Bangladesh State. 11 Including, critically, civil society. 12 Bangsho: lineage, understood hierarchically in the allocation of social status. 13 Zamindars: the old landlord class, used to describe anyone (men) behaving like traditional patrons. 14 The term ‘mastaan’ refers to an underworld of gangsters and brokers, perhaps akin to the mafia. They are normally associated with criminality, but recent work has recognised their role as intermediaries (e.g. between slums and municipalities) for their clients, enabling informal routes of access to services, employment opportunities and protection. Thus they represent a kind of social capital, though not quite as intended by Putnam. See Khan (2000a) for a detailed exposition of this argument in a Dhaka slum. 15 The economy is, after all, growing through the expansion of the garments industry and other globalised economy activity, providing an expanded base for rent seeking. 16 This process of co-opted inclusion has interesting implications for the limited access thesis since it represents a mechanism for the informal extension of access.


G. Wood

17 North, Wallis and Weingast (2009) differentiate between states in terms of their sponsorship of organisations which manage access in some form. For the contrast between Gramscian and de Tocquevillian perspectives, see Davis and McGregor (2000); see also Woldring (2000) on de Tocqueville’s approach to state and civil society. 18 Parsons’ pattern variables again. 19 A useful distinction made by Harry Blair (2010) between the immediacy and intimacy of short loop in contrast to the depersonalised remoteness of long loop accountability between providers of services and the claimants of rights and entitlements. 20 A thinly disguised acronym to provide some anonymity. 21 Bringing Resources Across Communities (formally the Bangladesh Rural Advancement Committee). 22 This judgement was shared by the more critical NGOs, though Lewis (2010, 2011) has a more benign view of this relationship, reporting that ‘it is quite common for an NGO to employ at least one reasonably senior former government official in order ensure that problems in working relationships can be solved by using personal ties and channels of communication’ (Lewis 2011, p. 121). From this PMUS case and other examples, such appointments are more likely to be double-edged! 23 I know many other NGOs where these relationships also apply. 24 See note 15 above. (NB We tend not to pluralise words from Bangla, so this can be singular or plural, and also refer to an institution.) 25 He once told me of the necessity to put ‘muscle behind the policy’. 26 Another large regional NGO experienced similar investigations, including directly by its donors. 27 A more Anglo-Saxon term for ‘thugs’. 28 Which has now been achieved after several attempts. 29 This a reference to the founder of the Grameen Bank, who has twice tried to enter the political arena on a technocrat ticket. The first time, with two other well established senior NGO players, was towards the end of the Ershad period, offering in effect a lifeline to a continuation of the military-bureaucratic oligarchy. More recently towards the end of the Interim Government in late 2007 he tried again to form a party of technocrats, but gave up after two months of very negative media coverage. His move attracted the deep hostility of political incumbents who have subsequently managed to remove him from the leadership of the Grameen Bank and undermine his reputation through allegations of financial impropriety. In other words, like the leader of PMUs, he has been punished for trying, in Northian terms, to widen the limited access coalition. 30 This title is a play on Barrington-Moore’s famous Social Origins of Dictatorship and Democracy (1966). 31 Though the continuation of the Nehru dynasty in India and the precarious survival of the Bhutto one in Pakistan makes one less optimistic about this aspect of transition in Bangladesh. 32 BRAC, the Daily Star and TIB come to mind. 33 Including NGOs and trades unions. 34 Including the media and other industries. 35 Ha-Joon Chang from LSE writes about different forms of capitalism reflecting different histories and socio-political circumstances (e.g. in the Guardian, 16 November 2011) rather like Wood and Gough’s historically rooted and specific welfare regimes (2006). 36 Remember, with an extended mastaan structure, the personality issue exists all the way down to the grassroots in daily face-to-face encounters as a problem of succession. 37 Though contested between the principles of Bengali nationalism and Islamic fundamentalism. 38 From the prison.


When things go wrong in NGOs What can be learned from cases of organizational breakdown and ‘failure’? David Lewis

Introduction Non-governmental development organizations such as Gonoshahajjo Sangstha, Proshika, Samata and Saptogram were once regarded as leading examples of dynamic, innovative Bangladeshi NGOs, but each of them, often for complex combinations of reasons, ran into difficulty and collapsed. This chapter explores the issue of organizational ‘failure’, an aspect of development non-governmental organizations (NGOs) that seems to have received surprisingly little attention in the literature. While there has been considerable debate about the role of NGOs in Bangladesh, the focus has generally tended to be on externalities rather than on organization and management issues. This chapter begins by discussing the problematic concept of ‘failure’, and then briefly reviews some NGO cases that have been documented in Bangladesh. A short original case study of Comilla Proshika, a large NGO that failed spectacularly during the early 1990s, is then presented. By considering cases of organizational failure, and then linking these to organizational theory and politics, the chapter seeks a new perspective that may help us better understand (1) the role of hybrid forms of management that can allow an organization to build management systems that ensure the stability of internal systems while responding flexibly to its wider societal and ideological environment, and (2) the possible value of an ecological perspective on NGOs that recognizes the importance of some level of organizational ‘failure’ in ensuring the wellbeing of the sector as a whole. ‘Failure’ of course is a far from straightforward concept, and needs problematizing from the start. It can be usefully explored in two distinct ways – one that is at least ostensibly objective, and another that is more subjective. On the one hand, organizational failure is commonly understood in structural terms. As a result of poor leadership, political interference or changes in the funding environment, an organization implodes, is closed down, or it simply ceases operations. On the other, failure can be seen in more complex, socially constructed terms – as a label or political judgment that becomes applied to an organization for reasons that may, or may not, actually have to do with its performance or results. In this perspective, failure may not be an ending or rupture, but a state of being. Rather than signifying the end of an organization’s life, failure may take


D. Lewis

the form of a near-permanent condition, with an organization persisting in a sort of ‘death in life’ where it carries on existing and operating, but manages to achieve few if any of its stated goals. In relation to the first scenario, a set of structural reasons for NGO failure are regularly outlined in studies and reports that seek to analyse the management and regulatory challenges faced by NGOs. For example, Transparency International Bangladesh (TIB 2006) lists many such issues that are believed to be responsible for the poor performance or decline of organizations – including lack of financial sustainability, shortage of skilled staff, interference or lack of support from government, poor infrastructure, lack of professionalism. A World Bank (2006, p. i) report on NGOs summarized a range of commonly perceived weaknesses: the rapid growth and diversification of the NGO sector has … given rise to questions and concerns. These include the viability of a regulatory framework developed when the size and scope of NGOs was far more limited, the appropriate political and commercial spaces for NGO activities, trade-offs between NGO sustainability and pro-poor orientation, the impact and quality of NGO services as they have scaled up, NGO corporate governance, and the implications of different government-NGO partnerships. These concerns about regulation, accountability, governance and leadership have become familiar to those who follow debates about NGOs in Bangladesh and elsewhere, and they are regularly voiced. Policy makers, donors and civil society groups then propose solutions and these usually involve capacity building inputs for NGOs and new regulatory systems. Organizations such as Transparency International Bangladesh (TIB) have for many years raised awareness about many of these problems in the NGO sector and advocate appropriate remedies (TIB, 2006). Yet prescriptions for change tend to follow standard technical ‘organizational development’ approaches to improving organizational performance and accountability through interventions that seek to build new management systems, improved governance structures, tighter legal frameworks and greater government regulatory powers. Yet these initiatives have tended to meet with fairly limited results. Are we missing part of the picture? Are there new insights that might be learned from studying cases of full-blown NGO failures that might take us beyond these technical orthodoxies? Why have some development NGOs fallen apart or faded away? In relation to the second scenario, there are some potentially useful alternative perspectives on organizational failure. In particular, rationalist models of organizations that stress only technical factors are vulnerable to critique because notions of ‘success’ and ‘failure’ are at least in part social constructs, and because organizations need to be understood as operating within wider political environments. Clarke and Perrow (1999, p. 179) take a cognitive approach to analysing what goes wrong with organizations, observing that ‘organizations fail often and they fail in important ways’. Central to what these authors term

When things go wrong in NGOs 127 ‘prosaic organizational failure’ is the argument that symbolic factors contribute to organizational failure, because an organization may become unable to make adequate plans to deal with future problems that may arise in relation to its activities, or even its eventual survival as an organization. This, they argue, is because of two stages of problem: first, an organization comes to believe in its own representations and then, second, it becomes unable to take account of any information it finds that challenges or undermines the logic of this inaccurate representation. This constructivist perspective is also in some respects reminiscent of recent work within the anthropology of development. David Mosse has argued convincingly in his book Cultivating Development (2005) that development interventions do not fail but instead are failed on the basis of socially constructed judgements that reflect conditions where meanings fragment, social relationships shift and the overall coherence of a representation of ‘success’ can no longer be sustained. It is common within rational organizational models to view ‘failure’ as a temporary condition that reflects the inability of an organization’s stakeholders to secure enough power to control an organization in their interests, and that eventually under the right circumstances they may be able to do so. Seibel (1999) offers a different possibility. He instead points to cases where organizations continue to operate not because they are ‘successful’ in the conventional sense, but because they serve wider contextual political purposes – for example, they maintain the illusion that something is being done about an issue, they placate a potentially disruptive political constituency or they serve a particular ideological purpose. He dubs such organizations ‘successful failures’, because they may continue to go about their work despite their inability to meet formal or accepted standards of efficiency, effectiveness and accountability. There is another potentially interesting issue if we begin think more about NGO failure – the speed with which the few high-profile cases that have been documented now seem to have been largely forgotten, and the lack of lesson learning within the development industry. There seems on the whole to have been relatively little reflection within the sector, or within the government and donor community more widely, on why things may have happened the way they did, or what kinds of lessons might be learned from these conspicuous cases of NGO failure. Perhaps this is an unsurprising fact given that the wider world of international development policy remain faddish and largely uninterested in its own past.1 Development agencies and policy makers live in a ‘perpetual present’ in which ideas and plans constantly move on, and people are keen to leave behind the failed expectations or disappointments of earlier periods so that they can engage with the ‘next big thing’ (Lewis 2010).

NGOs in Bangladesh Since the 1980s, it would be fair to say that Bangladesh’s NGO sector has become one of the best known in the world, and one of the largest, being believed to contribute about 6–8 per cent to the country’s gross domestic product


D. Lewis

(Irish and Simon 2005). It has arguably influenced global development theory and practice in many ways, and it has given rise to a range of national and international admirers and critics (Choudhury 1989; Khan et al. 2009; Lewis 2011; White, 1999). There has probably been more written about Bangladesh’s development NGO sector than that of any other country. Studies of individual NGOs, their projects and work, and of the sector itself, began appearing from the late 1980s and have continued to emerge ever since. Over the years, the development donors have produced countless evaluation reports, while academics and researchers have also found much to engage their interests and energies. There is not the space in this chapter to do more than skim selectively over this wideranging landscape of work. For example, the context and long history of voluntarism, philanthropy and non-governmental work that provided a foundation for the modern NGO sector is usefully set out in Hasan et al. (1992). Hashemi (1990) documented the growth of development NGO activities in Bangladesh initially as relief and rehabilitation work by local activists in the wake of the war of liberation in 1971, and the famine in 1974, followed by the subsequent evolution of these organizations – with the assistance of international donors – into what became known as a distinctive ‘NGO approach’ to development based on grassroots group formation, the provision of various types of services and the construction of local and national level advocacy agendas. The sector can therefore be understood as ‘a locally determined formation of globally determined influences’ (Lewis 2011, p. 109). The institutional context in which the NGOs operated also received considerable attention from researchers. NGOs evolved alongside the Bangladesh state, in circumstances where each often viewed the other with unease. Work by Sanyal (1991) highlighted the recurrent tensions that characterized the relationship between the government and the NGO sector and identified patterns of cooperation and conflict inherent these relationships. Moving to NGO relationships with international development actors, Stiles (2002) explored the problematic relationships that emerged between NGOs and foreign donors, highlighting the sometimes unwelcome pressures generated by the aid relationship that led some NGOs with more radical origins to accommodate their work to mainstream development policies, and the rapid increases in funding that sometimes led to rapid organizational growth and change. Political analyses of NGOs focused on their roles as mediators between mainstream and counter-hegemonic visions of transformation. Some work has focused on the political contexts of ‘civil society’ and the ways development NGOs relate to this wider picture, such as Blair’s (2005) work on pro-poor advocacy, and Devine’s (2002) study of NGO campaigning and mobilization efforts to secure pro-poor land redistribution. Rahman (2006) discusses the ‘de-politicization’ of the NGO sector away from its more radical origins as a result of these external and domestic institutional imperatives, while Mannan (2005) argues that development NGOs can be understood as serving as agents of global capital to connect people in rural areas to modernity and market forces. The narrowing of the NGO sector’s focus to one that overwhelmingly focuses on microcredit approaches has also been the subject

When things go wrong in NGOs 129 of criticism (Kabeer et al. 2012), and the microcredit approach itself has come in for increasing criticism (Bateman 2010). For Wood (1997), there are concerns that the extensive roles of NGOs in the delivery of basic services weakens accountability between state and citizens, creating a form of privatization in which government delegates its responsibilities to non-state actors with unclear lines of accountability to service users – an arrangement that he has termed the ‘franchise state’. There have been fewer studies that have engaged directly with management and organizational issues in the NGO sector. Despite an openness to researchers interested in development issues, NGOs as organizations have tended to remain closed communities, generally protecting their organizational boundaries against researchers who wish to understand how the organizations themselves work. One exception was Smillie and Hailey’s (2001) rich account of a set of South Asian NGO leadership, which offered some useful insights into the organizational histories and leadership strategies and styles of organizations that included two of Bangladesh’s largest NGOs, Proshika and BRAC. Another was a comparative evaluative micro-level study by Michael Edwards (1999) that usefully explored the organizational characteristics that have produced a range of both successful and unsuccessful outcomes within four NGOs in Bangladesh and India. BRAC has also attracted lots of attention from writers interested in explaining its apparent success. Influential books based on organizational studies of BRAC by Lovell (1992) and more recently Smillie (2009), have indeed contributed to a better understanding of the management and strategy of what is now one of the few relatively well documented development NGOs and probably one of the best known NGOs in the world. Overall then, both critical and supportive voices emerged in relation to the NGO sector. For example, there has been widespread recognition of the sector’s success in tackling the problem of poverty and inequality in areas such as microfinance services, reproductive health, legal training, non-formal education and gender rights. Development NGOs for a while came to be seen as genuinely innovative organizations that could in some cases develop new creative approaches to addressing longstanding societal problems. Organizations such as BRAC, Nijera Kori and Grameen Bank – in their very different ways – have justifiably gained global reputations for undertaking innovative work that often challenges conventional or mainstream development assumptions and practices. Yet these largely positive views have also coexisted alongside a range of persistent concerns. For example, development NGOs have been variously criticized for having a relatively small scale of operation, low levels of accountability to their users, limited impacts on overall poverty levels and undermining the state and citizenship (Lewis and Kanji 2009). There have also been increasing criticisms of the way that the NGO sector became dominated by microcredit approaches to the exclusion of other types of approach, and of the often exaggerated claims around the use of microcredit itself as a tool for development. We are now a long way from the essentially positive 1990s view of NGOs as a ‘magic bullet’ that would help solve development problems, but NGOs have


D. Lewis

nevertheless remained a major set of players on the landscape of development and poverty reduction work (Edwards and Hulme 1995).

Narratives of NGO decline, disintegration and ‘failure’ There is considerable diversity to the organizations that make up Bangladesh’s NGO sector in terms of origins, size and overall approach. The fortunes of many of these organizations have waxed and waned over the years, with some expanding and thriving, and others surviving in more of a ‘hand to mouth’ existence. During the past two decades there have been a handful of examples of cases of ‘organizational failure’ within the sector, and these serve to illustrate some key organizational and contextual issues for NGOs.2 The aim of this section is to consider some of the lessons that emerge from this, taking into account both organizational issues and the wider social processes and embeddedness. The first of the major development NGOs to falter was Comilla Proshika, an organization that is barely remembered today but which during the 1980s had grown into a well respected medium-sized development NGO, funded primarily by the Canadian government’s International Development Agency (CIDA), and with an ambitious service delivery and grassroots empowerment agenda. Comilla Proshika was the result of a leadership split within the original Proshika that had been founded in 1976. Proshika had emerged at the interface between post-conflict nation building, student activism and the growing influence of international development agencies in the new country. Originally a Canadian University Service Overseas (CUSO) project (its two main founders had worked as CUSO project officers), it then evolved into an independent local NGO. Both co-founders had been inspired by Bangladesh’s national liberation in 1971, but felt constrained by the increasingly authoritarian style of Sheikh Mujibur Rahman’s government. One founder had once been set to go into the civil service, but wanted a more flexible base to pursue social change. The other, a leftist activist in his student days, had already briefly worked as a trainer for BRAC, another NGO. They formed Proshika Manobik Unnayan Kendra, which became known as Proshika – a Centre for Human Development. The name was an acronym of three Bangla words, proshikkhan (training), shikkha (education) and kaj (action). Proshika experimented with local-level organizing influenced by the radical politics of Paolo Freire, and engaged with the emerging ‘alternative’ development paradigms also coming from some of the international NGOs. As Smillie and Hailey (2001, p. 8) state, ‘Proshika grew out of a donor project which was also staffed, and later taken over, by young social activists’, and actively sought to link activist and developmental objectives. It initially began working in a few villages of Dhaka and Comilla districts, but grew steadily. However, differences gradually emerged between the two founders and this led to the NGO splitting – reasonably amicably – into two separate organizations in 1981. Throughout the 1980s, the resulting two large NGOs, known respectively as Proshika MUK (or ‘Dhaka Proshika’) and Comilla Proshika, operated a geographical division of

When things go wrong in NGOs 131 labour, with the latter operating mainly in the Comilla and Chittagong divisions and Proshika everywhere else. Yet while Proshika thrived, Comilla Proshika fell apart around the late 1980s and collapsed in the early 1990s (for a variety of reasons explored in a short case study presented below). The next high profile NGO to get into trouble was an organization called Gonoshahajjo Sangstha (GSS) in 1999. At the time, GSS was a large NGO with a radical Freirian empowerment philosophy and an innovative approach to village education programmes. Between 1995 and 1997, it had received more than US$25 million from donors including Britain, Sweden and the European Commission. GSS began to feature strongly in the literature around NGOs and mobilization during the 1990s, such as Hashemi’s (1995) work on GSS’ experience putting up candidates from among their client groups in local elections in Nilphamari in an effort to challenge the local power structure. Yet by 1999, the government had taken over the organization after complaints were made to the NGO Affairs Bureau by some of its employees about financial irregularities and employment rights, including some allegations of sexual harassment. The government then claimed to have uncovered serious financial irregularities and evidence of misuse of donor funds, and while a version of GSS still exists providing education services, the organization never recovered its status as a major NGO player. The case of Saptagram Nari Swanirvar Parishad is another interesting case. It has recently been documented in one of the few reflective retrospective studies of apparent organizational failure to be written (Kabeer and Huq 2010). Saptagram was established in 1976 by political activist Rokeya Rahman Kabeer. It was a path-breaking NGO that was unique for its time, run by women primarily for women, with a focus on organizing women to fight against gender injustice through group formation, building a savings programme and providing legal training. In their analysis, Kabeer and Huq (2010) describe in detail the complex combination of factors that led to the decline of the organization. These included the charismatic, forceful leadership style of the founder that made it difficult for the organization to build second tier leadership, the tensions that existed among staff and members of different backgrounds, and the large-scale expansion of official aid donor flows to NGOs that followed the organization’s success and contributed to rapid and disruptive organizational change from the early 1990s onwards. In 2001 Dhaka Proshika, which had become Bangladesh’s second largest NGO having expanded greatly after its 1976 parting of the way, was now also beginning to run into serious difficulties. Its leadership had long taken a tough confrontational line with the Bangladesh Nationalist Party (BNP) that by the late 1990s was in alliance with the Jama’at-i-Islami Islamists. The Proshika president had become became well known for being aligned with the Awami League party, and it was rumoured at the time that he might even stand as a member of parliament. He was also chair of the Association of Development Agencies of Bangladesh (ADAB), the country’s NGO umbrella organization, and was accused in some quarters of politicizing its work. When the BNP and its allies


D. Lewis

were elected, the new government pursued what many of the NGO’s staff regarded as a vendetta against the organization. Complaints were also raised among some of its donors that the organization had infringed the understanding that development NGOs should remain politically neutral and should not become involved in party politics (Lewis 2010). Attacked by both government and by sections of the development community, and further undermined by an increasingly autocratic and unaccountable leadership style, the organization split into opposing factions that went into ongoing and apparently unresolvable conflict, effectively destroying the organization and its work in the process. Samata, a leading NGO that emerged out of the social mobilization tradition, has been working for many years in the northwest of Bangladesh. Samata Samaj Samity – the word samata means equality – had been established in the 1976 as a village association by left activists, local youths and social workers, and slowly grew into a formal organization that registered with government in 1983. It aims to empower poor people to recover khas and other publicly owned land and water resources, and to put pressure on the government to better address people’s needs and rights in the organization’s working areas. Devine’s (2002) study analyses the organization’s success in the allocation of land, but also points to the ‘underlying logic of patron-client relations’ within the organization’s make up, and the ways these both facilitate and constrain its activities. Although it had a strong record in reclaiming land for the landless, the NGO became involved in a serious crisis in 2008, facing cancellation of its funding from UK, Norway and Sweden after the findings of an independent forensic audit showed evidence of misuse and misappropriation of funds among its leadership. During my own fieldwork in Eastern Bangladesh during the late 1980s, I encountered Comilla Proshika as one of the leading NGOs operating locally at that time. Drawing on my field notes from the period, and from interviews carried out subsequently with former staff members, I decided it would be useful given the paucity of literature to produce a rudimentary, reconstructed case study of the demise of this organization.

Case study: Comilla Proshika As we have seen, Proshika’s first organizational crisis was the clash of its two original founders in 1981. The result was a split into two different wings that soon became two separate organizations known as Dhaka Proshika and Comilla Proshika (CP), loosely corresponding to a division of the country into two spheres of influence. Each took one of the co-founders as its executive director. Ostensibly the split was based on ideological differences over the relative roles of service delivery work and grassroots mobilization approaches, with each founder claiming the need for a more radical approach, but in practice such ideological differences were secondary to conflicts that had emerged over leadership styles and the effort to deliver resources, primarily in the form of employment, to clients in the leaders’ social and geographical networks on which their status and power to lead depended. CP then developed on a separate path from Dhaka

When things go wrong in NGOs 133 Proshika, although relations mainly remained cordial. It had four operational regions and provided a combination of social mobilization and service delivery development inputs. After the heavy flooding of 1988, the organization took on many new members and continued to expand. CP grew to be a leading and well respected medium-sized national NGO employed more than 200 field staff, serving around 65,000 people organized into more than 1200 rural groups in the eastern areas of Bangladesh. Behind the scenes, all was not well. One senior manager in charge of the Chittagong Office, and close to the CP executive director, decided to stand as a candidate in the 1988 subdistrict election for the ruling military government’s widely discredited party, and resigned. Colleagues then found inconsistencies in his office accounts, fuelling the belief that some of the NGO’s resources had been misused to support a political career by, for example, siting tube wells on the land of rich rural elites. By the end of the decade, CP had also run into serious difficulties with its main donor. A CIDA evaluation team were finding that many of its activities were ineffective and its governance systems compromised, with power having become overwhelming concentrated in the hands of its charismatic, authoritarian executive director. There was no effective management structure that could bring in new ideas or skills. The original 14-person Governing Body that had been established in 1981 had now only seven members, and five of these were the same staff who made up the NGO’s Management Committee. The resulting evaluation report stated that ‘leadership within the MC and the GB appears to be largely provided by one person – the ED’ (p. 12). While it noted a positive ‘familial organizational style’ and generally high morale (all the staff were middle class, and personally recruited by the executive director), the report found a noticeable basic lack of specialized expertise. The NGO’s earlier dual approach that claimed to combine empowerment work with services had become a monolithic credit-based service. There was ‘little evidence that groups are now-a-days actively seeking to identify the causes of their poverty’ (p. 111), and the NGO ‘follows an elitist rather than a participatory model of development’ (p. 119). Power had become concentrated in the hands of a very small number of officials ‘who by controlling both the GB and the MC, in effect become accountable only to themselves’ (p. 119). There were no regulations for appointing or removing the executive director, who was in a unique position based on setting up the organization, recruiting staff, inducting them and dealing with the donors. In line with the thinking of the time, NGOs were being encouraged – in some cases even pressured – by donors to acquire commercial assets to generate income that would replace dependency on funding. A garment factory, financed with a CIDA loan, was acquired as an income generation venture but it had quickly failed. Yet few lessons seemed to be learned from this by the leadership. Rather than hiring new qualified people, the executive director had decided that the factory was to be managed by a selfappointed three-person managing group composed of the executive director, one other employee of CP and a third person – a local teacher – whom they hired: ‘a


D. Lewis

meeting of two persons passed a resolution to benefit one of the persons present out of the operations of a company that had essentially been financed by CIDA’ (p. 155). After several attempts to develop joint strategies with the organization to improve its organizational wellbeing, CIDA pulled the plug on its finding. By 1992, with funding no longer available, several factions became apparent within the organization, with a dissident group of staff and employees loyal to them set against the ED and his supporters. Eventually, BRAC’s executive director F.H. Abed was reportedly able to broker a settlement to the dispute. Dissident staff then formed a small new local organization that became registered with the government’s Social Welfare Department in 1993 and began receiving small-scale government funding, but the new group was largely ignored by the international donor community and never recovered its earlier momentum.

