UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Water policy and governance for the empowerment of river basin communities in rural Bangladesh Hossen, Mohammad Anwar 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2014_november_hossen_mohammad.pdf [ 2.61MB ]
JSON: 24-1.0074384.json
JSON-LD: 24-1.0074384-ld.json
RDF/XML (Pretty): 24-1.0074384-rdf.xml
RDF/JSON: 24-1.0074384-rdf.json
Turtle: 24-1.0074384-turtle.txt
N-Triples: 24-1.0074384-rdf-ntriples.txt
Original Record: 24-1.0074384-source.json
Full Text

Full Text

 WATER POLICY AND GOVERNANCE FOR THE EMPOWERMENT OF RIVER BASIN COMMUNITIES IN RURAL BANGLADESH by Mohammad Anwar Hossen M.S.S., University of Dhaka, 1994 M.A., Carleton University, 2008  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF Doctor of Philosophy in THE COLLEGE OF GRADUATE STUDIES Interdisciplinary Studies  THE UNIVERSITY OF BRITISH COLUMBIA (Okanagan)  September 2014 © Mohammad Anwar Hossen, 2014   ii  Abstract Agricultural communities in the Ganges Dependent Area of Bangladesh are facing extreme hardships due to current water management practices, environmental degradation, and agricultural modernization programs that benefit the elite but disadvantage the majority. India is a major contributor to these problems because of its diversion of large amounts of water from the Ganges River before it reaches Bangladesh. The governance systems within both Bangladesh and India are best understood as ecocracies, that is, highly centralized and bureaucratic systems in which resources are controlled by elite groups following neoliberal development goals. This style of governance has marginalized the majority of farming households who now find themselves unable to make their voices heard. The hardships they are now facing are severe enough to be classified as human rights violations.  In this study I describe the effects of regional hydropolitics on water management, focusing on three large engineering projects, the Farakka Barrage built by India on the Ganges River, and the Ganges-Kobodak Gorai River Restoration Projects in Bangladesh. I also describe the traditional livelihood strategies and local ecological knowledge of farming households in the community of Chapra, Kushtia District, Bangladesh, and the rapid displacement of traditional practices by a commoditized system created through government intervention. I gathered information about local knowledge and farming practices during a year of fieldwork in 2011-12, through focus group discussions, in-depth case studies of four farming households and a survey in which 259 households participated.  In addition to this primary data, I collected extensive secondary documents from the government of Bangladesh pertaining to water management, agricultural modernization and institutional structures. iii  My field data reveals that the right to water of the vast majority of Chapra farming households, as defined by United Nations Conventions and international customary law, is being systematically violated and that this violation in turn has led to multiple human rights violations in the areas of employment, food, education, health care and housing. I conclude with recommendations for how these problems could be solved through application of human rights principles and greater community inclusion in governance processes at international, national, and local levels.                    iv  Preface This research project was approved by the Behavioral Research Ethics Board (BREB) of the University of British Columbia (Okanagan Campus), Certificate H11-01129. I am solely responsible for the design and conduct of the research project, the analysis of the data and the writing of this dissertation.  Content from chapter one on hydropolitics has been published as Bilateral Hydro-hegemony in the Ganges-Brahmaputra Basin Oriental Geographer 53 (1 & 2) 2012. The organization and wording of the published article is significantly different, however, from the comparable sections of chapter one. I included content from my concluding chapter in a conference paper presented at the City University of Hong Kong with the title of “The Ganges Basin Management Review for Community Empowerment” on 20-21 May 2014. This conference paper has been accepted for publication in the Springer Journal Bandung: Journal of the Global South. The wording and content is not identical, however, to the concluding chapter.           v   Table of Contents Abstract .......................................................................................................................................... ii  Preface ........................................................................................................................................... iv  Table of Contents .......................................................................................................................... v  List of Tables ................................................................................................................................ ix  List of Figures ............................................................................................................................... xi  Glossary of Bengali Terms ......................................................................................................... xii  List of Acronyms ....................................................................................................................... xvii  Acknowledgements .................................................................................................................... xix  CHAPTER One:  Introduction .................................................................................................... 1                    The Research Question ..................................................................................................... 3  Ecological Characteristics of the Ganges Basin ............................................................... 4  Hydropolitics in South Asia .............................................................................................. 7  Hydropolitics in Bangladesh ........................................................................................... 10  Theoretical Approach ..................................................................................................... 11    Development Theory ............................................................................................. 11   Political Ecology ................................................................................................... 16   Human Rights ........................................................................................................ 18   Water Governance ................................................................................................. 21  Organization of the dissertation ...................................................................................... 23  CHAPTER Two:   Research Methodology ............................................................................... 26                      Method of Data Collection ............................................................................................. 33  Primary Data Collection ................................................................................................. 35   Focus Groups ......................................................................................................... 35   In-Depth Case Studies and Participant Observation ............................................. 39       Tanvir ................................................................................................................ 40       Tofajjal .............................................................................................................. 42       Joardar ............................................................................................................... 42       Billal .................................................................................................................. 43   Chapra Household Survey ..................................................................................... 45  Secondary Data Collection ............................................................................................. 46  Ethical Issues and Limitations of My Dataset ................................................................ 48 vi    CHAPTER Three:  The Ganges Dependent Area in Bangladesh .......................................... 53   The Dependency of the Gorai Bank Communities on Ganges   Hydro-ecology ................................................................................................................ 54  The Farakka Barrage ....................................................................................................... 64  Political Culture .............................................................................................................. 72  Water Policies in the GDA ............................................................................................. 76   CHAPTER Four:  The Gorai River Restoration and the Ganges-Kobodak                                       Projects...................................................................................................... 80                          The Gorai River Restoration Project .............................................................................. 81   Elite Domination at Chapra ................................................................................... 83   Sedimentation and Charland Domination ............................................................. 88   River Bank Erosion and Embankment Failures .................................................... 89   Flood Vulnerabilities ............................................................................................. 94   Drought Vulnerabilities ......................................................................................... 96  The Ganges-Kobodak Project ......................................................................................... 98   Outcomes of the GK Project ............................................................................... 103   Construction Phase Conflict and Mismanagement ............................................. 105   Operational Phase Conflict and Mismanagement ............................................... 107   Lack of Accountability and Transparency .......................................................... 110  CHAPTER Five:  Agricultural Transformation at Chapra ................................................ .116                          Cropping Patterns ......................................................................................................... 117  Crops and Seeds ............................................................................................................ 122    Indigenous Crop Varieties ................................................................................... 122    Introduced Crops ................................................................................................. 124  Water Commodification ............................................................................................... 129  Fisheries Commodification ........................................................................................... 132  Land and Equipment Costs ........................................................................................... 134  Fertilizers and Pesticides .............................................................................................. 137    Natural Fertilizers ................................................................................................ 137    Chemical Fertilizers ............................................................................................ 139  Domestic Animals ........................................................................................................ 142  Employment .................................................................................................................. 145    New Employment Patterns .................................................................................. 148  Transformation through Commodification ................................................................... 151   CHAPTER Six:  Local Governments’ Roles to Protect the Marginalized                                  Communities ............................................................................................... 153                        Power Structure of Local Governments ....................................................................... 154  Water Management ....................................................................................................... 156 vii   Seed Management ......................................................................................................... 159  Fertilizer Management .................................................................................................. 163  Food Security Program Management ........................................................................... 164  Employment Programs Management ........................................................................... 168  Education Program Management ................................................................................. 173  Health Care Program Management .............................................................................. 176  Housing Program Management .................................................................................... 181  The Consistent Failure of Safety Net Programs ........................................................... 184                        CHAPTER Seven:  Human Rights at Chapra ....................................................................... 185                          Water Rights Violations ............................................................................................... 188   Water Access for Basic Needs ................................................................................. 189   State Obligations in Regard to Shared Watercourses .............................................. 190   Vulnerable Communities and the Rights to Participate in Decision Making     Processes ............................................................................................................. 192   Ecological Integrity .................................................................................................. 193   The Right to Employment ............................................................................................ 196  The Right to Food ......................................................................................................... 203  Education Rights Violations ......................................................................................... 209  The Right to Health ...................................................................................................... 213  The Right to Housing.................................................................................................... 219  CHAPTER Eight:  Political Resistance, Conflict, and Social Movements .......................... 225                                 The Political Climate in Bangladesh ............................................................................ 226  Struggles Over Resources and the Environment .......................................................... 230   Farakka Protests ....................................................................................................... 230   Local Protests ........................................................................................................... 235  Struggles for Democractization and Political Inclusion ............................................... 241     CHAPTER Nine:  Conclusion ................................................................................................. 246                                                         The Human Right to Water .......................................................................................... 246  Problems  ................................................................................................................... 249  Causes  ................................................................................................................... 252   Ecocracy ................................................................................................................... 252   Farakka Barrage, the Bilateral Agreement, and Hydropolitics ................................ 254   Technocentric Water Management .......................................................................... 254   Elite Control of Local Institutions ........................................................................... 255  Solutions  ................................................................................................................... 257   Recommendation for an Elected Basin-Wide Institution ........................................ 259   Recommendation for the Reform of the JRC ........................................................... 261    Recommendation for a New Ward-Level Water Governance Institution ................ 263   Recommendations for the Reform of Local Level Government.............................. 265 viii   Water Democracy ......................................................................................................... 267   Reference Cited ......................................................................................................................... 269  Appendices ................................................................................................................................. 295   Appendix A: Ethics Board Approval Certificate  ......................................................... 295                                       ix    List of Tables Table 2.1: Focus Group Participants (pseudonyms) ................................................................ 38  Table 3.1: Bengali Calendar, Seasons and Cropping Periods .................................................. 59  Table 3.2: Ganges River Flow Data Before and After Construction of the Farakka  Barrage .................................................................................................................... 66  Table 3.3: Annexure 1 of the 1996 Ganges Treaty for Sharing Ganges River Flow ............... 69  Table 4.1: GK Project Infrastructures .................................................................................... 101  Table 4.2: Cropland Areas and Population Served by the GK Project .................................. 102  Table 4.3: GK Project Irrigated Acreage at Kumarkhali Sub-district .................................... 105  Table 5.1: Privatization of Seed Production in Bangladesh ............................................................... 126  Table 5.2: Comparison of HYV and Indigenous Crop Production (metric tonness) between 1967-70 and 1999-00 in Kushtia District .......................................................................... 128  Table 5.3: Fertilizer Usage in Kushtia ................................................................................................ 140    Table 6.1: Irrigation Management Committees at District and Upazila Levels ................................. 157  Table 6.2: Seed and Fertilizer Management Committees in Local Governments .............................. 160  Table 6.3: Agriculture Rehabilitation Committees in Local Governments ........................................ 162  Table 6.4: Food For Work or KABIKHA Committees in Local Government ................................... 166  Table 6.5: The EGPP Committees at the District, Upazila and Union Levels in Local  Governments ..................................................................................................................... 170  Table 7.1: Human Rights Violations at Chapra .................................................................................. 187  Table 7.2:  Water Rights Violations at Chapra .................................................................................... 188  Table 7.3: Harmful Ecological Effects as Reported by Case Study Participants ............................... 196  Table 7.4: Employment Rights Violations at Chapra ........................................................................  197  Table 7.5: Food Rights ....................................................................................................................... 204  Table 7.6: Monthly Average per Capita Food Expenditures at Chapra in 2010 ...............................  205  x  Table 7.7: Categories of People Most at Risk of Food Violations in Kumarkhali   Sub-district ........................................................................................................................ 207  Table 7.8: Education Rights Violations at Chapra ............................................................................. 211  Table 7.9: Health Rights Violations at Chapra ................................................................................... 214  Table 7.10: Estimated per Capita Expenditures of Marginalized Person for One Month .................... 218  Table 7.11:  Housing Rights Violations at Chapra ................................................................................ 219  Table 7.12: Average Housing Expenditures of a Marginalized Rural Household in 2010 ................   221  Table 7.13: Housing Types in Kumarkhali Sub-District (Percentage) ........................................  222    xi  List of Figures Figure 1.1:  The Ganges-Brahmaputra Basin in India, China, Nepal, Bhutan, and Bangladesh ................................................................................................................ 5  Figure 2.1:  Map of Ganges Dependent Area and Fieldwork Site at Chapra ............................. 29  Figure 3.1:  The Major Hydrological Zones in Bangladesh (in square kilometers) ................... 54  Figure 3.2:  Average Ganges Flow Before and After the Farakka Barrage ............................... 67  Figure 4.1:  The GK Project Map  .............................................................................................. 99  Figure 5.1:  Cropping Areas (in thousands of hectares) in 2004-05 in Kushtia District .........  120  Figure 5.2: Two Seed Production Program Budgets in Bangladesh ......................................  125  Figure 5.3:  Subsidized Seed Costs (US$) for One Acre of Cropland in 2005-06 ................... 127  Figure 5.4: Irrigation Costs (US$) for One Acre of Cropland in 2005-06 in Kushtia ............. 131  Figure 5.5: Land Rental Costs (US$) for One Acre of Cropland in 2005-06 .......................... 135  Figure 5.6: Land Preparation Costs (US$) with Plow Machine for one Acre of Cropland in 2005-06 ............................................................................................. 136  Figure 5.7: Fertilizer Costs (US$) for One Acre of Cropland in 2005-06 ............................... 141  Figure 5.8: Pesticide Costs (US$) for One Acre of Cropland in 2005-06 ............................... 142  Figure 5.9: Labor Costs (US$) for One Acre of Cropland Cultivation in 2005-06 ................. 