Making sense of ‘things going wrong’ Drawing on this short case study, and on the brief stories of NGO collapse that precede it, how far is it now possible to move forward with an analysis of ‘failure’ and try to identify some common causal factors? It is usually believed that an NGO loses its organizational coherence and drifts towards failure when it finds itself unable to ensure adequate leadership accountability, resource sustainability or public legitimacy. Mannan (2010) suggests that cases of NGO ‘collapse’ such as GSS and Saptagram were largely ‘due to poor management and leadership’. This is partly the case, but it tells only part of the story. We also need to consider contextual factors, such as the role of donor resources and the wider politics of institutional life in Bangladesh. We also need to recognize that all organizations are unique and a full understanding of the difficulties experienced by each of these NGOs would no doubt require a more detailed, nuanced analysis than available data will currently allow (Biggs and Lewis 2009). Indeed, it needs to be recognized that understanding NGOs – or indeed any other type of organization – requires an approach that sees them not as distinctive generalizable normative entities, but as contingent on prevailing societal values and structures. The precise reasons for ‘failure’ differ in each of the examples above, deriving from complex combinations of internal and external factors. Nevertheless, there are some common issues that can be drawn out from the discussion, starting with internal factors but then broadening the analysis out into wider issues of context, politics and institutions. The problem of poor leadership and management is often the result of overdominant charismatic founder leaders, which is a leadership style that offers benefits to organizations struggling to survive in a hostile environment. Following from this is the problem that even if such leadership is effective for a period, it lacks sustainability in the longer term. One set of ‘organizational’ perspectives on failure focuses on the organizational life cycle, for example, and the inability of leaders to manage transition effectively from one stage of an organization’s growth to another. For NGOs, the problem of charismatic and dominant founderleaders and the difficulties associated with such transitions are well known

When things go wrong in NGOs 135 within the third sector, and regularly create problems of stability and succession (Lewis 2001). At the same time, the issue of leadership cannot be separated from the political context outlined above. In order to survive, the leaders of development NGOs are to some extent forced to play these unruly actors ‘at their own game’, leading to strategies to manage organizational risk that are themselves highly risky. Robert Michels (1962) analysed intra-organizational political dynamics in the form of his ‘iron law of oligarchy’ that emphasized the formation of elite groups that seek to gain command and control of organizations. The ‘iron law’ stated that as an organization develops, it adopts a rational system of hierarchy designed to make its operations more predictable, which inevitably creates tendencies towards oligarchy. This happens because leaders become more able to gain access to resources such as information and the power to frame issues and problems. This gives them greater power of decision making than the general members, who are rendered ‘incompetent’ and less able to participate. Michels linked these processes with intra-elite conflicts and contests for power. Oligarchic groups within organizations tend not to be cohesive groups but are instead subject to leadership struggles as old leaders contend with new ones for overall control. Organizations may therefore become weakened by the activities of contending groups or factions who compete for power. Moving to the wider context, a second set of problems centre on the role of international funding and the flows from international development donors to NGOs. This has sometimes been explored using elements of ‘resource dependence’ perspective (see Pfeffer and Salancik 1978, and the work of Hudock 1999 for its application to NGOs). Bangladesh’s location within the international system of development institutions, resources and interests has been a strong influence on the evolution of the NGO sector as a whole and on the fortunes of individual organizations. The power of externally provided development aid resources is strong, influencing both the organizational emergence and subsequent direction taken by development NGOs. In analysing how well organizations fare, there is an emphasis within such theory to focus on the capacity of organizations to mobilize adequate resources, and their ability to demonstrate and measure performance in relation to the work that they do. For example, Linz and Stepan (1978) argue that the survival of an organization can be seen to depend essentially on its ability to maintain legitimacy – based on its capacity to formulate solutions to problems (efficacy) and its ability to implement such solutions (efficiency). This approach is normally at the heart of practical efforts to enhance organizational capacity through technical inputs, or the current donor fashion for impact measurement and management by results. A contributory factor in many of the examples discussed above is the rapid growth of donor funding, leading to rapid, stressful and ultimately unsustainable growth. In a sense, some of these NGOs are victims of their own success, or at least of their perceived success. But most NGOs are operating within the same broad donor funding environment, so it still becomes necessary to explain why some NGOs appear to manage their relationships and resource flows satisfactorily while others do not.


D. Lewis

This leads on to a third set of issues around wider societal context. Organizations in Bangladesh need strategies to manage very high levels of risk, since they face a predatory state, clientelistic institutions and political parties and religious groups that often resort to violence. The wider context is dominated by a set of complex interpersonal social structures based around the power of competing elites, and this also feeds into an unstable, bi-polar political system based around two competing political parties. Class interests and patron-client networks combine to create instability in the country’s political institutions. As Mushtaq Khan (2000b) has argued, these become intertwined in the competing efforts of dominant groups to claim and control resources within a pyramidal structure. Political parties form the top of the pyramid, and their efforts to lead and organize a range of other groups and classes through a complex series of patronclient networks forms the base, and its foundations have roots that stretch out to penetrate all levels and sectors of society. This analysis can be linked to elite theory in order to explore how NGOs, as organizations that serve as vehicles for elites, experience change. Italian sociologist and economist Vilfredo Pareto (Tarascio 1968) famously wrote of ‘elite circulation’, seeing an elite as a minority group that manages to occupy and control what he termed the formal ‘command posts’ of a society. He developed the concept of a ‘governing elite’ that draws political power from its control of the ‘means of administration’, rather than simply of the means of production that was emphasized within Marx’s conventional analysis of power (Bottomore 1993). For Pareto, governing elites were not a stable group or class, but were characterized by processes of ‘circulation’ that changed their composition over time. Social change, he argued, could be explained as an outcome of the continual political processes of intra-elite struggles and as arising from these processes of replacement. As well as intra-elite circulation, Pareto also argued that individuals may circulate ‘between the elite and non-elite’, with members of lower strata gaining access to positions of formal authority, sometimes forming ‘new elite groups’ that then engaged in political struggles with dominant elites. As we have seen, NGOs in a country like Bangladesh are, alongside government and business organizations, areas of the commanding heights worthy of attention and control by elites. As NGOs have come to be seen as the leading actors in civil society in Bangladesh, they serve as powerful magnets for political actors and become points of capture, giving them a distinctive structural position in the context of political struggles for resources and power in Bangladesh. NGOs may also be the vehicles through which non-elite groups try to gain elite status and compete with older elite groups. Finally, we know that NGOs are forced to operate within particularly hostile and difficult institutional and political environments in Bangladesh. German sociologist Georg Simmel’s work emphasized the importance of societal harmony and conflict, and resulting levels of integration and disassociation, for the understanding of social structures. It follows from this insight that an organization is more likely to fragment when the relative balance between conflict and support relationships, or affect and indifference, is disrupted (Anheier and

When things go wrong in NGOs 137 Moulton 1999). A disjuncture between normative vision and reality can be maintained up to a point (and ‘cognitive dissonance’ is arguably an important aspect of life in most forms of development organization), but once such relationships and perceptions become unbalanced an organization is likely to go into crisis. Recognition of the importance of leaders’ capacities to balance these conflict and support relationships and prevent their fragmentation may provide a fruitful approach for exploring further why some NGO leaders manage to perform boundary spanning roles between the organization and its environment more effectively than others. Finally, we can also consider organizational ecology approaches to understanding change, which take a macro-level perspective on organizational fields rather than particular organizations. There is a focus on populations of organizations within a given environment, whose interrelations compose a system at the level of the field as a whole (Trist 1977). Like the resource dependency perspective, this approach links organizational survival with the ability to operate within a given material resource environment (McAdam and Scott 2005). Some forms of ecological theory go on to emphasize ideas of Darwinian competition and the selectivity of particular ‘successful’ organizations as an explanation of why others fail. Other views within organizational ecology, influenced by the ideas of Russian philosopher Peter Kropotkin, place less of an emphasis on competition and more on cooperation, evolution and mutual aid in the creation of shared organizational futures within a wider organizational field.

What can we learn from all this? So why do some organizations persist and others fall apart? This is clearly a complex question, but one thing that seems clear is that there is a tendency within development policy and practice literature to focus too much on technical organizational issues and efficiency-oriented explanations. Instead, we must also examine wider sets of organizational actors, networks and rules – the societal context – and engage with the importance of politics in organizational life. Development, as Brett (2009) has argued, is centrally concerned with the politics of structural institutional change. The performance of non-state actors depends on institutional arrangements that create rules and incentives governing cooperation and competition. He suggests that neither conventional left or right ideologies, nor market-based or statist orthodoxies, can in the end offer useful enough guides to transformation. Instead it is necessary to try to build ‘viable hybrid solutions’ (p. 8) to development problems that can both recognize local needs and capacities, and simultaneously commit to strengthening citizenship rights. Solidaristic organizations such as NGOs tend to be formed and operated by individuals ‘driven by affective and ethical needs rather than material selfinterest’ (Brett 2009, p. 11). Yet for them to contribute effectively to processes of social and economic transformation, such organizations nevertheless need to acquire certain characteristics in the form of effective hierarchies that can deliver performance, reciprocal systems that go beyond altruism (i.e. using professional


D. Lewis

and appropriately paid staff ), and forms of accountability secured via marketbased systems that can allow resources to be allocated as fairly as possible. How far does the absence of such characteristics explain what went wrong in the Comilla Proshika case? Building effective hierarchies The case study highlights the problem that effective hierarchies did not emerge in the organization, which remained dominated by a single leader. The organization resembled a classic oligarchy of the type described in Michels’ (1962) thesis. Within a charismatic leadership structure, decision-making input was unavailable to the majority of staff, and important key skills were lacking in the organization. Fisher (1994) once argued that in the case of some development NGOs ‘the iron law of oligarchy is rusting away in the third world’, due to the strength of their egalitarian principles and their use of more participatory forms of organization. The evidence from the Comilla Proshika case challenges such an optimistic view with its concentration of authority in a single leader, and in fact it would be quite difficult to find such egalitarian principles even within relatively successful NGOs. Within Comilla Proshika, the development of an effective hierarchy was further undermined by processes of elite circulation, which had also contributed to the original split in 1981. As we have seen, theories of elite circulation attempt to explain political change as taking place from above rather than from below. Within such change processes, the role of ordinary people – in this case staff in the NGO and members of grassroots groups – is strongly limited either as initiators of action or as principal actors. They become subordinate groups who can only be followers or supporters of one elite faction or another. Operating principles and professionalism Clientelistic forms of organizational politics and relationships mean that while NGOs may contain committed staff, they may also face barriers to building professional competences, since recruitment patterns tend to draw on social and political criteria above technical ones. Interviewees reflecting back on the Comilla Proshika story each remarked on the appointment of persons who did not possess the necessary skills or competence to manage effectively. As one senior former employee explained, the director was a leader with charismatic qualities that were effective at mobilizing people and resources, but he nevertheless had poor leadership skills and limited administrative control over the organization: The organisation had no clear vision or mission, no real programmes apart from credit programmes (which had a poor repayment rate), no properly trained staff and no real ethic of professionalism. I didn’t have the skills or the training to manage an effective micro-credit programme.

When things go wrong in NGOs 139 Another informant described the donor’s efforts to encourage the NGO to improve sustainability of the organization by doing a detailed cost-benefit analysis. When a consultant carried out the study, it showed that the NGO was not at all cost effective. As a result, some staff became interested in discussing ways of improving performance. However, the ED merely took this as a form of disloyalty: The ED thought that I was going against Comilla Proshika by working with CIDA, and he began to marginalise me.… The ED’s attitude was to argue that ‘we are working for the poor, so why do we need a cost-benefit analysis?’ Eventually it became clear that a set of organizational changes would need to be made. The consultant’s plan was to request shedding two-thirds of the jobs in an effort to improve cost-effectiveness within a severely over-staffed organization, but this informant reported that it was impossible for the ED to agree to this because of commitments to the friends and relations he had employed. Accountability principles In Bangladesh, political inclusion occurs within a context in which horizontal forms of civil society such as trade unions and peasant associations are relatively weak (Westergaard and Hossain 2005, p. 209). As we have seen in the discussion above, incorporation instead takes place through clientelist processes and vertical networks where individualized conflict over resources and spoils tends to discourage or displace the emergence of structural or class-based identities and action. This makes it difficult to build proper accountability systems based on market logics of, for example, partnerships for service provision that can be based on effectiveness and efficiency calculations. Relationships are more easily structured and resources allocated according to the logic of allegiance and interest. While it makes sense for NGOs to try to go about their work by ‘keeping politically neutral and negotiating with whoever is in power’ (Appadurai 2001, p. 23), the context of clientelism also produces incentives for leaders and factions to link up with wider politics. In the case of Comilla Proshika, a contributory factor in its decline was when a senior NGO staff member was elected as a Jatiya party candidate in the discredited 1988 upazila election organized by General Ershad, the country’s unelected military leader. This helped to destabilize the NGO and further weaken its morale. The organization’s legitimacy suffered when he was forced out two years later when the reviled regime of General Ershad was brought down, in part by a civil society-led popular movement, and a democratic system was restored. These days there are more and more calls for NGOs to demonstrate better accountability through measurement of results and impacts. Yet performance measurement is not necessarily an effective response to dealing with organizational failure. Transparency, efficiency and effectiveness are not always prereq-


D. Lewis

uisites for the survival of an organization. Transparency may actually make an organization more vulnerable in this context, and organizations may continue to mobilize resources and carry out activities in ways that do not at all correspond with prevailing notions of economic or technical rationality.

Conclusion: ‘failure’ as process and progress One of the assumptions framing the argument in this chapter is the need to look back as well as forward in development, and learn lessons from cases where things go wrong. Although development policy makers and NGO practitioners may often be reluctant to do so, we need to learn from ‘failure’ as well as from ‘success’. The problems experienced by NGOs mentioned here that led to disastrous outcomes are of course matters of enormous concern and regret both for those who have worked hard to serve their organization, and more importantly still, for people in local communities struggling to improve livelihoods and strengthen rights. But we should not avoid confronting and learning from them, and they should not be forgotten. If we now look back on NGO ‘failure’ in Bangladesh, we learn both about internal management issues, and also about the ways that NGOs are embedded in wider social and institutional processes. A form of hybrid management that combines traditional internal organizational functioning with managing the complex challenges and threats within the external environment – including what are sometimes termed ‘boundary spanning’ roles – is central to any organization’s ability to manage internally and externally, and to its capacity for political agility in dealing with the challenges of predation and capture common in the Bangladesh setting. When leaders fail to manage boundaries with formal politics effectively, organizations implode, often because there is a crisis of legitimacy in which they are no longer able to signal ‘success’, and even if, as Seibel suggests, that ‘success’ may only have been illusionary. The boundary spanning role is made more difficult in resource-scarce, unstable contexts in which elite circulation and elite capture are difficult to separate from NGO organizational life. In addition to the formal organizational characteristics that give functional shape to NGOs such as those found in Bangladesh, we also need to recognize that there is an important role that is played by ‘informality’ in organizations (see also Chapter 7 in this volume). In building successful hybridity, NGO managers deploy informal social and political relationships in pursuit of both organizational and personal ends. Elements of informality are essential to resilience of organizations because they allow them to function in a difficult environment. Managing informality is essential, but as the case study suggests, when this is managed in ways that do not respect the crucial processes reciprocity and hierarchy set out in Brett’s framework, a lack of alignment is produced between an organization and its environment that easily becomes damaging. An ecological approach offers some useful insights. For example, if we look at the NGO sector as a whole, it seems reasonable that some form of selection may take place in helpful ways, with certain organizations falling by the wayside

When things go wrong in NGOs 141 as a result either of internal contradictions, external resource barriers or legitimate forms of regulation. At the same time, Kropotkin’s view of solidarity and evolution within organizational populations may be relevant if we wish to examine further the potential learning and support that may also take place within organizational fields. Clarke and Perrow (1999) are correct in their view that ‘success’ and ‘failure’ are not straightforward labels for organizations, since apparently functioning organizations may be regularly failing in all sorts of important ways. Following from this, failure should be embraced and perhaps even celebrated, because it is an essential element of the operation of any organization – for example in informing processes of organizational learning. Conversely, as Kabeer and Huq (2010) show in their case study of Saptagram, an NGO that had declined and ‘failed’ by the turn of the millennium was nevertheless found to possess a surprising afterlife when it was successfully reconstituted as a smaller, still effective organization by some of the loyal staff and group members whose sense of personal loyalty to the organization and its values has proved enduring. Organizational ‘failure’ may therefore actually be positive for the NGO sector as a whole. First, there are wider lessons that need to be learned from failure, especially if its elements are actively remembered and openly discussed. Following the argument of Rondinelli (1993), NGOs, like development projects, can be seen as ‘policy experiments’ in which for some to succeed others must fail. Second, cases of failure form part of, and contribute further to, the diversity that is essential to the functioning of any country’s third sector. This echoes arguments made by writers such as Kendall (2003) that the third sector should of course be regulated but not too tightly controlled or specified. He refers to the analogy of Henry James’ view of the novel as a ‘loose baggy monster’ – difficult to define, and taking varied forms, but with this lack of specificity a key strength. Third, we can perhaps also see failure in a Darwinian sense, as weeding out less effective organizations in ways that contribute to or allow for progress among other organizations and within the sector as a whole. Fourth, we need to recognize that development is a difficult and slow process, in which long-standing deep-rooted problems are confronted and ‘chipped away at’ by organizations that may only be able to make a small impression. Yet it is important for society to be seen to be addressing key problems. NGOs that are ‘successful failures’ may signal that something is being done, even if the response is as yet an inadequate one. This does not excuse poor performance, but it at least recognizes the difficult challenges that exist. In the end, the capacity of NGOs, or the lack thereof, cannot be reduced to a simple technical problem that requires fixing through new legal frameworks or training in ‘capacity enhancement’. Nor should the increasingly fashionable marketbased or ‘social enterprise’ model of the NGO form be seen as the only viable hope for the future. Instead, hybridity is key – NGOs need a management structure that ensures the stability of internal mechanisms, but at the same time provides the flexibility to respond to opportunities and threats within the wider societal environment. The organizations that make up the NGO sector in Bangladesh are forced to engage


D. Lewis

with forms of unruly politics, uncertainty and informality, but only by moving away from an instrumentalist, technical view of organizations and management towards a more politically informed perspective can we fully understand the usefulness of engaging with cases of breakdown and disintegration as being in its long-term interests.

Acknowledgments Earlier drafts of this chapter benefited greatly from incisive comments provided by two anonymous reviewers, as well as from some very helpful comments from Khushi Kabir. These inputs helped me greatly to lend shape to a rough early draft of this chapter, but the views expressed and any errors contained here remain my own.

Notes 1 One exception to the lack of attention given to engaging with failure in the NGO sector is the ‘Admitting Failure’ website that has been set up by the international NGO Engineers Without Borders at www.admittingfailure.com/browse (accessed 3 August 2012). 2 At the same time, it is noteworthy that there seem to have been surprisingly few highprofile cases of major NGOs running into trouble, given the difficult social, institutional and political contexts in which NGOs operate in Bangladesh. But this picture may of course conceal the less conspicuous cases of smaller organizations that appear and disappear around the country on a regular basis, and this would perhaps be a fruitful topic for further research.

7 The significance of unruly politics in Bangladesh Naomi Hossain

Introduction: why political rule- breaking matters Calls for good governance in Bangladesh invariably remind me of my young cousin’s views on rules: in general, he feels they are important – crucial, even; yet he does not see why they should apply to him. His views on rules were articulated in the context of some trouble with campus authorities, yet they nicely expressed salient features of Bangladeshi popular political culture: faith in the value of laws and formal rules is coupled, apparently without irony, with a strong sense of exceptionalism and routinised rule-breaking by political actors. Thus the desire for order in a fractious setting competes with – and is often overridden by – powerful political interests that cannot be articulated or contained within existing rules. That rule-breaking is characteristic of elite politics is well-known and has been convincingly interpreted within a political economy framework, most notably by Mushtaq Khan; it has also been the recent subject of institutional analyses by the Centre for Policy Dialogue and the Institute of Governance Studies’ State of Governance series, and been explored in donor-commissioned country governance assessments and similar analytical exercises. With Khan’s idea of competitive clientelism to guide our thinking, we now have a strong understanding of the sources of the fragility of the Bangladeshi elite settlement.1 With the gaze of political economy and institutional analyses fixed firmly on the politics of the patron, far less attention has been paid to the politics of those whose position is mainly that of client. In fact, we know less about the contemporary ‘politics of the governed’ than of the immediate post-Independence period, because the village study tradition of class and power analysis that enriched 1970s and 1980s political sociology appears to have now more or less disappeared; the few and limited analyses of popular politics available suggest, however, that clients are not total prisoners.2 This chapter looks at both elite and mass modes of politics. Unruliness – framed variously as informality, fragility etc. – is the substantive governance issue in Bangladesh, and political unruliness is at its nucleus. That rule-breaking powers political action in Bangladesh has been recently reaffirmed by the unprecedentedly large cross-class Shahbag movement protesting the war trials


N. Hossain

verdicts in 2013, and the violent response it provoked by the Jamaat-e-Islami.3 Both the movement and the response are examples of rule-breaking to protest rule-breaking and in order to establish the legitimacy or authority of a different rule. This chapter attempts to explain why Bangladeshi political culture is so dependent on rule- breaking for its power, and implicitly, to reflect on what that means for the ‘good governance’ agenda associated with rule-bound transparent administration.4 The chapter does not address all the possible angles. It approaches these issues mainly out of an interest in popular political culture, and is partly an exercise in conceptual ground-clearing in preparation for comparative empirical research on unruly politics in Bangladesh. ‘Unruly politics’ are a distinctive form of popular contemporary political action, which with colleagues at the Institute of Development Studies I have been thinking about since the global wave of protests and mass movements started in 2008. The core proposition is that recent instances of unruly political action derive their power precisely from their collective rulebreaking and what it means for the workings of power. This is because under particular ‘states of exception’, often threats to the moral economy, otherwise unorganised and relatively powerless people (political society) can find the power within themselves to draw attention to and legitimate their claims by collectively breaking the rules of politics that generally favour the powerful. Many such acts are ignored or repressed. But acts of defiance can momentarily fracture the surface of power relations, creating possibilities of change that were otherwise absent. Many such acts also elicit responses: some course correction for a policy deemed to have gone off the rails; a temporary or partial victory in line with the unruly demands; or transformed rules of engagement with path-changing effects. The paper tests these ideas in the fertile Bangladesh context, taking first a look at the issue of hartal, and second at the garments workers’ revolt, 2006–11. The chapter is organised as follows. The next section sets out some of the motivations for the chapter, sketching the main debates about governance in Bangladesh with which the chapter seeks to engage. The third section sets out in more detail what is meant by ‘unruly politics’. Section 4 applies these to the issue of hartal, taking a preliminary look at how it has changed over time and at its position within political culture. Section 5 applies the analysis to the readymade garments (RMG) workers’ protests, which started in 2006 and continue episodically to the present. The concluding section draws together some thoughts about the contribution of an ‘unruly politics’ frame, and what it means for the analysis of politics and governance in Bangladesh.

Motivations Any analysis of politics and governance in Bangladesh necessarily speaks to the dominant ‘good governance’ debates. This chapter was written in the ‘upside down’ spirit of looking not at the distance between actual conditions and ideal governance norms of transparency, Weberian bureaucracy and predictable, rulebound behaviour, but instead with a relatively open gaze at how actual conditions

The significance of unruly politics 145 govern. Here we briefly review the dominant debates about politics and governance in Bangladesh to show the gaps a stronger understanding of political rulebreaking might fill. 5

The unruly political settlement in Bangladesh A motivation for analysing political rule-breaking is that it may shed light on the conditions of governance more generally, a determining feature of which, it is agreed, is the unruliness or tendency to rupture of the main elite political settlement. The political elites cannot agree to the rules by which political power is shared and distributed, chiefly because the stakes are so high in a winner- takesall system. From this core instability stem various manifestations of weak political governance which have marked the period since 1991: the uncertainty of political transitions, the inability of incumbent parties to stage credible electoral processes, comparatively high levels of political violence, the routine use of hartal (strike) as a key component of political competition, the abuse of official powers to punish political opponents – all arise out of a failure by political parties to develop self-enforcing arrangements to police their own behaviour in relation to electoral competition.6 This failure to make useable rules for elections and political transitions is institutionalised in the caretaker government system, an institution designed precisely to obviate the need for political parties to agree on the rules of democratic transition. It thus ensures that the parties do not develop habits of playing fair, and so never develop the necessary internal institutions. And as the military-backed takeover in 2006 showed, there was no guarantee that the rules of the caretaker administration would be adhered to, either.7 This failure to self-regulate is an institutional failure of the party system; the pressures for victory within a ‘winner-take-all’ system further fracture the watchdog and administrative institutions (the election commission, the upper courts, the civil service) along clientelist-partisan lines. These and other defining characteristics of the Bangladesh political economy arise, we know, from the deeper structures of patron-clientelism. Wood’s earlier characterisation of Bangladeshi political economy as a ‘total institution’ illustrated the vital interests of ‘prisoners’ in retaining their client status in the short term, which he elaborated further as a ‘Faustian bargain’ (Wood 2003). This is linked in turn to the permeability between formal and informal institutions, and that in turn again to the limited legitimacy and authority of formal institutions (Wood 2000). For Khan, the political economy of Bangladesh as a whole can be seen as driven by the interaction of broad-based clientelist political factions amidst changing economic opportunities and contexts. Bangladesh’s particular brand of ‘competitive clientelism’ features short time horizons and an enduring intensity of political antagonism that mitigate against more developmental trade-offs between growth and stability and smooth political transitions:


N. Hossain Compared to the other variants of a clientelist political settlement, the competitive clientelism variant has … the most adverse growth-stability tradeoff. This is both because the permanent presence of significant excluded factions that are always building up their oppositional strength serves to reduce the time horizons of incumbents and focuses their minds on making quick money to fight the next election. In addition, the competitive context increases the power of lower-level factions within the ruling coalition and makes policy implementation difficult and the enforcement of formal institutions relatively poor. And yet, given a large number of powerful primary factions, this may be the only viable type of political settlement in countries like Bangladesh. (Khan 2010a: 79)

Khan shows how and why the settlement fell apart to bring about the extended interim caretaker regime of 2007–09, and why attempts to make political parties operate in more rule-bound ‘good governance’ mode fail (Khan 2010a, 2010b). Competitive clientelism is resilient and protean in this context; many powerful interests are not aligned behind good governance-type reforms. In Wood’s application of North et al.’s (2009) idea of limited access orders to Bangladesh, the games played by political elites successively widen the circle of interested groups to include state and other powerful groups, but not other non-elite groups. The ‘doorstep conditions’ needed to move Bangladesh to a more open access society and polity are not yet in evidence (see Chaper 5 in this volume). Less powerful groups may remain prisoners of the system, but there are signs that clientelist forms no longer exhaust the possibilities of mass political culture and/or mobilisation. An example of this is the more clearly class-based mobilisation by garments workers since 2006 which we will look at below. But while we have a good understanding of the logics of elites within this system, our focus on the base of the patron-client pyramid is more fuzzy. What, politically speaking, does it mean to be a client in this system? How does competitive clientelism shape capacities to organise around interests that go against those of their patrons? From Khan, we know that the political settlement does not only generate resources and power for the elites, but has pay-offs and interactions throughout the lower tiers of the system. The base of the pyramid remains strong mainly to the extent that a critical mass of poorer and less powerful groups are individually incorporated and rewarded within this capillary model of resource distribution. My own earlier work gave evidence of an elite settlement on the basic economic rules in the late 1990s: a minimum consensus on the need for broad-based export-oriented growth to reduce poverty, alongside investment in basic education and health, space for non-governmental organisation or NGO action and improvements in women’s status, and support, strongest among local elites, for social safety nets for the poorest (Hossain 2005). The substance of the elite agreement will have changed with the change in the composition of the Bangladeshi elite over the past two or three decades: it is now a strikingly more private

The significance of unruly politics 147 than public sector elite, compared with the 1980s. And even in the 1990s this was no redistributive or developmentalist agenda but the minimum necessary to sustain the social peace and a modest pace of growth. Yet economic change shifts interests and organisational power, as the Khan framework set out. The elite settlement on the minimal social contract may or may not have changed, but the aspirations and expectations of the people are likely to have done. The period since 2006 has seen fast economic change, growth but also more volatility, with some signs of new forms of political organisation among less powerful groups that do not fit the clientelist logic. The signs of this are emerging, but as yet we do not know how to read them. Understanding ‘the power within’ A second motivation for investigating unruly politics is that one reason we misread or miss the politics of ‘the escapee’ is the limited nature of research into political culture in Bangladesh. This means a profound deafness to the alternative logics and interests that are emerging and being articulated in what could be termed emergent ideological perspectives of proto-escapees. This gap is being filled. Recent comparative research by Simeen Mahmud, Lopita Huq, Naila Kabeer and others have made important contributions to understandings of popular political agency, and the role of NGOs and social mobilization within that. Maheen Sultan and Sohela Nazneen have been working on the women’s movement and women’s political engagement more broadly. Work on wellbeing in Bangladesh as part of the Wellbeing in Developing Countries (WeD) programme also contributes a sense of what is important in political culture. Together these give a far stronger sense of the agency of less powerful groups, and help us to think more clearly about how ‘the power within’ may grow.8 To make the connections between elite and mass politics more meaningful we need concepts that will connect this growing knowledge of the politics of the governed with the more clearly defined elite political settlement. Fortunately, political sociology provides us with an entire battery of concepts that have to date been underused in Bangladesh. Political analysis for Bangladesh needs to shift its focus from agency and structure to state-society interaction, to ask questions like: how do less powerful groups see the state (Corbridge et al. 2005)? Does the growth of governmentality create new spaces for marginal groups to engage (Chatterjee 2004)? When do interests align behind broad-based developmental goals, and what might coalitions of interest look like (Houtzager 2003)? Who contends in the spaces that exist, with what repertoires and to what effect (Tarrow 2011; C. Tilly 2008a)? Mundane and magical power Political rule-breaking may also play a role in course correction, enabling less powerful groups to hold more powerful to a rough kind of account. Overall, we know there needs to be a reasonably functional alignment of interests to


N. Hossain

power for the kind of equilibrium and minimal order implied by a notion of ‘settlement’: The achievement of a social order or political settlement requires in turn formal and informal institutions (such as property rights or informal rules of redistribution) that create benefits in line with the relative power of powerful groups. … [A] political settlement implies an institutional structure that creates benefits for different classes and groups in line with their relative power. The commonsensical understanding of a political settlement as a stable agreement between elites (or a social order) is therefore only likely to be viable if it is underpinned at a deeper level by a viable combination of institutions and a distribution of power between organizationally powerful groups in that society. (Khan 2010a: 20) This is, as Khan points out, a broader point about the bases of political authority: if not at the point of a gun or jackboot, societies need to consent to rule; for that to happen the political and economic structures need to be able to reproduce themselves (Acemoglu and Robinson 2006; Moore 1978). A distinctive feature of Bangladeshi political culture is that political authority is not only in the modern democratic period but also historically mundane and material, lacking the symbolic cultural power of traditional or religious authority (Price 1989). This is a distinctive feature of the clientelist settlement in Bangladesh (Ali and Hossain 2005; Hossain 2005; Khan 2010b; Ruud 2011). As the anthropologist B.K. Jahangir bluntly explains: ‘[i]n the rural areas of Bangladesh the perception of a good government depends on: whether a government is able/unable to feed its people in times of crisis’ (Jahangir 1995: 93). Popular understanding of paternalist responsibilities of political leaders is the basis for all hierarchical relationships, based on high cultural value assigned to virtues of indulgence and benevolence (Greenough 1982, 1983). The legitimacy and authority of those in positions of power derives from how they behave more than who they are, and because power and legitimacy are based not on inherited status or massive wealth but on demonstrating this intention and ability, this needs frequent confirmation. At the base level of delivery of minimal basic necessities, the state acts as the patron across factional divisions (Davis and McGregor 2000; McGregor 1989). This is an imperative – an agreement to provide a base standard of protection to the poor – that has broadly united national and local elites in Bangladesh (Hossain 2005). To some extent, then, a basic level of material security has been the part of the political settlement in Bangladesh that connects the powerful to the poor and powerless.

The significance of unruly politics 149 The Bangladesh paradox Khan reminds us of the central importance of changing economic structures and opportunities in political analysis (Khan 2010a, 2010b). We know that in Bangladesh the past half decade has been a period of both strong growth and surprisingly strong progress on income poverty reduction (BBS 2011). It is not clear that the pace of social and human development progress of the previous decade has been sustained, but there have been promising gains in equity and the more stubborn social problems like maternal health. For all the concerns about its governance failings, the pathways of development in Bangladesh are clearly yielding impressive, measurable gains for poor people and women. This is all the more remarkable because this has also been a period of considerable global economic volatility: this was moderate or possibly lagged compared to many countries in the region and elsewhere (CPD 2012a, 2012b; M. Rahman et al. 2009, 2010; Rahman, Moazzem and Hossain 2009; Rashid et al. 2012). However, standard economic indicators rarely pick up lagged effects (e.g. of nutrition on cognitive development) or those on ‘non-economic’ sectors, e.g. women’s care work (Espey et al. 2010; Heltberg et al. 2012). There are good reasons to believe the period of food and fuel price volatility (2007–11 so far), the global downturn (2008–10), and cyclones in 2009 and 2010 have all had an effect on broader dimensions of wellbeing in Bangladesh (Hossain and McGregor 2011). The political and policy significance of these are generally lost, particularly to aid donors, who increasingly fixate on big numbers to tell them about people’s realities. To understand and analyse the broader social impacts of global economic volatility, wellbeing-based approaches appear to be the most promising methodologically and conceptually (Camfield et al. 2009; McGregor et al. 2009; White 2009). It is in this context of important but unmeasured social costs of economic change that what my colleague Alex Shankland terms ‘unruly politics’ appears to become more important in shifting power towards the poor. As this chapter will try to show, the garments workers’ revolt is an example of why unruly politics shifts power to the powerless in a context of global economic turbulence. It shows that it is not only powerful groups who can use rule-breaking to their advantage. But to what extent and under what conditions are unruly forms of politics empowering for poor and powerless people – particularly for women – in contemporary Bangladesh? And given the importance of formal civil society in Bangladesh’s development to date, what role does it play in taking on and standing alongside such struggles? In 2006 the State of Governance report noted that: [T]he recent explosion of ‘unruly civil society’, particularly militant Islamist organisations and spontaneous single-issue protest groups (power sector, RMG workers), strongly suggests a gap between what NGOs and CSOs have to offer, and what people appear to respond to and need. (IGS 2006: 99)


N. Hossain

The Shahbag movement is yet another instance of ‘unruly civil society’. The motivation for focusing on these issues here is that closer examination of phenomena such as popular protests and industrial riots may provide conceptual handles for some of the mysteries of Bangladesh’s development, including why Bangladesh succeeds in income poverty reduction when its disorderly and unpredictable governance institutions appear to create barriers to this. In examining these issues, this chapter extends some of my earlier work on the importance of ‘rude accountability’, or the idea that poor and powerless people, particularly women, rely on informal strategies to hold public officials to account, including manipulation of status and reputation and the vague threat of the angry crowd. These informal pressures are comparatively effective in a context in which formal mechanisms of public accountability systematically fail poor powerless people (particularly women) and wholesale public sector reform is painfully slow (Hossain 2010). Just as poor and powerless people use shame and fear relatively effectively to ensure public service delivery where there is little scope for more formal accountability mechanisms, a culture of political unruliness may create more space for marginalised political groups to mobilise in a context in which formal political channels preclude this. As with the ‘rude’ mechanisms of accountability on which the poor and powerless rely, unruly politics are risky politics. But they may afford an alternative channel of effective political expression for groups whose concerns otherwise lack a sufficiently loud voice.

Some ideas about unruly politics9 ‘Unruly politics’ is a set of ideas about what is so distinctive and potent about the wave of protests and revolts since 2009. These have been taking place in the global public sphere of the internet and against a backdrop of seismic volatility in financial and commodity markets and associated political crises. A group of colleagues at the Institute of Development Studies came together over a shared interest in these issues, as researchers, activists or both in what Partha Chatterjee calls ‘the politics of the governed’ – the struggles of marginalised groups to secure rights and resist oppression, beyond the institutionalised public spaces of governance and civil society. Individually, we have had access to parts of those localised political struggles in Angola, Bangladesh, Brazil, Egypt, Ethiopia, India, Kenya, Mozambique and Yemen. Our entry points varied from public services, recognition of cultural or sexual identity, food price spikes, freedom from violence, to the basic democratic right to choose leaders. We shared an epistemological bent towards exploring these issues from the perspectives of the relatively powerless – poor people, minority groups, women, slum-dwellers, sexual dissidents, workers and their own political ideologies and organisational strategies. For the purposes of this chapter, our discussions about these twenty- first century protest phenomena identified four dimensions of this wave of protests and revolts that made them distinctive: (1) timing, or historic specificity; (2) form; (3) content or ideology; (4) agency or the actors involved.10

The significance of unruly politics 151 Timing/historical specificity The concentration of unruly political episodes and events gave rise to a question about timing: was there something distinctive about the present moment, i.e. the early twenty- first century, that was making this happen? The conclusion most observers seemed to have reached was that the political moment grew out of the deep structural crisis in global capitalism, specifically its fast- growing inequalities over the recent past, sharp economic shocks with uneven impacts and new waves of connection and communication between groups and across contexts, enabled by ICTs (information and communication technologies) and social networks.11 Our joint thinking was initially triggered by the wave of popular unrest during the 2008 commodity price spike. These ‘food riots’ (more properly also fuel riots) drew our immediate attention because they were so widespread, gave the appearance of spontaneity and often seemed to succeed in sending a message that was both loud and clear, which resulted in some action by government (Arezki and Bruckner 2011; Arora et al. 2011; Berazneva and Lee 2011; Brinkman and Hendrix 2011; Bush 2010; Schneider 2008). At the same time, food riots were generally framed (e.g. in the media) as the natural defensive spasms of hungry people, and not as properly political acts (Hossain 2009). The 2008 wave of food riots was followed by other public protests around the time of the 2009 global financial crash, and then by ‘austerity protests’ against the public sector cuts that followed in 2010 and 2011, particularly in Europe. The 2011 Global Peace Index, which measures ‘domestic and international conflict, safety and security in society and militarisation in 153 countries’ registered ‘overall score increases for several indicators, the largest of which were in the potential for terrorist acts and the likelihood of violent demonstrations’ (IEP 2011: 3). After a protest- filled winter of 2010–11, UK observers started talk of a ‘new protest politics’. Time Magazine famously declared ‘The Protestor’ as the Person of the Year. Many of the protests looked and felt qualitatively different from those in the immediate past, drawing in a new generation of protesters as well as new techniques of protest (e.g. using SMS and social networks like Twitter and Facebook), and reflecting a conscious sense of originality thanks to the power of the modern forces that have propelled its birth. These give credibility to its double wager of defiance: that what the state, the government, and the corporate media offer to the country and especially its young as our fate is unacceptable, and that the claim which accompanies it, that there is no viable alternative, is a lie. (Barnett 2011: 14–15) It is possible that as Barnett suggests, aspects of these protests mark a new repertoire of contentious performances (C. Tilly 2008a), but it is primarily their apparent contagiousness that makes them stand out. In its synchronicity, the


N. Hossain

early twenty- first century wave of protests looks very like the contemporary version of the ‘structural adjustment riots’ of the 1970s and 1980s: One effect of greater integration and synchronicity within the world economy is a greater degree of simultaneity and similarity in what might broadly be referred to as ‘class struggles’ around the processes of capital accumulation and around government measures to resolve crises in that process and to promote renewed accumulation. It becomes less surprising then that a ‘wave’ of popular protest accompanies the process of structural adjustment and government austerity measures as it takes places across the developing world (and even the developed world) during the last two decades.… It is the crisis and process of reconstruction of global capitalism that began in the late 1960s and early 1970s that has generated the economic policies of liberalization and austerity that have themselves given rise to the upsurge of popular unrest across the developing world. (Walton and Seddon 2008: 21–22) At first, the Arab uprising seemed to be part of the family of protests relating to economic crisis and fiscal adjustment, with the price of bread explicitly raised in Egypt, Algeria, Tunisia and Yemen, among other countries. Later in 2011 it became clear that what the structural adjustment protests of the 1970s and 1980s were to the food, crisis and austerity protests of 2008–11, the collapse of Communism and the break-up of the Soviet Union was to the Arab Spring. These went beyond a complaint against economic policy into fundamental political upheavals, fought on the grounds of basic liberties. That there were such grounds on which to stage a revolution was clear in the Arab context. One member of our unruly politics group, Mariz Tadros, noted that two things needed to happen to enable the Arab uprising: (1) the fear barrier had to be broken, including fear of the brutal security forces; and (2) citizens had to reach a breaking point, in which the cumulative impact of injustice would make people feel that there was nothing to lose.12 Another way of putting this is that people had to become conscious of ‘the power within’. Perry Anderson took the view that ‘volcanic social pressures’ built up, to be freed by the political fissure created by the defiant self- immolation of Mohamed Bouazizi: The odious cast of the regimes in place has persisted unaltered for decades, without triggering mass revolts against them. The timing of the uprisings is not to be explained by their aims. Nor can it plausibly be attributed just to novel channels of communication: the reach of Al-Jazeera, the arrival of Facebook or Twitter have facilitated but could not have founded a new spirit of insurgency. The single spark that started the prairie fire suggests the answer. Everything began with the death in despair of a pauperized vegetable vendor, in a small provincial town in the hinterland of Tunisia. Beneath

The significance of unruly politics 153 the commotion now shaking the Arab world have been volcanic social pressures: polarization of incomes, rising food prices, lack of dwellings, massive unemployment of educated – and uneducated – youth, amid a demographic pyramid without parallel in the world. In few other regions is the underlying crisis of society so acute, nor the lack of any credible model of development, capable of integrating new generations, so plain. (Anderson 2011: 10) The baguette waving and bread-helmets of protestors were unforgettable, triggering a barrage of debate about the connections between food prices and protests. Scientific American asked: ‘Are high food prices triggering revolution in Egypt?’13 Barings economists updated a historical analysis of bread prices and revolutions to find a neat correlation between food prices and the Arab Spring. The BBC’s Paul Mason explained: There were, undoubtedly, deeper seated causes to the Jasmine Wave than the price of a loaf of bread. But to paraphrase Berger and Spoerer’s [the historians from the initial study] investigation into the economic causes of the revolutions that rocked Europe in 1848, while food inflation does not provide the brains, it does supply the brawn.14 This view is of food riots as ‘natural’ responses to threats to basics needs, or what Thompson called a ‘spasmodic’ view of riot (Thompson 1971). Within the unruly politics group, our view was that while the economic crisis provided the conditions of the moment, political acts such as these are inadequately explained as physiological reactions to deprivation. More properly political explanations included that food and austerity riots are part of a repertoire of contentious politics with a distinct moral justification, functionally separate from more profound revolutionary protests. Food riots are familiar to the governing classes, who know their texts and frequently respond accordingly (with export bans, subsidies, price controls). For the protestors in Egypt, Tunisia and elsewhere, perhaps registering their defiance in this more rehearsed form meant breaking Tadros’ ‘fear factor’, even while the sense of growing hardship contributed to the sense that there was ‘nothing more to lose’. In other words, the question of timing and historical specificity was resolved, but required us to look more closely at the form and the content of the protests, as well as at who was protesting. When are politics ‘unruly’? Forms and frames There are several ways in which the forms taken by unruly politics shift our thinking about the politics of the governed. First, there are implications for thinking about civil society and citizenship, and the work of NGOs, coalitions and social movements, and other aid programming related to governance and politics. The normative preference for civil, ruly power-sharing typical of good


N. Hossain

governance programming is at odds with rude, rule-breaking protests. Might powerless people have to break the rules if they are to gain power? Given that formal invited participatory spaces often fail to empower (Cornwall and Coelho 2007), how ‘civil’ can shifts in power ever really be? In a ‘results’-focused aid industry, we need also to grapple with why unruly forms of politics so often seem to deliver ‘results’ for poor people. We can safely say that unruly politics evade or are at most on the periphery of what is known as ‘civil society’ in developing countries, away from formal seats at the policy table and at a more frictional outer edge. Focusing on unruly politics is part of the shift in focus from civil back to political society (Chatterjee 2004; Houtzager 2003). A second aspect of the form of unruly politics that matters conceptually and methodologically is their interactive nature. What we have termed unruly politics are a contemporary sub-species of contentious politics, which is itself defined as the interactive making of claims that call for actions that would affect one party’s interests (for better or worse), and including the concepts of events, performance and repertoire (Tarrow 2011; C. Tilly 2008a). From the analyses of food riots in European history (Bohstedt 2010; Thompson 1971; L.A. Tilly 1971), it was clear that riots were potent to the extent that they claimed action by public authorities, and in doing so drew on shared or at least legitimate understandings about entitlements and responsibilities (or the ‘moral economy’). The significance of this point is that unruly politics must be situated within relationships – people do not merely erupt when denied affordable bread. The third and most visible aspect of the forms of these politics is their disruptive quality. This is in many ways their most defining quality: that they break, dismiss or challenge laws, rules or norms of political behaviour and practice. At their most extreme they include violence and vandalism and the ever-present threat of ‘the crowd’ (Rudé 1981), elsewhere the public shaming and criticism as a classic ‘weapon of the weak’ (J. C. Scott 1977, 1987, 1992). Humour appears to have made a welcome return to the political activist scene during this unruly period. Possibly the new means through which images and ‘memes’ (such as the bread- head man of Cairo) can be instantly flashed around the world or groups convened flashmob- style lend themselves to creativity in protest styles. The desire for online audiences, particularly in the western country protests, means that comedic defiance is not only increasingly theatrical but also a potent means of defying power. Khanna retells an encounter between the police and the Clandestine Insurgent Rebel Clown Army, a group whose aim is to ‘make clowning dangerous again’ during a UK anti-austerity protest. During an episode of ‘kettling’ (a police encircling containment technique), a senior officer approached the lead clown for ‘a serious word’. The clown reportedly responded: ‘Encyclopaedia: is that a serious word?’ Khanna interprets this as disruptive of established grammars of politics and political behaviour. In southern Europe the indignados, and in the various Occupy movements, theatre, art and spectacle were all good humouredly used to break the older molds of politics (Khanna 2012). This defiance is the kernel of a possibility of a shift in power, a potential turning point. In Domination and the Arts of Resistance, Scott