150  Figure 6.1: The central government’s Food Safety Program Budgets (US$) .......................... 167  Figure 6.2: The Eight Scholarship and Grant Program Budgets ............................................. 174  Figure 6.3: Percentage of Health Care Sources in 2010 in Rural Bangladesh ........................ 178  Figure 6.4: The central government’s Health Care Programs from 2009-10 to 2012-13 ........ 180  Figure 6.5: The Governments’ Housing Safety Net Budgets (in US$) ................................... 182  Figure 7.1: Minimum Expenditures (US$) of a Marginalized Household at Chapra .............. 199   xii  Glossary of Bengali Terms Ain O Salish Kendra name of a human rights NGO     Aman the most important local rice variety grown during the wet season  Andalon social movement Ashrayan a government housing program; the literal meaning of the name is “place of shelter” Aus the most important rice variety grown at the end of the Spring season Baacha morar lorai do or die struggle Beel wetland Beel Dakatia Andolon Dakatia Wetlands Movement; a local movement in Khulna District Bigha a unit of land measurement; in Kushtia District 1 bigha equals 0.3 acres Bhukha michhil demonstration against food shortage Bonna extreme flood Bonsher baati line of inheritance Boro name of a rice variety now commonly grown in place of aus and aman due to continuous ecosystem failures Bosenter kokil a spring bird Charland a build-up of sand in a river bed Chashi traditional farmer xiii  Chor thief Dadagiree an elder brother’s authoritative practice Deki paddy husking pedal Dharmaghat strike or withdrawal of service as a form of protest Dhubba grasses that grow in between rows of crops Dhup fragrance of a particular herb Ejara a lease Gajar snakehead fish Gucchagram a government housing program; the literal meaning of the name is “cluster of houses”  Eid-ul-Azha Muslim festival of sacrifice celebrated on the tenth day of the month of Jilhajj in the Bengali calendar Eid-ul-Fitr Muslim festival on the first day of the month of Shawal to celebrate the end of Ramadan (fasting) Gom wheat Gorurgaree bullock cart Ha-do-do a Bengali community outdoor sport  Haor oxbow lake  Harijon scheduled caste Harichaada traditional toll collection Hartal a street demonstration that involves the closure of shops, offices and transportation systems Imam Muslim preacher xiv  Jaigidar food and accommodation in exchange for private tutoring Joler dame incredibly lower price  Joke blood sucking worm  Kaachi scissors  Kakla crocodile fish  Kalijera local name for rice paddy  Kamla day laborer  Kata diyea kata tola set a thief to catch another thief Kecho earthworm  Khal a canal  Khal knonon kormochuchi a canal renovation program  Kharif 1 agricultural cropping period from February to April Kharif 2 agricultural cropping period from May to August  Khora dry season that occurs normally from February to May but the term can be applied to dry periods during the summer months as well Kola bang bull frog Kolosh earthenware jar Kot traditional economic transaction system Krishak modern farmer  Kuichya eel  Kutcha house built with straw and mud xv  Kula winnow fan Kuya ring-well Langol traditional plow pulled by bullocks Maacha a raised platform made with bamboo and rope Matobbar a village chief Mohajon a traditional money lender Nagorik Oikya citizens’ forum Nerica-10 hybrid rice variety that requires a strictly strictly regulated water supply Nouka bice boat race Noya Krishak Andolon new farmers’ movement Odhikar name of a human rights NGO  Paisa one unit of Bangladesh currency; 100 paisas equals one Bangladesh Taka Pakhi a unit of land measurement; one pakhi equals 0.6 acres Paribesh Andolon environmental movement Pitha traditional cake or sweetmeat  Polo bice community fishing sport Pucca house made with brick and cement  Puja mandir Hindu prayer house Purdah Muslim religious practice of seclusion according to which the female body is covered from head to toe Puti minnow Rakhal yearly contract laborer xvi  Rakkhosh monster Robi Agricultural cropping period from October to January Sada sar white fertilizer or urea  Shaheb a form of address for an adult male; Mr. Semipucca house made with brick, corrugated iron, and mud Shako locally made bridge over a small river or canal Shikor root Sonar Bangla Golden Bengal Songkot crisis Sorkarer Dalal agent of government Swanirvar a government employment program for marginalized people’s empowerment   Ufsi aus hybrid rice variety that requires a strictly regulated water supply  Union Council lowest level of local government that includes 12 wards made up of 20 to 30 villages Unnoti development Upazila Parishad sub-district level of local government Vaate marbe paanite Marbe local people will be killed by rice and water deprivation Vater-maar a scum or liquid substance made from boiled rice paddy Vita raised bed for growing crops Jee hujur yes sir Zamindar a landlord xvii  List of Acronyms ASK Ain O Salish Kendra ASP Ammonium Sulfate Phosphate  BADC Bangladesh Agricultural Development Corporation  BARC Bangladesh Agricultural Research Council  BARD Bangladesh Academy for Rural Development  BBS Bangladesh Bureau of Statistics   BDT Bangladesh Taka BFA Bangladesh Fertilizer Association  BFRI Bangladesh Fishery Research Institute  BLAST Bangladesh Legal Aid and Services Trust  BRDB  Bangladesh Rural Development Board BSEHR Bangladesh Society for the Enforcement of Human Rights BSF Border Security Force BWDB Bangladesh Water Development Board DAE Department of Agricultural Extension DAP Diammonium Phosphate DoE Department of Environment DoF Department of Fisheries EGPP Employment Generation Program for the Poorest FAP Flood Action Plan FCDI Flood Control, Drainage and Irrigation FFW Food For Work (KABIKHA)  GBB Ganges-Brahmaputra Basin GDA Ganges Dependent Area GK Ganges-Kobodak (Project) GR Gratuity Relief GRRP Gorai River Restoration Project HYV High Yielding Variety IFC International Farakka Committee IMF International Monetary Fund IWRM Integrated Water Resource Management  JRC Joint River Commission xviii  KCERP Khulna Coastal Embankment Rehabilitation Project  KIDRP Khula-Jessore Drainage Rehabilitation Project LGED Local Government and Engineering Department LGRD Local Government for Rural Development MDGs Millennium Development Goals MoA Ministry of Agriculture MoF Ministry of Finance MoDMR Ministry of Disaster Management  and Relief MoFDM  Ministry of Food and Disaster Management MoHFW  Ministry of Health and Family Welfare  MoWR Ministry of Water Resources MP Member of Parliament NGO Non-Government Organization NRLP National River Linkage Project NWMP National Water Management Plan NWP National Water Policy PDB Power Development Board  PIC Project Implementation Committee PRSP Poverty Reduction Strategy Paper RAB Rapid Action Battalion SSP Single Super Phosphate TR Test Relief TSP Triple Super Phosphate UBINIG    Unnayan Bikalper Nitinirdharoni Gobeshona (Policy Research for Development Alternatives) UC Union Council UHC Upazila Health Complex  UNO Upazila Nirbahi Officer UP Upazila Parishad VGD Vulnerable Group Development VGF Vulnerable Group Feeding WARPO   Water Resource Planning Organization xix  Acknowledgements I would first like to thank my supervisor Dr. John Wagner for his extensive support throughout my studies at the University of British Columbia Okanagan. His critical comments during my comprehensive examinations, research proposal writing, data collection and analysis reshaped my theoretical approach and my understanding of water governance. Additionally, he visited my fieldwork site in Bangladesh to assist in my data collection. I am also grateful to his wife, Dr Jennifer Gustar, who always supported me during my student life at UBCO. Dr. Naomi McPherson, as a member of my PhD committee and my instructor in a research methods course, helped me develop my academic skills and write a preliminary research proposal. She also provided very helpful feedback during my comprehensive examinations, data analysis and the writing of this dissertation. Dr. Jon Corbett, another committee member, was also very helpful during my comprehensive examinations, research proposal writing, data analysis, and the writing of this dissertation. A course I took with Dr. Corbett helped deepen my understanding of the geographic aspects of the Ganges Basin. Dr. Jordan Stouck, Director of the Center for Scholarly Communications at UBCO, provided me with invaluable support with my written communication skills. Dr. Barry Myers of Carleton University, Ottawa, provided me with excellence guidance as I adjusted to the Canadian university system as an MA student.   In Bangladesh, I am grateful to Professor A.A.M.S. Arefin Siddique, Vice Chancellor of Dhaka University for his enormous support during my fieldwork. Professor Ainun Nishat, Vice Chancellor of BRAC University, helped me select my fieldwork site. I received important insights from Professor S. Aminul Islam and A.I.U. Mahbub Ahmed during my fieldwork. Muhammad Ruhul Quddus, Joint Secretary of the Ministry of Water Resources, Saleh Ahmed, Executive Engineer and Salma Begum, Deputy Chief of the Bangladesh Water Development xx  Board, Md. Siddiqur Rahman, Principal Scientific Officer and A.KM. Khusrul Amin, Scientific Officer, of the Water Resources Planning Organization and A.K.M. Jahangir Hossain, Upazila Agriculture Officer of Department of Agricultural Extension also assisted my secondary data collection. In Kushtia, I am very grateful to Professor Anwarul Karim, Sheikh Mohammad Abu Shamim and Mst. Parveen Sultana for their kind support. Abdul Wahid, Deputy Director of the Ganges-Kobodak Project office in Kushtia also helped me with my collection of secondary data. I also wish to thank my research assistants, especially Anwar, Raihan, Selim and Tanvir, without whom it would have been very difficult to complete my fieldwork. I owe more than I can ever express to the fieldwork participants at Chapra who provided me with the information described in this dissertation and who supported me voluntarily despite their busy schedules.  I am very grateful to the International Development Research Centre, Ottawa for awarding me a Doctoral Research Award during my year of fieldwork. I am also grateful to the UBC Centre for India and South Asia Research for awarding me the Nehru Humanitarian Award in 2010. I also wish to acknowledge the funding support provided to me by UBC Okanagan Graduate Fellowships and Research Assistantships with Dr. Wagner.  I wish to extend my heart-felt gratitude to my fellow graduate students especially Markandu Anputhas, Eva-Marie Kovacs-Kowalke, and Michelle Walks. We overcome many professional challenges together. I am indebted to my wife, Uwana Akhand, for her enormous support during my fieldwork and data analysis. Sajjad Haider, Deputy Director of the Information Ministry, Md. Shariful Islam, Associate Professor of Dhaka University, and Md. Rafiqul Islam, my youngest brother also supported me by making my social life easier during my PhD studies.1   Chapter One INTRODUCTION  Bangladesh is a delta country dominated by an agrarian society. Two of the largest rivers in Asia, the Ganges and Brahmaputra, meet in Bangladesh before emptying into the Bay of Bengal. The Ganges River and its many distributaries dominate the southwest of the country where this study is situated. Rural communities in this region produce agricultural crops on a year around basis, following seasonal patterns and the variable flow of water in the Ganges River. Agricultural production provides the basis for the local economy and contributes significantly to the national economy of Bangladesh. Croplands turn into wetlands during the wet season and this makes them ideal for the rice paddy production that is the basis for the whole year’s food security as rice is the staple food in Bangladesh. After the wet season, croplands are suitable for a wide variety of other crops like mustard oil and wheat and local people are able to produce as many as three crops in a single year. The Ganges Basin ecosystems on which cropping patterns depend also produce a rich supply of common property resources like wild fish and water spinach. The rivers deposit silt on croplands during the monsoon season, enriching soil nutrients and as a result farmers do not need to use chemical fertilizers. They are also able to manufacture many essential tools and other agricultural production materials from freely available natural resources like bamboo and wild tree species. The traditional system of agricultural production is technologically simple but well suited to local conditions. Farmers till their croplands with domestic oxen and ploughs made from local materials. They produce their own crop seeds every year.   2  However, this eco-agricultural system is now undergoing major disruptions due to a number of factors including regional hydropolitics and neoliberal and highly centralized approaches to water resource management that follow the principles of  “ecocracy” as that term has been used by Sachs (1992) and Escobar (1996). The central governments of both India and Bangladesh follow much the same principles as they seek to manage environmental resources like rivers through a combination of engineering projects, bureaucratization, and the application of economic liberalization policies. But, unfortunately for Bangladesh, India holds the upper hand when it comes to their sometimes competing interests over water resources. The central government in India, for instance, constructed the Farakka Barrage unilaterally on the Ganges River, very near the Bangladesh border, reducing flows to agricultural communities in Bangladesh at critical times in their cropping cycles. The barrage was built in order to divert water to Kolkata, in order to enhance navigation and shipping and provide the city with fresh water. The central government in Bangladesh has been unable to restore this flow to a sufficient level and subsequently has intensified its own top-down water management approaches. These approaches fail to recognize the full range of ecosystem services on which most people rely and thus displace natural systems with multiple ecocracies.  Currently, the Ganges River flow is very unpredictable which creates major ecosystem failures; some croplands face flooding or water stagnation for longer than normal periods in a year and other croplands, at other times of the year, face prolonged drought. River levels can no longer be predicted on the basis of natural seasonal changes but depend also on the mercy of the Farakka Barrage diversion. The diversion now determines whether wet and dry seasons will be longer or shorter than normal. These ecosystem changes create enormous vulnerabilities for most local farming households and are responsible for major environmental, agricultural and 3  employment practice failures in the community of Chapra, my primary research site. The poorer farming households that constitute the vast majority of the population of the Chapra region and the Ganges Dependent Area as a whole now find themselves more and more vulnerable to livelihood failures. These failures result in what can be considered human rights abuses, since they mean that many households are now unable to secure basic human rights, as defined by the United Nations, to food, water, education, housing, and health care. Based on the United Nations (2002) definition of access to water as human right, I argue in this dissertation that the Ganges River water right, including the right to the ecosystem services it provides, is the prerequisite for maintaining other human rights. The situation at Chapra is complicated by the fact that hardships are unevenly distributed. A small number of farming households are relatively well off by comparison to the majority and constitute an elite group with strong connections to government. These connections allow them to dominate local water management institutions and benefit from capital intensive agricultural modernization programs. The elite farmers are market oriented and increasingly less dependent on the ecosystem services and common property resources on which the majority depend.  The Research Question The human rights violations occurring in the Ganges Dependent Area are unacceptable from any perspective but existing research is not sufficient to identify suitable remedies. Research has identified the need for empowerment of local democratic institutions and practices but it has not clearly documented the links between ecosystems and community livelihoods or provided in-depth accounts of livelihood strategies at the household level. My goal in this study is therefore to document local practices, livelihood challenges and human rights violations and, on that basis, 4  identify the changes in water policy and governance necessary to alleviate current hardships. In order to determine what improvements to the governance system would be practical in this setting, I seek to respond the following questions: 1) how can local ecological knowledge be incorporated into national water policy in Bangladesh; 2) what strategies and reforms are required at the international watershed governance level; and, 3) how can human rights principles, including the principle of water as a human right, be used to formulate more effective water policy and governance principles?  I will not attempt to summarize my answers to these questions until my concluding chapter, but it is worth noting here that a human rights approach introduces new and challenging questions into the governance debate. Given the hydropolitics of South Asia as a whole and the current dominance throughout the region of neoliberal approaches to resource management, is it feasible to think that a marriage (a healthy relationship) between ecosystems and economic development in the Ganges Basin could be achieved? A marriage of this type, at a minimum, will require greater respect for and incorporation of local ecological knowledge, and a coordinated, multi-scalar governance approach involving all the countries of the Ganges Basin. Without these fundamental changes to existing policies, it will not be possible to secure human rights for the majority of farming households living in the Ganges Basin in Bangladesh.  Ecological Characteristics of the Ganges Basin The Ganges Basin is best understood as part of the larger Ganges-Brahmaputra Basin that originates in the Himalayan Mountains, flows through China, Nepal, India, and Bangladesh, and ends in the Bay of Bengal (Figure 1.1). The Alakananda and Bhagirthi Rivers are the two major headwater sources of the Ganges where it arises in the western Himalayas in India at an elevation 5  of 23,000 feet. After merging, these rivers flow south to the Ganges Plain, north of New Delhi, and then continue eastward across the plain, traversing almost the entire country before reaching Bangladesh. Along the way it is fed by several other major Himalayan tributaries, including the Yamuna, the Gomati and the Rama Ganga, Ghaghara, and the Gandak and Kosi Rivers which originate in Nepal (Parua 2010:267-268). The headwaters of the Brahmaputra, on the other hand, arise on the northern side of the Himalayas and flow east first across the Tibetan Plateau, before turning south to India and then west to Bangladesh. The Ganges River in Bangladesh is known as the Padma and the Brahmaputra is known as the Jamuna, but throughout this dissertation I will use their more commonly recognized names.     Figure 1.1: The Ganges-Brahmaputra Basin in India, China, Nepal, Bhutan, and Bangladesh. Source: Wikimedia 2013.   6  The Ganges-Brahmaputra Basin in Bangladesh includes 57 international and 310 local rivers (Bangladesh Bureau of Statistics 2010). About 94 percent of the water flow within these rivers in Bangladesh originates from outside the country in India (Elhance 1999:158). Most of Bangladesh is less than six meters above mean sea level (Gupta et al. 2005:387) which contributes to the fact that more than one million of the country’s total 14 million hectare surface area is covered with local water bodies (Bangladesh Bureau of Statistics 2010), like the Chapaigachi oxbow lakes, Shinda and Shaota canals, and the Lahineepara and Chapra wetlands at Chapra which are replenished with water from the Gorai River, a distributary of the Ganges.      Agricultural practices at Chapra have developed over many centuries in response to these conditions and the seasonal characteristics of the region. The two main seasons are known as borsha (wet season), which occurs during four months from June to September and khora (dry season) which occurs between February and May, with some variation, depending on the volume of water in the Ganges River. These seasonal patterns occur within what is overall a moderately hot and humid climate, and this combination of climate, seasonal variation, low elevation, and abundant wetlands, provides the foundation for traditional agricultural and employment practices at Chapra.  Water in this setting, when analyzed in relation to  social, economic and cultural practices, constitutes  what Orlove and Caton, relying on Mauss (1990), have referred to as a “total social fact” (Orlove and Caton 2010:402). Every aspect of people’s lives at Chapra is bound up with the movement of water through the environment and their ability to adapt to its seasonal characteristics. Traditionally, every household had egalitarian access to the flow of water during the borsha season for agricultural (Islam and Atkins 2007:131) and other uses (Zaman 1993:995). Water can thus be considered an essential actor within this socioecological 7  system, possessing its own form of agency (Latour 1995; Wagner 2013). In the chapters that follow I argue that the realization of fundamental human rights is dependent on the characteristics of this socioecological system and on the movement of water through it.   Hydropolitics in South Asia India is the most rapidly developing and most water scarce nation in South Asia but also the most powerful politically and economically. Bangladesh suffers from this power difference since it is located downstream from India with respect to most major shared rivers including the Ganges and Brahmaputra. Political conflict in the region is also an outcome of colonial legacies. British colonial policies laid the groundwork for the division of South Asia into the independent states that exist there now and current political tensions in the region owe a great deal to this history. The British colonial legacy has also affected the ways in which modernization processes have unfolded and the way in which statecraft and governmentality have evolved (Pels 1997). In this setting as in many others, the colonial period has also strongly affected the relationship of contemporary governing elites to local communities (see for example Davidson-Harden et al. 2007; Escobar 1991; Ferguson 1994; Gardner and Lewis 1996:16; Kirby 2006:636; Langford 2005:274; Mehta 2005; Smith 2004:504; Torry et al. 1979:523, 31; Turner 1997:13).  The British colonizers established a development discourse that described the colonized margins as backward, illiterate and traditional people in need of western style economic, educational and bureaucratic systems (Gardner and Lewis 1996:5). With the arrival of the postcolonial state, the people on the margins were ruled by the elite, bourgeoisie groups that had risen to power within the colonial bureaucracy or retained their authority as Zamindars (wealthy landowners) (e.g., Das and Poole 2004).  