The significance of unruly politics 155 (1992: 202) writes of a moment of ‘charged political impact’ when ‘the frontier between the hidden and the public transcripts’ is breached. What such defiance actually achieves is a question for empirical investigation. It is clear that in the contemporary period, unruly political acts tilt the power balance – possibly temporarily – but introduce something new into the relationship between ruler and ruled. The classic response to defiant acts is hostility and repression, but If the withdrawal of acceptance, cooperation, and obedience can be maintained in the face of the rulers’ punishments, then the end of the regime is in sight. Thus, all rulers are dependent for their positions and political power upon the cooperation of their subjects. The theory that power derives from violence, and that victory necessarily goes to the side with the greater capacity for violence, is false. Instead, the will to defy and the capacity to resist become central. (Sharp 1994: 6) It was through the withdrawal of cooperation opened by the initial demonstrations, and the recognition that first the Tunisian and then the Egyptian state were unable to face down widespread defiance that revolutions there became possible. It is possible too that the unruly political act is defined by its denial of the legitimacy of public authority, a temporary disruption of power. For akshay khanna, form is content: disagreement with established modes of doing politics expresses rejection of the institutions and practices of politics in general (Khanna 2012). Fourth, and despite their disruptive qualities, unruly political acts come within a range of repertoires, which change relatively slowly. These include how claims or protests are voiced, with which objects and with which language, The use of social media, first properly noted in the Iranian Green Movement in 2009, seems to mark a paradigm shift in repertoire through new means of organising and tools for communicating about the protests through Twitter, Facebook etc., which may have drawn in new actors and debates. However, while it is the slightly shocking nature of the individual unruly political act that gives it its power, it only really takes on recognisably political characteristics within wellworn repertoires of action, which innovate slowly and iteratively over similar actions (C. Tilly 2008a). Content: moral economies and laws of necessity These fragmented responsive events often look as though they share a loose underlying bare-bones political logic – a common ideological thread about the responsibilities of a reasonable state (Apter 1964). If so, what might that common thread include? One possibility is that popular protest in an era of economic crisis and adjustment shares a political logic with the claims of a ‘moral economy’ – a consensus about the right to adequate provision of basic needs, in contrast to the claims of


N. Hossain

the market economy, which was claimed at particular historical moments when these were most threatened (Bohstedt 1992, 2010; Thompson 1971). The idea of a moral economy may be an invented tradition, in that such ideas were rarely explicitly articulated until people’s entitlements to affordable food were threatened by policies that prioritised rights to profit over rights to consume. It would be interesting to explore whether an implicit or ‘natural right’ is invoked in contemporary protests. Women were often prominent protestors, reflecting both their roles in provisioning and their lack of political voice (J. Bohstedt 1988; J.W. Scott and Tilly 1975; Taylor 1996; L.A. Tilly 1987). During periods of economic adjustment, how governments responded to such mobilisation could be constitutive of public authority, legitimacy and order (Bohstedt 2010; L.A. Tilly 1983; Walton and Seddon 2008; also Acemoglu and Robinson 2006; Berger and Spoerer 2001). The political logic for political unruliness was in the past and may now be based to some degree on a ‘law of necessity’ (Bohstedt 2010), or the idea that threats to basic human needs are prior to all other laws. This idea of a law of necessity, rooted in (interpretations of ) people’s basic needs, does not entail accepting a hardship or ‘spasmodic’ view of the material bases of protest as the natural, physiological revolt of the hungry angry masses. As Bohstedt (2010: 9) notes of English food riots in the sixteenth to nineteenth centuries: Certainly hunger, combined with moral outrage over profiteering from deaths, powered popular protests. But there is a nearly-universal core to the ‘moral economy’ of acute hunger; it did not depend simply on the English precedents and paternalist regulations marshalled in Thompson’s essay on the ‘moral economy’. Found in many different polities and periods, that core is the ‘law of necessity’; in emergencies, when survival hangs in the balance, human subsistence must take precedence over property rights. Unruly politics seem to derive their distinctive quality and power from the contradictions between bodies – people’s basic material being – and the law. In Bohstedt’s law of necessity, the claims of human subsistence trump legal property title, the very foundation of the capitalist order. He calls it a ‘law’ but it is really its opposite: a situation in which the law is suspended and a natural right is substituted for a man-made technology of rule. The relationship between bodies and the law parallels Agamben’s (2005) ‘state of exception’, which is a time or space in which the law is suspended, through the power of the sovereign to declare it such. For Agamben, the state of exception creates an increasingly indistinct zone between the properly political realm and that of ‘bare life’, or life beyond the scope of the law; the growth of modern state power means the convergence of the techniques of government or policies and state practices on the one hand, and the ‘technologies of the self ’ through which people are disciplined, forming their identities and sense of selves. In a separate discussion of exceptionalism, Chatterjee explains that modern or Western political theory’s powerful ‘norm-exception construct’

The significance of unruly politics 157 enabled it to reconcile, for example, liberal democratic principles with practices of imperial rule. The grounds for declaring the exception here derive from ‘the empirical deviation of prevailing institutions and practices’ from a universally desirable norm. The scope for ‘improving’ the population, the White Man’s Burden, marked an empirical deviation worthy of the exception (Chatterjee 2011: 9). It is beyond the present scope to explore the implications of bio-politics and exceptionalism for political unruliness in detail, but there are several ways in which bodies and exceptions to the law meet in unruly politics (but see Khanna 2012). Instead of thinking about politics as laws and institutions, it helps to see it as the concrete practices through which bodies are disciplined (Agamben 1998). To do so sheds light on the content of unruly politics in several ways. First, because its modes are directly confrontational: bodies in the wrong places and times, acting in ways that break the disciplines of the classroom or factory. The rejection of specific disciplines makes the acts so potent – as noted above, the medium is the message. But unruly political acts are also a claim to sovereign power, to the extent that they lay claim to declare their own ‘state of exception’. This is very much the case with the hartal, as we will see below. And unruly politics do so by reference to the politics of bare life, a space beyond, but perhaps therefore also above, the law. Unruly actors If the basic unit is the event, this distinguishes what we are looking at from a focus on the activist groups more characteristic of social movement analysis. Unruly politics in which we are interested are more diverse than social movements, less continuous and more patchily connected. Who are these unruly politicians? Are they part of Guy Standing’s (2011) ‘precariat’, chronically insecure low-income workers comprising ‘the new dangerous class’? Methodologically, we need to ask: what kinds of occupations, communities and networks do they belong to, what brings them together, who constitute the ‘faces in the crowd’ (Rudé 1981: 11)? Chatterjee’s concept of political society helps to distinguish from the activism of a civil society that presumes equitable standing and full citizenship: The rural poor who mobilize to claim the benefits of various governmental programs do not do so as members of civil society. To effectively direct those benefits toward them, they must succeed in applying the right pressure at the right places in the governmental machinery. This would frequently mean the bending or stretching of rules, because existing procedures have historically worked to exclude or marginalise them. They must, therefore, succeed in mobilizing population groups to produce a local political consensus that can effectively work against the distribution of power in society as a whole. This possibility is opened up by the working of political society. When school teachers gain the trust of the rural community … they do not


N. Hossain embody the trust generated among equal members of a civic community. On the contrary, they mediate between domains that are differentiated by deep and historically entrenched inequalities of power. They mediate between those who govern and those who are governed. (Chatterjee 2004: 66)

The defiance of unruly politicians distances them from the rule- bound civil society engagement promoted by aid donors, although there may be some overlaps in individual actors. When political society succeeds, Chatterjee suggests it is more transformative of poor people’s lives than civil society organisation. Might unruly acts emerge from the failures of civil society to represent the classes of ‘the governed’?

Unruly politics in Bangladesh With this necessarily long conceptual preamble out of the way, we now look at some patterns of political rule-breaking in Bangladesh, to test these ideas empirically. The slow evolution of the hartal repertoire The contemporary Bangladeshi hartal lacks many of the qualities of unruly politics as set out above: these days, hartal is not the defiant political act of the governed, but the routine mode of competition by big powerful parties. Yet it has not always been that way: hartal has a long and honourable tradition of earning powerless groups power, famously in the Gandhi model through non-violent means. Although nowhere else is it as ubiquitous as in Bangladesh, it is still an institution within South Asian politics. In its original repertoire, hartal worked because it threatened public authority on moral grounds, not because of the economic threat or the violence it mobilised to ensure fear and the semblance of popular support. There has been little sustained public discourse about the phenomenon since the UNDP’s 2005 Beyond Hartals study. An internet search for literature on hartal in Bangladesh found few scholarly analyses of the phenomenon, yet it is widely and expertly discussed in the blogosphere and in social media.15 Moniruzzaman (2009) found that the use of the hartal has increased over time (see Figure 7.1). Hartal has thus clearly been a potent form of a specifically anti- colonial mode of unruly politics, most notably since the Civil Disobedience movement against the Raj in 1930s’ India. In more recent times, the long hartal campaigns of the democratic movement in the late 1980s and the mid-1990s campaign to institutionalise the caretaker government system both had credible claims to a higher purpose in their attempts to restore democratic order. Yet in its contemporary adoption by political parties within competition-asusual, its moral claims to be asserting the law of necessity have been lost, as are its claims to be subverting or defying public authority. Without a credible claim

The significance of unruly politics 159 350

Number of hartals

300 250 200 150 100 50

2 00



–9 19


4 –9

95 19

0 91 19


–9 87

–8 19

2 –8

83 19

8 79 19


–7 75

–7 19

0 71

–7 19

6 –6

67 19

2 63 19


–6 59

–5 19

4 –5

55 19


47 19





Figure 7.1 The number of hartals in democratic and military regimes, 1947–2002. Source: Moniruzzaman (2009), based on analysis of various sources. Notes The dark grey columns denote democratic regimes, light grey denote military periods (the categorisation is from Moniruzzaman 2009).

to be in defence of the public interest, the hartal is now widely recognised as the routine performance of the political opposition, and indeed the repertoire of politics for all major parties. The gap between the hartal’s moral origins and its venal political uses in the present is filled by a lingering attachment by both parties to hartal as a mode of political action. It has been a successful tactic in the past. Their party structures and cultures are aligned to deliver an effective hartal, particularly their youth and students wings, creating strong internal institutional imperatives to stick with it. But there is wide ambivalence about how it is used. In an article about a July 2011 hartal, the Prime Minister Sheikh Hasina vacillates over the issue of hartal per se: she wants to protect it as a means of protest in the public interest, but also to decry its use as a crude political tactic.16 The article sheds valuable light on the political mind of the most important politician in the country. The Prime Minister argues that the hartal called for the next day lacks moral purpose in that it is not ‘in the people’s interests’, but in the particular and close family interests of the leader of the opposition. She goes on to make two additional points of interest: first, that the people need to respond if the movement is to work – else the violence of ‘damaging or torching a few vehicles’ will be pointless. Second, she rounds off her statement by listing her government’s achievements in food security and similar policy areas, presumably to signal her


N. Hossain

expectation that the people will not respond to the hartal because they are well cared for. Thus a failure to protect the moral economy is anticipated – and rejected as groundless – as a justification for unruly politics. In the absence of robust quantitative analysis of the repertoire over time we cannot measure the change, but by any qualitative assessment, the objectives and role of hartal in articulating political opposition have changed over time.17 The form has been notably more common during ‘political’ or multi-party elected regimes, than under military or unelected rule in the years since independence, according to analysis by Moniruzzaman (2009) (see Figure 7.1). This is a pattern likely to have been continued in the years since the early 2000s, because the caretaker government of 2006–08 banned most protests, whereas the ‘democratic’ periods before and since have been marked by a growing ferocity in routine politics.18 As they tend to close down businesses and much of the urban economy, hartals tend to be costly for the Bangladesh economy. Estimates suggest 3–5 per cent of GDP was lost to hartal annually in the 1990s (Khundker 2005). Less attention tends to be paid to the human experience of hartal: unpredictability and disruption to work, school and travel; acute hardship for some; boredom and frustration for all. It is not surprising that hartals are unpopular. What is of interest is the extent to which Bangladeshis nevertheless think it is important to protect the right to call hartals. The UNDP perceptions survey in 2003 found that while 95 per cent of respondents thought hartals were very or somewhat damaging for the economy, 73 per cent nevertheless thought they were a legitimate means of protest. More recent discussions in the blogosphere suggest a similar ambivalence, but also a waning patience with political party antics. Below is a selection of responses to a 2010 article by Sazid Khan on ‘Does Hartal Deliver in Democracy?’ on the London-based e-Bangladesh portal www.ebangladesh.com/2867. I found these to be of particular interest because they were posted around the time of the hartal itself and so give a sense of the frustrations in real-time. In some respects the hartal has been institutionalised so as to be a set piece in the theatre of political competition. Suykens and Islam (2013) see the political performance of hartal as enabling local leaders to showcase their power: the idea of audience is critical to this performance, and in their analysis of its micro effects, the hartal is a success, which is also why it is such a resilient aspect of the political repertoire. In its form it now uses fear and violence more than the suspension of authority – the creation of a ‘state of exception’ – to achieve its ends. The apparently successful use of hartal by Islamic groups protesting the National Women’s Act earlier in 2011 suggests that new actors are adopting the model to contest issues around which they are able to mobilise sufficiently large support, where the hartal had traditionally been the domain of the big parties. In 2013, the Jamaate-Islami again called a series of hartals against life sentences received by their leaders for 1971 war crimes. These make the interesting political argument that the state was wrong to interfere with judicial process – in effect, that the law

The significance of unruly politics 161 must be adhered to. Yet the Jamaat hartals were themselves the response to the Shahbag movement which also demanded the law be properly adhered to and that convicted criminals receive the full punishment of the law. The Shahbag protestors were also certain that the judicial process had been interfered with to assure a lenient verdict, and that should convicted war criminals not be hanged, the verdicts would be overturned with the next government. It should be a matter of great interest to Bangladeshi political sociology that it is through this mélange of demand for rules and rule-breaking that our politics are performed. There is a sense that people are less willing to comply than in the past, and certainly some – such as a 2011 BNP-sponsored strike about the rising price of fuel and food – have failed. Success in this context is judged by the extent of disruption to everyday and particularly economic life – failure would imply little disruption. That people failed to strike against high food prices is surprising given that food riots have been frequent in this context; yet it also reflects the fact that in times of rising prices, people must work, and harder than before: there can be little sympathy with a political strategy that hurts the poorest first and foremost. Until the Jamaat started to use it more routinely, the hartal had appeared to be in its dying days as part of a contentious political repertoire. A repeat of the 2003 UNDP survey on perceptions of hartal may well now reveal a degree of unpopularity that could persuade the parties to settle the matter and restrict its use. But the hartal has had effects in the past, and is part of the political culture of oppositional all-out hostility between the parties. The tactics are widely known and may inform those of other actors and popular mobilisation. So it remains interesting to think about why such a repertoire persists within more routine politics, and how the underlying logic of hartal might change. Garments workers’ protests A campaign for higher wages (and other improvements in labour rights) erupted in the garments industry in 2006, and has persisted even after the agreement of an almost doubling of the minimum wage in 2010. On this issue, there is relatively little empirical data on which to conduct a meaningful analysis. To date, there has been no notable effort to study the issue as an instance of political mobilisation. For this section, I will draw on existing literature and background material gathered for a paper on garment employment as women’s empowerment (Hossain 2012b). This is another instance of contentious politics that has received little serious attention. The RMG sector has seen an increase in the size, seriousness and frequency of protests, typically violent, in Dhaka and the Ashulia-Savar belt since 2006. These became particularly long-running and violent campaigns in 2006 and then again in 2010. Even the agreement to almost double the minimum wage rate, which took place in early 2011, failed to fully defuse the tension, as many workers’ groups continue to demand higher wages still. A catalogue of the protests that covers the full period does not yet exist, and the struggle does not yet appear to have been fully resolved; what follows


N. Hossain

depends on the results of a simple keyword search of the Daily Star news archive and secondary sources (cited in Hossain 2012b).19 At the time of writing, we lack basic facts about the mobilization in its entirety, which prevents us from gaining a sense of its scale and dynamics. During a seven-month period in July 2010, 340 people were injured, significant investments in plant and vehicles were damaged and substantial volumes of production and transport were disrupted (Daily Star, ‘Contemplating sustainable solutions to garments sector unrest’, 10 July 2010). For the remainder of the year, between August 2010 and January 2011, and using a different metric, Rahaman et al. (2011) record a further 30 disputes or incidents of labour unrest in the garments industry. From these and other newspaper accounts, the issues at stake revolved around unfair labour practices and the wider issue of low wages, but were in many instances sparked off by particular factory-level events – the closure of a unit without notice or payment of back- wages; abuses on the factory floor; late wages; unfair dismissals, etc. A rapid review of Bangladesh news articles of other RMG protests since 2006 suggests this is a fair sample of the points of contention. Place the garments workers’ protests within their global and historical context, however, and things look different. A new study of food riots and rightto-food movements in 2007–12 in four countries (including Bangladesh) has provided an opportunity to start to explore the garments revolt in more detail and in comparative perspective. The international press has a different take: what were industrial disputes in the Bangladesh media were often – and somewhat lazily – labeled ‘food riots’ by the international press, as these were taking place at a time when similar cost of living-related unruly political events were erupting around the world. It may be telling that references made in the Bangladesh media to ‘food riots’ at this time were mainly to events in other countries, including India and Nepal; our Indian research partners tell us that similarly while the Indian press reported no food riots at home, it dwelt at length on the Bangladesh ‘food riots’. Two points are worth noting: first, that a food riot seems to be something that happens elsewhere. But this is probably because, second, looked at closely, the contexts, the motives and the actors involved in unruly acts are more complex than the term ‘food riot’ appears to allow for. In the case of the garments revolt, the protests often clearly articulated food price rises as a vital concern; this reflects an accurate reading of the protestors’ object: to raise their wages, not to argue (no doubt pointlessly) for lower food prices. The events catalogue on which this preliminary analysis of the garments revolt draws is a work in progress, yet it is already clear that the article is unusual in the amount of space it gives to protestors’ views, and in how it certifies their concerns with ‘expert’ evidence. The elite view of the garments revolt, it should be noted, is of a conspiracy by India to destroy Bangladesh’s RMG sector in order to boost its own market share. The idea that external forces are stirring up the masses is common enough among elites who are reluctant to accept that their modes of rule may be unpopular (see Rudé 1981 on this during the French Revolution). There is rather more middle-class sympathy with garments workers, by contrast.

The significance of unruly politics 163 The garments workers were initially contending with the industry owners to secure the basics: better working conditions and higher wages, at a time of rapidly increasing food and other prices. As the conflict spread in the flagship economic sector, it became a national law and order priority. The minimum wage agreement reached in 2011 marks a certification of the problem (C. Tilly 2008a): the state recognised that there was a problem here, which only it was capable of addressing. This has also meant a move towards more formal recognition of the need for worker representation. Several questions arise about the garments revolt as an instance of unruly politics. One is that of form: why did the struggle become so violent? Contentious politics offer many opportunities for conflict, but the level of violence used on both sides and the security responses have been both polarising and painful. Indisputably, one of the key problems the garments workers face is outright hostility to labour organisation. This is one of the lessons of the post-Independence state, and public sector service delivery continues to feel the pain of overpowerful union interests among doctors, teachers and civil servants. The absence of any means of incorporating the interests of factory workers virtually ensures an unruly response from this group. The intensity of the struggle also reflects the historical moment and the nature of the actors: RMG workers have been fully exposed to the perfect storm of global economic volatility that struck in 2007, bearing the brunt of food price inflation, economic downturn and ‘flexibilisation’. They are also members of Guy Standing’s precariat. The high level of violence also expresses defiance. In the absence of any rights to representation, workers lacked legitimate and effective means of articulating a powerful sense of grievance. Blocking traffic and closing factories suggest the logic of the moral economy at work here: basic material security must be guaranteed as a precondition for political authority. These acts may not be intended as destructive but to correct public authority when it fails to ensure a fair share for all. Possibly it is the shared moral sense of when it is right to protest – which appears to have been present during the revolt – which stimulates ‘the power within’ and provides a shared target for its action. As noted above, public discourse suggests that middle-class people were comparatively sympathetic with the garments revolt, even though they widely decried the violence and vandalism. This may be because the middle class have also felt the pain of recent inflation, and can also see the fast- growing gap between factory owners and the rest of Bangladesh. It is possible that middleclass sympathies reflected the moral power of organising around rights to food in this context, and a recognition of ‘the law of necessity’. This long period of political contention will have shifted power relations, but labour remains poorly organised, repressed and excluded from policy spaces and relationships of power. Yet there is agreement that the workers won key concessions through their struggle, and so have more power than before. One area in which the workers have made progress is in shifting the perception of them as docile. Men workers are thought to have played an increasingly important role in


N. Hossain

protests as the knitwear and sweater sub-sectors grew and more men were hired; however, women workers were very prominent, and the images of women workers fighting with the police were seen all around the world.20 To conclude the discussion of the garments workers’ revolt, we refer back to the Khan framework of competitive clientelism. The object of the workers’ contention – wages set by factory owners and regulated by the state – meant their political target was spread across the political factions, and so was largely separate from party competition. Yet garment factory owners are numerous on both sides of parliament, and better represented than any other profession. There have been factional and party connections to the workers’ groups, and suspicion that external agents stir up worker discontent. But the garments workers’ revolt is the most significant class- based mobilisation seen in Bangladesh for some time, and likely to impact on the way politics is popularly conceived and staged. The political settlement framework also sheds light on the global sources of the garments workers’ revolt. Workers riot partly because world food prices and export volumes fluctuate. The government of Bangladesh is generally responsive on acute food insecurity and responds generally better than in many places. But its sphere of action is limited when global food and fuel prices soar or global markets for Bangladeshi garments shrink (or grow). So the global economy intrudes into the moral economy of the nation. This suggests that the ultimate object of the worker protests is or should be at least partly at the global level (the UN, the G20?). The political settlement is destabilised here by the limited strength of national rule over global economic forces.

Conclusions: what unruliness means for the governance agenda This chapter asked: why is politics so indisciplined – violent, lawless, irregular, defiant – in Bangladesh? What, if anything, does it matter? The why is answered partly with the help of recent political economy analysis which has clarified the roots of the fragile political settlement in Bangladesh: understanding the unruly political culture should shed light on the core problems of elite politics. These issues have been amply documented and are widely understood. We know less of what political unruliness means to struggles by the non-elite. But there are good historical reasons to believe that defiance can be a powerful weapon in the political armoury of the governed. This contribution set out some ideas about what is interesting and distinctive about the unruly politics that have featured across the world since 2007. All of this matters because political rule- breaking ultimately defines the governance problem in Bangladesh. Over-reliance on the destructive unpopular hartal in routine party politics raises the political temperature; violence elicits repression and ratchets the game up. The repertoire of the hartal – the objects and tools of contention – has changed from a strategy of public interest and civil disobedience, to a tactic of holding the nation hostage to poorly concealed sectional interests. Yet popular political culture is ambivalent: hartal is deeply and widely

The significance of unruly politics 165 unpopular in fact. But in theory, there is a desire to protect the right to defy authority, learned, no doubt, from the protracted and repeated bouts of unelected rule. There is a shared suspicion of a form of politics in which defiance is not permitted. Popular unruliness may also correct public policy in minimally pro-poor directions. We say this partly on the back of a small but persuasive literature that indicates a growth of ‘the power within’ – a sense of courage, solidarity and strategy within popular political praxis in Bangladesh. Much of this evidence comes from the rising political awareness of women. Yet as Mushtaq Khan’s work predicts, and close- grained accounts of political action confirm, political struggle remains substantially organised around factional rather than class interests, which remain fragmented and indistinct. But do the factional interests of the client still exhaust the possibilities of organisation by poor, powerless groups? Although to date unregistered as such, the garments revolt has redrawn the lines of class conflict. Unlike industrial unrest in the public sector, classically factional, the garments revolt has been transparently a class conflict. For the industrial working class of Bangladesh, substantially female, it is as workers in a globalised capitalism that they have experienced recent economic volatility. The garments workers of Bangladesh are ‘flexibilised’, non- unionised, predominantly young and female, precariously exposed to volatile global commodity markets as consumers, and equally uncertain global export markets as producers. Their political significance has almost certainly been under-played because of their youth and sex. The Bangladeshi public is generally well informed about the conditions in which garments workers live, yet the perception lingers that women garments workers cannot have a serious grievance, as these nimble fingers work merely for pin money. What if anything does our loosely formulated ‘unruly politics’ framing add? One contribution is that we are reminded of the importance of consent to public authority, and therefore of the combustible quality of defiance in that context. This is particularly relevant in Bangladesh because of the mundane quality of public authority, the absolute levels of poverty and deprivation, the political history of protest, the absence of class-based mobilisation and the expectations of a patron class. How these factors combine in the Bangladesh context explains the portion of the political settlement that incorporates the poor and marginal. With its focus on the moral power of defiance within unruly acts, an unruly political framing sets out a more realistically capillary and relational model of power than a resource model of power (you have it or you do not). Unruly politics helps understand power in ways that not only appear to fit the realities better, but also makes it more possible to imagine a more inclusive polity. The emphasis on time, form, content and agency also remind us that these protests are not random or unconnected: the contentions of the garments workers of Bangladesh is only partly in the control of their immediate bosses and state; their wages and cost of living complaints are at least partly a complaint about their positioning within the global economy. Finally, the focus on unruly actors and events shifts the gaze on to political society and away from the aid-funded managerialism of


N. Hossain

development NGOs and civil society groups. This is a far less mediated form of action but with far greater potential for political empowerment and pro-poor change.