8  Large scale water management projects were a fundamental aspect of the British colonial period. Technocentric irrigation systems in support of commercial agriculture were introduced to the Indian Subcontinent beginning with the East India Company in 1760s (Altinbilek 2002:10; Baviskar 2007:1; D’Souza 2006:621; Gilmartin 1994; Stone 1985; Verghese et al. 1994). In 1854 the British Government in India completed the construction of 2,298 miles of canals and distributaries in the upper Ganges alone (Khan 1996).  D’Souza (2006:621) refers to these massive historical water interventions as “colonial hydrology” and the continuation of this approach today has been termed “hydrological nationalism” by Bandyopadhyay (2006:2). According to Bandyopadhyay, political leaders win elections on the basis of their ability to exploit basin hydrological resources in the name of national socioeconomic development. India, as an independent state, had built about 4,300 dams by 1994 and is considered one of the major dam building countries of the world (Hill 2006:149). This hydronationalism continues to inform India’s approach to shared rivers, which is characterised by a strong emphasis on unilateral actions or, when agreements are reached with neighboring countries, on a bilateral rather than multilateral approach (Crow and Singh 2000:1910; Gyawali and Dixit 2001).   India utilizes a bilateral approach in order to limit the application of general principles that could work to its disadvantage (Elhance 1999:172; McGregor 2000). As a downstream country, for instance, India argues that upstream China, Nepal or Bhutan should not execute any action that can harm their interests as a downstream country. As part of their bilateral agreements with Nepal and Bhutan, India has funded the construction of several hydro-projects in Nepal and Bhutan that provide them with hydro-power at a very reasonable cost. However, upstream India built the Farakka Barrage on the Ganges River unilaterally, without consideration of the interests of downstream Bangladesh. They subsequently concluded a series of bilateral treaties with 9  Bangladesh concerning the management of the Farakka Barrage and shared rivers more generally but did so on very unequal terms that fail to promote reasonable and equitable water sharing (Brichieri-Colombi and Bradnock 2003:53)   Despite the many well documented problems associated with the Indian approach to international waters, the government in India avoids participation in conflict resolution processes (Elhance 1999; Giordano et al. 2002; Gleick 2009; Haftendorn 2000; Wolf 1997), and they avoid invitations to develop a multilateral approach for basin-wide water management (Bandyopadhyay and Ghosh 2009; Brichieri-Colombi and Bradnock 2003; Gyawali and Dixit 2001). Indian hydro-domination in South Asia is thus increasing (Hossen 2012), with China as the only country in a position to challenge India’s authority. The well documented effects of this particular approach to water management include loss of croplands, forest, fisheries, natural vegetation and habitat, due to increasing incidences of flooding, drought, river bank erosion, salinity and water stagnation (Bharati and Jayakody 2011; Hanchett 1997; Islam and Gnauck 2011; Islam and Karim 2005; Karim 2004; Mirza 1997; Mirza and Sarkar 2004; Nilsson et al. 2005; Paul 1999; Potkin 2004; Swain 1996). The effects in Bangladesh are not caused only by the Farakka Barrage, but also by a series of other large barrages and structures that are capable of diverting 100,000 cubic feet per second (cusecs) from the Ganges flow before it reaches Bangladesh (Khan 1996). Currently, these water structures are used for commercial water and agricultural production (Altinbilek 2002:10; Baviskar 2007:1; Biswas and Tortajada 2001; Cox 1987; Khush 2001; Selby 2007; Van der Pijl 1998; Verghese et al. 1994).  10  Hydropolitics in Bangladesh The effects in Bangladesh of the Farakka Barrage and other Ganges River diversions in India have been magnified by a series of actions taken within Bangladesh itself. As a result of flood damages experienced in 1954 and 1955 the Bangladesh Government, then East Pakistan, developed an agricultural modernization and water management plan with the goal of increasing cropland acreage by removing water from local wetlands and oxbow lakes (Alexander 1998:1). The Flood Control, Drainage and Irrigation (FCDI) project, implemented in the early 1960s, included projects in several districts including the Ganges-Kobadak project in Kushtia, which I describe in detail in chapter four. FCDI projects were credited by some as worsening rather than preventing the floods of 1988 (Zaman 1993:989) and eventually FCDI was replaced by an even larger scale program of government intervention, the World Bank sponsored Flood Action Plan. The Gorai River Restoration Project, which I describe in more detail in chapter four, was implemented as part of the National Flood Control Plan and, like FCDI, is now criticized as having caused great hardship and environmental degradation and as riddled with corrupt practices.   In coordination with its water development projects, the government in Bangladesh has also supported a series of agricultural modernization and food security projects beginning in 1960s (Amin 1974; Sobhan 1981:328). The Green Revolution was one major component of this modernization approach (Cleaver 1972:184), promoting new technologies like chemical fertilizers, pesticides, HYV crops and controlled irrigation. The Ford Foundation and other international agencies provided funding support for this revolution and the Bangladesh Academy for Rural Development (BARD) worked with them to develop new seeds (Blair 1978:65-6).  11  Theoretical Approach  I rely on four closely interlinked bodies of theoretical literature in my approach to these issues. Literature from the field of development theory (Escobar 1996; Ferguson 1994; Gardner and Lewis 1996; Goldman 2007; Hanchett 1997; Sachs 1992; Scott 1998; Sillitoe 2000) informs my analysis of government ideology and practice, the postcolonial state, and the foreign aid machinery that so dominates government practice in Bangladesh. The field of political ecology informs my analysis of environmental change in the region and the relationship of this change to the political structure, class relations, local knowledge and culture (Agarwal 1992; Blaikie 2012; Latour 2004; Mann 2009; Robbins 2012). I rely on human rights literature in my analysis of the hardships being faced by farming households in Chapra, and in developing recommendations for policy changes to mitigate these hardships (Irujo 2007; Khadka 2010; Klawitter and Qazzaz 2005; Langford 2005; O’Manique 1992; Pamukcu 2005; Sultana and Loftus 2012). Finally, I rely on water governance literature to determine the type of governance system best suited to protecting the human rights of all residents of the study area, while also sustaining the critical ecological characteristics of the river basin (Agrawal 1995; de Loë et al. 2009; Lebel et al. 2005; Orlove and Caton 2010; Wagner 2009, 2013).   Development Theory In their historical review of development ideologies, Gardner and Lewis (1996) discuss the opposition that emerged during the 1960s between modernization and dependency theories. Contemporary debates differ from those of the 1960s in many ways, but carry forward some of the fundamental characteristics of that early debate. The modernization approach remains intact and continues to promote social transformation through industrialization and urbanization 12  (Gardner and Lewis 1996:13). Level of industrialization is still used as one basic indicator of modernization following Rostow’s model of economic growth. This approach continues to ignore the historical and political factors that cause underdevelopment (Gardner and Lewis 1996:15) and, arguably, the tendency of modernization approaches to cause underdevelopment has increased since the 1980s with the growing strength of neoliberal theory. The argument by dependency and world system theorists that colonialism laid the groundwork for underdevelopment in the global South (Amin 2003; Frank 1969; Prebisch 1963; Wallerstein 2004: x) remains relevant in the contemporary postcolonial era in which development agencies continue to depict the people of less developed nations as ‘backward’ and ‘traditional.’ The continuing domination of the modernization paradigm with its perceived opposition of modernity and tradition creates epistemological differences that are often expressed today as an opposition between scientific knowledge and local knowledge. Many scholars have pointed out the ways in which this is a false dichotomy but they have also drawn attention to the fact that the disappearance of certain forms of local knowledge can serve the interests of powerful outside groups as well as local elites (Agrawal 1995; Brosius 1999; Latour 1987; Leach et al. 1999:225; Mosse 2013; Nygren 1999; Peet and Watts 1996b:11-16; Scott 1998:115). Another perceived form of opposition is that between top-down development approaches dependent on western scientific knowledge, imported technologies and development bank loans, and bottom-up development approaches using local knowledge. The most important message to be learned from an examination of these opposing discourses is that each works in support of or in opposition to a specific power structure (Gardner and Lewis 1996:25).  Ferguson has provided an insightful analysis of how the discursive strategies of the Canadian International Development Agency (CIDA) and the World Bank (WB) informed the 13  implementation of the Thaba-Tseka Development Project in 1975-84 in Lesotho. The project identified “backwardness,” “lack of knowledge,” “lack of credit” and the “traditional land tenure system” as key development limitations in this setting (Ferguson 1994:51). One indicator of “backwardness” was that households were the basic unit of production and the only source of agricultural labour. The local exchange of labor, ploughs, cattle and seeds was also perceived as a limitation. The World Bank termed these limitations as transitions from traditional isolation to integration in a national economy. Through the massive Thaba-Tseka Development Project, they sought to transform the region through intensification of cash crop production and integration of agricultural and non-agricultural issues through changes to local administrative, educational, economic and health institutions (Ferguson 1994:81). As Ferguson documented, these structural changes increased inequality, decreased self-sufficiency, broke down local social and cultural rhythms of production and left people more vulnerable to ‘natural’ disasters, specifically drought.   Scott (1998) describes a similar process in Tanzania that he characterizes as following a western “high-modernist” agenda. The high-modernist agenda reshapes the rules and regulations of statecraft, reconfiguring local practices to support this agenda. Following a utilitarian discourse and rational scientific forest management, state agencies reorganized the “seeding, planting, and cutting” of forest resources so that they were easier “to count, manipulate, measure and assess” (Scott 1998:13). This approach commercializes forest resources while ignoring locally embedded relationships between humans and forest that maintained complex, coordinated and sustainable ecological relationships.  Goldman (2007) demonstrates the ways in which the high modernist agenda of the World Bank is linked specifically to water commodification on a worldwide basis. During his five year 14  (1968-73) tenure as the president of the World Bank, Robert McNamara financed more projects, 760, and loans, $13.4 billion1, than that of the combined investments of the Bank during all previous regimes. This new approach provided essential support for the green revolution in the global South based on northern technologies, including irrigation technologies. Goldman (2007) describes this approach as green neoliberalism. The bank justifies its approach as providing support for food security and poverty reduction, while ignoring questions about its own self-interest and evidence about the negative outcomes of technological interventions and water commodification in project areas. Evidence of such negative effects is already well documented for Bangladesh, for instance. World Bank supported programs like the Flood Action Plan have created survival crises for many people (Alexander et al. 1998; Banerjee 2010; Boyce 1990; Brammer 1990; Hanchett 1997; Mirza and Ericksen 1996; Paul 1999, 1995). Haque and Zaman (1993) have documented these negative effects for the Ganges Dependent Area specifically. Hanchett and Akhter (1998:2) have documented the fact that, rural women are the most vulnerable and bear the heaviest burdens in the case of these negative effects.  Within the field of development theory, the idea of ecocracy emerged during my study as a key concept that captures many of the arguments noted above. Sachs (1992) uses the term to describe the dominant approach of governments and international agencies such as the United Nations to the global crises of natural resource depletion and pollution. This approach is consistent with modernization theory, in that it relies on capital, bureaucracy and science to achieve better management and resolve these environmental crises. As problems escalate around the world, the “modernists” invent new technologies and programs with advanced models and integrated planning but with the ultimate goal of continuing business as usual. Escobar (1996)                                                           1 All dollar figures in this dissertation are given in US dollars unless otherwise stated. 15  also uses the term ecocracy and, like Sachs, constructs it as part of his critique of the concept of sustainable development. Following the publication of Our Common Future, in 1987, the final report of the UN World Commission on Environment and Development, chaired by Gro Halem Brundtland, the term sustainable development came into prominence within the discourse of both environmentalism and development.  But, according to Escobar and Sachs, the report promotes rapid economic growth and a technocentric vision of natural resource management. According to the report itself: “The concept of sustainable development does imply limits    not absolute limits but limitations imposed by the present state of technology and social organization on environmental resources and by the ability of the biosphere to absorb the effects of human activities. But technology and social organization can be both managed and improved to make way for a new era of economic growth” (World Commission on Environment and Development 1987:8). The idea of ecocracy, as I use it in this dissertation, draws on theory from all four of the theoretical fields I discuss in this chapter. Though developed initially as a critique of development, it bears heavily on the issue of governance and the relationship of human rights to environmental governance in particular.  While the above critiques of the dominant development paradigm provide insight into its many failures, they provide less insight into alternative approaches. Hanchett (1997) and Sillitoe (2000), however, do address the issue of alternative approaches and do so specifically in context of studies they have conducted in Bangladesh. On the basis of her analysis of the negative outcomes of the Flood Action Plan, Hanchett (1997) argues that greater community participation in planning and implementation is necessary in order to avoid these negative outcomes. However, Hanchett argues, the current political and bureaucratic forms of controlled participation cannot change these outcomes. She thinks that the World Bank and government 16  should 'consult' with local people at the planning stages of a water development program. NGOs can be a helpful partner in this process.  Sillitoe (2000) emphasizes the importance of local and indigenous knowledge but is careful not to blame science and technology for the negative outcomes of the Flood Action Plan. He proposes that development agendas should be developed within communities. Community members are the experts when it comes to local environmental resources like soils, crops and water. Their knowledge is systematic and comprehensive and essential to local development. Based on close investigation in Bangladesh, Sillitoe (2000:4) argues that local knowledge is not stagnant and close-minded; rather it is “flexible, adaptable and innovative.”    Political Ecology Several of the development theorists discussed above, Escobar in particular, are also well known for their contributions to political ecology. As Robbins (2012) points out, the field of political ecology has arisen from the confluence of several areas of study including cultural ecology, environmental history, political economy and development studies. Unlike development studies, however, political ecology necessarily involves an emphasis on the ecological as well as the political.   Piers Blaikie is a leading figure in the world of political ecology and is especially well known for his work on environmental degradation (Blaikie and Brookfield 1987). Of more immediate relevance to my research project, however, is his more recent work on environmental justice movements. Blaikie notes the complex relationship among policy makers, politicians, consultancy firms, aid agencies, NGOs and civil society and the tendency for their collective efforts to exclude the voices of marginalized people from decision-making processes (Blaikie 17  2012). As a consequence, environmental justice movements have arisen, in the global south especially, and political ecologists have increasingly been drawn to document and support these movements. Blaikie cites the work of Robbins (1998) on the environmental aspects of development in Rajasthan, India and his argument that engagement between southern and northern activists can produce a better outcome of than purely local movements.  Some, though not all, political ecologists also closely examine the ways in which different social groups “construct” their ideas of “nature” and examine how those constructions support or undermine political structures and processes (Robbins 2012:122-138). Bruno Latour, however, in Politics of Nature: How to Bring the Sciences into Democracy, criticizes the field of political ecology for its tendency to promote an idea of nature that has been constructed through western science. Latour proposes the renewal of political ecology through the “depoliticization of science” (2004:7). This depoliticization will only be possible, he argues, when science comes out of modernism and negotiates the debate of hierarchy (Latour 2004: 27-29).  Mann’s approach to the construction of nature is consistent with Latour’s but more productive for my study. Mann (2009) argues that political ecology should be more Gramscian in approach in order to incorporate the “ethico-political” element of Gramscian analysis. Ethico-politics in this context refers to the moral struggles that inform political debates. Mann (2009:336) thus argues that, “what is of interest is more than the discursive production of nature; it is a nothing less than a moral ecology”. Mann proposes that the moral ecologies that oppose hegemonic processes become part of the historical process by which environmental change occurs and that political ecologists should align themselves with the moral ecologies that support environmental justice movements and oppose processes of marginalization. This approach is thus 18  very consistent with that of Escobar where ecocracy is the most likely outcome the current hegemonic processes informing the sustainable development agenda.  My theoretical approach is also informed by the contributons of political ecologists to the analysis of environmental conflict and the protest movements that arise from environmental conflict. I did not originally intend to study protest movements at my fieldwork site but they were such a constant feature of life in Chapra that I could not ignore them. Protest movements arise in opposition to the hegemonic development processes I describe above but take many forms as accounted for in other settings by Scott (1985), Kapoor (2001), Turner (1997), and Escobar (1992). Scott’s theories about “everyday forms of peasant resistance” and “weapons of the weak” (1985:49) are especially relevant to the pattern of resistance in Chapra. However, the pattern of resistance does not always correspond to what Scott has described for Malaysia. In many cases, people do organize themselves into formal movements, participating in street demonstrations, signing petitions, and creating enduring organizations. Some of these movements in Bangladesh can be described in terms of Turner’s (1997) parapolitical action, referring to actions “oriented above and beyond the political system of the state towards universal social, ethical and cultural values.” Many of these movements also correspond to Escobar’s (1992:27) description of “localized, pluralistic grassroots movements” (1992:27) that have many goals, attract people with diverse objectives and distrust conventional political parties and organizations. I discuss these theories in more depth in chapter eight.   Human Rights Historically, the achievement of human rights is considered foundational to the progress of human civilization. The English Magna Carta (1215), the United States Declaration of 19  Independence (1776), the French Declaration of the Rights of Man and the Citizen (1789), and the United Nations Universal Declaration of Human Rights (1948) are among the most important human rights documents (Khadka 2010). In addition to the Universal Declaration of Human Rights, the UN General Assembly has also passed declarations establishing the human right to water and also to food, work, health and housing. In line with other researchers who use UN definitions to evaluate the wellbeing of particular societies (Davidson-Harden et al. 2007; Khadka 2010:38; Langford 2005: 275), my argument in this study is that access to water, specifically to Ganges River water, is the prerequisite for other human rights.    Article 25 of the Universal Declaration of Rights focuses on standards of living and well-being and specifies the need for universal access to food, clothing, housing, health care and social services. The human right to water is first mentioned in the 1966 UN Covenant on Economic, Social and Cultural rights and in 2002 articles 11 and 12 of the Covenant were revised to include guidelines for protecting basic water rights (Dennis 2003; Khadka 2010; Klawitter and Qazzaz 2005:253; Irujo 2007; Langford 2005). International water law agreements such as the Helsinki Rules (1966),  the UN Watercourses Convention (1997), and Berlin rules (2004) also lend support to the water as a human rights argument and provide detailed recommendations in support of the improved governance of international waters (Khadka 2010:42). Most recently, in 2010, the UN General Assembly adopted a further resolution recognizing “the right to safe and clean drinking water and sanitation as a human right that is essential for the full enjoyment of life and all other human rights” (General Assembly Resolution 64/292 as cited in Sultana and Loftus 2012:1).  UN resolutions have limited legal authority even within signatory countries and, as Bakker (2012) and Sultana and Loftus (2012:10) emphasize, the right to water can mean very 20  different things to different people. Nevertheless, a growing number of scholars are now evaluating the actions of governments and corporations in relation to the international standards outlined in international water agreements. Documentation of human rights failures, at a minimum, lends moral authority and political weight to the arguments of community organizations and social movements advocating for improved water access.  Langford, for instance, notes that all households in South Africa have the right to receive 25 liters of free water, in accordance with the principle that water is a human right. Similarly, in Chile, following World Bank advice, water companies charge clients the full price for water but the government provides subsidies to water companies so they can give a basic amount of water free to low income families. Langford also notes, however, that implementation of these policies has been problematic in both countries (2005:277). In Ghana, by contrast, Langford (2005: 277) informs us that water prices doubled overnight, for rich and poor alike, when a public water utility was privatized, a move made by  the national government in order to make the system attractive for foreign investors.      Irujo (2007) argues that water rights protection is a major prerequisite for other human rights based on the United Nations (2002) principles of respect, protect, and fulfill. Klawitter and Qazzaz (2005:253) focus on the status of human rights in several Middle Eastern countries focusing on water quantity, quality and price for domestic use. They argue that the right to water should be supported because of its importance to human survival, security and capability construction (Klawitter and Qazzaz 2005:255). Water management should be set up, they argue, so as to ensure these freedoms based on an equality and social justice approach. From this perspective governments are also responsible for protecting their citizens from the influence of a 21  third party like the International Monetary Fund or the World Bank. Neoliberal economic policy and structural adjustment programs should not be allowed to diminish the right to water.  My approach in this study extends the arguments discussed above to emphasize the government’s obligation to care for and preserve the ecosystems on which communities like Chapra depend. Also, as Pamukcu (2005) argues, the government obligation to protect water access and all the rights that depend on it, extends to international rivers and implies the need for cooperation among all countries sharing a river basin.  Water Governance Environmental governance theory broadens the notion of how resources are managed to include much more than the government regulations created as part of a formal management system. This broader framework includes the roles of community members, advisory groups, interest groups, lobbyists and NGOs (de Loë et al. 2009; Wagner 2009). Governance theory also recognizes the role of values in shaping decision-making processes and the fact that different governance approaches articulate with different types of human/water relationships. Wagner (2013:7) argues that, “if our goal is to live sustainably within the watersheds we inhabit, then we need to understand watersheds as socioecological systems, as whole systems, not systems that are sometimes social and sometimes ecological, and not always both at the same time.” Hastrup (2013), similarly, argues that “waterworlds,” come into being based on the connections among river flows, seasonal patterns, vegetation and habitat and cultural practices. A socioecological approach to governance is relevant to my research because it could help promote community voices in water management and support an ecosystem-based approach. The ecosystem concept has its limitations, as many scholars have pointed out, but it is appropriate for 22  my research because of the deep dependence of marginalized households in Chapra on the ecosystem services of the Ganges and Gorai River basins. An ecosystem approach that is inclusive of the social relations embedded within it is consistent with the observation by Orlove and Caton (2010:402) that water is a “total social fact” that needs to be understood holistically. On that basis they critically analyze the Integrated Water Resources Management (IWRM) approach because of its tendency to integrate only specific resource sectors in management processes and to limit IWRM boundaries to national boundaries (Orlove and Caton 2010). Criticizing overly technocentric and bureaucratic approaches, they argue for inclusion of cultural factors and the integration of different sectors and groups at regional, national and international levels. Rather than the World Bank’s evaluation of water as a commodity, Orlove and Caton evaluate water as a human right and emphasize the social justice aspect of water policy and governance.  Wagner (2009) recommends a distributive, multi-level water governance approach that facilitates the involvement of both formal and informal institutions in decision-making processes. He argues that this approach will be more effective than centralized, state-dominated water management systems if the goal is to preserve ecological resilience as well as economic sustainability. Distributing authority among institutions at different levels within the total system, he argues, is also more democratic and can create a system of checks and balance among competing groups. Based on his research in the Okanagan Valley in British Columbia, Wagner notes that governance systems, which are highly centralized in formal terms, can behave more like distributed systems in practice, but that legislative changes should be made to acknowledge the principle of joint authority by institutions at different levels and to discourage unilateral decision-making by central authorities (2009:5).  23  Lebel et al. (2005) focus more explicitly than Wagner, on the question of scale in their study of Mekong Delta water governance, noting the simultaneous presence of spatial, temporal and jurisdictional scales. Multi-scalar institutional networks generate their own unique political dynamics and different actors’ interests “constrain, create and shift scales and levels” (Lebel et al. 2005:2). Governance networks are strongly influenced by local development history, by international development institutions like the World Bank, class relations and by hydrological and ecological factors including climate change. Ideally, a multi-scalar system will include mechanisms for including marginalized groups at different scales and will provide them with real power in decision making processes. In conclusion, the governance approaches that are most informative for my research are those that focus on social justice and human rights issues and on the principles of equity and environmental sustainability, as opposed to the more narrow approaches that focus on water as an economic resource alone.    Organization of the Dissertation   In the following chapter, I describe the research methods I used to gather the data necessary to answer my research questions. I then present the evidence I gathered in chapters three through eight. In chapter three, I describe the impact of reduced flows of Ganges River water on the local ecoystem and on livelihood practices at Chapra. My evidence confirms that the diversion of water by India at the Farakka Barrage is reducing local natural resources and increasing vulnerability to flooding, drought and riverbank erosion. In chapter three, I also focus on the way in which the political culture in Bangladesh causes further hardships for local communities by ignoring their voices in favor of a top-down approach that serves elite interests.   24  In order to fully document the effects of top-down water management by the Bangladesh national government, I analyze two major water management programs in chapter four: the Gorai River Restoration Project and the Ganges-Kobodak Project. I explain how and why the restoration project has failed to overcome the negative effects of the Farakka diversion and has contributed to additional environmental degradation of the Ganges Dependent Area. I also explain why the Ganges-Kobodak Project has not been able to realize its goal of securing sufficient water supply for agriculture.  Currently, local people at Chapra are experiencing a radical transformation of their agricultural practices. All aspects of the productive cycle, including the supply of seeds, fertilizer, land and water are subject to new forms of commodification controlled by state agencies and local elites. I describe these processes in Chapter five. The government of Bangladesh attempts to mitigate the hardships and social inequality caused by these forms of commodification through safety net programs that subsidize the cost of new agricultural materials and technologies, or provide employment, housing, education and health benefits to the poorest households. In chapter six, I describe the role of local governments in distributing these safety net program benefits and the extent to which benefits are captured by local elites.  In chapter seven, I evaluate the extent to which the hardships now being experienced by Chapra residents constitute human rights violations based on international customary law and a series of conventions and agreements created by the United Nations. The marginalized households experiencing these hardships are raising their voices and resisting their marginalization in many ways including participation in local environmental movements which I describe in chapter eight. These movements are occurring at local, national and international levels. In chapter nine, my concluding chapter, I summarize my data and return to several of the 25  theoretical questions discussed above in order to formulate recommendations for how water governance in the Ganges Dependent Area in Bangladesh can be improved so as to protect the human rights of marginalized households and the critical environmental characteristics on which their livelihoods depend.  26  Chapter Two  RESEARCH METHODOLOGY  As noted in chapter 1, the main goal of this research project is to document the hardships currently being faced by agricultural communities in the Ganges Basin in Bangladesh and identify the impact of those hardships on the human rights of farming households. A second important goal is to develop policy recommendations to improve water governance within the Ganges Basin in Bangladesh and thereby reduce the vulnerability of farming communities to human rights abuses and loss of livelihood. In order to accomplish these goals I first needed to acquire in-depth knowledge of the farming and livelihood practices of basin communities. Traditional farming practices in this region are deeply embedded in local culture, history and ecology and I thus employed ethnographic methods, informed by the fields of environmental anthropology and political ecology, to gather information about traditional farming practices, water management, local ecological knowledge, the local economy, human rights issues, and the opinions of farming households about recent changes to their way of life and current hardships. This was done through a combination of participant-observation, interviews, focus group discussions, in-depth case studies and household surveys, all of which were carried out in the community of Chapra.  Since water governance in the Ganges Basin in Bangladesh is now controlled almost entirely by national government agencies and the international donors on which they depend, and since their actions over recent decades have powerfully affected almost all domains of village life, I also interviewed government officials whenever possible and gathered secondary data from government reports and published literature. Given the impact of the Farakka Barrage on 27  the flow of water into Bangladesh, extensive secondary data was gathered about the barrage and about international water law from Indian, Bangladeshi and international sources.  The policy recommendations I make in my concluding chapter are based on my   intention to give voice to marginalized farming households in the Ganges Basin as well as to apply international theories about best practices in water governance and the protection of human rights.  Residents of Chapra vary significantly in socio-economic status based on the size of their landholdings. Since the vast majority of people at Chapra and in many nearby communities fall into the marginalized category, it is possible to speak of marginalized communities as well as marginalized households. Marginalized households include those with landholdings insufficient to meet their economic needs and those without land, who work as agricultural day laborers, blacksmiths, boatmen or as unskilled labourers. This category also includes the disabled and widowed. Whether they possess their own land or not, marginalized people are all highly dependent on the ecosystem services of the Ganges Basin and extremely vulnerable to disruptions in Ganges Basin due to human interventions such as the Farakka diversion (described in chapter 3), and the Gorai River Restoration Project and Ganges-Kokabak Project (described in chapter 4). On the other hand, the better off farmers are able to overcome the negative effects of these technologies based on their surplus economic resources. Historically, the Ganges Basin is the site of one of the first ancient civilizations, dating to 2500 BCE and located at Mahasthangar in Bangladesh and Gaud in West Bengal, India (Novak 1993:58). The basin was referred to as the “Paradise of the World” and was the major source of wealth for the Mogul Empire and later the British Empire (Rahaman 2009:2). Local farming communities thus, have been subject to interference for centuries though never on the scale experienced today by communities in Bangladesh. The British Colonial Government in India 28  constructed 2,298 miles of canals and distributaries as part of the upper Ganges Canal project in 1854 and used them to irrigate 1.2 million acres of croplands. After independence in 1947, the government of India constructed a series of large barrages and diversions on the Ganges for irrigation and flood control purposes. Only after 1975, however, when the government constructed the Farakka Barrage did this type of large scale intervention in Ganges Basin hydrology have a major impact in Bangladesh (Jackson and Marmulla 2001; Mamun 2010:923; Minkin and Boyce 1994).   Given the large area of impact within Bangladesh, one that extends from the border of India and Bangladesh to the point where the now much diminished Ganges River joins the larger flow of Brahmaputra, it was necessary to select a representative but delimited area of study. For the reasons described below, I selected Chapra, a village in Kushtia District, as suitable for my research goals. Chapra is located on the Gorai River, a main branch of the Ganges River which, after its origin in Kushtia District, flows 199 kilometers downstream to the Bay of Bengal through the Madhumati and Baleswar Rivers. It provides essential freshwater flow to the coastal Sundarbans mangrove forest, controlling salinity intrusion of this unique coastal area for at least the past 500 years (Islam and Gnauck 2001). Although my study focuses on the local effects of government interventions at Chapra, the scale of the interventions has necessarily required some consideration of more distant impacts.   Chapra Village I selected Chapra as my fieldwork site because of its location on the Gorai River and because it has experienced significant environmental impacts during both wet and dry seasons due to the 29  Farakka diversion. Chapra is only five kilometers away from the Ganges River itself (Figure 2.1).     Figure 2.1: Map of Ganges Dependent Area and Fieldwork Site at Chapra. Source: Adapted from Banglapedia 2012.  30  The area is also subject to the influences of two large scale water engineering projects implemented by the government of Bangladesh to mitigate the effects of the Farakka diversion. Farming households at Chapra are currently dependent for much of their irrigation water on the Ganges-Kobodak (GK) Project, the largest irrigation project in Bangladesh occupying an area of about 1700 square kilometers in four districts (Bangladesh Water Development Board n.d.). In some locations GK Project canals are located only five hundred feet away from the site of another major project, the Gorai River Restoration Project. Chapra is thus subject to multiple large-scale water engineering projects that exemplify the scientific and technology-driven approach of both the Indian and Bangladeshi governments and the impacts of these projects on hydro-ecology, community livelihoods and human rights is broadly representative of what is happening throughout the region.  Chapra is a rural village that is composed of six paras or sub-villages: Bohla Govindpur, Dekipara, Koburat, Madulia, Charpara and Purbopara, and Sindoh. With a population of 4,331 people, Chapra is the largest of several villages under the jurisdiction of the Chapra Union Council and it falls within the Kumarkhali Upazila Parishad (Sub-district) within Kushtia District. Chapra village is about 254 kilometers away from the Farakka Barrage in India and about 11 kilometers downstream from the Hardinge Bridge in Kushtia (see Figure 2.1, page 29) where the central government in Bangladesh has located one of its major Ganges River water data collection stations. The Ganges River enters into Bangladesh in Rajshahi District and passes through Kushtia District before it merges with the Brahmaputra River at Shibalya Upazila in Manikgonj District. There is a population of 1.8 million people in Kushtia District, with an average density of 1,073 people per square kilometer (Bangladesh Bureau of Statistics 2011). In 31  Kushtia, 96 percent of the people are Muslim and four percent are Hindus or adherents of other religions.  About 70 percent of the population of Chapra (963 of 1387 households) make their living through farming or farm labour (Bangladesh Election Commission 2011). The Bangladesh Bureau of Statistics (2005) classifies farming households as large, medium, small and marginalized. Throughout this study, I describe the large and medium farmers as the rich and intermediate farmers respectively and, for reasons I explain below, I often use the term marginalized to refer to both small and marginalized groups as they are classified by the BBS. Rich farmers are those who own 7.50 acres or more agricultural land; intermediate farmers own between 2.50 and 7.49 acres; small farmers own between 0.50 and 2.49 acres; marginalized farmers own less than 0.49 acres of cropland. Rich farmers make up one percent of the population; intermediate farmers make up eight percent; small farmers 51 percent and marginalized farmers 41 percent (Bangladesh Bureau of Statistics 2005). Fourteen percent of the marginalized farmers own no farm land at all but depend on farm labour for all or most of their income. Based on my own research, I refer to both small and marginalized farmers as marginalized throughout this study since 2.5 acres of land are not enough for household self-sufficiency. Small farmers must therefore rely on a combination of off-farm activities to support themselves; these include working as labourers for rich and intermediate farmers, sharecropping and the gathering of freely available natural resources. Small farmers are therefore extremely vulnerable to loss of any of these livelihood resources and, like those with no land or less than half an acre, are significantly marginalized within the local socioeconomic environment.  32  Class differences at Chapra are typical of those that exist throughout Kushtia District and elsewhere in Bangladesh and they are highly significant for this study. It should be noted, however, that systems of reciprocity and mutual support cut across local class hierarchies to some extent while at the same time providing the basis for strong community bonds among the marginalized majority. Those who are better off often help poor people with donations of food, vegetables and fruits. Rich and poor participate together in Muslim religious celebrations and recite similar religious prayers to ask for success in agricultural production. Muslims and Hindus also enjoy community activities like polo bice and nouka bice together.  Polo bice is a community activity at the end of the borsha season. Nouka bice is a community boat race in the middle of the borsha season. Women of all classes, as well as men, perform agricultural work or carry out domestic tasks in support of household agricultural production. Mostly, Bangladesh is a patrilineal society where sons own two-thirds and daughters own one-third of family assets. After marriage, daughters sell their assets and bring their cash endowments with them to their husband’s and father-in-law’s houses. Although women do gender specific forms of agricultural work they mainly rule inside household resources, while men take the main responsibility for outside activities such as agricultural production.  A good deal of local ecological knowledge is also common to all classes and is transmitted from elder to younger generations through stories and proverbs. This includes knowledge about how to grow crops but also knowledge about the manufacture of tools such as ploughshares from local sources. It also includes knowledge about local housing and transportation strategies adapted to seasonal conditions.  During the rainy season, people travel by locally made boats to markets and schools, to visit relatives and participate in traditional community boating activities locally called nouka bice. During the dry season, they travel the 33  local mud roads in bullock carts and on foot. Community fishing, locally called polo bice, promotes community solidarity and develops consciousness about conservation.  Class differences are of fundamental importance in this setting, however, when assessing the relative importance of local natural resources to household livelihoods. Like the villages in Rajasthan, India, described by Rita Brara (2006), the marginalized households at Chapra are deeply dependent on common property resources as well as on the farm lands and water resources to which they hold formal, individual property rights. Although rich and intermediate farmers may also access these resources, they are much less vulnerable to their disappearance. As noted above, these commons include river, lake and wetland fisheries, many of which are seasonal in nature, dependent on both local rainfall patterns and the level of water in the Gorai River. Common property resources also include a wide variety of wild plants that are harvested as food, used to make tools, used as fuel or as building supplies, and as compost. Siltation carried by the river during flood season also enriches the farm soils. Traditional agricultural practices require extensive knowledge and use of such resources and each season is distinct for the resources it provides. I developed the ethnographic approach I describe below largely in order to document the richness of local ecological knowledge but also the relative importance of local ecosystem services resources to marginalized households by contrast to those that are better off.   