Notes 1 In particular, Khan 2010a, 2010b; On the State of Governance Series, see Institute of Governance Studies 2006, 2008. 2 The ‘politics of the governed’ is Partha Chatterjee’s expression (Chatterjee 2004). There have been a few recent efforts to revive or revisit village studies in Bangladesh: Lewis and Hossain 2008; Siddiqi 2000; Westergaard and Hossain 2005. Other recent efforts to explore the politics of the masses include Ali and Hossain 2005; Kabeer and Kabir 2009; Ruud 2011. The ‘prisoner’ metaphor refers to Wood 2000. The idea that Bangladeshis, and in particular poor and powerless people, are ‘prisoners’ within a system with deep structural roots in patron-client organisational forms has probably been the most influential overall framing of the governance challenge in the 2000s. 3 This is an ongoing situation at the time of writing, and any summary of Shahbag and its aftermath would be instantly out of date. It is such a heatedly political debate that it is hard to find any detached or objective analysis, in the absence of which, see The Economist’s view: www.economist.com/news/asia/21571941-huge-protest-capitalagainst-islamist-party-and-its-leaders-mass-dissatisfaction; and my own blog: http:// participationpower.wordpress.com/2013/02/18/bangladesh-is-revolting-again (both accessed 12 March 2013). For the debates from among progressive young intellectuals, see blogspot http://alalodulal.org. For news coverage, see http://bdnews24.com. 4 See Centre for the Future State 2010. 5 See Centre for the Future State 2010. 6 For a selection of governance assessments that cover this territory, see Duncan 2010 (which is a meta-assessment, as it synthesises a range of recent governance assessments); also Institute of Governance Studies 2006, 2008). 7 The point that the caretaker government institution has meant a failure by parties to develop self-enforcing rules of the game is from Mirza Hassan. 8 For publications on Bangladesh from the Development Research Centre (DRC) on Citizenship, Participation and Accountability, see www.drc-citizenship.org/ search?country_ids=841464564 (accessed 12 March 2013). Relevant research from the Bangladesh section of the Pathways of Women’s Empowerment Research Programme Consortium includes Nazneen 2009; Nazneen and Sultan 2009, 2010). On the WeD research, see in particular the Shammo case study in Devine et al. 2008 and Wood 2007). 9 This section draws from discussions with my ‘unrulista’ colleagues akshay khanna, Patta Scott-Villiers, Alex Shankland, Mariz Tadros and Joanna Wheeler. See Shankland et al. 2011. Other related writings that use the notion include short blog pieces on the Occupy movement: www.opendemocracy.net/ourkingdom/alexshankland/occupylsx-unruly-politics-and-subversive-ruliness; the Arab Spring: www. unrisd.org/80256B3C005BE6B5/(httpNews)/6CACEA99340950AAC125795D00581 C33?OpenDocument; and other blogs http://participationpower.wordpress.com/category/unruly-politics-2 (accessed 13 March 2013). For a more detailed account of the theoretical underpinnings of the idea of unruly politics, see Khanna 2012. 10 Our initial literature review has touched on some of the main texts on contentious politics, social movements, rebellion and revolutions, authority, resistance and power, provision politics, food riots and the moral economy, non-violent action, political space, citizenship and civil society, and democracy, sovereignty and representation, political society and the politics of the governed, or ‘the crowd’. These will be referred to in the following section. Not all of the group will agree

The significance of unruly politics 167 11

12 13 14 15 16 17 18 19 20

with this categorization or the terminology used here; see Khanna 2012 for a less empiricist perspective on unruly politics. Several articles in New Left Review connect the unruly politics of the period to the particularities of the economic-political-technological-social nexus, often pointing also to the reluctance of more mainstream and centre-right commentators to draw such obvious connections. See, for example, Anderson 2011; M. Davis 2011; Žižek 2010). See also Tadros 2012. See Johnstone and Mazo 2011; Lagi et al. 2011. The original article was Berger and Spoerer 2001; www.bbc.co.uk/blogs/newsnight/ paulmason/2011/04/revolutions_and_the_price_of_b.html (accessed 30 June 2011). Moniruzzaman (2009) provides a valuable recent macro analysis of the phenomenon on which I draw here. Suykens and Islam (2013) give one of the first close- grained account of the micro politics of the ‘complex political phenomenon’ of hartal. Daily Star, 7 July 2011, ‘Hartal not called for people Says PM’, www.thedailystar. net/newDesign/news-details.php?nid=193198 (accessed 5 November 2011). For an account of how to analyse and measure political events like hartal, see C. Tilly 2008b. See Ahmed 2010 for a blow-by-blow account of the ‘Fakhruddin interregnum’. The Daily Star is the leading English language daily newspaper in Bangladesh (www. thedailystar.net). Anyone who believes that Bangladeshi women garment workers were docile victims should see the photographic series by Reuters’ Andrew Biraj of the garments revolts in 2010, called ‘The Fashion Victims’. See also images on www.guardian.co.uk/ world/gallery/2010/jun/30/bangladesh-protest#/?picture=364397796&index=6.


Governance challenges in Bangladesh Old wine in not so new bottles? Joe Devine, Ipshita Basu and Geof Wood

Introduction Since the early 1990s governance has occupied a more central position in the international development policy lexicon. Whilst some dismiss the idea of governance and indeed criticise it as a political tool to extend a Western model of a liberal, democratic and pluralist state, it has become an important reference point for an analysis of the way states and citizens interact, and how this interaction is tied to the performance and behaviour of key institutions, duty-bearers and decision-makers in society. In other words, governance allows for an analysis of the quality of political life in society and how this is both shaped by and shapes the structure and dynamics of power. In the original meeting which led to the publication of this book, we were acutely aware of the conceptual weakness of the idea of governance and the lack of agreement about how to measure it. As a result, we resisted the temptation to impose a single guiding definition. Each of the contributing authors, however, shared a concern about the quality of political life in Bangladesh and how this impacted on different sectors of society. Contributing authors were also convinced that political developments in Bangladesh were fluid, uncertain and exposed to a wide range of influences and pressures originating at both national and international levels. In this, authors came close to Rose’s (1991: 21) understanding of governance as an emergent pattern or order of a social system, arising out of complex negotiations and exchanges between ‘intermediate’ social actors, groups, forces and organisations, public and semi-public institutions in which state organizations are only one – and not necessarily the most significant – amongst many others seeking to steer or manage these relations. Bangladesh is a fascinating and arguably unique country in which to explore these governance dynamics. For most of its relatively short history, the country has demonstrated a resilient commitment to a modern, inclusive and prosperous democracy. There have been of course major challenges to this commitment, all of which have profoundly shaped the way democracy has taken shape as well as

Governance challenges in Bangladesh 169 the way citizens in Bangladesh think about democracy (see Devine et al.’s Chapter 8 in this volume). More recently, one of the most common observations made about political life in Bangladesh contrasts the country’s ‘exceptional’ performance in socio-economic terms (Asdullah et al. 2014) with its poor performance in political or governance terms. This observation leads to the construction of yet another ‘Bangladesh paradox’ in which the country seems to be making good headway in its aspirations to be a middle income country at the same time as the quality of its political life seems to be deteriorating. There is no shortage of evidence that this deterioration is real and current. Thus in the 2014 Corruption Perceptions Index, Bangladesh slipped from 136th place in 2013 to 145th place in the global rankings. Its score was also the worst in South Asia with the exception of Afghanistan (Transparency International 2014).1 Meanwhile, in their analysis of the 2013 Worldwide Governance Indicators (WGI), Hasan et al. (2015) illustrate the extent of the contrast between Bangladesh’s ambition to secure middle income status and its performance in governance terms. The authors analysed and compared the WGIs of Bangladesh against the average performance of Lower Middle Income Countries (LMICs) and Low Income Countries (LICs). Table 8.1 summarises their findings. For four of the six WGIs, Bangladesh falls above the average for LICs, but none of the indicators exceeds the average for LMICs. Of immediate relevance to the contributions of this book, the indicator where Bangladesh falls significantly below the average for both LMICs and LICs is political stability and the absence of violence.

Using governance to understand political developments in Bangladesh When we first embarked on the idea of putting together the book, we were driven by three overarching and connected ideas. It is worth briefly summarising the ideas here. First, we drew inspiration from work carried out by colleagues at the Institute for Governance Studies2 and, in particular, their pioneering State of Governance Reports (IGS 2006, 2008a, 2009a, 2011a, 2012, 2013). The overall aim of the reports was to provide a detailed empirical analysis of key aspects of Table 8.1 Comparison of Bangladesh’s WGI performance Worldwide governance indicator

Lower middle income countries

Low income countries

Control of Corruption Government Effectiveness Political Stability & Absence of Violence Rule of Law Voice and Accountability Regulatory Quality

Below Below Below Below Below Below

Above Above Below Above Above Below

Source: Hasan et al. (2015).


J. Devine et al.

political life in Bangladesh. Refreshingly, the reports moved away from a rather sterile description of governance in Bangladesh to highlight and question the significance of key underlying socio-political processes. In this way, the reports demonstrated and interrogated, through rigorous analysis, the co-existence of relatively advanced, formal and progressive governance characteristics with informal processes and actions which seemed to be inconsistent with, if not contradictory to, some of the ideals of good governance. Rather than rush to conclusions based on some ‘substractivist epistemology’3 which ends up ‘pathologising’ the many dysfunctions of the state or political system, the reports invited a more nuanced analysis of how and why everyday governance is shaped and reproduced in the way that it is. We have tried to remain faithful to this general approach by taking ‘the quality of everyday politics’ as our central focus instead of starting with a list of ‘governance deficits’ of Bangladesh and comparing them with some preconceived notion about the formal requirements or processes of governance (however important these may be). In so doing, we wanted to push our analyses to identify key forces which influence or are influenced by the governance arrangements in Bangladesh, and then to highlight the significance of the negotiations and exchanges that take place between them, either explicitly or implicitly. As many of the contributions to this book demonstrate, the terms on which people are excluded from or included in these exchanges have a significant impact on the way they come to experience the effects of governance in their lives. This is an important observation. At a time when it appears that citizens in Bangladesh are more empowered and have more voice than ever before (Hasan et al. 2015; Kabeer et al. 2012), there is growing evidence of a concentration of power in the hands of a few political and economic elite and the continued disenfranchisement of the majority (see Hossain’s Chapter 7 and Wood’s Chapter 5 in this volume). The interaction of these forces is central to the pattern of governance which manifests itself today in Bangladesh. This brings power to the centre of our analysis and understanding of governance. The second idea which inspired the book is closely related to the first. One of the common criticisms of governance analysis is that it is weighed down by heavy normative expectations. In policy terms, this tends to take one of two forms. On the one hand, governance and even more so ‘good governance’ reforms are aimed at instilling positive characteristics, such as transparency or accountability, that will help improve how government is carried out. Second governance reforms focus on the kinds of activities governments should prioritise such as poverty reduction, protection of rights and so forth. In the context of Bangladesh, like many other countries in the Global South, these normative concerns are ultimately derived from abstract, Western-derived formulas usually consisting of a commitment to open access democracy in which states are ideally held accountable by a vibrant citizenry. The ‘governance agenda’ therefore effectively consists of a political programme established to help failing or lagging countries develop more modern political systems. Paradoxically this political approach is not wired to look at the real political practices that underpin specific political settlements in society. Perhaps not unsurprisingly, these practices are stubbornly complex and indeed – as

Governance challenges in Bangladesh 171 the chapters in this book demonstrate – the closer you get, the more complex they prove to be. Given such complexity, the teleology and linear assumptions underpinning conventional governance paradigms start to look tired and rigid. Removed from their ideological load therefore governance concerns are much more wrapped up in uncertainty, complexity and fluidity. It is at this level where we begin to engage with more relevant questions about ‘how’ and ‘why’ particular governance arrangements emerge and are reproduced. This injects into the governance debate in Bangladesh a much-needed attempt at explanation as opposed to a simple description of what appears to exist. The third starting point for this book highlighted the importance of time, history and context. In governance analysis, a great deal of emphasis is given to identifying specific variables that lead to particular outcomes. If it is accepted that there is no single magic bullet to fix governance woes, it has not stopped researchers and practitioners from trying to find a number of bullets (variables) which, if manipulated properly, will lead to better governance outcomes. All too often, however, research into governance variables treats them outside of their temporal context. As such the complex dynamics of change, or the reasons behind a lack of change, are sidelined. What we end up with is a motionless governance snapshot, usually in the form of a list of deficits, instead of a moving picture of governance dynamics. In this book, we have made a very conscious effort to anchor our analysis in time and context. Why is this important? First of all, we are now aware of the fact that ideas, institutions and beliefs all evolve over time, and are both path and context dependent (North 2005). It goes without saying therefore, even if we ignore it in practice, that the chances of governance reform success will to a great extent be determined by historical circumstances and developments. In affirming this, the chapters in this volume have sought to move beyond functionalist approaches to governance to locate particular outcomes in specific structured patterns or processes (see Landell Mills’ Chapter 2 in this volume). Second, the constellation of governance arrangements in Bangladesh today is the product of processes of negotiation and interaction carried out over extended periods of time. It is important to acknowledge the content and legitimacy of these processes even if we are then critical or uncertain of the outcomes (Woolcock et al. 2011). Even if there were a global consensus on what governance should look like (and there is not), experience from around the world tells us that there are many paths to achieve particular governance thresholds. Understanding the historical trajectory of the Bangladesh governance pathway is therefore crucial if we are to engage in policy terms with its future direction. Finally, in discussing papers for this book, many of the authors wrestled with ideas emanating from the work of Douglas North and colleagues on Violence and Social Order. One of the key messages from this seminal work confirms the durability of social structure and the ‘slow pace’ of social change. This is particularly germane to discussions on governance in Bangladesh where endless reforms and apparent favourable conditions, especially socio-economic ones, have not resulted in the desired governance changes. Rather than rush to see this as a sign of a dysfunctional or pathological political system, North el al.’s


J. Devine et al.

analysis invites us to reflect on the motivation and behaviour of actors who influence the direction and speed of political change. In this, the role of elites in crafting and managing ‘social orders’ is crucial (North et al. 2009) – an observation again confirmed in various contributions of this book.

Bangladesh: a brief governance update In the introduction to the book, we attempted to give a very brief overview of the different regimes that have come to rule in Bangladesh. Here we give an equally cursory overview of recent relevant developments under the current regime. A quick glance at commentaries, media reports and research literature on Bangladesh’s very recent political situation reveals a remarkably familiar picture – or in Blair’s (Chapter 1 in this volume) terminology: a situation of statis. Once again Bangladesh seems to be at a political crossroads with a democratic project which has been described as illusive cosmetic (Parnini and Othman 2014), toxic (Islam 2013) and dysfunctional (Riaz 2013). The pessimism found in recent assessments received fresh impetus with the parliamentary elections of 2014 and their aftermath. In early 2014, the Awami League resumed the mantle of political power after an election that was never really an election. The main rival party, the Bangladesh Nationalist Party (BNP), had declared that it would boycott the election in protest against Awami League’s decision to dispense with the caretaker government formula. The boycott meant that the run-up to the elections was mired with street protests, hartals and blockades. Furthermore, the leader of the Jatiyo Party (the country’s third largest party) was detained and prevented from taking part in the elections because he had supported the boycott, and the Jamaat-e-Islami party had been banned from electoral participation because its politics were deemed to contradict the country’s commitment to secularism and democracy. With no credible opposition and an uncharacteristically low voter turn out (estimated at 40 per cent by the Election Commission and 10 per cent by the Fair Election Monitoring Alliance (FEMA)4), the Awami League secured just under 80 per cent of parliamentary seats and a two-thirds majority in Parliament. The elections were marred by accusations of vote rigging and violence, with the Human Rights Watch describing them as the bloodiest in Bangladesh’s history.5 On the day the 2014 election results were announced, the legitimacy of the winning party was already being questioned. Bangladesh had entered yet another new era of political crises. Although many of the post-1990 elections have been nominally overseen by a caretaker government, political transitions in Bangladesh usually occur because of social protest and mobilisation. So far the post-2014 election era has been faithful to this script. The BNP is absent from Parliament and plays out its opposition role ‘in the streets’, trying to de-legitimise the 2014 elections and force new elections. This has meant mobilising its support base to participate in nationwide blockades and protests, many of which have turned violent. In response, the ruling party has sanctioned the arrest of many BNP activists and

Governance challenges in Bangladesh 173 went as far as effectively detaining its leader to ‘house arrest’. Incidences of violence have escalated, with conservative estimates that over 100 people have been killed and over 10,000 opposition activists have been arrested since the 2014 elections. In April 2015, the BNP leader was attacked and shot at whilst campaigning during the mayoral elections. If the BNP has intensified its commitment to de-legitimising parliament and implementing its opposition role in the streets (Jahan 2015), the ruling party seems determined not only to sideline the BNP but to destroy it and remove its claim to be a legitimate political party. Attempts to destroy political opposition do not end with the BNP. The Jammat-e-Islami party has also come under intense pressure over the past few years. Many of the Jammat-e-Islami leaders have been arrested and tried under the International Crimes Tribune (ICT). With a huge majority in parliament, the Awami League is in a perfect position to force the agenda in relation to Jammate-Islami. Since January 2015, the ICT has issued death sentences to 16 individuals, of whom 12 are Jammat-e-Islami leaders. There is a growing concern about the ICT and its credibility, with many worried that it has been hijacked and politicised by the ruling party. This would be consistent with evidence that the ruling party has further politicised the judiciary (see below). Concerns about a possible worsening governance situation in Bangladesh are reinforced when we consider accusations that the ruling party has introduced measures to silence dissent and protest in society. The government, for example, has taken steps to tighten control of online websites, internet messaging services such as Viber and Tango, social media and blogs. Although formally these measures have been introduced to prevent the growth of extremism, there are concerns that the government is simply closing down social media outlets that are popular with or have been used by opposition activists. Print media outlets accused of disseminating criticism of the government meanwhile have also been shut down and their editors charged with sedition. The government has also moved to curtail the room for manoeuvre of the country’s vast NGO sector. A new draft law (Foreign Donations (Voluntary Activities) Regulation Act 2014) has already been approved by the Cabinet and is now awaiting the approval of Parliament. The new law will give the NGO Bureau greater powers to monitor any civil society activity funded by foreign donations, replicating moves by Modi’s government in neighbouring India. This has been seen as a move to curb NGO activity, especially activity considered to be critical of the government (ICG 2015). The new act, however, reflects a more general deterioration of relations between civil society and government. This sentiment was summed up in a recent remark by the Food Minister who described civil society members in Bangladesh as ‘cancer and fathers of 1/11’.6 One of the ironies of the governance agenda is that it assumes the existence of a vibrant civil society capable of holding government to account. The prospects of this happening in Bangladesh at this moment are not great (see Chapter 5 by Wood and then Chapter 6 by Lewis in this volume). Given the relatively restricted room for manoeuvre available to opposition political parties and civil society actors, the one sector which might retain some


J. Devine et al.

autonomy from government is the judiciary and the justice system. In the introduction, we referred to the judiciary as possibly the last bastion of any semblance of open access democracy in Bangladesh. In the 2014 elections, however, Awami League again secured a two-thirds majority in Parliament, giving it an uncontestable political mandate and the authority to introduce changes without the need for coalition support. In its previous tenure and with a similar majority in Parliament, Awami League controversially removed the caretaker government. In 2014, it passed the Sixteenth Constitutional Amendment giving Parliament the power to remove Supreme Court judges if they are deemed to be ‘incapable or acting improperly’ (Hasan et al. 2015). Many commentators have been quick to point out this amendment effectively ‘disarms’ an important institution partly responsible for ensuring the State is accountable in its actions. Even from a cursory reading of recent developments in Bangladesh therefore, it is clear that there are two fundamental and mutually reinforcing forces at work. On the one hand there is evidence of an increasing concentration of power in the hands of the ruling party. To be more precise, it seems that state power has been personalised and is now controlled by the Prime Minister’s Office, which includes the increasingly influential unelected advisers to the Prime Minister whose power now seems to eclipse that of Ministers. On the other hand, the ruling party has actively sought to limit the room for manoeuvre for protesting sections of society including what many consider to be legitimate opposition political parties (a legitimacy which the ruling party does not acknowledge). When necessary or when these restrictions have been less effective, the government has not shied away from deploying its expanding law enforcement agencies, including the controversial Rapid Action Battalion.

Peeking into the future Where does this analysis leave us? There are of course no crystal balls to predict what will happen in Bangladesh in the future. We are also aware of the perils of overburdening governance reform policies with too many unrealistic expectations and demands. For this reason, we would like to conclude reflecting on three possible and quite broad scenarios. These are intended to simply encourage further reflection rather than act as specific predictions. First, one of the paradoxes of Bangladesh’s current position is that its economic performance and profile are significantly stronger than its political one. This is important because although much of the governance discourse sees improved governance as a condition of further growth and development, the reality is that many economically advanced countries managed to prosper well before they had achieved ‘good governance’ (Chang 2002). So in one future scenario it is quite possible that Bangladesh continues to build political legitimacy primarily via economic growth and not governance reforms. This would take Bangladesh close to an ‘authoritarian developmentalist’ model outlined by Khondker in Chapter 3 in this book. There are, however, fundamental challenges in this scenario. On the one hand, it assumes continued and sufficient

Governance challenges in Bangladesh 175 economic growth and capacity in the future. While there is good reason to believe that this can be achieved, it should not be taken for granted. Bangladesh’s economic growth to date has been premised on consistently strong GDP performance, but its domestic and foreign direct investments remain relatively low. This simply adds an element of uncertainty to assumptions about future economic growth. On the other hand, the implementation of an authoritarian model to some extent requires the tacit support of the majority of its citizens. Here also lies an important unknown for Bangladesh. Any uncertainty about citizen support or tolerance for an authoritarian regime is heavily influenced by the regime’s ability to deliver core services and goods (Ong 1999). The State of Governance Reports are full of examples where citizens’ tolerance of poorquality and yet rent-inducing, core services is under strain (see also Chapter 7 by Hossain in this volume). The need to deliver quality services reinforces the fragility of over-relying on an ‘economic growth’ solution to governance in Bangladesh. A second scenario can be simply constructed from Bangladesh’s recent past. In this scenario, we expect that at some point in the future we will have a regime change but one in which the fundamental nature of the polity does not change (see Blair, Khondker and Wood in this volume). This has been the overriding characteristic of Bangladesh politics since at least the early 1990s and takes us back to arguments rehearsed in the introduction about how political and organisational behaviour are rooted in deeper structures and as such are continuous across regimes. As the mantle of power seesaws between the two main parties in Bangladesh, it simply creates opportunities for the ‘winner of the moment’ to take advantage and strengthen their own positions. The reality is that so far in Bangladesh, regime change on its own has not produced meaningful governance reforms. There is of course an important sub-theme to all of this, which focuses attention on how regime change comes about. In the recent ‘democratic’ past the role of the caretaker government has been central to the process of regime change. Recently, however, it has been the role of the military which has come to the fore, especially during the 2006–08 political crises. Over the past 20–25 years, all of the leading political parties have managed their relations with the military by allowing it to take advantage of growth opportunities and developing considerable employment, business and commercial interests. This means that now the military has an even larger and decidedly ‘self-interested’ stake in the governance of the country. It may feel it needs to intervene in politics and support or prevent regime change, not just to enforce law and order but also to protect its own interests. A third scenario is derived from the various contributions of this book and is built around a more nuanced analysis of, first, how power is actually exercised in society; and, second, how that exercise of power might be changed. The different contributions of the book draw attention to a myriad of local social institutions and arrangements as well as cultural attachments, which evolve over time and which impinge on the architecture and dynamics of governance. Many of these institutions and attachments operate informally (see especially Hossain in


J. Devine et al.

this volume) and may even reflect values associated with the deeper structures alluded to earlier. The third scenario therefore is committed to improved governance arrangements but is more cognisant of hybrid processes and the need to support ways of reforming that are locally meaningful (Chapter 4 by Basu, Brown and Devine in this volume). Deep structures are problematic but we would argue that the same structures offer practical prospects for social change and development. This establishes a policy commitment to reform that accepts a priori that governance takes time to work itself out, and that the process of reform will be ambiguous, complex and non-linear. We accept that this analytical stance approach stands in contrast to the prescriptions associated with conventional governance paradigms but we also hope that this stance will enable a more meaningful engagement with the way governance arrangements are actually reproduced in Bangladesh and will ultimately be more responsive to the aspirations of its citizens.

Notes 1 From 2001 till 2005, Bangladesh was at the bottom of the global rankings list. Over the 2005–13 period, it progressively rose to 16th bottom position. The 2014 ranking therefore represents a reverse trend. 2 The Institute has now merged with the BRAC Development Institute to form the BRAC Institute of Governance and Development. 3 Substractivist epistemology refers to an epistemology which has been derived from outside the socio-economic and political conditions and culture of the object of analysis. The idea is rehearsed by Khan (2012) in his analysis of public sector performance in Pakistan, and was first suggested to us by Professor Graham Brown. 4 FEMA estimates were widely reported in the press on 6 January 2014. 5 “Bangladesh: Elections Scarred by Violence”, Human Rights Watch, 29 April 2014. See www.hrw.org/news/2014/04/29/bangladesh-elections-scarred-violence (accessed 18 May 2017). 6 The term ‘1/11’ is popularly used to refer to 11 January 2007, the day the military intervened to support the caretaker government of President Iajuddin. The remark on cancer and 1/11 was made on 12 February 2015 and reported widely in the national press. See www.dhakatribune.com/politics/2015/feb/13/quamrul-civil-society-instigateskhaleda-burn-people (accessed 18 May 2017).


Acemoglu, D. and J.A. Robinson (2006). Economic Origins of Dictatorship and Democracy. New York: Cambridge University Press. Agamben, G. (1998). Homo Sacer: Sovereign Power and Bare Life. Redwood, CA: Stanford University Press. Agamben, G. (2005). State of Exception. Trans. K. Attell. Chicago: University of Chicago Press. Ahmad, M. (2009). ‘Views from the Madrasa: Islamic Education in Bangladesh’, in M. Ahmad and M.J. Nelson, Islamic Education in Bangladesh and Pakistan: Trends in Tertiary Institutions. Washington, D.C.: National Bureau of Asian Research. Ahmed, R. (1988). ‘Contradictions in Bengali Islam: Problems of Change and Adjustment’, in K.P. Ewing (ed.), Shariat and Ambiguity in South Asian Islam. Berkeley: University of California Press. Alamgir, J. (2009). ‘Lessons in Democracy from Bangladesh’, The Nation, 16 March 2009. Available: www.thenation.com/article/lessons-democracy-bangladesh (accessed 29 August 2012). Alavi, H. (1972). ‘The State in Post-Colonial Societies: Pakistan and Bangladesh’, New Left Review, 1(74), July–August. Ali, T. and N. Hossain (2005). Popular Expectations of Government: Findings from Three Areas in Bangladesh. PRCPB Working Paper. Dhaka. Anderson, P. (2011). ‘On the Concatenation in the Arab World’, New Left Review, 68(5): 5–15. Anheier, H.K. and L. Moulton (1999). ‘Organizational Failures, Breakdowns and Bankruptcies: An Introduction’, in H.K. Anheier (ed.), When Things Go Wrong: Organizational Failures and Breakdowns. London: Sage, pp. 3–16. Appadurai, A. (2001). ‘Deep Democracy: Urban Governmentality and the Horizon of Politics’, Environment and Urbanization, 13(2): 23–43. Apter, D. (1964). ‘Introduction: Ideology and Discontent’, in D. Apter, Ideology and Discontent. International Yearbook of Political Behavior Research, Vol. 5. New York and London: The Free Press & Collier-Macmillan. Arezki, R. and M. Bruckner (2011). Food Prices and Political Instability. International Monetary Fund. Arora, A., J. Swinnen and M. Verpoorten (2011). Food Prices, Social Unrest and the Facebook Generation. LICOS Centre for Institutions and Economic Performance, University of Leuven. Asadullah, M.N. and Chaudhury, N. (2009). ‘Holy Alliances: Public Subsidies, Islamic High Schools, and Female Schooling in Bangladesh’, Education Economics (Special



issue ‘Quality Education for All in South Asia’), 17(3): 377–394. doi:10.1080/0964529 0903142593. Asian Development Bank (2011). Asian Development Outlook Bank. Manila, Philippines Barnett, A. (2011). ‘Britain’s Winter of Protest: A Fight Back for the Future’. Open Democracy UK. Available: www.opendemocracy.net/ourkingdom/anthony-barnett/ britains-winter-of-protest-fight-back-for-future (accessed 6 February 2011). Bateman, M. (2010). Why Doesn’t Microfinance Work? The Destructive Rise of Local Neoliberalism. London: Zed Books. Bates, R.H. (2001). Prosperity and Violence: The Political Economy of Development. New York and London: W.W. Norton & Co. BBS (Bangladesh Bureau of Statistics) (2011). Household Income and Expenditure Survey (HIES) 2010. Dhaka. Benhabib, S. (2001). ‘Deliberative Democracy, the Public Sphere and the Internet’, Philosophy and Social Criticism, 27: 21–29. Berazneva, J. and D.R. Lee (2011). Explaining the African Food Riots of 2007–2008: An Empirical Analysis. Charles H. Dyson School of Applied Economics and Management, Cornell University, Ithaca. Berger, H. and M. Spoerer (2001). ‘Economic Crises and the European Revolutions of 1848’, The Journal of Economic History, 61: 293–326. Bethel, L. (2000). ‘Politics in Brazil: From Elections without Democracy to Democracy without Citizenship’, Daedalus, 129, Spring: 1–27. Biggs, S. and D. Lewis (2009). ‘Fair Trade and Organizational Innovation in Nepal: Lessons from 25 Years of Growth of the Association of Craft Producers (ACP)’, European Journal of Development Research, 23: 1–20. Blair, H. (1971). ‘Sheikh Mujib and déjà vu in East Bengal: The Tragedies of March 25’, Economic and Political Weekly, 6(52), 25 December: 2555–2562. Blair, H. (2005) ‘Civil Society and Pro-Poor Initiatives in Rural Bangladesh: Finding a Workable Strategy’, World Development, 33(6): 921–936. Blair, H. (2010). ‘Party Overinstitutionalization, Contestation, and Democratic Degradation in Bangladesh’, in P. Brass (ed.), Handbook of South Asian Politics. London: Routledge, pp. 98–117. Blair, H. (2011). ‘Social Violence and Democratization in South Asia: The General Case and Bangladesh as a Case Study’. Paper for Workshop to Honor Paul Brass and His Work, Seattle, 4 September. Blair, H., R. Charlick, R. Haque, M. Hasan and N. Kalimullah (2004). Democracy and Governance Strategic Assessment of Bangladesh. Report for USAID. Burlington, VT: ARD. Blood, A.K. (2002). The Cruel Birth of Bangladesh: Memoirs of an American Diplomat. Dhaka: University Press. Bohstedt, J. (1988). ‘Gender, Household and Community Politics: Women in English Riots 1790–1810’, Past and Present, 120: 88–122. Bohstedt, J. (1992). ‘The Moral Economy and the Discipline of Historical Context’, Journal of Social History, 26(2), 1 December: 265–284. doi:10.2307/3788421. Bohstedt, J. (2010). The Politics of Provisions: Food Riots, Moral Economy, and Market Transition in England, c.1550–1850. Farnham, Surrey and Burlington, VT: Ashgate. Bose, S. (2011). Dead Reckoning: Memories of the 1971 Bangladesh War. New York: Columbia University Press. Bottomore, T. (1993). Elites and Society. London: Routledge. Bratton, M. and R. Mattes (2001). Support for Democracy in Africa: Intrinsic or Instrumental? British Journal of Political Science, 31: 447–474.

References 179 Brett, E.A. (2009). Reconstructing Development Theory: International Inequality, Institutional Reform and Social Emancipation. Basingstoke: Macmillan. Brinkman, H.-J. and C.S. Hendrix (2011). Food Insecurity and Violent Conflict: Causes, Consequences, and Addressing the Challenges. Occasional Paper 24. Rome: World Food Programme. Bush, R. (2010). ‘Food Riots: Poverty, Power and Protest’, Journal of Agrarian Change, 10(1): 119–129. Camfield, L., G. Crivello and M. Woodhead (2009). ‘Wellbeing Research in Developing Countries: Reviewing the Role of Qualitative Methods’, Social Indicators Research, 90(1): 5–31. doi:10.1007/s11205-008-9310-z. CAMPE (Campaign for Popular Education) (2005). Education Watch Report 2005. Dhaka. Centre for the Future State (2010). An Upside Down View of Governance. Brighton: Institute of Development Studies. Chang, H. (2002). Kicking Away the Ladder: Development Strategy in Historical Perspective. London: Anthem Press. Chang, H.-J. (2011) ‘Anti-capitalist? Too simple. Occupy can be the catalyst for a radical rethink’, Guardian, 15 November. Chatterjee, P. (2004). The Politics of the Governed: Reflections on Popular Politics in Most of the World. New York: Columbia University Press. Chatterjee, P. (2011). Lineages of Political Society. New York: Columbia University Press. Choudhury, A.N. (1989). Let Grassroots Speak: People’s Participation Self-Help Groups and NGOs in Bangladesh. Dhaka: University Press. Chu, Y.H., L. Diamond, A. Nathan and D.C. Shin (eds) (2008). How East Asians View Democracy. New York: Columbia University Press Chua, B.H. (1995). Communitarian Ideology and Democracy in Singapore. London: Routledge. CIDA (Canadian International Development Agency) (1990). Comilla Proshika Evaluation Report. Prepared by J.C. Cairns, N. Ilahi and M. Vigoda. Unpublished consultancy report. Ottawa. Clarke, L. and C. Perrow (1999). ‘Prosaic Organizational Failure’, in H.K. Anheier (ed.), When Things Go Wrong: Organizational Failures and Breakdowns. London: Sage, pp. 179–196. Cooke, B. and U. Kothari (2001). Participation: The New Tyranny. London: Zed Books. Corbridge, S., G. Williams, M. Srivastava and R. Véron (2005). Seeing the State: Governance and Governmentality in India. Vol. 10. Cambridge: Cambridge University Press. Cornwall, A. and V. Schatten Coelho (2007). Spaces for Change? The Politics of Citizen Participation in New Democratic Arenas. London: Zed Books. Court Correspondent (2011). ‘Money-Laundering Trial of Tarique Begins’, Daily Star, 12 September. CPD (Centre for Policy Dialogue) (2012a). State of the Bangladesh Economy in FY2009–10 (Second Reading). Dhaka. CPD (Centre for Policy Dialogue) (2012b). Analytical Review of Bangladesh’s Macroeconomic Performance in FY 2011–12 (Second Reading). CPD Working Paper. Dhaka. Davis, M. (2011). ‘Spring Confronts Winter’, New Left Review, 72: 5–15. Davis, P.R. and J.A. McGregor (2000). ‘Civil Society, International Donors and Poverty in Bangladesh’, Journal of Commonwealth and Comparative Politics, 38(1): 47–64.



De Souza, P.R., Palshikar, S. and Yadav, Y. (2008) ‘The Democracy Barometers: Surveying South Asia’, Journal of Democracy, 19(1): 84–96. Devine, J. (2002). ‘Ethnography of a Policy Process: A Case Study of Land Redistribution in Bangladesh’, Public Administration and Development, 22: 403–414. Devine, J. (2007). Governance, Democracy and the Politics of Wellbeing. ESRC Research Group on Wellbeing in Developing Countries, Working Paper 36. University of Bath. Devine, J. and White, S.C. (2013). ‘Religion, Politics and the Everyday Social Order in Bangladesh’, Journal of Contemporary Asia, 43(1): 127–147. Devine, J., L. Camfield and I. Gough (2008). ‘Autonomy or Dependence – or Both? Perspectives from Bangladesh’, Journal of Happiness Studies, 9(1): 105–138. Devine, J., D. Gould, C. Tan, A. Emamgholi and S. Sadeghi (2013). Attracting Foreign Direct Investment: What Can South Asia’s Lack of Success Teach Other Developing Countries? Policy Research Working Paper; no. WPS 6696. Washington, D.C.: The Worldbank. Available: http://documents.worldbank.org/curated/en/2013/11/18491804/ attracting-foreign-direct-investment-can-south-asias-lack-success-teach-other-developingcountries (accessed 18 May 2017). DFID (Department for International Development) (2011). ‘Governance Portfolio Summary Review’. Available: www.dfid.gov.uk/Documents/publications1/governanceportfolio.pdf (accessed 18 May 2017). Diamond, L. (2011). ‘Why Democracies Survive’, Journal of Democracy, 22(1): 17–30. Duncan, A. (2003) ‘Drivers of Change: Reflections of Experience to Date’, Paper for Department for International Development workshop, Oxford, 23 June 2003, Oxford Policy Management. Duncan, A. (2010). Bangladesh: Multi-donor Country Governance Analysis. Dhaka: Local Consultative Group on Governance. Duncan, A. and G. Williams (2012). ‘Making Development Assistance More Effective Through Using Political-economy Analysis: What Has Been Done and What Have We Learned?’ Development Policy Review, 30(2): 133–148. Economist, The (2012). ‘Bangladesh and Development: The Path through the Fields’, 3 November. Edwards, M. (1999). ‘NGO Performance: What Breeds Success?’, World Development, 27(2): 361–374 . Edwards, M. and D. Hulme (eds) (1995). Non-Governmental Organisations – Performance and Accountability: Beyond the Magic Bullet. London: Earthscan. Eicher, P., Z. Alam and J. Eckstein (2010). ‘Elections in Bangladesh 2006–2009: Transforming Failure into Success’. Dhaka: United Nations Development Programme. Available: www.undp.org.bd/info/pub/election in bangladesh.pdf (accessed 29 September 2011). Elliott, L. (2012). ‘New Wave Economies Going for Growth’, Guardian, 18 December. Espey, J., C. Harper and N. Jones (2010). ‘Crisis, Care and Childhood: The Impact of Economic Crisis on Care Work in Poor Households in the Developing World’, Gender and Development, 18(2): 291–307. Esping-Andersen, G. (1990). The Three Worlds of Welfare Capitalism. Cambridge and Oxford: Polity Press. Esping-Andersen, G. (1999). Social Foundations of Postindustrial Economies. Oxford: Oxford University Press. EU (European Union) (2009). Election Observation Mission, ‘People’s Republic of Bangladesh: Final Report, Parliamentary Elections, 29 December 2008’. Available:

References 181 www.eueom.eu/files/dmfile/EUEOMBangladeshFinalReport2008.pdf (accessed 29 September 2011). Fisher, J. (1994) ‘Is the Iron Law of Oligarchy Rusting away in the Third World?’ World Development, 22(2): 129–143. Fukuyama, F. (1995). Trust: The Social Virtues and the Creation of Prosperity. New York: Free Press. Fukuyama, F. (2011). The Origins of Political Order. New York: Farrer, Straus and Giroux. Gerschenkron, A. (1962). Economic Backwardness in Historical Perspective: A Book of Essays. Cambridge, MA: Harvard University Press. Giddens, A. (1984). The Constitution of Society: An Outline of the Theory of Structuration. Cambridge: Polity Press. Goffman, E. (1971). Presentation of Self in Everyday Life. New York: Pelican. Greenough, P.R. (1982). Prosperity and Misery in Modern Bengal: The Famine of 1943–1944. Oxford: Oxford University Press. Greenough, P.R. (1983). ‘Indulgence and Abundance as Asian Peasant Values: A Bengali Case in Point’, The Journal of Asian Studies, 42(4): 831–850. Griffin, J. and R. Judge (2010). Civil Society Policy and Practice in Donor Agencies, Overview Report commissioned by DFID and INTRAC. Governance and Social Development Resource Centre. Gutman, A. and D. Thompson (2002). Why Deliberative Democracy? Princeton: Princeton University Press. Halim, S. (2011). ‘Right to Information as a Tool to Fight Corruption in Bangladesh’, New Age (Dhaka), 4 July. Hardt, M. (2003). ‘Porto Alegre: Today’s Bandung?’, New Left Review, 14 (March/ April). Hasan, M. (2011). ‘Democracy and Political Islam in Bangladesh’, South Asia Research, 31(97): 97–117. Hasan, M., J. Rose and S. Khain (2015). Background Paper for the 7th Five Year Plan of the Government of Bangladesh: Governance and Justice. Available: www.plancomm. gov.bd/wp-content/uploads/2015/02/12_Governance-and-Justice-_Final-Draft.pdf (accessed 18 May 2017). Hasan, S., G. Mulamoottil and J.E. Kersell (1992). ‘Voluntary Organizations in Bangladesh: A Profile’, Environment and Urbanization, 4(2): 196–206. Hashemi, S.M. (1990). NGOs in Bangladesh: Development Alternative or Alternative Rhetoric? Unpublished research report, Institute of Development Policy and Management (IDPM), University of Manchester. Hashemi, S.M. (1995). ‘NGO Accountability in Bangladesh: Beneficiaries, Donors and the State’, in M. Edwards and D. Hulme (eds), Beyond the Magic Bullet. London: Earthscan, pp. 103–110. Heltberg, R., N. Hossain, A. Reva and C. Turk (2012). ‘Coping and Resilience during the Food, Fuel, and Financial Crises’, The Journal of Development Studies, 49(5): 705–718. doi:10.1080/00220388.2012.746668. Hossain, N. (2005). Elite Perceptions of Poverty in Bangladesh. Dhaka: University Press. Hossain, N. (2009). ‘Reading Political Responses to Food, Fuel and Financial Crises: The Return of the Moral Economy?’ Development, 52(3), September: 329–333. doi:10.1057/dev.2009.34. Hossain, N. (2010). ‘Rude Accountability: Informal Pressures on Frontline Bureaucrats in Bangladesh’, Development and Change, 41: 907–928.



Hossain, N. (2012a). The Significance of Unruly Politics in Bangladesh. Dhaka: Institute of Governance Studies. Hossain, N. (2012b). Women’s Empowerment Revisited: From Individual to Collective Power among the Export Sector Workers of Bangladesh. Institute of Development Studies, University of Sussex. Hossain, N. and J.A. McGregor (2011). ‘A “Lost Generation”? Impacts of Complex Compound Crises on Children and Young People’, Development Policy Review, 29(5): 565–584. Houtzager, P. (2003). ‘Introduction: From Polycentrism to the Polity’, in P. Houtzager and M. Moore (eds), Changing Paths: The New Politics of Inclusion. Ann Arbor: University of Michigan Press, pp. 1–31. Available: www.press.umich.edu/pdf/0472113224ch1.pdf (accessed 20 May 2017). Hudock, A. (1999). NGOs and Civil Society: Democracy by Proxy? Cambridge: Polity Press. Huntington, S.P. (1991). The Third Wave: Democratization in the Late Twentieth Century. Norman, OK: University of Oklahoma Press. Huq, A. (ed.) (1976). Exploitation and the Rural Poor. Comilla: BARD. ICG (International Crisis Group) (2015). Mapping Bangladesh’s Political Crisis. Asia Report No. 264. Brussels, 9 February. Available: www.crisisgroup.org/~/media/Files/ asia/south-asia/bangladesh/264-mapping-bangladesh-s-political-crisis.pdf (accessed 18 May 2017). ICG (International Crisis Group) (2016). Political conflict, extremism and criminal justice in Bangladesh. Asia Report No. 277. Brussels, 11 April. IEP (Institute for Economics and Peace) (2011). Global Peace Index 2011. Available: http://economicsandpeace.org/wp-content/uploads/2015/08/2011-GPI-Results-Report. pdf (accessed 18 May 2017). IGS (Institute of Governance Studies) (2006). The State of Governance in Bangladesh 2006: Knowledge, Perceptions, Reality. Dhaka: Institute of Governance Studies, BRAC University. IGS (Institute of Governance Studies) (2007). The State of Governance in Bangladesh 2006. Dhaka: Institute of Governance Studies, BRAC University. IGS (Institute of Governance Studies) (2008a). The State of Governance in Bangladesh 2007: Confrontation, Competition, Accountability. Dhaka: Institute of Governance Studies, BRAC University. IGS (Institute of Governance Studies) (2008a). The Public Service Commission: Policy Note. Institutions of Accountability. Dhaka: Institute of Governance Studies, BRAC University. IGS (Institute of Governance Studies) (2009a). The State of Governance in Bangladesh 2008. Dhaka: Institute of Governance Studies, BRAC University. IGS (Institute of Governance Studies) (2009b). The Judiciary: Policy Note. Institutions of Accountability Series. Dhaka: Institute of Governance Studies, BRAC University. IGS (Institute of Governance Studies) (2011a). The State of Governance in Bangladesh. Dhaka: Institute of Governance Studies, BRAC University. IGS (Institute of Governance Studies) (2011b). Governance Barometer Survey Bangladesh 2010. Available: www.igs-bracu.ac.bd/images/stories/pdfs/governancebarometer 2010.pdf (accessed 18 May 2017). IGS (Institute of Governance Studies) (2012). Regions, Representation, Disparity 2012. Available: http://bigd.bracu.ac.bd/images/SoG/2.%20sog-2012%20full%20book.pdf (accessed 18 May 2017). IGS (Institute of Governance Studies) (2013). The State of Governance Bangladesh 2013:

References 183 Democracy Party Politics. Available: http://bigd.bracu.ac.bd/images/SoG/1.%20sog% 20report-%202013-%20democracy%20party%20politics.pdf (accessed 18 May 2017). IGS (Institute of Governance Studies) (forthcoming). State of Dhaka City. Dhaka: Institute of Governance Studies, BRAC University. Inglehart, R.F. and C. Welzel (2003). ‘The Theory of Human Development: A CrossCultural Analysis’, European Journal of Political Research, 42(3): 341–379. Irish, L.E. and Simon, K.W. (2005). NGOs in Bangladesh: Legal and Regulatory Environment. New York: International Centre for Civil Society Law. Islam, A.S. (2002). ‘Political Parties and the Future of Democracy’, in A.M. Chowdhury and A. Fakhrul (eds), Bangladesh on the Threshold of the Twenty-First Century. Dhaka: Asiatic Society of Bangladesh. Islam, M. (2013). ‘The Toxic Politics of Bangladesh: A Bipolar Competitive Neopatrimonial State?’, Asian Journal of Political Science, 21(2): 148–168. Johnson, C. (1982). MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925–1975. Stanford: Stanford University Press. Juberee, A. (2011). ‘Khaleda Threatens Dhaka Siege’, New Age, 18 October. Jahan, R. (2015). ‘The Parliament of Bangladesh: Representation and Accountability’, The Journal of Legislative Studies, 21(2): 250–269. Jahangir, B.K. (1995). ‘Famine and Politics’, Journal of Social Studies, 68: 91–101. Johnstone, S. and J. Mazo (2011). ‘Global Warming and the Arab Spring’, Survival, 53(2): 11–17. Kabeer, N. with A.H. Kabir, and T.Y. Huq (2009). Quantifying the Impact of Social Mobilisation in Rural Bangladesh: Donors, Civil Society and The Road not Taken. Working Paper 333. Institute of Development Studies, University of Sussex. Kabeer, N. (2011). Citizenship Narratives in the Face of Bad Governance: The Voices of the Working Poor in Bangladesh’, Journal of Peasant Studies, 38(2): 325–353. Kabeer, N. and L. Huq (2010). ‘The Power of Relationships: Love and Solidarity in a Landless Women’s Organisation in Rural Bangladesh’, Institute of Development Studies Bulletin, 41(2): 79–87. Kabeer, N. and A.H. Kabir (2009). Citizenship Narratives in the Absence of Good Governance: Voices of the Working Poor in Bangladesh. Working Paper 331. Institute of Development Studies, University of Sussex. Kabeer, N., S. Mahmud and J.G. Isaza Castro (2012). ‘NGOs and the Political Empowerment of Poor People in Rural Bangladesh: Cultivating the Habits of Democracy?’, World Development, 40(10): 2044–2062. Kapoor, K., A. Rahman and G. Kasunko (2000). Estimating the Effects of Corruption: Implications for Bangladesh. Policy Research Working Paper 2479. Washington, D.C.: World Bank. Kaufmann, D., A. Kraay and P. Zoido-Lobaton (1999). Governance Matters. Washington, D.C.: World Bank. Kaufmann, D., A. Kraay and P. Zoido-Lobaton (2002). Governance Matters II: Updated Indicators for 2000/01. Policy Research Working Paper 2772. Washington, D.C.: World Bank. Kendall, J. (2003). The Voluntary Sector. London: Routledge. Khan, F.C., A. Ahmad and M. Quddus (eds) (2009). Recreating the Commons? NGOs in Bangladesh. Dhaka: University Press. Khan, I.A. (2000a). The Struggle for Survival: Networks and Relationships in a Bangladesh Slum. PhD Thesis, University of Bath, UK. Khan, I. (2011). ‘Right to Information: Making State Work for the Poor’, Daily Star, 13 June.