Methods of Data Collection My research project focuses on empowerment for marginalized basin communities by bringing their voices forward into water policy making venues (Bebbington 1993; Long and Long 1992; Schrijvers 1991; Thompson and Scoones 2009:66-7). The basin community at Chapra is the center of my investigation with ethnographic methods as the main data collection technique. 34  Since my intention was to document human rights concerns as well as livelihood practices, much of my ethnographic fieldwork focused on non-agricultural activities such as education, health care and housing (Johnston 2010:10). An extensive body of secondary data about water management practices, policies and projects was gathered from government records and publications at local, regional and national levels. This was supplemented by secondary data concerning international water law and international initiatives in support of the principle of water as a human right and the relationship of water to other human rights.   Although I created a list of interview questions prior to leaving for the field, as required by the UBC Okanagan Ethics Review Board, I did not finalize any of my interview questions until after I had visited my field site and introduced myself to local farmers, water managers and local government officials. Mistakes collecting local knowledge are normal in the research process (Berkes et al. 2000:1252; Fife 2005:72) and, even though I grew up in and previously conducted research on water management and agriculture in a nearby area of Bangladesh (Hossen 2009), I wanted my research process to be as fully informed as possible by local interests and local knowledge. Following a series of conversations with my initial contacts and after several tours of the area I developed a set of questions for use with focus groups and invited local participants on the basis of advice provided to me by my contacts. My project was thus community-based to some extent though constrained by the fact that my community of study is deeply divided by class, socioeconomic status and political power. Although my primary goal was to give voice to marginalized farming households, I also attempted to document the opinions and livelihood practices of dominant groups. On the basis of the focus group responses, I was able to create appropriate criteria for selecting several households for participation in more in-depth case study research. The focus group and case study outcomes then supported my 35  development of a survey that I was able to administer to a large sample of households with proportional representation from each of the three socio-economic groups I identify below. Although I was not able to live in the community itself during my time of study, I secured lodgings in the nearby city of Kushtia, commuting daily to Chapra to conduct interviews, accompany household members during their daily activities and, as a result, over a period of one year, I was able to spend many hours in the community as a participant-observer, attending social gatherings and learning about the political, social and environmental aspects of community water and agricultural practices. These experiences have guided my fieldwork activities, my development of a personal social network within the community and my attempt in this dissertation to represent marginalized households in ways that could be empowering for them (Berkes et al. 2000:1258; Chambers 1994; Sillitoe 2000).    Primary Data Collection  Focus Groups After gaining as much familiarity with the research site as possible, I convened four focus groups, each stratified by socioeconomic position, to gather qualitative data about local water practices, seasonal patterns, basin history and community livelihoods at Chapra based on the methodological approaches of Chambers (1994:1437) and Sillitoe (1998:215). The village people at Chapra use local ecological knowledge for livelihood practices. I developed questions to explore the linkage between the Ganges Basin ecosystems and community livelihood practices on the premise that, success in accessing this basin flow is necessary to secure success in agricultural and employment practices, which in turn can promote freedom and autonomy of social, cultural and economic aspects of individual and community livelihood (Molle et al. 36  2008:1; Orlove and Caton 2010:401; Van Eijck and Roth 2007:934). Many people have lived in Chapra for seven generations or more and the focus group allowed me to gather information about these generational water practices and present challenges. The target groups were local farmers, of varied socio-economic status, and agricultural laborers whose livelihoods are dependent on the basin ecosystems (Adnan 1991; Boyce 1990; Wood 1999:731). A significant number of women attended the focus group with their male household members. I made sure to include individuals over 40 years of age since they are more likely to have in-depth knowledge about past water practices, ecosystems and local livelihoods. I excluded children who were below 18 years of age from the focus group, although several children did accompany their parents to the meeting.  Local research assistants helped me to arrange these meetings, collect preliminary data to develop the focus group questions and to invite appropriate participants. The mother-in-law of one of my local research assistants resides at Chapra and her family guided me to select the focus group site and provided facilities like mats and chairs for the meeting. I was careful to select a meeting place that was suitable for villagers from diverse backgrounds and different classes. The site was a fruit garden situated on the Gorai River bank. The garden trees provided us shade to avoid extreme sunlight or rain. At the end of the focus group, participants were offered food and drinks. In addition to local farmers and farm labourers, I invited village leaders and their followers to attend the focus group meeting. Local school teachers, government administrators, research assistants and volunteers helped me to organize the focus group successfully. Local government officials in Kushtia, the Union Council (UC), Kushtia Government College and Kushtia University provided me with suggestions. Professors from the sociology departments at 37  Kushtia Government College and Kushtia University provided me with local information and encouraged their students to participate as research assistants. I hired five students from Kushtia Government College and one student from Kushtia University. Two female research assistants were hired to collect data from female respondents while four male assistants were hired to collect data from male respondents. These research assistants had social science research backgrounds and knowledge about local culture and tradition.    Forty-four male and female household heads attended the focus group meeting, including seven rich farmers, nine intermediate farmers and 28 marginalized farmers, 12 of whom were small farmers and 16 day labourers (see Table 2.1). Based on my previously mentioned statistical data of the different socioeconomic groups of people, I wished to ensure proportionate representation of the rich, intermediate and marginalized farmers. However, given the relatively small number of rich households (1% of all households), I included all who were willing to attend in order to avoid small sample bias. I also included a few more intermediate farmers than justified on a strictly proportionate basis for the same reason. The percentage of marginalized farmers in my focus group was thus significantly lower than in the general population but large enough to effectively represent this group. Given the difficulty of distinguishing rich and intermediate, and marginal and landless farmers prior to my focus group meeting, I was satisfied with this outcome.     There were three major parts to the focus group meeting which was held on 11 November, 2011: participants were first asked to answer a few questions about their household’s demographic characteristics and socioeconomic status, for example, full name, age, marital status, extent of land ownership, number of children and educational level. This initial, written, semi-structured interview included 20 questions and took two hours to complete. Once I and my 38  research assistants had collected participant responses to the first 20 questions, we divided them on the basis of the information they provided into four socioeconomic groups based on the size of their landholdings as described above. Each group then sat down on a separate mat in a separate area of the fruit garden. I assigned one research assistant to the group of seven rich farmers, one to the nine intermediate farmers, and two assistants to each of the larger groups of 12 small farmers and 16 day laborers. Although these latter two groups were larger than ideal for a focus group, they were able to set aside sufficient time to complete the task and everyone who wished to speak had a chance to do so.   Each focus group was asked to respond to a set of 49 questions which focused on past water practices and current challenges. These sessions took four hours to complete on average.   Table 2.1: Focus Group Participants (pseudonyms) Rich Farmers Intermediate Farmers Marginalized farmers Small Farmers Day Laborers Tanvir Ibrahim Faisal Khaleque Dulal Mokles Naser Tofajjal Jalil Unus Jafor Hanif Taher Barek Abul Sorkar   Joardar Kafil Malek Faruk Sabina Harun Morjina Kabir Suman Sabbir Tarun Billal Kamal Fajal Keramot Laily Kusum Jobbar Alamgir Bakar Mofiz Fazlu Abonti Gedi Bimal 07 09 12 16 39    We used the Bengali language, with local dialects, as the main means of communication. With their verbal permission, following Research Ethics Board guidelines of the University of British Columbia Okanagan, I audio recorded their conversations and took some photographs. The third part of my focus group meeting involved the selection of four community members for in-depth case studies. I describe the selection process in the following section.   In-Depth Case Studies and Participant Observation My focus group data provided the foundation for developing a more extensive set of questions for in-depth case studies of four households, one from each of the above groups. The in-depth studies were conducted partly in the homes of participants and partly during site visits to their agricultural lands, the river, nearby canals or other local water bodies, thus affording me the opportunity to observe as well as participate in local activities. I used this participant-observation and in-depth case studies to explore seasonal aspects of borsha (wet) and khora (dry) seasons in coordination with the seasonal cropping patterns which require year-round fieldwork (Mamun 2010; Paul 1984:7). The case study respondents provided me with detailed information about locally embedded water, ecosystem service and livelihood practices during a year of fieldwork conducted from December 2011 to April 2012. In addition to spending informal observation periods with them, I also prepared a set of 122 interview questions in order to systematically gather demographic information, knowledge about local ecosystem service changes, the outcomes of government sponsored water management projects, agricultural practices, employment patterns, education, health care and housing. These questions guided my exploration of linkages between the Ganges River flow, local water management traditions, land use patterns 40  and people’s current challenges. These traditional ecosystem service practices are cultural texts of locally adapted conservation and livelihood systems.  The selection of four community representatives was a challenging task. I used some specific criteria to select these representatives. In order to gain as much insight as possible into traditional agricultural practices, farming history and environmental history, I selected older farmers and labourers who showed leadership qualities and extensive local knowledge. Based on these criteria, I selected Tanvir, Tofajjal, Joardar and Billal, who represented the rich, intermediate, marginalized and traditional day labourer socioeconomic groups respectively. Through these four respondents, I wanted to understand the different socioeconomic group voices regarding local water practices and challenges for agricultural and employment practices based on local knowledge and cultural practices (Berkes et al. 2000:1251; Biersack 1999). In my case studies, I also included the voices of female household members in order to understand their distinct contributions and the gendered nature of work in families and the community.   Tanvir Tanvir is a sixty-year old rich farmer, who has lived in Chapra all his life. He completed the Higher Secondary Certificate (HSC) educational degree but did not apply for an office job because he received 23 acres of cropland from his parents and another seven acres from his wife’s parents. He married Sohali in 1970. Tanvir proudly informed me that his father did not take any dowry from his father-in-law. One brother of Sohali is a national leader of the Bangladesh National Party (BNP) and the two brothers are bureaucrats of the central government.  41  Currently, Tanvir makes profits on croplands, fruit gardens, fisheries, commercial agricultural production and agribusiness. He has five fisheries projects and four fruit gardens. He uses leased land from the government sponsored Ganges-Kobodak project for two of his fisheries projects and low cropland of his own for the other three. Most of his assets are within four kilometers of his house, the Gorai River, and the Ganges- Kobodak project. He also has two irrigation machines, two plow machines, two mechanized vans, eight mechanized paddy machines and three crop storage facilities. Moreover, he has three residences, one each in Chapra, the nearby city of Kushtia and the capital Dhaka. His houses in Kushtia and Dhaka are four and three floor buildings respectively.  Tanvir and Sohali have three sons, Kofil, Masud and Firoj, and two daughters, Shapla and Shaon. Kofil is 40 years old, holds a Bachelor of Science degree and owns an agricultural business farm under Tanvir’s supervision. He sells chemical fertilizers, HYV seeds, and pesticides in this farm. He is also a district level political leader of the Bangladesh Awami League, the current governing party. Kofil’s wife, Rakhi, is a Bachelor of Arts degree holder and lives with Kofil in Kushtia. Tanvir’s second son, Masud, is thirty-five years old, a government officer with a Master of Arts degree. Masud’s wife, Trishna, is a banker and lives with him in Dhaka. Firoj, the third son, is 31 years old and holds a Master of Science degree. He has migrated to Australia with his wife. Shapla, Tanvir’s oldest daughter, is 27 years of age, is married to a lawyer and lives in Kushtia. Shaon, the youngest daughter, is 23 years old, has completed the School Secondary Certificate degree and is married to a local poultry businessman who is also a local political leader of the Bangladesh Awami League.         42  Tofajjal   Tofajjal is a fifty-nine year old intermediate farmer who has lived in Chapra since his birth. In addition to his current house, he has a second house in Chapra and has built a three floor apartment building in Kushtia. He has completed a School Secondary Certificate degree but his parents did not allow him to apply for an office job. He received four acres of cropland from his parents. He married Salma in 1979 and obtained another three acres from his father-in-law. One of Salma’s brothers is a government officer in Dhaka and another is a non-government organization (NGO) officer in Kushtia. Tofajjal has two fish ponds and three fruit gardens. He has a shallow tube-well, one plow machine and two mechanized paddy machines. Currently, Tofajjal is successful in maintaining a decent livelihood based on his own resources. Tofajjal and Salma have two daughters, Shahida and Karima, and two sons, Mamun and Kabir. Shahida, age 30 years, holds a BA degree and is married to a police officer. Karima, age 28 years, has an MSc and is a government officer in the rank of assistant secretary. Mamun is 25 years old and also holds an MA and, during the time of my research, was looking for a job. Kabir, age 21, is currently a BSc student at Kushtia Government College.   Joardar  Sixty-five year old Joardar is a marginalized farmer whose family has lived in Chapra for more than six generations. He dropped out from school in the ninth grade and now owns one half an acre of cropland. His cropland is one kilometer away from his house and the Gorai River. He married Suma Khatun in 1971 and his wife’s family has also lived in Chapra for more than four generations. One of his wife’s brothers is a labour leader in the Kushtia Keri Sugar Mill and a sister’s husband is a rickshaw association leader in Dhaka. 43  Joardar and Suma have three daughters, Amina, Joshna and Shifaly, and two sons, Jakir and Karim. Joardar’s mother and one brother also live with his family. Amina is thirty-one years old; she married at the age of eighteen and does not have formal school education. Rahim has completed a School Secondary degree and works as an NGO employee in Chapra. Joshna is twenty-eight years old, also holds an SSc degree and works for the same NGO.  Twenty-five year old Shifaly is studying at the BA level at Kushtia Government College. Jakir is twenty-three years and has migrated to Kuwait for an employment opportunity. Karim is twenty years old and works with his father on his own land and obtains outside employment when it is available.     Billal  Billal is a fifty-two year old, traditional agricultural day laborer whose family has lived in Chapra for three generations. He has neither a formal school education nor a piece of cropland. Agricultural day labour is his main source of income. His house is 20 meters from the Gorai River. His wife, Parvin, has also lived in Chapra since her birth but her father is originally from another district.  Billal and Parvin have four daughters, Jesmin, Kamrun, Shampa and Shakina, and three sons, Jamal, Lablu and Kamal. Jesmin is twenty-three years old and does not have a formal school education. She is married to Kubbat who works as a van driver. Kamrun, twenty-five years old, dropped out from school after the sixth grade and is married to Khokon who works as an agricultural worker. Shampa, age 23, dropped out of school after the fifth grade and is married to Sabbir, who works as an agricultural worker and fisherman. Jamal, age 21, has migrated to Dhaka, where he works as a garment worker. Kamal, 18 years of age works for Tanvir.  44  Tanvir, Tofajjal, Joardar, and Billal represent the four different socioeconomic groups who depend on the Ganges Basin ecosystems differently. These group identities originate from disparities in patrilineal land ownership, which appear to have been increasing over the past few decades. Through interviews and by accompanying these four respondents during their daily activities, I explored culturally embedded water practices and their connections with social, economic, political and bureaucratic aspects of local resource management. As a participant observer, I visited crop fields, government managed irrigation and water diversion projects, the fisheries projects, and wetlands with my in-depth case respondents. I also visited the Ganges Basin at Taalbaria and Gorai River mouth periodically during summer and rainy seasons to observe the basin flow, sedimentation, charland and river bank erosion. The Gorai River originated from the basin at Taalbaria.  For these activities, my male research assistants and I were closely attached to the male case respondents and my female research assistants were closely attached to the female case respondents (Hanchett et al. 1998). A male researcher is not welcome to meet with female respondents and, thus, the female research assistants developed good relationships with these women to collect information about their water practices and challenges. They met during times when the women were cooking, socializing, carrying water to their houses, caring for domestic animals and collecting wild vegetables.  Local people accepted me positively on the occasions when I accompanied case study participants on their rounds of activities. I kept records of all activities and took photographs of the basin flow, the Gorai River’s shallow condition, river bank erosion, embankment failures, fisheries projects, agricultural production and local water bodies.  45  Chapra Household Survey Based on the results of my focus group and in-depth case study interviews, I developed a household survey in order to collect quantitative data from a larger, more representative and proportionate sample of farmers on current water practices, agricultural challenges and human rights conditions at Chapra (Mamun 2010:926; Olsson and Folke 2001). This survey included 185 questions organized in the same nine sections as the case study questions but included more closed-ended questions as opposed to the more open-ended style of the case study interviews.  I used the Bangladesh Election Commission voter list for Chapra to select the household survey respondents. As noted above, Chapra has 1,387 households, 963 of which are agricultural households. I selected a total of 259 agricultural households based on simple random sampling using a lottery system without replacement to select them (26.9% of all agricultural households). Among these households, there were 34 women respondents who originated from the marginalized households. According to Bangladesh Bureau of Statistics (2005) data mentioned above, 51 percent and 41 percent of the agricultural population own less than 2.49 and 0.49 acres of cropland respectively. As noted above, I categorize these two groups as marginalized farmers. My random sample included 115 marginalized households (44.4 percent of my sample and 23.0 percent of all marginal households in Chapra) and 89 day laborer households (34.4 percent of my sample and 22.5 percent of all day laborers households in Chapra). The percentages of these households in my sample are slightly lower than their percentages within the Chapra population as a whole but, rather than employ a stratified random sampling technique in order to exactly match percentage figures, I decided to slightly expand the number of rich and intermediate households in my sample, due to their low numbers in the community. I selected seven rich farming households (2.7% of my sample but 100% of the rich farmers in Chapra) and fourteen 46  intermediate farming households (5.4% of my sample but 22.4% of all intermediate farmers in Chapra) for my household survey. I did not apply the sampling technique to select the rich farmers given their limited number. I use this technique for all other socioeconomic categories which included some participants of the focus group discussion as survey respondents.   