Khan, M. (2003). ‘State of Governance in Bangladesh’, The Round Table: The Commonwealth Journal of International Affairs, 92(370): 391–405. Khan, M.H., (2005). Markets, States and Democracy: Patron-Client Networks and the Case of Democracy in Developing Countries’, Democratization, 12(5): 704–724. Khan, M.H. (2000b). ‘Class, clientelism and communal politics in contemporary Bangladesh’, in K.N. Panikkar, T.J. Byres and U. Patnaik (eds), The Making of History: Essays Presented to Irfan Habib. New Delhi: Tulika. Khan, M. (2010a). Political Settlements and the Governance of Growth-enhancing Institutions. Available: https://core.ac.uk/download/pdf/2792198.pdf (accessed 18 May 2017). Khan, M. (2010b). Bangladesh: Partitions, Nationalisms and Legacies for State-Building. Available: http://eprints.soas.ac.uk/11685/1/Bangladesh_paper.pdf (accessed 18 May 2017). Khan, M.H. (2010c). Bangladesh: Economic Growth in a Vulnerable LAO. Paper presented at the World Bank, 4 June 2010. Khan, S. (2010) ‘Does Hartal Deliver in Democracy?’ on the London-based e-Bangladesh portal. Khan, S. (2012). Analysing Public Sector Performance in Pakistan: A Study of Interface with State Bureaucracy and Citizens. Unpublished PhD, University of Bath, UK. Khanna, A. (2012). ‘Seeing Citizen Action through an “Unruly” Lens’, Development, 55(2): 162–172. Khondker, H.H. (2007). ‘Class and State in Bangladesh: Political Economy of Authoritarian Democracy’, in M.F. Islam and S. Andaleeb (eds), Development Issues of Bangladesh III, Dhaka: University Press, pp. 83–110. Khondker, H.H. (2010). ‘The Curious Case of Secularism in Bangladesh: What is the Relevance for the Muslim Majority Democracies?’, Totalitarian Movements and Political Religions, 11(2): 185–201. Khundker, N. (2005). ‘The Price of Hartals: Impact on Economy’, in Beyond Hartal. Dhaka: United Nations Development Programme. Kochanek, S.A. (1993). Patron-Client Politics and Business in Bangladesh. New Delhi: Sage. Krugman, P. (2011). ‘Oligarchy, American Style’, New York Times, 3 November. Kulkarni, A. (2008). ‘Governance and the Right to Information in Maharashtra’, Economic and Political Weekly, 43(35): 15–17. Lagi, M., K. Bertrand and Y. Bar-Yam (2011). The Food Crises and Political Instability in North Africa and the Middle East. Available: https://papers.ssrn.com/sol3/papers. cfm?abstract_id=1910031 (accessed 23 May 2017). Landell-Mills, P. (2002). Taming Leviathan: Reforming Governance in Bangladesh. An Institutional Review. Washington, D.C.: World Bank. Lateef, S. (2006). Why South Asia Grows Despite Weak Public Accountability: Understanding the Bangladesh Conundrum. Washington, D.C.: World Bank. Latif Mondal, M.A. (2011). ‘Election before Dissolution of Parliament’, Daily Star, 7 July. Lewis, D. (2010). ‘Disciplined Activists, Unruly Brokers? Exploring the Boundaries between Non-Governmental Organisations, Donors and the State in Bangladesh’, in D.N. Gellner (ed.), Varieties of Activist Experience: Civil Society in South Asia. New Delhi: Sage Publications, pp. 159–180. Lewis, D. (2011). Bangladesh: Politics, Economy and Civil Society. Cambridge: Cambridge University Press.

References 185 Lewis, D. and A. Hossain (2008). Understanding the Local Power Structure in Rural Bangladesh. Stockholm: Sida. Lewis, D. and Kanji, N. (2009). Non-Governmental Organisations and Development (Routledge Perspectives on Development). London: Routledge. Linz, J.J. and A. Stepan (1978). The Breakdown of Democratic Regimes. Baltimore: Johns Hopkins University Press. Liton, S. (2010). ‘Boycott Culture Crippling Parliament’, Daily Star, 23 February. Liton, S. (2011). ‘JS Dodging to Get Minimized’, Daily Star, 30 March. Liton, S. and S. Ashraf (2011a). ‘ “Minus 2” Met Messy Fate’, Daily Star, 16 September. Liton, S. and S. Ashraf (2011b). ‘Tarique Symbol of Violent Politics’, Daily Star, 9 September. Liton, S. and R. Hasan (2011). ‘Caretaker System Abolished; JS Passes Constitution’s 15th Amendment Bill amid BNP’s Absence’, Daily Star, 1 July. Long, N. (1992). ‘From Paradigm Lost to Paradigm Regained: The Case for an ActorOriented Sociology of Development’, in N. Long and A. Long (eds), Battlefields of Knowledge: The Interlocking of Theory and Practice in Social Research and Development. London and New York: Routledge, pp. 16–47. Lovell, C.H. (1992). Breaking the Cycle of Poverty: The BRAC Strategy. West Hartford, CT: Kumarian Press. Ludden, D. (2011). ‘The Politics of Independence in Bangladesh’, Economic and Political Weekly, 46(35): 79–85. Lukes, S. (1974). Power: A Radical View. 2nd edn, 2004. Basingstoke: Palgrave MacMillan. Mahathir, Dr Bin Mohamad (2012). Speech at University Santo Tomas’s Special Convocation and Conferment of Honorary Professorship and the Neo-Centennial Lecture in Manila, Philippines on 11 June, 2012. Available: www.kln.gov.my/web/phl_manila/ s2012/-/asset_publisher/ME2g/blog/yabhg-tun-dr-mahathir-mohamad-neo-centennialkeynote-lecture-universiti-of-santo-tomas?redirect=%2Fweb%2Fphl_manila%2Fs2012 (accessed 23 May 2017). Mahbubani, K. (1992). ‘The West and the Rest’, in National Interest, 28, Summer: 3–12. Mannan, M. (2005). ‘Rural Power Structures and Evolving Market Forces in Bangladesh’, in K.B. Ghimire (ed.), Civil Society and the Market Question. London: Palgrave. Mannan, M. (2010). Enslaving Development: An Anthropological Enquiry into the World of NGOs. Doctoral thesis, Durham University. Marcuse, H. (1969). ‘Repressive Tolerance’, in R.P. Wolff, B. Moore, Jr. and H. Marcuse (eds), A Critique of Pure Tolerance. Boston: Beacon Press, pp. 95–137. Markoff, J. (1996). Waves of Democracy. Thousand Oaks, CA: Pine Forge Press. Mattes, R. and M. Bratton (2007). ‘Learning about Democracy in Africa: Performance, Awareness and Experience’, American Journal of Political Science, 51(1): 192–217. McAdam, D. and W.R. Scott (2005). ‘Organizations and Movements’, in G.F. Davis, D. McAdam, W.R. Scott and M.N. Zald (eds), Social Movements and Organization Theory. Cambridge: Cambridge University Press, pp. 4–40. McGillivray, M., S. Feeny, N. Hermes and R. Lensink (2005). It Works; It Doesn’t; It Can, but That Depends … 50 Years of Controversy over the Macroeconomic Impact of Development Aid. Research Paper No. 2005/54. Helsinki: United Nations UniversityWider. McGregor, J.A. (1989). ‘Towards a Better Understanding of Credit in Rural Development. The Case of Bangladesh: The Patron State’, Journal of International Development, 1(4): 467–486.



McGregor, J.A., L. Camfield and A. Woodcock (2009). ‘Needs, Wants and Goals: Wellbeing, Quality of Life and Public Policy’, Applied Research in Quality of Life, 4(2): 135–154. doi:10.1007/s11482-009-9069-7. Michels, R. (1962). Political Parties: A Sociological Study of Oligarchical Tendencies of Modern Democracy. New York: Free Press. Moniruzzaman, M. (2009). ‘Party Politics and Political Violence in Bangladesh: Issues, Manifestation and Consequences’, South Asian Survey, 16(1): 81–99. Moore, B. Jr (1966). Social Origins of Dictatorship and Democracy. Harmondsworth: Penguin Books. Moore, B. (1978). Injustice: The Social Bases of Obedience and Revolt. White Plains, NY: M.E. Sharpe. Mosse, D. (2005). Cultivating Development. London: Pluto Press. Mouffe, C. (2005). On the Political. New York: Routledge. Murray, J., B. Tshabangu and N. Erlank (2010). ‘Enhancing Participatory Governance and Fostering Active Citizenship: An Overview of Local and International Best Practices’, Politikon, 7(1): 45–66. Myrdal, G. (1968). Asian Drama. New York: Pantheon Books. Nasr, S.V.R. (2005). ‘The Rise of “Muslim Democracy” ’, Journal of Democracy, 16(2): 13–27. Nazneen, S. (2009). ‘ “Something is Better than Nothing?” Political Party Discourses on Women’s Empowerment in Bangladesh’, South Asian Journal, 24 June: 44–52. Nazneen, S. and M. Sultan (2009). ‘Struggling for Survival and Autonomy: Impact of NGO-ization on Women’s Organizations in Bangladesh’, Development, 52(2): 193–199. Nazneen, S. and M. Sultan (2010). ‘Reciprocity, Distancing, and Opportunistic Overtures: Women’s Organisations Negotiating Legitimacy and Space in Bangladesh’, IDS Bulletin, 41(2): 70–78. NDI (National Democratic Institute) (2009). International Observation Mission, ‘2008/09 Bangladesh Elections, Final Report’. Washington, June. Available: www.ndi.org/files/ Bangladesh_Elections_Report_June2009.pdf (accessed 29 September 2011). Nijera Kori (2011). Website on training activities. Available: www.nijerakori.org/index. php?option=com_content&task=view&id=13&Itemid=27 (accessed 11 October 2011). North, D. (1990). Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. North, D. (2005). Understanding the Process of Economic Change. Princeton: Princeton University Press. North, D., J.J. Wallis and B.R. Weingast (2009). Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. New York: Cambridge University Press. O’Donnell, G. (1986). Bureaucratic Authoritarianism: Argentina 1966–1973 in Comparative Perspective. Oakland, CA: University of California Press. O’Donnell, G. (2004). ‘An Exploration of the Development of Democracy’, in D. Caputo (ed.), Democracy in Latin America: Towards a Citizen’s Democracy. United Nations Development Program. Offe, C. (2011). ‘Crisis and Innovation of Liberal Democracy: Can Deliberation be Institutionalized?’, Czech Sociological Review, 27(3): 447–472. Oliver de Sardan, J.-P. (2008). Researching the Practical Norms of Real Governance in Africa. Discussion Paper No. 5. London: Overseas Development Institute (Africa Power and Politics Programme), December.

References 187 Ong, A. (1999). ‘Clash of Civilizations or Asian Liberalism? An Anthropology of the State and Citizenship’, in H. Moore (ed.), Anthropological Theory Today. Oxford: Polity Press, pp. 48–72. Osman, F.A. (2010). ‘Bangladesh Politics: Confrontation, Monopoly and Crisis in Governance’, Asian Journal of Political Science, 18(3): 310–333. Parnini, S. and M.D. Othman (2014). ‘Democratic Consolidation and Credibility of Governance Institutions in Bangladesh’, Journal of Asian and African Studies, 49(1): 34–48. Pfeffer, J. and G. Salancik (1978). The External Control of Organizations: A Resource Dependence Perspective. New York: Harper and Row. Polanyi, K. (1944). The Great Transformation. New York: Rinehart. Price, P.G. (1989). ‘Kingly Models in Indian Political Behavior: Culture as a Medium of History’, Asian Survey, 29(6): 559–572. Quadir, F. (2010). Consolidating Democracy without Trust: Bangladesh’s Breakdown of Consensus in 2007, The Round Table: The Commonwealth Journal of International Affairs, 99(406): 65–73. Rahaman, M.M. (2007). ‘Origins and Pitfalls of Confrontational Politics in Bangladesh’, South Asian Survey, 14: 101–115. Rahaman, M.M., M.R. Islam, M. Abdullah and S.J. Rayhan (2011). ‘Female Workers’ Attitudes on Industrial Disputes at RMG Sector in Bangladesh: An Empirical Analysis’, Journal of Economics and Sustainable Development, 2(3): 76–86. Rahman, H.Z. (2009). Unbundling Governance: Indices, Institutions, Solutions. Dhaka: Power and Participation Research Centre. Rahman, M., D. Bhattacharya, M.A. Iqbal, T.I. Khan and T.K. Paul (2009). Global Financial Crisis Discussion Series, Paper 1: Bangladesh. London: ODI. Rahman, M., A. Iqbal, T.I. Khan and S. Dasgupta (2010). Global Financial Crisis Discussion Series, Paper 12, Bangladesh Phase 2. London: ODI. Rahman, M., K.G. Moazzem and S.S. Hossain (2009). Impact of the Global Economic Crisis on the Employment and Labour Market of Bangladesh: A Preliminary Assessment. Dhaka: Centre for Policy Dialogue. Rahman, S. (2006). ‘Development, Democracy and the NGO Sector: Theory and Evidence from Bangladesh’, Journal of Developing Societies, 22(4): 451–473. Rashid, M., B. Hasan and N. Hossain (2012). ‘Bangladesh: Pathways and Impacts of the Global Economic Shocks’, in R. Heltberg, N. Hossain and A. Reva (eds), Living Through Crises: How the Food, Fuel and Financial Shocks Affect the Poor. New Frontiers of Social Policy. Washington, D.C.: World Bank, pp. 61–89. Riaz, A. (2004). ‘The Politics of Vengeance and the Erosion of Democracy’, Asian Survey, 45(1): 112–118. Riaz, A. (2008). Islamic Militancy in Bangladesh: A Complex Web. London and New York: Routledge. Riaz, A. (2013). ‘Democracy in Bangladesh: A Report Card’, South Asia Journal, 7. Available: http://southasiajournal.net/democracy-in-bangladesh-a-report-card (accessed 24 May 2017). Roberts, A. (2010). ‘A Great and Revolutionary Law? The First Four Years of India’s Right to Information Act’, Public Administration Review, 70(6), November/December: 925–933. Rondinelli, D. (1993). Development Projects as Policy Experiments. London: Routledge.



Rose, N. (1999). Power of Freedom: Reframing Political Thought. Cambridge: Cambridge University Press. Rudé, G.F.E. (1981). The Crowd in History: A Study of Popular Disturbances in France and England, 1730–1848. London: Lawrence and Wishart . Ruud, A.E. (2011). ‘Democracy in Bangladesh: A Village View’, in S.T. Madsen, K.B. Nielsen and U. Skoda (eds), Trysts with Democracy: Political Practice in South Asia. London, New York, Delhi: Anthem Press, pp. 45–70. Sanyal, B. (1991). ‘Antagonistic Cooperation: A Case Study of Nongovernmental Organisations’, Government and Donors’ Relationships in Income-Generating Projects in Bangladesh’, World Development, 19(10): 1367–1379. Sarkar, A. (2011a). ‘Caretaker System Declared Illegal’, Daily Star, 11 May. Sarkar, A. (2011b). ‘Caretaker Illegal Right after Verdict’, Daily Star, 17 June. Sarkar, A. and S. Liton. (2011). ‘Caretaker Govt Verdict Not Contradictory: Just-Retired CJ Defends Scope for Next 2 Polls under the ‘Illegal’ System’, Daily Star, 18 May. Schneider, M. (2008). ‘We Are Hungry!’ A Summary Report of Food Riots, Government Responses, and States of Democracy in 2008. Ithaca, NY: Cornell University Schwab, K., Sala-i-Martin, X. and Brende, B. (2013). The Global Competitiveness Report 2012–2013. Geneva: World Economic Forum. Available: www3.weforum.org/docs/ WEF_GlobalCompetitivenessReport_2012-13.pdf (accessed 23 May 2017). Scott, J.C. (1977). The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia. New Haven, CT: Yale University Press. Scott, J.C. (1987). Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven, CT: Yale University Press. Scott, J.C. (1992). Domination and the Arts of Resistance: Hidden Transcripts. New Haven, CT: Yale University Press. Scott, J.W. and L.A. Tilly (1975). ‘Women’s Work and the Family in Nineteenth-Century Europe’, Comparative Studies in Society and History, 17(1): 36–64. doi:10.1017/ S0010417500007635. Seibel, W. (1999). ‘Successful Failure: An Alternative View of Organizational Coping’, in H.K. Anheier (ed.), When Things Go Wrong: Organizational Failures and Breakdowns. London: Sage, pp. 91–104. Selznick, P. (1953). TVA and the Grassroots:A Study of Politics and Organization. Berkeley and Los Angeles: University of California Press. Sen, A. (1997). ‘Human Rights and Asian Values’, The New Republic, 14–21 July. Shankland, A., M. Tadros, N. Hossain, P. Scott-Villers and A. Khanna (2011). Unruly Politics: A Manifesto. Institute of Development Studies, University of Sussex (mimeo). Sharp, G. (1994). The Role of Power in Nonviolent Struggle. East Boston, MA: Albert Einstein Institution. Shastri S. and S. Palshikar (2010). ‚Democratizing the Meaning of Democracy’, Paper presented at Global Barometer Surveys Conference, 15–16 October 2010, Taipei. Siddiqi, K. (2000). Jagatpur 1977–97: Poverty and Social Change in Rural Bangladesh. Dhaka: The University Press. Smillie, I. (2009). Freedom from Want: The Remarkable Success of BRAC, the Global Grassroots Organization That’s Winning the Fight against Poverty. Sterling, VA: Kumarian Press. Smillie, I. and J. Hailey (2001). Managing for Change: Leadership, Strategy and Management in Asian NGOs. London: Earthscan. Sobhan, S. (2011). ‘RTI Act, 2009: Present Status and Scope’, Daily Star, Supplement on 40 Years of Independence, 7 April.

References 189 Staff Correspondent (2009). ‘MPs Retain Power over UZ Parishad: JS Passes Law Making Legislators’ Advice Mandatory for the Local Govt Body’, Daily Star, 7 April. Staff Correspondent (2011a). ‘Bhuiyan Wanted to Replace Khaleda’, New Age, 13 September. Staff Correspondent (2011b). ‘Khaleda was Asked to Go into Exile’, New Age, 13 September. Staff Correspondent (2011c). ‘Hasina, Khaleda Arrested for Not Giving up Politics’, New Age, 14 September. Staff Correspondent (2011d). ‘Caretaker System Abolished’, New Age, 1 July. Standing, G. (2011). The Precariat: The New Dangerous Class. London: Bloomsbury Academic. Star Report (2011a) ‘King’s Party Circus’, Daily Star, 16 September. Star Report (2011b). ‘Of Miscarriage of Reform, Its Victims’, Daily Star, 16 September. Star Report (2011c). ‘Advisers Acted Differently’, Daily Star, 17 September. Star Report (2011d). ‘Army Pressed for Nat’l Unity Gov’t’, Daily Star, 18 September. Stiles, K. (2002). ‘International Support for NGOs in Bangladesh: Some Unintended Consequences’, World Development, 30(5): 835–846. Suman, R.H. (2009). ‘Suggestions of Experts, UZ Chairmen Ignored’, Daily Star, 7 April. Suman, R.H. (2011). ‘No Caretaker, No Polls’, Daily Star, 11 October. Suykens, B. and A. Islam (2013). ‘Hartal as a Complex Political Performance: General Strikes and the Organisation of (Local) Power in Bangladesh’, Contributions to Indian Sociology, 47(1), 1 February: 61–83. Tadros, M. (2012). ‘Introduction: The Pulse of the Arab Revolt’, IDS Bulletin, 43(1): 1–15. Tarascio, V.J. (1968). Pareto’s Methodological Approach to Economics: A Study in the History of Some Scientific Aspects of Economic Thought. Chapel Hill, NC: University of North Carolina Press. Tarrow, S.G. (2011). Power in Movement: Social Movements and Contentious Politics. Cambridge: Cambridge University Press. Taylor, L. (1996). ‘Food Riots Revisited’, Journal of Social History, 30(2), 1 December: 483–496. doi:10.2307/3789390. Thompson, E.P. (1971). ‘The Moral Economy of the English Crowd in the Eighteenth Century’, Past and Present, 50: 76–136. TIB (Transparency International Bangladesh) (2006). Problems of Governance in the NGO Sector: The Way Out. Dhaka. Tilly, C. (2008a). Contentious Performances. Cambridge: Cambridge University Press. Tilly, C. (2008b). ‘Describing, Measuring, and Explaining Struggle’, Qualitative Sociology, 31(1), March: 1–13. Tilly, L.A. (1971). ‘The Food Riot as a Form of Political Conflict in France’, Journal of Interdisciplinary History, 2(1), 1 July: 23–57. doi:10.2307/202442. Tilly, L.A. (1983). ‘Food Entitlement, Famine, and Conflict’, Journal of Interdisciplinary History, 14(2), 1 October: 333–349. doi:10.2307/203708. Tilly, L.A. (1987). Women, Work and Family. London: Routledge. Transparency International (2014). 2014 Corruption Perceptions Index. Available: www. transparency.org/cpi2014 (accessed 24 May 2017). Trist, E. (1977). ‘A Concept of Organizational Ecology’, Australian Journal of Management, 2: 161–175. Uddin, S. (2006). Constructing Bangladesh: Religion, Ethnicity, and Language in an Islamic Nation, Chapel Hill, NC: University of North Carolina Press.



United Nations (1986). Declaration on the Right to Development, A/RES/41/128. www. un.org/documents/ga/res/41/a41r128.htm (accessed 24 May 2017). UNDP (United Nations Development Programme) (2005). Beyond Hartals: Towards Democratic Dialogue in Bangladesh. Dhaka. Walton, J.K. and D. Seddon (2008). Free Markets and Food Riots: The Politics of Global Adjustment. Hoboken, NJ: Wiley-Blackwell. Weber, M. (1919). ‘Politics as a Vocation’. Available: http://anthropos-lab.net/wp/wpcontent/uploads/2011/12/Weber-Politics-as-a-Vocation.pdf (accessed 24 May 2017). Westergaard, K. and A. Hossain (2005). Boringram Revisited: Persistent Power Structure and Agricultural Growth in a Bangladesh Village. Dhaka: The University Press. Westney, D.E. (1987). Imitation and Innovation: The Transfer of Western Organizational Patters to Meiji Japan. Cambridge, MA: Harvard University Press. White, S.C. (1996) ‘Depoliticising Development: The Uses and Abuses and Participation’, Development in Practice Vol. 6(1) 6–15. White, S.C. (1999). ‘NGOs, Civil Society, and the State in Bangladesh: The Politics of Representing the Poor’, Development and Change, 30(3): 307–326. White, S.C. (2009). Bringing Wellbeing into Development Practice. University of Bath Wellbeing in Developing Countries Research Group. Woldring, H.E.S. (2000). ‘State and Civil Society in the Political Philosophy of Alexis de Tocqueville’, Voluntas: International Journal of Voluntary and Non-Profit Organisations, 9(4): 363–373. Wood, G.D. (1981). ‘Rural Class Formation in Bangladesh 1940–80’, Bulletin of Concerned Asian Scholars, 13(4): 2–17. Wood, G.D. (1985). Labelling in Development Policy: Essays in Honour of Bernard Schaffer. London: Sage Publications. Wood, G.D. (1997) ‘States without Citizens: The Problem of the Franchise State’, in D. Hulme and M. Edwards (eds), Too Close for Comfort? NGOs, States and Donors. London: Macmillan, Chapter 5. Wood, G.D. (2000). ‘Prisoners and Escapees: Improving the Institutional Responsibility Square in Bangladesh’, Public Administration and Development, 20(3): 221–237. Wood, G.D. (2003). ‘Staying Secure, Staying Poor: The “Faustian Bargain” ’, World Development, 31(3): 455–471. Wood, G.D. (2007a). ‘Labels, Welfare Regimes and Intermediation: Contesting Formal Power’, in J. Moncrieffe and R. Eyben (eds), The Power of Labelling: How People are Categorized and Why it Matters. London: Earthscan, pp. 17–32. Wood, G.D. (2007b). ‘Clashing Values in Bangladesh: Secularism, NGOs and the Ummah’, in F.C. Khan, A. Ahmad and M. Quddus (eds), Recreating the Commons? NGOs in Bangladesh. Dhaka: The University Press, pp. 43–78. Wood, G.D. and I.R. Gough (2006). ‘A Comparative Welfare Regime Approach to Global Social Policy’, World Development, 34(10): 1696–1712. Woolcock, M., S. Szreter and V. Rao (2011). ‘How and Why does History Matter for Development Policy?’, Journal of Development Studies, 47(1): 70–96. World Bank (1983). World Development Report 1983. New York: Oxford University Press. Available: https://openknowledge.worldbank.org/handle/10986/5966 (accessed 18 May 2013). World Bank (1989). Sub-Sahara African: From Crisis to Sustainable Development. Washington, D.C.: World Bank. World Bank (1997). World Development Report 1997: The State in a Changing World. New York: Oxford University Press.

References 191 World Bank (2006). Economics and Governance of Nongovernmental Organizations in Bangladesh. Bangladesh Development Series, Paper No. 11. Dhaka. World Bank (2010). Bangladesh Country Assistance Strategy 2012–2014. Dhaka. World Bank (2011). World Development Report 2011. New York: Oxford University Press. World Bank (2012). Migration and Remittance Data. Available: http://go.worldbank. org/092X1CHHD0 (accessed 14 February 2013). World Bank (2013). Bangladesh Poverty Assessment: Assessing a Decade of Progress in Reducing Poverty, 2000–2010. Bangladesh Development Series, Paper No. 31. Dhaka. World Bank Institute (2011). The Power of Using the Right to Information Act in Bangladesh: Experiences from the Ground. Available: http://wbi.worldbank.org/wbi/Data/ wbi/wbicms/files/drupal-acquia/wbi/Final%20version%20-%20The%20Benefits%20 of%20Using%20the%20RTI%20Act%20in%20Bangladesh.pdf (accessed 19 October 2011). Žižek, S. (2010). ‘A Permanent Economic Emergency’, New Left Review, 64: 85–95.


Page numbers in italics denote tables, those in bold denote figures accountability: accountability principles, NGOs 139–40; negotiated 58; ‘rude’ 59, 150; short loop 2 Acemoglu, D. 121 Afrobarometer 88, 89 Agamben, G. 156 agency, security of 116 agrarian legacy 5–6 Ahmed, Fakruddin 7 Ahmed, General Moeen Uddin 23 aid agencies 54, 60n21 AL see Awami League (AL) Alavi, H. 113 ambiguity 93–8; ambiguous democrats 96, 97, 98; ambiguous theocrats 96, 98; around governance 99, 100–1, 102 analytical relativism 2, 107 Anderson, Perry 152–3 Anti-Corruption Commission (ACC): governance reform drivers 42, 45, 52; and party dysfunction 24, 28, 29, 31, 36 Anticorruption Drive and the Way Forward, roundtable discussion 60n16 Arab Spring 152 Asian barometers 89 Asian financial crisis 77 “Asian Tigers” 68, 74 Association of Development Agencies in Bangladesh (ADAB) 118, 131 authoritarian democracy, Bangladeshi model as 14, 40 Awami League (AL) 4, 5, 7, 8, 9, 89, 114, 119, 131, 172; evidence from Bangladesh 89, 91; and party dysfunction 19, 22, 24, 26, 27, 28, 30, 36; state building paradigm 42, 43; see also BAKSAL

BAKSAL (coalition) 7, 9 Bangladesh: agrarian legacy 5–6; compared to other high-performing Asian economies 14; delta inhabitants 48; elections in see elections in Bangladesh; independence from Pakistan 1; NGOs in 127–30; party dysfunction see dysfunction, party; political stability 32–3; postindependence period 43; ‘prison’ metaphor 4, 10–12, 13, 15, 120; recent political history 43; state building paradigm 42–6; unruly politics in 158–64; upazila parishads (UZPs) see upazila parishads (UZPs) Bangladesh Environmental Lawyers Association 34 Bangladesh Garment Manufacturers’ and Exporters’ Association 60n15 Bangladesh Legal Aid Society Trust 34 Bangladesh Nationalist Party (BNP) 4, 7, 8, 43, 119, 131, 172; evidence from Bangladesh 89, 90, 91; formation 9; and governance reform drivers 44; and party dysfunction 19, 22, 24, 28, 31 bangsho (lineage) 6, 16n1, 114, 123n12 bargadars (sharecroppers) 8 Barnett, A. 151 barricades, opposition to 21 Basu, Ipshita 14–15 Bhuiyan, Abdul Mannan (BNP Secretary General) 37n10 bhumihin (landless) 8 biological sciences, homeostasis in 17 Blair, Harry 10, 13–14, 128 BNP see Bangladesh Nationalist Party (BNP)

Index Bouazizi, Mohamed 152–3 boycotts, parliamentary 25 Bratton, M. 88, 89 Brazil 70 Brett, E.A. 137 BRIC (Brazil, Russia, India and China) 62, 65, 77, 82 British Conservative Party 24 British Labour Party 24 Brown, Graham K., 14–15 Bureau of Anti-Corruption 28 Burke, Edmund 123n1 ‘business as usual’ 10 Cambodia 58 Canadian Tories 24 Canadian University Service Overseas (CUSO) 130 capitalism 17, 122, 124n35; crony 77; global 151, 152, 165 Caretaker Government (CTG) 12, 20, 25, 32, 37, 42; abolition 36; caretaker interregnum and elections of 2008 22–4; interim (2006–7) 7, 8, 10; militarybacked (2007–8), 7, 14, 18, 52; Right to Information (RTI) ordinance 24, 29, 33–5, 34, 58 Carlyle, Thomas 123n1 Centre for Policy Dialogue 50, 143 Changi International Airport 76 Chatterjee, Partha 150, 157, 166n2 China 73, 77 Chittagong Port 52 circulating elites 112 citizenship democracy 67 citizenship rights 69 civic disruption 25; see also hartals (strikes) Civil Disobedience 158 Civil Liberties see Political Rights and Civil Liberties scores; Freedom House civil liberties 80–1 civil society 33–5, 49, 57; demand side role 116–17; natural state, significance for 108–10; surreal context for 114–15; whether free of the state 15 civil society organisations (CSOs) 16n9, 25, 33, 34, 82; advocacy 53; governance reform drivers 47, 50, 51, 52 Clarke, L. 126, 141 Clientelism 138, 164; see also patronclientelism colonialism, British 5, 114


Comilla Proshikha (NGO) 15, 130, 132–4, 138, 139 Committees of Concerned Citizens 52 constituency development funds 26–7 Constitution 1972 28, 29 co-opted inclusion 123n16 Corruption Perception Index for Bangladesh 60n28, 75, 169 crony capitalism 77 Cuba 16n7 Cultivating Development (Mosse) 127 Dahl, Robert 2 Daily Star newspaper 14, 16n14, 25, 93, 162 decentralisation 84 Declaration of the Right to Development, UN (1986) 69 deep structures 3, 4, 10–12, 13, 14; social origins 113–14 democracy: ambiguity 93–8, 99, 100–1, 102; authoritarian, Bangladeshi model as 14; barometers 88; citizenship 67; demand for 86–106; and development 79–82; electoral 14, 20, 67; ethical conceptualisation 103; quality of 67; support for 93–8; trajectory in Bangladesh 18–20 democracy index, South Asia 66 Democracy Poll survey 14 Department for International Development (DFID) 70 De Souza, P. R. 96, 103 De Tocqueville, Alexis 15, 109 developing countries 40 development, and democracy 79–82 Development Research Centre (DRC) 166n8 development studies 2 Devine, Joe 14–15, 128 Dhaka 23; elites in 5, 8; meeting in late 2011 2 Dhaka–Comilla belt 5, 114 Dhaka Proshika (NGO) 131–2 Domination and the Arts of Resistance (Scott) 154–5 donors, support for governance reform 54–8 dysfunction, party 17–29; caretaker interregnum and elections of 2008 22–4; civic disruption 25; instability and stability 13–14; “minus two” strategy 22, 23; parliamentary boycotts 25; prognosis 31–2; Rules of the Game 14,



dysfunction, party continued 18, 20–2, 28; and stability 32–3; upazila parishads (UZPs) 25–9; volatile twoparty system in Bangladesh 17, 19; see also homeostasis East Bengal 1, 5, 30 East Indian Company 5, 114 East Pakistan 38n22 Economic Environment 34–5 economic growth 14, 36–7, 62–85 Economic Origins of Dictatorship and Democracy (Acemoglu and Robinson) 121 Economist Intelligence Unit (EUI) 80 Edwards, Michael 129 Election Commission 52, 172 elections in Bangladesh: bogus election of February 1996 27–8; caretaker interregnum and elections of 2008 22–4; electoral democracy 14, 20, 67 elites 15; circulating 112; of Dhaka 5, 8; of Punjab 5 epistemological perspective 4–5 Ershad, Mohammad 7, 8, 9, 19, 24, 25–6, 43, 81, 92 Establishments Division, central government 11, 46 export manufacturing 46 factor loadings in indices 99 failed states 68 Fair Election Monitoring Alliance (FEMA) 172 Faustian bargain arguments 4, 11, 145 Feedback Unit, Ministry of Community Development 76 Fifteenth Constitutional Amendment 92 food riots 151 Freedom House 18, 19, 20, 34; Freedom of the Press report 36 Freire, Paolo 130 Friend of Bengal (Bangabondhu) 19 Fukuyama, Francis 76 gangsterism, in public life 21 garment workers, protests of (2006–2011) 144, 161–4 Gemeinschaft and Gesellschaft 110, 115 Global Competitiveness Report (2012–2013) 65, 74 see Anti-Corruption Commission (ACC) Global South 170 globalisation, 46

Gonoshahajjo Sangstha (GSS) 131 good governance 1, 5, 11, 116; agenda 45, 104, 144; challenges 170, 174; evidence from Bangladesh 64, 68, 69, 74, 86, 89; governance reform drivers 45, 51, 52, 59; reforms 146, 170; unruliness in politics 143, 144, 146 governance of Bangladesh 86–106; agenda 164–6; assessment of current position 172–4; background 1–16; challenges 168–76; deep structures 10–12; factor loadings in indices 99; frustration and hope, between 89–92; good governance see good governance; institutions and organisations 6–7; methods 93; participatory governance 67, 69; poor 14, 55; potential future developments 174–6; probit regressions 102; realityrhetoric gap 6–7; reform, drivers of see governance reform drivers; support for democracy 93–8; terms of contestation of power since liberation 7–10; using to understand political developments 169–72 governance reform drivers 14, 39–61; business groups/businessmen 47, 48; challenges 39; donor support for reform 54–8; and economic progress 41; independent media 50; NGOs 50–1, 56; paradigm for state building 39–46; policy research centres 50; potential 47–54; professional organisations 49; supply-side reform 12, 55; trade unions 49 Grameen Bank 46, 82, 129 Gramsci, Antonio 15 Hailey, J. 129 hartals (strikes) 19, 20, 25, 30, 36; evidence from Bangladesh 90, 91; and governance reform drivers 44; hartal repertoire, slow evolution of 158–61; Rules of the Game 21–2; unruliness in politics 144, 145, 158–9, 160, 161, 164–5 Hasan, S. 128, 169 Hashemi, S.M. 128, 131 Hasinia, Sheikh 23, 159 High Court 35 high-performing Asian countries 67 homeostasis metaphor: breaking away from homeostasis 29–30; concept of homeostasis 17; relevance to Bangladesh 17–18; use of 37n2;

Index using to describe recent politics in Bangladesh 13–14; whether homeostasis misperceived 30–1; see also dysfunction; party Hossain, Naomi 2, 10, 13, 15–16 Hossian, Kamal 84 Human Development Indices (HDI) 62, 63 Human Resources (HR) 10 Huntington “two turnover test” 20, 87 Huq, Lopita 131, 141, 147 India 70, 73, 75, 77; Member of Parliament Local Area Development Scheme (MPLADS) 27 Indonesia 70, 73, 74, 77 industrial action see hartals (strikes) Information and Communication Technology Act 2013 36 Information Commissioner 72 Institute of Development Studies 150 Institute of Governance Studies (IGS) 143; State of Governance (SOG) Reports see State of Governance Reports (BRAC Institute of Governance) institutional development 41 institutional responsibility 4, 111 institutions and organisations 6–7 International Crimes Tribune (ICT) 173 International Development Agency (CIDA), Canada 130, 133, 134 Iranian Green Movement 155 Islam 91, 92, 95, 97; see also Jamaat-eIslami (fundamentalist party) Islam, A. 160 Islamic State 95, 96, 100, 105 Jamaat-e-Islami (fundamentalist party) 7, 89, 92, 173 James, Henry 141 Jatiyo Party 8, 9, 27, 89 Jatiyo Sangsad (parliament) 6, 8, 24, 115, 119 jotedars (rich peasantry) 8 judiciary, two-track 21 Kabeer, Naila 131, 141, 147 Kendall, J. 141 ‘kettling’ 154 Khan, Mushtaq 43, 136, 143, 146–9 Khan, Yahya 31 Khanna, A. 154 Khondker, Habibul Haque 14, 15 kinship 4 Kissinger, Henry 39


Kropotkin, Peter 137 Krugman, Paul 80 Landell-Mills, Pierre 10, 13, 14, 113 Latin America 108 Latinobarometer 89 “Lee hypothesis” 64 Lee Kuan Yew 64 Legal Environment 35 Lewis, David 3, 13, 15, 120 liberation 5, 7 liberation military forces 9 limited access orders 13, 15, 39 limited access states 111–13 Linz, J.J. 135 local governance as patronage mechanism 21 Lovell, C.H. 129 Low Income Countries (LICs) 169 Lower Middle Income Countries (LMICs) 169 Lukes, Stephen 2, 109 Mahathir, Mohamed 72 Mahila Parishad (women’s organisation) 52 Mahmud, Simeen 147 Malaysia 72, 74, 75 Mannan, M. 128, 134 Manusher Jonno Foundation (MJF) 52 Mason, Paul 153 mass movements 58 mastaan (underworld gang leaders) 3, 6, 11, 16n3, 46, 60n10, 123n14; mastaanisation 6, 57, 114; mastaanised processes 122; and natural state 118, 119, 121; and party dysfunction 20, 21 Mattes, R. 88 Member of Parliament Local Area Development Scheme (MPLADS), India 27 Members of Parliament (MPs) 26, 27, 28 Michels, Robert 135, 138 military coups 40 Millennium Development Goals (MDGs) 67, 86 Mills, C. Wright 112 Ministry of Local Government, Rural Development and Cooperatives 26 Moniruzzaman, M. 158, 160 moral economies and laws of necessity 155–7 Mosse, David 127 Multi-Fibre Agreement 43



Musharraf, Parvez 23 Myrdal, Gunnar 68 Nari Pokkho (radical organisation) 52 nationalism, Bengali 8 National Women’s Act 2011 160 natural state 107–24; apparent and hidden behaviour 110–11; deep structures 113–14; demand side, sustaining 116–17; holistic and marginal risk 116–17; limited access states 111–13; permeability concept 110; PUMUS case 117–20; significance of, for civil society 108–10; surreal context for civil society 114–15 Nazneen, Sohela 147 Nepal 58, 73 neta-karma (leader-worker) relations 90 NGOs (non-governmental organisations) 3, 9, 29, 33, 117; accountability principles 139–40; in Bangladesh 127–30; development 56; effective hierarchies building 138; failure 15, 125–42; governance reform drivers 50–1; lessons to be learnt 137–40; making sense of ‘things going wrong’ 134–7; narratives of decline, disintegration and ‘failure’ 130–2; NGO Affairs Bureau 33, 118; operating principles and professionalism 138–9; organizational breakdown and ‘failure’ 125–42; process and progress ‘failure’ as 140–2; social enterprise model of NGO form 141 Nielson 14 Nijera Kori (NGO) 34, 51, 129 Ninth Parliament 24 non-state actors 129, 137; governance reform drivers 41, 47, 56, 57, 59, 60 North, Douglas 3, 4, 13, 39, 40, 60n4, 108–9, 111, 113, 121, 171 North–Wallis–Weingast paradigm 43, 44, 45, 46, 53, 107 ‘obstacles to development’ 2 Occupy movements 154 O’Donnell, Guillermo 67 Offe, Claus 64 OLS regression 100 open access orders 15, 40 open access societies 4, 5, 112 organised life 21 Overseas Filipino workers (OFW) 78 Pakistan 73, 75; East Pakistan 9, 38n22;

independence of Bangladesh from 1; Pakistani identity 8 Pakistan Army 31 Pakistan Day holiday 31 Palli Karma Sahayak Foundation (PKSF) 7 Pareto, Vilfredo, 136 Paris Accord 58 Paris Commune (1870) 17 participatory governance 67, 69 paternalistic politics 83 patronage 45 patron-clientelism 4, 5, 13, 145 peasants 5, 8, 9 People’s Independence Day 31 permeability concept 15, 110 Perrow, C. 126, 141 Peru 58 Philippines 58, 70, 72, 74, 77–8 Police Reform Project 56 policy research centres 50 Policy Research Institute 50 political competition 90 Political Environment 35 political rights 80–1 Political Rights and Civil Liberties scores, Freedom House 18, 20, 37n4 political society 157 post-independence period 43 post-war reconstruction 108 poverty 11, 52 pragmatism 101 Prime Minister’s office (PMO) 12 print media freedom, relative 21 ‘prison’ metaphor 4, 10–12, 13, 15, 120 probit regressions 102 professional organisations 25, 49 protest marches 36; see also garment workers; protests of Public Procurement Rules 2007 52 Public Service Commission 11, 52 PUMUS case 117–20 Punjab, elites of 5 Punjabi identity 8 pyramid political structure 11 Rabbani, Golam 84 Rahman, S. 128 Rahman, Sheikh Hasina 7 Rahman, Sheikh Mujibur 7, 9, 19, 30, 31, 42, 92, 130 Rahman, Tarique 28 Rahman, Ziaur (Major-General) 7, 9, 19, 92

Index raiyats (landlords) 8 Rapid Action Battalion (RAB) 36 Rapporteurs Sans Frontières (watchdog organization) 35, 36 ready-made garments (RMGs) 144 reality-rhetoric gap 6–7 Rebel Clown Army 154 regression analysis 100 rent-seeking behaviour 4, 9, 10, 11 Right to Information (RTI) ordinance 24, 29, 33–5, 34, 58 Right to Information Act 2009 29, 72 Robinson, J.A. 121 Rondinelli, D. 141 Rule of Law 35 Rules of the Game 14, 18, 20–2, 44; all power to election winners 21; barricades, opposition to 21; election winners, all power to 21; electoral democracy 20; electoral renewal 21; gangsterism, in public life 21; local governance as patronage mechanism 21; organised life commandeered 21; relative print media freedom 21; safety mechanism 21; two-track judiciary 21 Samata (NGO) 132 Sanyal, B. 128 Saptagram Nari Swanirvar Parishad 131, 134, 141 Scott, J.C. 154–5 ‘Secular Bengali Nationalism’ 92 secularism 101 security forces 41 Seddon, D. 152 Selznick, P. 118 Sen, Amartya 64 separation of powers 28 Shahbag movement 143–4, 150 Shankland, Alex 149 Sharp, G. 155 Simmel, Georg 136 Singapore 14, 72–7 Smillie, I. 129 Soban, Farooq 48 social deference 11 social media 58 SOG Reports see State of Governance Reports (BRAC Institute of Governance) South Africa 70, 84 South Korea 70, 74 Soviet Union, collapse 68, 152 Sri Lanka 33, 73


stake-holders 69 Standing, Guy 157, 163 stasis thesis 10 state building paradigm 39–42; Bangladesh fitting 42–6 State of Democracy in South Asia project 93–4 State of Governance Reports (BRAC Institute of Governance) 10, 11, 12, 46, 88, 113, 169 Stepan, A. 135 Stiles, K. 128 Sultan, Maheen 147 Supreme Court 27, 31, 35, 45 Suykens, B. 160 Tadros, Mariz 152 Taiwan 74 Taliban 107 Tanzania 58 technocrats 42 teleological analysis 3 Thailand 73, 74, 75 The Economist Intelligence Unit (EIU) 67 theocrats 93–8 Thirteenth Constitutional Amendment 27, 91 Thompson, E.P. 153 trade unions 49 Transparency International Bangladesh (TIB) 51–2, 53, 60n9, 113, 126 TVA and the Grassroots (Selznick) 118 two-party-plus system, in Bangladesh 17, 19 Ummah 110 UN Human Development Report 75 unruliness in politics 13, 15–16, 143–67; in Bangladesh 3, 149–50, 158–64; content 155–7; forms and frames 153–5, 165–6; garment workers, protests of (2006–2011), 144, 161–4; and governance agenda 164–6; ideas 150–8; moral economies and laws of necessity 155–7; motivations 144–50; mundane and magical power 147–8; paradox, in Bangladesh 149–50, 169; political rulebreaking, reason for importance 143–4; popular 165; ‘power within,’ understanding 147; slow evolution of the hartal repertoire 158–61; timing/ historical specificity 151–3; unruly actors 157–8; unruly political settlement in Bangladesh 145–7



upazila parishads (UZPs) (local councils) 23, 24, 25–9, 35, 37n11; Caretaker Government (CTG), 27–8; constituency development funds 26–7; judicial separation 28–9; Right to Information (RTI) 29 uprisings 40 Vietnam 73, 74 Violence and Social Orders (North, Wallis and Weingast) 4, 13, 107, 109, 111, 171 Voice and Accountability 70, 72 Vulnerable Group Feeding programmes 11, 46 Wajid, Sheikh Hasina 19 Wallis, J.J. 13; see also North–Wallis– Weingast paradigm Walton. J.K. 152 ‘warder’ class, political 4 Weber, Max 60n4, 83

Weingast, B.R. 13; see also North–Wallis– Weingast paradigm Wellbeing in Developing Countries (WeD) 147 Wikileaks 37n9 Wood, Geof D. 3, 6, 10, 15, 45, 48, 87, 113, 145 World Bank 36, 123n4; Africa Report 54; Council of African Advisers 54; “Political Stability and Absence of Violence” indicator 32; World Development Report (1997) 68 World Values Survey (2001) 93 Worldwide Governance Indicators (WGI) 32, 38n28, 70 Yunus, Muhammad 46 Zaman, Iftekhar 53 zamindars (Hindu landlords) 6, 8, 16n2 Zia, Begum 7, 19 Zia, Khalida 19, 23, 28