Secondary Data Collection  In addition to my primary data sources, I collected secondary data at the local, national and international levels. Although I had originally intended to conduct interviews with government officials as well as farming households, time did not allow for more than the collection of secondary data in the form of statistical reports, fact sheets and grey literature. At the local level, I collected local water and agricultural data from Chapra Union Council, Kumarkhali Upazila (Sub-District) office and the Ganges-Kobodak Project office. The Chapra Union Council chairman provided me local voter lists of the Bangladesh Election Commission, Chapra community and union area maps and information about the roles of Union Council officials in local resource management and safety net programs. The Kumarkhali Upazila officials provided me local socioeconomic and demographic data as well as data from the Upazila on local flooding, drought, river bank erosion, embankment failures and water stagnation. I also gathered information about roles of the Upazila Parishad officials in local resource management and safety net programs. I visited the Roads and Highways Department at the Upazila level to access their road construction data on Chapra. Again, the Department of Agricultural Extension officials at the Upazila level provided me local HYV crop production data. The Upazila Education officials gave me educational data about local schools, students and drop out rates and the Upazila Health Complex gave me health care service data in Kumarkhali Upazila. The GK 47  Project office in Bheramara and Kushtia provided me with detailed secondary data such as published booklets, reports and maps about their history, irrigation supplies and challenges.  At the national level, I collected secondary data directly from several government ministries including the Ministry of Water Resources (MoWR), Bangladesh Water Development Board (BWDB), Department of Agricultural Extension (DAE), Bangladesh Agricultural Research Council (BARC), Ministry of Agriculture (MoA), Bangladesh Bureau of Statistics (BBS), Water Resources Planning Organization (WARPO), and Ministry of Disaster Management and Relief (MoDMR). Of special importance were documents from the MoWR on National Water Policy and the Feasibility Study and Detailed Engineering Plan for a proposed Ganges River Barrage Project (described in chapter 4). The BWDB provided me with the Ganges River flow data at the Hardinge Bridge Station, downstream from the Farakka Barrage in India, for 1960-2011 and detailed documentation concerning the Gorai River Restoration Project and the Ganges-Kobodak project. WARPO gave me National Water Management Plan documents outlining their contribution to the development of water resource management systems in Bangladesh.  Ministry of Agriculture provided agricultural policy documents and the agricultural extension office provided agricultural modernization data for HYV crops, chemical fertilizers, deep tube-wells and shallow tube-wells. Bangladesh Bureau of Statistics provided me with detailed socioeconomic data about food, employment, education, health care, and housing. The MoDMR provided data about government planning in respect to disaster relief following episodes of flooding and drought. Dhaka University, Bangladesh Agricultural Research Council and University of British Columbia libraries were also major sources of secondary data on water and agricultural activities.  48  Other sources of secondary data were gathered from the websites of the Indo-Bangladesh Joint River Commission, the United Nations, the World Bank, and from Bangladeshi non-governmental organizations including Shushanar Janniya Nagorik, Transparency International Bangladesh and the Centre for Policy Dialogue. Online newspapers at the national level such as, Daily Star and Daily Amardesh in Bangladesh, the Times of India in India, the Economist and New York Times at the international level, gave me detailed data about South Asian politics, history, and water resources. The government websites in Bangladesh, for example, the Bangladesh Agricultural Research Council, were valuable sources of information about recent agricultural research on HYV crops, chemical fertilizers and pesticides. I accessed the Millennium Development Goals and human rights documents from United Nations websites. The World Bank websites provided documents on programs which they helped develop and fund including the Poverty Reduction Strategy Paper and the Gorai River Restoration Project.   Ethical Issues and Limitations of My Dataset  Given the influence of external “elites” in this setting and their role in implementing top-down policies, I had a moral responsibility to represent local communities’ concerns fairly at the central government level (Caduff 2011; Fassin 2008). Value judgments were necessary at every stage of research and sound judgments were critical to the construction and representation of local knowledge. During my fieldwork at Chapra, I had enormous freedom in smelling, tasting, seeing and experiencing everything I wanted while conducting focus group, case study and participant observation fieldwork. Continual self-reflexivity was essential to the gradual process through which I constructed my understanding of local knowledge (Fife 2005:150). Using a standard anthropological approach, I explored community water practices through observing, 49  recording and analyzing based on research accountability, trust and intimacy. This engagement reduced the gap between ‘I’ and ‘them.’  My fieldwork experience always reminded me that knowledge is a major issue between “I” and “them.” Fife (2005:124), for instance, has argued that local knowledge is a form of private property owned by local communities. Sometimes local knowledge can be exploited by corporate elites to promote their own agendas, which can and often do disempower marginalized communities. I was always conscious of these agendas and sought, instead, to represent local knowledge so as to protect community interests and support their empowerment (Sillitoe 1998:208).  Local elites created a major challenge for me when accessing the voices of members of marginalized households at Chapra. I always encountered local elites’ surveillance when I visited the Gorai River dredging and the Ganges-Kobodak Project irrigation sites. Because of this surveillance and the distrust it implied, I resided with an elite family from Chapra who now live in a nearby area, Lahineepara, to secure their trust in my fieldwork activities although my original plan had been to reside with a marginalized family. Staying with an elite family also provided me with increased security in a region where radical political leaders and extremist (Marxist) groups occasionally kidnap rich people for ransom or execute them for suspected bribery. During the day, when conducting research in Chapra, I always travelled with my in-depth case respondents or research assistants to ensure my personal security. Living with an elite family could have impacted negatively on my ability to gather reliable information from marginalized households, however, I took specifric actions to limit these impacts. Two of my former students at Dhaka University are also residents of Chapra and their parents were traditionally boatmen and fishermen. These students had already known me 50  for more than four years and were well aware of my previous advocacy for the rights of marginalized communities through my teaching, research, and activism. These contacts helped me obtain local community trust and I also took care to explain my research purposes carefully to all potential participants and to assure them of confidentiality. As much as possible, given my own background as an elite member of a rural community in another district of Bangladesh, I also followed a non-hierarchical approach when interacting with non-elite community members. My anthropological training assisted me with this but during my boyhood I also had the experience of sharing my living room, food, and transportation with neighbors from marginalized households when they were displaced because of extreme flooding. This personal background in addition to my professional training helped me to overcome the hierarchical relationships that might otherwise have limited my ability to establish a trust relationship with all socioeconomic groups within the Chapra community.  My ethnographic fieldwork at Chapra is able to account for only a small portion of the total effects of the Farakka Barrage and other government intervention projects. I was able to document the vulnerability of Chapra farmers to drought, flooding, river bank erosion and embankment failures. However, I did not attempt to document more distant effects such as salinity problems in the coastal region and water stagnation in Sathkhira and Jessore districts.  My focus on water experiences at the community level brings essential new knowledge forward but more work will be necessary to correlate the outcomes of this research with the activities of dominant external agents, like the Farakka Barrage authority, the World Bank and the central government in Bangladesh. The Ganges Basin flow is now controlled through a complex, hydro-political and technological network, with components that extend well beyond the scope of this project (Mehta et al. 2001:3). I need also point out that this study did not 51  involve the gathering of first-hand information about the many other dams, barrages, irrigation projects and hydroelectric dams built throughout the Ganges Basin in India. On the basis of a thorough study of the literature, however, in addition to my own study findings, I will address macro-level policy issues in my concluding chapter. Climate change is another dominating factor behind water flow variability in Bangladesh but I was not able to focus on this issue during my data collection. Currently, the central government in India exploits this issue to divert attention from their own actions in respect to the Farakka diversion and Ganges flow. The central government in Bangladesh also exploits the climate change issue to divert attention away from their policy failures and criticism of their top-down technology-dominated approach to water and agricultural management (Cosgrove and Rijberman 2000; Mollinga 2008:9). Again, the corporate elites exploit the issue of climate vulnerable people to promote their neoliberal economic agenda. I did not focus on any of the climate change factors during my data collection. I faced major difficulties obtaining local government cooperation. Local government officials are surrounded by political leaders who visit them regularly to monitor and control local development and safety net programs. In this context, local government staff seemed to me very busy when I visited them for my secondary data collection. At the central government, I visited the Ministry of Water Resources for data about water and agricultural policy documents, a challenging task, but I was able to use my personal networks to get an entry permit from the Ministry. I faced another level of bureaucratic complexity in this Ministry from security staff, front desk clerks and ministry officers when trying to access secondary data. I faced similar challenges when attempting to gain access to documents held by the Water Resource Planning Organization (WARPO) and Department of Agriculture Extension (DAE). Although I have 52  confidence in the accuracy of the secondary data I gathered, I cannot claim to have gained access to all significant documents.  These secondary data, in conjunction with the primary data, provide the foundation for my research argument: access to the Ganges River water at Chapra is essential to protect other socioeconomic rights like food, work and education. I use these data to explain current river flow conditions and their effects on community livelihoods at Chapra. For this purpose, I begin my analyses, in the following chapter, with a detailed account of the hydroecology and hydropolitics of the Ganges Dependent Area and the Chapra region.  53  Chapter Three THE GANGES DEPENDENT AREA IN BANGLADESH  In this chapter, I describe the basic hydrological and environmental characteristics of the Ganges Dependent Area (GDA) in which Chapra is located, noting how government interventions such as the Farakka Barrage and the Flood Action Plan have undermined the ability of the GDA to provide essential ecosystem services to the majority of its residents. Traditional agricultural livelihoods in this region are well adapted to seasonal variations in rainfall and to regular episodes of borsha (wet) and khora (dry), which produce a variety of freely available ecological services. These services include river fisheries, the seasonal availability of wild foods, animal forage, soil replenishment through siltation and construction materials, such as wood and bamboo. The political culture and centralized political structure of Bangladesh allows elites at the local, national and international levels to establish programs that give them control over natural resources that promote scientific and technological knowledge at the expense of local knowledge, and promote the privatization and commodification of water and agriculture. Instead of a predictable and manageable variation of rainfall during each season, farmers now face more extreme and less predictable episodes of bonna (extreme flooding) and drought. The data I present in this chapter demonstrates that the current government approach is not sustainable when evaluated in relationship to the ecosystems and services on which the majority of households depend for their livelihoods.    54  The Dependency of the Gorai Bank Communities on Ganges Hydro-ecology Bangladesh is a riverine country with more than ninety four percent of its river water originating from international rivers of India. The Ganges Dependent Area (GDA) occupies the entire southwest area of Bangladesh and is classified by government agencies as one of eight major hydrological zones in the country (Figure 3.1). The southwest hydrological zone is dependent on the water that originates from as far as 2600 kilometres away in the Himalayan Mountains and flows through the four countries of China, Nepal, India and Bangladesh before reaching the Bay of Bengal. The Ganges Basin as a whole has a catchment area of 33,520, 147,480, 860,000 and 46,300 square kilometers respectively within each of these countries (Ministry of Water Resources n.d.). In this study, I focus on the GDA but more specifically on the Gorai River, one outlet of the Ganges in the southwest zone, and on Chapra village in Kushtia, which is located alongside the Gorai.      Figure 3.1: The Major Hydrological Zones in Bangladesh (in square kilometers). Source:  Ministry of Water Resources 2001: 28.  South West, 26,100 South Central, 16236 North West, 31,607 North Central, 15,950 North East, 20,061 South East, 10,275 Eastern Hills, 19,910 Rivers & Estuaries,     8,431 55   The GDA in southwestern Bangladesh includes river floodplains, peat basins, estuaries and the mangrove dominated Sundarbans (Figure 2.1, page 29). The mangrove forests of the Sundarbans depend on the Ganges Basin freshwater flow from the Gorai River and provide support for a unique regional ecosystem. The Sundarbans are one of the largest mangrove regions of the world and are dominated by freshwater swamp forests. About 80 percent of Bangladeshi people, including those who live in the GDA, live in rural areas and are dependent on water and agricultural practices for their livelihoods. In communities like Chapra, the rainy season brings regular water flow locally called borsha from the Ganges Basin. The borsha or wet season transports siltation onto agricultural lands that is helpful for promoting soil fertility and agricultural production. Local farmers do not need to use chemical fertilizer because of this natural fertility source. This is the most productive time of the year for both crops and for other natural resources, like the water-borne wild vegetables, fisheries and vegetation that are also a major foundation of livelihood practices. Local people do not need to buy these from markets as they are freely available from wild sources. Before the borsha season, local communities get khora or a dry season that is helpful for producing different crops and receiving different wild-vegetables, fish and fruits. However, the Farakka Barrage in India is controlled in such a way that extra water is released during the rainy season which causes bonna or flooding and drives away a regular borsha season. Khora is also made much worse during the dry, pre-monsoon spring season when water is diverted for use in India and the downstream flow to Bangladesh is drastically reduced (see chapter 4 for a more detailed account of this issue). The national government of Bangladesh has historically failed to secure a proper share of the basin flow due to power differences with India; to compensate, they have developed a technology based water and agriculture management system which I describe in chapters four and five. 56  However, this technology-driven system has created further vulnerabilities for the marginalized basin communities at the southwest hydrological region in Bangladesh including Chapra. The Gorai River is the major distributary of the Ganges River in Bangladesh. This river originates in Kushtia District and provides ecosystem services like fresh water to the southwest hydrological region and has protected the Sundarbans mangrove forests from salinity intrusion for about 500 years (Islam and Gnauck 2001). The Gorai bank communities at Chapra are seven kilometers away from the Ganges River and are entirely dependent on the Ganges and Gorai Rivers for protecting agricultural practices and livelihoods.  The Gorai River flow area is 199 kilometers in length with a catchment area of 15,160 square kilometers that includes the districts of Pabna, Chuadanga, Kushtia, Rajbari, Faridpur, Gopalganji, Jessore, Jhenaidah, Magura, Narail, Perojpur, Barguna, Bagerhat, Khulna and Satkhira (Figure 2.1, page 29). Islam and Gnauck (2001) describe five different sections of the Gorai River: the Gorai-1 to Gorai-4 sections include the non-tidal watercourses and Gorai-5 is the tidal lower watercourse. The tidal and non-tidal sections of the Gorai River discharge into the Bay of Bengal via the Madhumati and Baleswar Rivers.  The Ganges Basin provides major water services for Chapra households including my case study respondents, Tanvir, Tofajjal, Joardar and Billal, who represent rich, intermediate, small, and landless laborers respectively. According to Joardar, the basin ecosystem provides “pukur vora mach, gola bhora dhan, goal bhora goru” (a pond full of fish, a household store room full of rice paddy and a domestic animal shelter full of animals). This basin flow provides ecosystem support to local water bodies like the Chapaigachi haor (oxbow lake), Chapra and Lahineepara beels (wetlands), Lahineepara and Shaota khals (canals) at Chapra. These waterbodies are naturally developed based on the Ganges-Gorai River flow. The borsha or rainy 57  season flow transports water from the basin to these water bodies and ensures a healthy environment for the whole year. Based on this flow, the better off farmers, Tofajjal and Tanvir, produce commercial agricultural crops and the marginalized farmers, Billal and Joardar, grow some of their own food and receive employment opportunities and ecosystem services. Parvin, wife of Billal, and Suma, wife of Joardar, use local water bodies for domestic activities. As noted previously, the basin flow is the prerequisite for basic livelihood activities like producing crops and catching fish but is also essential to a range of social activities such as sailing boats, visiting relatives, organizing water sports and providing resources for celebrations. Irrespective of their socioeconomic conditions, everybody at Chapra has equal rights to use the basin flow and natural resources without any discrimination or conflict. In Bangladesh, local communities enjoy historical rights to river water and resources without any fee or restriction. Chapra residents understand the seasonal dynamics of winter, summer and rainy seasons and their connection with the Ganges Basin flow. Based on these seasonal patterns, they practice the three major cropping patterns of kharif-1, kharif-2 and robi for agricultural production (see Table 3.1). Kharif-1 occurs during the period from March to May, from mid-spring to mid-summer. Kharif-2 occurs during the period from June to October, from the mid-summer rainy or monsoon season to autumn. The robi cropping season is the period from November to February, that is, from late autumn, through winter to early spring. The Chapra community cultivates, for example, onion crops during the winter season as a robi crop because this plant cannot be grown during the rainy season. Again, they cultivate aman paddy during the borsha season because this crop requires a higher level of water availability. They developed this system of crop scheduling over several generations and the knowledge it requires creates a deep bond among people, river, water and land.   58  Marginalized farmers have a deep understanding of the local ecosystem services and use this knowledge to ready crop lands for planting. They visit crop lands immediately after decreases in the borsha flow and evaluate the siltation level. They pick up a piece of soil and test it to understand moisture levels and readiness for starting crop production. They can ready cropland earlier or later by putting water hyacinth on the land. The water hyacinth is a local plant borne in water bodies and can be used to preserve cropland humidity for a longer time. They can also determine what type of crop production is suitable based on this soil type. They cultivate crops like pulse in muddy areas and start other crops based on land conditions.                 59  Table 3.1: Bengali Calendar, Seasons and Cropping Periods Months - Western Calendar Western Seasons Months - Bengali Calendar2 (dates are for the current year) Bengali Seasons Cropping periods January Winter Poush (Dec.17-Jan.15) Sheet  Robi February Magh  (Jan.16-Feb.13)   Kharif-1 March Spring Phalgun  (Feb.14-Mar.15) Basanta April Chaitra  (Mar.16-Apr.14) May Baishakh  (Apr. 15-May 15) Khora or Grisma     Kharif-2 June Summer Jyaistha  (May 16-Jun. 15) July Asadh  (Jun.16-Jul.17) Borsha  August Shraban  (Jul.18-Aug.17) September Fall Bhadra  (Aug.18-Sept.17) Sharat  October Ashwin  (Sept.18-Oct.18)  Robi November Kartik  (Oct.19-Nov.17) Hemanta December Winter Agrahayan  (Nov.18-Dec.16)   My focus group respondents, Billal, Rahim, Laily and Suman, explained to me how local land use patterns vary on the basis of cropland elevation. Mid-level lands are naturally developed croplands that differ from the higher and lower croplands. This land can produce single, double or triple harvests in a year depending on the basin flow. Based on the borsha seasonal flow, the higher land can also produce triple crops during kharif-1, kharif-2 and robi. The term vita is used to refer to the practice of mounding up soils to overcome challenges of water stagnation and                                                           2 The Bengali calendar was created during the reign of the Mughal Emperor, Akbar, in the sixteenth century. It was introduced as a replacement for the Islamic Hijra calendar which did not match the harvest seasons in Bengal and was therefore less useful for taxation purposes. The Bengali calendar combines lunar and solar elements and is considered to begin in mid-April with the Baishakh season. All Muslims and Hindu religious and cultural practices and holidays are incorporated into this calendar year. The year 2014 in the western calendar is counted as year 1421 in the Bengali calendar.   60  flooding. The term maacha is used to refer to another technique for agricultural production that involves construction of an infrastructure on higher elevation lands with bamboo sticks, rope and other local materials; vegetables like cucumbers or potatoes are then planted within this structure. My survey data indicated that 80 percent of the people at Chapra use these techniques to maximize their agricultural production.  The Ganges River is a major source of fertilizer for farmers at Chapra. According to Tofajjal, the regular borsha flow produces siltation and algae in croplands and develops sustainable environments. The borsha flow washes away exhausted top soil from crop fields. Again, the flow produces water hyacinth in local water bodies. Joardar deposits water hyacinths on croplands during the borsha season, as compost, to promote cropland fertility. Also, for almost eleven months in a year, Joardar and his wife, Suma, collect domestic animal dung and decompose it in a pit at his homestead. Joardar transfers this fertilizer onto his croplands immediately after the borsha season for the winter and summer seasons’ crop production. Tanvir, the land owner, hires Billal, a traditional day labourer to prepare this fertilizer and hires additional labourers to spread them on his crop fields. Moreover, they receive earth-worm generated fertilizers. This is a long worm locally called kecho that burrows into croplands at the beginning of rainy season; the burrowing mechanism improves cropland fertility.   My focus group respondents, Joardar, Suma and Tanvir, provided me with elaborate descriptions of how they produce their own seeds based on local ecosystems. As a marginalized farmer, Joardar scrutinizes a specific cropland’s elevation, temperature and soil quality for producing crops like rice, onion, ginger and turmeric. He does not need to spend much money to plant these crops since he is able to produce the seeds himself through careful application of his own local knowledge. The female household members like Suma cultivate local crops like bean 61  or watermelon. She also contributes by preserving some of the seeds from these crops. Joardar and Suma demonstrate deep ecological knowledge about temperature and humidity through their seed preservation practices. They preserve onion seeds in an earthenware jar or kolosh to ensure an ecologically friendly environment that allows no extreme hot or cold. Suma inspects temperature and seed condition every fifteen days. Sometimes, they exchange these seeds with trusted neighbors or relatives and develop community bonds. As a rich farmer, Tanvir supervises seed production and, Sohali, the wife of Tanvir, supervises seed preservation. Based on his supervision, day laborers harvest the seeds when they are ready. After bringing the seeds home, Sohali supervises her maid servants who preserve them in a suitable place following local knowledge.    Chapra community people husband domestic animals like cattle, hens, chickens, ducks, goats and sheep based on river-dependent ecosystem services resources. Suma and her daughters Amina, Joshna and Shifaly look after domestic animals like chickens and ducks. Suma puts them in crates every evening and brings them out every morning. They send ducks to the Chapra beels or wetlands every day. Joardar’s sons Jakir and Karim take care of cows and calves, feeding them local plants. Tanvir hires Billal to take care of his domestic animals. Moreover, Tanvir’s domestic help, Parvin, takes care of these animals under the supervision of Tanvir’s wife Sohali. During the borsha season, local people at Chapra use reserved paddy straw, water hyacinths, banana plants and bamboo leaves as fodder for bullocks or goats. Based on these efforts, they regularly get eggs and meat from chickens and ducks and produce young chicks every four months. They cultivate croplands with bullocks and get fertilizer, biogas and fuels from these domestic animals. The cows provide milk every day and produce calves every year. They sell extra domestic animals, eggs or milk, which helps to pay for school fees and household items. 62  Joardar gets bullocks from their domestic cows and successfully produces agricultural crops.  They use cattle bones, horns and hooves for making agricultural materials like plows, building houses and for healing purposes.   The focus group respondents, Billal, Parvin and Kabir, informed me that the occupational groups like boatmen, fishermen, blacksmiths, potters, thatchers and basket makers also depend on agricultural production. The boatmen transport agricultural goods and services, fishermen provide fish and blacksmiths make agricultural materials like hoes for farmers. The potters make household cooking materials and children’s toys. The thatcher builds homes for community members. The basket makers provide different types of baskets for community people. Parvin uses these baskets for gardening and collecting fuel woods. Much of this local knowledge transmits from one generation to the next. Children from these families begin learning this knowledge in early childhood when they see older generations’ occupational practices. Grandfathers tell many stories for teaching grandchildren knowledge about seasonal patterns, local cropping practices, agricultural production materials, housing and transportatation practices. According to Billal, young people would traditionally become knowledge experts themselves by the age of fourteen.  My focus group respondents, Rahim, Parvin, Suma and Soheli, specified that local natural resources are required to produce a variety of essential agricultural materials. They use bullock carts locally called gorur garee for transporting field crops during the summer season. This bullock cart is made with local natural resources like bamboo, wood and jute. During the rainy season, Joardar uses a gaab, a boat made with locally available natural resources like trees, bamboo and coating materials for transportation. When field crops are harvested, Suma contributes by paddy husking and winnowing. As a marginalized woman, Parvin works at 63  Tanvir’s house under Sohali’s supervision to winnow the rice plant, boil them and make rice. The winnowing fan is locally called kula and is made with bamboo. The paddy husking machine is locally called deki or ‘husking pedal.’ They make ploughshares, frames, ladders and sticks from freely available natural sources like bamboo and trees. They also make other agricultural materials like hoes, sickles and cleavers from the same natural resources.  Wild vegetables are a traditional natural resource in Chapra. Some of the wild vegetables are water lilies, marsh herbs, water spinaches, hyacinth beans and ferns available at the local water bodies like Chapra and Lahineepara beels. Parvin, wife of Billal, regularly collects water lilies and their fruits for family consumption. Sometimes, their children Jamal, Lablu, Kamrun and Shampa collect them from local water bodies. During times of food shortages, they collect and eat water lily fruits to overcome starvation. Wild banana trees are also readily available close to their homesteads, in forested areas and along the Gorai River banks. They also eat wild bananas from roadside trees and nobody complains about these practices. They collect midribs and inflorescences from the wild banana trees for household vegetable consumption and also the leaves and roots of arum plants. In addition to these vegetables, Billal gets other vegetables like jute leaves, onion flowers and pulse leaves from Tanvir’s crop fields.        The borsha season also provides more than sufficient fisheries in local water bodies like Chapra and Lahineepara beels. These fisheries are freely accessible to all and community members promote bonds by sharing them. For example, Billal’s father caught a nine-kilogram carp fish and distributed the major portion to neighbors. Billal informed me that this fish sharing develops community bonds and overcomes the risk of loss through rotting. This is a common practice for Chapra communities. Billal is welcome to catch fish in Tofajjal’s ponds like the other neighbors. Tanvir prepares ponds using a sheltering mechanism to promote different types 64  of fisheries. He puts water hyacinths at a side of the pond so that, for example, catfish and snakehead fish can find suitable habitat. Another side of the pond is covered with water lilies for fish like carp. The community people preserve local species alive, like catfish, after catching them from local water bodies. They keep these fish in earthenware jars. Suma takes care of these reserved fish and cooks some every week for household consumption. The Ganges Basin ecosystems are a major source of these fish, which are a core feature of socioeconomic foundations at Chapra.  However, these ecosystems are being profoundly disrupted by the Farakka Barrage diversion in India.               The Farakka Barrage The Ganges Basin countries have failed to develop effective basin management agreements and have thereby created major ecosystem failures and livelihood challenges for basin communities such as Chapra. The central Government in India was able to build the Farakka Barrage unilaterally in 1975 because of their hydropolitical dominance of the region (Turton and Henwood 2002; Zeitoun and Warner 2006). The barrage was constructed mainly in order to divert additional water into the Hoogly River, a branch of the Ganges that flows through Calcutta and empties in the Bay of Bengal. The additional water ensures that the port of Calcutta can remain open year round rather than suffer closures due to sedimentation. India has also built a small hydroelectric facility at the barrage and makes some of the stored water available for industrial and irrigation uses within West Bengal (Khan 1996; Iyer 1997:4; Swain 1996). Iyer (1997:4) notes that, “the primary purpose of the Farakka Barrage was the diversion of a part of the waters of Ganges to the Bhagirathi/Hooghly arm to arrest the deterioration of Calcutta Port. The secondary purpose was to protect Calcutta’s drinking and industrial water supplies from the 65  incursion of salinity.” The central government in Bangladesh was not able to prevent the construction of this barrage due to disparities of geographic size, economic capability and political and military strength.  The governments of India and Bangladesh have concluded two water treaties, one in 1977 and the other in 1996 and two Memoranda of Understanding, in 1983 and 1985. The goal of the most recent treaty, the Ganges Treaty of 1996, signed for a thirty-year period, is to enable water sharing of the Ganges River, especially during the dry season. However, the treaty does not address the specific ecological concerns reported above regarding borsha failures, bonna, drought, river bank erosion, water stagnation and embankment failure. And, despite its stated intent, the 1996 Ganges Treaty has also not resulted in positive outcomes for khora (dry season) flow, as I will explain below.  Ganges River flow data measured in Bangladesh at the Hardinge Bridge Station on the Ganges, seven kilometers distance from the mouth of the Gorai River and eleven kilometers distance from Chapra, indicates how the basin flow changes during the rainy and dry summer seasons (Table 3.2 and Figure 3.2). During the summer season the flow of water in the Ganges is reduced significantly due to less rainfall and flow reductions at its place of origin in the Himalaya Mountains. The Farakka Barrage authority diverts enough flow for India’s needs into the Hoogly River during the dry season but is able to limit the flow of water into the Hoogly during the wet season. This limits flooding in West Bengal but worsens it in Bangladesh. Table 3.2 shows that the average daily water flow into Bangladesh during the dry season has been reduced two-thirds since the Farakka Barrage began operating in 1975 (see also Figure 3.2). Currently, the Gorai River is at the verge of extinction due to this basin flow reduction.  66  Joardar and Billal argued that reductions to river flows during the borsha season are the most common cause of crop production and employment failures for Chapra. The borsha season occurs every year during the months of July, August and September, with water levels reaching their peak in August. I calculated the average flow from July 25 to September 10 every year for the last 50 years, from 1960 to 2010, to demonstrate historical changes of flow during this season. Table 3.2 data demonstrates that the minimum flows are significantly lower on average during the rainy season. This reduction causes failures of aman rice production, a staple crop on which households depend for the whole year and which cannot mature without sufficient water. This seasonal flow reduction also causes ecological service failures of the kinds described above.    Table 3.2: Ganges River Flow Data Before and After Construction of the Farakka Barrage   Flow rates at Hardinge Bridge in Bangladesh in cubic feet per second (cusecs) Year Khora (March-April) seasonal flow (cusecs) Borsha (Jul-Sept) seasonal flow (cusecs) Average Max Month Min Month Average Max Month Min Month 1960 2326 2480 Mar 07  2170 Apr 30 39794 48000 Sept 04 25800 Aug 02 1964 2320 2600 Mar 01 2180 Apr 15 39733 48300 Sep 10 29200 Jul 28 1970 2360 2640 Mar 11 2030 Apr-24 32917 40800 Aug 18 24400 Aug 30 1976 780 1130 Mar 01 657 Mar 29 33569 50000 Aug 31 19100 Jul 31 1980 927 962 Mar 02 874 Mar 30 48350 57800 Aug 22 37500 Aug 02 1985 823 1020 Mar 05 701 Apr 04 37373 48000 Aug 29 24200 Jul 25 1990 828 1030 Apr 29 698 Mar 01 40444 51000 Aug 21 23100 Sep 09 1995 509 769 Mar 01 363 Apr 26 35219 48800 Aug 19 18800 Jul 30 2001 766 997 Apr 27 456 Apr 21 34390 44004 Aug 30 22095 Aug 19 2006 828 1092 Apr 07 418 Apr 22 26271 35079 Aug 31 21887 Aug 15 2010 743 970 Mar 20 475 Mar 31 25190 40276 Sep 25 8701 Jul 05 Source: Bangladesh Water Development Board 2012  67   Figure 3.2: Average Ganges Flow Before and After the Farakka Barrage. Source: Bangladesh Water Development Board 2012.  The khora season occurs during the months of February, March, April and May every year. It has its own distinct growing season, which begins in January and ends in May; water scarcity begins in February, reaches a peak in April and ends in May. Based on the months of March and April, I calculated an average flow for the river every year for the khora season. This flow provides the basis for understanding water crises and their consequences for dry season agricultural production, employment and livelihood practices.  Table 3.2 indicates major discrepancies in the Ganges flow before and after construction of the Farakka Barrage. In 1960, the average flow was 2,326 and 39,794 cubic feet per second (cusecs) during the summer and winter seasons respectively. There was no major deviation of the flow for the next fourteen years from 1960 to 1974. In 1974, the average flow was 2,393 and 43,369 cusecs during the summer and rainy seasons respectively. However, the average flow dropped as low as 780 cusecs during the summer season in 1976, immediately after the Farakka Barrage began diverting water to India (Bangladesh Water Development Board 2012). This flow 050001000015000200002500030000350004000045000500001960 1964 1970 1976 1980 1985 1990 1995 2001 2006 20102326 2320 2360 780 927 823 828 509 766 828 743 39794 39733 32917 33569 48350 37373 40444 35219 34390 26271 25190 Khora (March-April Average) Bonna (July-August-September Average)68  reduction created a sudden water shock to agricultural production and employment in Bangladesh. The Ganges River flow has never returned to pre-Farakka levels and has, in fact, encountered increasing reductions over the last thirty-eight years. In 2010, the average basin flow was 743 and 25,190 cusecs per day during the summer and rainy seasons respectively (Table 3.2).  Based on the 1996 treaty, Bangladesh should receive more than 35,000 cusecs water in every alternate ten day period during the summer season depending on the basin flow conditions (Table 3.3). However, in the dry season, when flows are generally much lower than 70,000 cusecs, other provisions of the treaty are applied. Article-II (iii) informs us that in case of the river flow reduction below 50,000 cusecs in any ten-day period at the Farakka point, the governments of India and Bangladesh will work together on an emergency basis to ensure “equity, fair play and no harm” to either party (Government of Bangladesh 1996). However, when comparing the pre-Farakka flow rates to the most recent flow rates for the dry season, it is apparent that Bangladesh now receives only about one third of the former flow, suggesting that India diverts two thirds of the dry season flow for their own use. The major limitation to measuring the specific amount of the diversion is that the Joint River Commission fails to publish the Ganges Basin flow data at the Farakka Barrage point. The Indian members of the Commission do not cooperate with their Bangladeshi counterparts and do not provide this information to them or the government of Bangladesh.        69  Table 3.3: Annexure 1 of the 1996 Ganges Treaty for Sharing Ganges River Flow  Availability at Farakka  Share of India  Share of Bangladesh   70, 000 cusecs or less 50%  50% 70,000 cusecs - 75,000 cusecs Balance of flow 35,000 cusecs 75,000 cusecs or more 40,000 cusecs  Balance of flow  Source: Government of Bangladesh 1996   In addition to problems associated with the enforcement of its water sharing provisions, the treaty fails to describe guidelines for ecological concerns associated with borsha seasonal failures, flooding, river bank erosion and embankment failure (Bhattarai 2009; Brichieri-Colombi and Bradnock 2003:53; Gupta et al. 2005; McGregor 2000; Rahaman 2009). Article VIII focuses on the mechanisms of cooperation for different issues like flood management, irrigation, and hydropower based on the principles of “equitable” utilization and “no harm” that are described in articles IX and X (Government of Bangladesh 1996). Article IX extends these principles to other international rivers shared by India and Bangladesh, however, to date, none of the many river disputes between the countries have been resolved, including that over the Teesta River, which has been a major source of contention (Padmanabhan 2014; PTI 2014). The government of India has also violated the principles of the Ganges River Treaty by unilaterally creating their National River Linking Project (NRLP) that involves massive water transfer from the Brahmaputra to the Ganges Basin to service industrial growth in drought prone areas in India. The implementation of NRLP is likely to bring about environmental disasters in the rest of Bangladesh similar to those currently being encountered in the southwest (Bandyopadhyay and Ghosh 2009:54; Bhattarai 2009:4; Haftendorn 2000:64; Khalequzzaman 1994; Swain 1996). Despite the existence of the Treaty, the basin flow continues to diminish in Bangladesh. Shortly after the treaty was signed in 1996, the average flow dimished from 828 cusecs in 1990 70  to 766 cusecs in 2001 during the dry season (Table 3.2). In 1990, prior to signing the treaty, the average flow during the khora season was 828 cusecs, down from a slightly higher level in 1988 of 1,011 cusecs. The average flow fluctuated between 1,058 and 866 cusecs from 1998 to 1999 but by 2001 it had diminished to 766 cusecs. In 2005, the average flow remained low at 782 cusecs and then declined further in 2010 to 743 cusecs. Rainy seasonal flows since the signing of the treaty in 1996 also show a consistent reduction. In 1995, the basin flow was 35,219 cusecs during the rainy season but this diminished to a historic low of 25,190 cusecs in 2010 (Table 3.2). The 1996 Ganges Treaty has clearly failed to achieve positive outcomes for Ganges water sharing between India and Bangladesh.  Flow records for the borsha season since 1960 also indicate an increased incidence of bonna or severe flooding. Records for the 1960 to 1970 period indicate that maximum flows ranged from 40,800 to 48,000 cusecs. However, in 1976, 1980 and 1990, the maximum flows ranged from 50,000 to 57,800 cusecs. The unpredictability of water flows is as damaging for basin communities as either floods or shortages since it makes it impossible to effectively plan ahead.       The unpredictability of water flows also has direct environmental impacts. In 1982, the summer season average flow was 1,426 cusecs whereas this flow diminished to a low of 845 cusecs the very next year, in 1983 (Bangladesh Water Development Board 2012). The average flow reached a low of 558 cusecs in 1989 and increased to 828 in 1990. Again, this average flow shrank to 358 cusecs in 1993 and increased to 1,058 cusecs in 1998 (Bangladesh Water Development Board 2012). These flow fluctuations generate sedimentation throughout the basin and in local water bodies as water loses its capacity to carry sedimentation downstream and distribute it in a predictable manner. On the other hand, variations in the rainy seasonal flow, 71  especially those associated with extreme floods, cause severe river bank erosion and embankment failures. Most Chapra households have no means to overcome or mitigate these concerns and face major damage and loss of crop production, household infrastructure and agricultural materials.  While a certain range of fluctuation was normal before the construction of the Farakka diversion, the Farakka operation has magnified these fluctuations to a destructive and unmanageable level. Sometimes, the length of the khora season is shorter or longer than the regular season due to domination of the Farakka diversion. Sometimes, the khora season begins in December or March and ends in March or July. For example, the khora season began on 31 March in 2009 with the flow of 602 cusecs and ended on 25 May with the flow of 948 cusecs. In 1979, the khora season began in January and continued about six months till June. The basin flow was 1,520 and 1,090 cusecs on January 29 and June 18 respectively in 1979, which caused major changes to the summer season. In 2009, the khora season began in January and ended in June (Bangladesh Water Development Board 2012). Again, this fluctuation occurs during the borsha season as well as the khora season. Currently, there is no regular borsha season and it gets longer or shorter depending on the Farakka diversion from the Ganges Basin flow.   These ecosystem failures create major challenges for community sustainability in the GDA.  They also pose significant challenges to the political and institutional culture of Bangladesh as it seeks to mitigate these problems while pursuing its own development agenda. As I explain below, the political culture of Bangladesh and the policy framework within which they attempt to manage water resources, tends to replicate and exacerbate, rather than mitigate, the problems created by India through construction of the Farakka Barrage.  72  Political Culture  The Peoples’ Republic of Bangladesh is a parliamentary democracy similar in structure to many other British Commonwealth nations. Unlike Canada and India, however, it is not divided into Provinces or States that each elects their own legislative assemblies. It is divided into seven administrative divisions and those divisions are further sub-divided into 64 districts. Local elections are held to populate some of the positions on sub-district level (upazila) councils and the union councils that represent a group of villages, but other members of these councils are appointed by the central government. There are 64 District Councils, 483 Upazila Parishads and 4500 Union Councils in Bangladesh (Bangladesh Bureau of Statistics 2010). Below the union level there is also a ward level with nine wards in very Union Council. Every ward in turn is made up of 2 to 4 villages. There are no offical government institutions at ward and village levels but villages especially organize a large number of events through traditional informal mechanisms. They organize religious events, sporting events and various kinds of meetings to discuss events that might require collective action such as an erosion problem or asisstance to a neighbor in need.  Since the authority of upazila, union and village councils is very limited by comparison to that of the central government, and since no formal government exists below the union level, the system is structurally one that allows for a great deal of top-down control. The political culture that has emerged in Bangladesh since independence in 1971 is partly an outcome of the structural characteristics of this system as well as the other historical and cultural factors identified in chapter one.   Members of Parliament (MPs) are given appointments to the different ministries like water resources, agriculture and finance depending on their personal relationship with the Prime 73  Minister. Senior bureaucrats within each ministry are also appointed by the Prime Minister, from the governing political party. Ministers will appoint less senior bureaucrats and they will also be appointed on the basis of their service within the governing party, and their personal relationship to the Minister. The Water Resource Minister thus heads a bureaucratic organization that extends from the central government to the sub-district level and is dominated at all levels by members of the governing party who only retain their positions for as long as they demonstrate loyalty to those above them. Government Ministers thus tend to be insulated within a closed circle of political leaders at national and local levels who mobilize power structures based on their own agendas. These leaders constitute a political-economic elite with very little accountability to the grassroots interests of local communities like Chapra.  MPs, in addition to having the opportunity to be appointed to a ministerial position, may also be nominated for an advisory position at a local resource management committee at the district and sub-district levels. Based on this official position, they control local power structures based on their political party organization. They nominate party candidates for local government elections at Upazila Parishad (UP) or Union Council (UC) levels. These local elections run independently and are held at different times of the year than national elections.  The Zila Parishads or District Councils are composed of both elected and appointed members. The elected members include all the MPs from that district, the Upazila Parishad Chairmen, the Union Council Chairmen and the Mayors of all municipalities within the district. The Zila Parshad has equal representation from both urban and rural areas. Urban municipalities are governed by mayors and councils while rural areas are governed by Upazila Parishads and Union Councils. The Chair of the Zila Parishad is appointed by the national government based on political loyalty and the parishad also appoints additional members from the local area, 74  usually members of the local elite. The national government also appoints a Deputy Commissioner, selected from the Bangladesh civil service, to manage the administrative work of each Zila Parishad. Members of the civil service cannot belong to a political party but are chosen on the basis of their loyalty to the ruling political ideology as well as their formal qualifications.  Upazila Parishads of Sub-district Councils consist of one Chair and two Vice-Chairs who are elected directly to their positions, and all the Union Council Chairs within that sub-district. The Union Councils consist of one Chair and twelve other members who are also elected directly by local citizens. Generally these positions are all filled by elite members of the local communities.  The government party is thus able to control the actions of District Councils by appointing only party supporters to the non-elected positions. This is not the case at the sub-district and union council level but even when a majority of the members of those councils are from opposition parties, they tend not to directly oppose the government over local development activities but rather cooperate as best they can, protecting their class position while waiting for the next national election. All of the major political parties like the Bangladesh Nationalist Party (BNP) and the Awami League (AL), who currently form the government, have party organizations from the national to the village levels. These parties have top-down hierarchies and the Party Chairperson (the Prime Minister) dominates the system. Those closest to the Party Chairperson are appointed to the party’s Executive Committee (Ahmed 2013). These executive committee members are political leaders, businessmen, and former bureaucrats and military officers. Every Executive Committee Member has control over a specific geographic area like a division or districts in Bangladesh. This Executive Member dominates their party organizations and election candidate 75  nominations at upazila and union levels. In many cases, this member dominates nominations of Members of Parliament candidate within the geographic area. After a parliament election, the Members of Parliament under the leadership of this Executive Member work as a powerful group inside the party or government.   In addition to their extraordinary power over local levels of government, the central government MPs and Ministers are major power brokers for the foreign development agencies and corporations that introduce new technologies into the water and agricultural management system. Billal as a focus group respondent described these political leaders as joke (blood sucking worms), based on their exploitative activities. These political leaders develop close relationships with military leaders, as well as with civil bureaucrats, professional groups, financiers and fund raisers to promote their own business interests. During elections they behave like business contractors trying to gain control of a particular geographic region or government institution. As one focus group respondent put it: “voter pore amder keo mone rakhena” (no elected official like the MP or Upazila Parishad Chairman recognizes our voices after the election). Mallet (2014) in a Financial Time article termed these practices of politicians in Bangladesh “as a way of seizing (government) control and making money.”  Focus group participants Sabbir and Tarun stated that most of these political leaders are urban elites and often expatriate Bangladeshis, living only occasionally in Bangladesh. According to them, these elites are called bosenter kokil, or ‘spring birds,’ who visit local communities for their own interests. These elites are able to profit by exploiting the Ganges flow reductions, as will be explained in more depth in later chapters. Some, for instance, make money by mining sand from the river at times of low flow and selling the sand to construction industries. The perceptions of my respondents are supported by the fact that more than fifty 76  percent of the MPs in 2009-13 were businessmen (Ahmed 2013). Focus group respondents reported that many MPs do not hesitate to kill local people who raise voices against their interests. They stated, for example, that one MP’s son, who is only 18 years old, has killed 12 people to fulfill his family’s personal interests. Many of the MPs work as lobbyists and others perform official duties in formulating water and agriculture policies. MPs and senior bureaucrats get opportunities like foreign tours, training and contracts based on this technological development. Many bureaucrats get assignments as directors to implement large scale water management projects. Furthermore, government staff get pleasure trip opportunities abroad informally from businessmen and contractors.      Water Policies in the GDA  Consistent with the highly centralized nature of the Bangladesh government, water policy within the GDA is developed on the basis of a National Water Management Plan (NWMP) and a set of national policy guidelines originally developed in 1999.  These guidelines incorporate many of the same principles as the Flood Action Plans of the 1990s which were developed in coordination with the external agencies, including the World Bank, United States Agency for International Development (USAID), and United Nations Development Program (UNDP), who provided most of the funds necessary to secure what they promoted as a “permanent” solution to flooding problems in Bangladesh (Custers 1993). This approach is based on the idea that all water management problems can be solved by large scale engineering projects in combination with neoliberal economic policies that favor the privatization and commodification of water resources and water-dependent economic sectors such as agriculture (Faber and McCarthy 2003; Nuruzzaman 2007). In Bangladesh, as in neighboring India, these projects are often grandiose in 77  scale and work in opposition to local knowledge and traditional systems of agricultural production.  In Chapter 4, I describe the outcomes of the Gorai River Restoration Project and the Ganges-Kododak Project, two large engineering projects within the GDA but, in this chapter I wish to focus on the government’s approach to policy development, emphasizing the scale and centralized, top-down nature of their planning. Given the negative effects of the Farakka diversion, the national government was especially interested in the GDA area and as a result the Bangladesh Water Development Board (BWDB), following the NWMP guidelines, executed a feasibility study for an integrated water development project for the region.  As a result of this study BWDB has formulated three major projects (Ministry of Water Resources 2001:306). Firstly, the BWDB has proposed a major dredging project immediately downstream from the Hardinge Bridge on the Ganges River to remove sedimentation and thereby increase the flow of water in northwestern rivers like the Boral and southwestern rivers like Mathabanga that depend on Ganges water. Secondly, they are planning to construct a major barrage on the Ganges River upstream from the headwaters of the Gorai River. This barrage will allow for the storage of water during the borsha season for use later during the winter and spring seasons. Thirdly, in addition to the barrage, they are planning the construction of a headwater structure at the head of the Gorai River. The major purpose of this headwater structure is to control flooding in the GDA during the rainy season and secure water supplies during the khora season. The headwater structure will be 1,870 meters wide with 84 radial gates, each 18 meters wide that will allow fish passage and controlled boat navigation (Ministry of Water Resources 2001:306). Based on the Ganges Barrage, BWDB has also proposed the construction of two linking channels (channels 78  numbered one and four) that will move water by gravity flow from the Ganges River and the upper reaches of the Gorai River to points further south.   Based on the proposed barrage, link channel one can move water from the Ganges River at a point immediately downstream from the Hardinge Bridge, parallel to the Gorai River west bank, to connect to the Hisna and Mathabanga Rivers. This channel will traverse the existing GK Project infrastructure (described in chapter four) but will be designed so as to avoid damage to it. This channel will move water all the way to the Bay of Bengal by connecting to the Nabaganga-Chitra and Bhairab-Kobadak-Betna Rivers and therefore, will reduce salinity problems caused by coastal saltwater intrusion (Ministry of Water Resources 2001:325). The central government also estimates that 7,324 square kilometers of agricultural land can be irrigated from this channel. Link Channel 4 will require construction of a barrage on the Ganges River at Tangorbari or Pangsha, on the east side of the Gorai River. This channel 4 will divert the basin flow to the Chandana River and Madaripur Beel. The government estimates that 3,596 square kilometers of net agricultural area will receive irrigation services from this channel.  In order to implement this ambitious set of goals the government also proposes a significant reorganization of the bureaucracy necessary to its planning, construction and operation. BWDB will construct the barrage, headwater structures and link channels, while local governments will build irrigation drainage canals. NGOs will take care of land acquisition and resettlement activities based on the World Bank’s Poverty Reduction Strategy Paper. The Ministry of Local Government and Rural Development established a Central Training Unit for local water managers and the staff of the Local Government Engineering Department and the Department of Public Health and Engineering. The Central Training Unit also trains the central government staff of the Water Resources Planning Organization, the Department of 79  Environment, the Disaster Management Bureau and the BWDB (Water Resources Planning Organization 2001). The plan of action described above also involves an expansion of the responsibilities of local water management institutions at the upazila (sub-district) level. Local institutions currently have authority over areas of 1000 ha but their mandates will be expanded to 5000 hectares. After completion of the projects, the central government will assign local water management responsibilities to community organizations or individual pump owners (Ministry of Water Resources 2001), thereby promoting private sector involvements in water resource management at union and village levels.   The history of large scale water management projects in the GDA suggests that the outcomes of this particular project will be very different and much less positive than those envisioned by project planners. In the next chapter I describe the outcomes of two previous, on-going projects, both of which have significantly transformed the ecological and hydrological characteristics of the GDA. The Gorai River Restoration Project and the Ganges-Kobadak Project have both created much greater hardship for the majority of agricultural households in Chapra, but have helped elite interests to consolidate and extend their control over water and agricultural resources.    80  Chapter Four THE GORAI RIVER RESTORATION AND THE GANGES-KOBODAK PROJECTS   The government of Bangladesh has attempted to mitigate the negative effects of the Farakka Barrage by implementing projects intended to distribute the available water in more efficient ways. As documented in chapter three, immediately after the construction of this barrage in India, the flow of water during the dry season lessened by 70 percent in Bangladesh and never returned to its previous level. The flow of water during the rainy season also became much more unpredictable with significantly less water most years but with extreme flooding in others. These effects have been magnified in the case of the Gorai River with disastrous consequences for local farmers. In this chapter, I describe two major government projects, the Gorai River Restoration Project and the Ganges-Kobodak Project, both examples of “high modernism” as that term is used by Scott (1998). According to Scott, high modernism as an ideology creates legitimacy for scientific technologies, justifying, for instance, the construction of massive water infrastructures and centralized transportation and communication systems but without consideration for the voices of the poor and marginalized. In Bangladesh, I argue that both of the projects failed to take local ecosystems into account and, therefore, they have worsened rather than improved the living conditions of the majority of farming households. I attribute these failures to the top-down approach of the central government, which gives local elites full control of the projects while excluding the majority of farming households from meaningful engagement in decision-making processes.    81  The Gorai River Restoration Project The Bangladesh Water Development Board (BWDB) implemented the Gorai River Restoration Project (GRRP) in 1998, with the assistance of the World Bank, to help restore normal flows during both borsha and khora seasons by dredging the river channel. The reduced flow of water during the khora season causes an increase in sedimentation in the Gorai River (Islam et al. 2001; Mirza 1998). This sedimentation is called charlands when it acquires relatively permanent form as extended shoreline or islands in the river (Chowdhury 1984; Lahiri-Dutt and Samanta 2013). The increasing level of charlands further reduces river flow and causes ecological service failures. Ninety-one percent of the households I surveyed at Chapra, reported major challenges for practicing traditional agricultural activities due to failures of water supply and ecological services from the Gorai River and local water bodies. These water bodies include Chapaigachi oxbow lakes, Shinda and Shaota canals and the Lahineepara and Chapra wetlands, which were normally replenished each year by the Gorai River during the borsha season but sedimentation and erosion patterns now interfere with this process.  The GRRP is a major example of corporate as well as governmental control over local water resource management in Bangladesh. It was implemented as part of the World Bank’s Flood Action Plan (Boyce 1990; Paul 1995; Thompson and Sultana 1996) and Integrated Water Resource Management program in Bangladesh (Ahmad 2003; Brammer 1990; CEGIS 2003; Gupta et al. 2005).  Phase one began in 1998 and continued until 2009. The dredging started at the point where the Gorai River branches off from the Ganges and continued 20 kilometers downstream, including the portion of the river that flows past Chapra (de Groot and Pieter 2001). Four foreign companies carried out this dredging at a total cost of $160 million (Khan 2012). Costs were covered by several external donors, including the Dutch government and the World 82  Bank (Ministry of Water Resources 2001; World Bank 1998). Phase two of the project began in 2010 and continued to 2013 with a target area of 36 kilometers from Kushtia city to Kumarkhali Sub-district. Phase two overlapped somewhat with phase one since new sedimentation and charland emerged in some phase one areas after only few months due to the continous basin flow reduction. The China Harbour Engineering Company (CHEC) executed the project under the supervision of BWDB. The total cost of the project was $220 million; $180 million was provided by the International Development Association and 40 million by the national government of Bangladesh (World Bank 1998).  Implementation of the GRRP required the creation of a complex bureaucratic network. The Ministry of Water Resources (MoWR) coordinates and monitors the GRRP and, under the direction of the Prime Minister of Bangladesh, it established a GRRP national implementation committee. The committee is comprised of a Project Director from the Bangladesh Water Development Board (BWDB) and some other members from the departments of fisheries, agriculture, environment, local government, inland navigation, planning, finance, economic relations and the Joint River Commission. Some local government departments at the upazila and sub-district levels, such as the Local Governments and Engineering Department (LGED), in coordination with Non-Government Organizations and private sector representatives, executed some tasks for local water bodies. Theoretically, based on the GRRP guidelines, the LGED is responsible, for instance, to restore linkages between the Gorai River and local water bodies in order to foster connections between water, crop lands and communities. Renovations of these water bodies are extremely important for restoring connection between croplands and the Ganges River. In addition to these LGED activities, the MoWR also formed a resettlement committee in coordination with NGOs to resettle the project affected people. NGOs get involved in this project 83  to support the Poverty Reduction Strategy Paper and Millennium Development Goals (International Monetary Fund 2013; Matin and Taher 2001). Most of the project, however, is carried out by engineering corporations under contract to one or another government agency. In this way, the central government is able to support corporate control as well as top-down management of local resources.  The corporate and foreign aid components of the GRRP are closely linked and as I will argue in the sections that follow, the foreign aid development approach is best understood as “a form of eco-imperialism and as a ‘debt trap’” as articulated by Wood (1984:703), and as perpetuating a vicious cycle of underdevelopment (Frank 1998; Ludden 2001:207; Prebisch 1963; Toye 1987:5; Wallerstein 2004: 17-8). The central government imports these foriegn dredging technologies based on economic liberalization policies (Baechler 1998:27; International Monetary Fund 2013; Nuruzzaman 2004) but, rather than realizing economic or social benefits, local populations experience mainly negative outcomes.  Elite Domination at Chapra The GRRP is a perfect example of the ways in which the political culture of Bangladesh and its top-down political processes tend to undermine local water management systems and promote corporate and elite control over local natural resources. The project was designed and implemented by the World Bank and the national government and, even when local government is involved, it is dominated by local elites. Local governments were responsible, for instance, to renovate local water bodies and connect them with the dredged area. Local governments were also responsible for ensuring proper working conditions and ensuring the security of dredging machines and staff. Many of the staff are foreigners and know nothing about local culture and tradition. They need a secure environment, food and lodging arrangements from local 84  governments and, in order to service these needs, the dredging company works closely with local government leaders but not with the communities those leaders are supposed to represent. Local political leaders are thus able to direct the project in ways that favor their interests but do not necessarily serve the interests of other community members.   Focus group respondents, Fajal, Keramot, Laily and Harun, stated that the local government officials make businesses from the GRRP development activities. In some cases the central government’s elites, like the Prime Minister, the Water Resource Minister and Secretary, the GRRP Director and their close circle of political allies, also gain personal economic benefits from this project. Many of them appear to gain these benefits by means of illegal activities. One major example is the Padma Bridge corruption allegation of the World Bank (T. J. 2013). The World Bank alleged that high level government staff and their relatives had accepted bribes from the international construction companies in exchange for the contract order. During the 2009 to 2013 project period, the illegal incomes of a nephew of the current Prime Minister and the National Parliament’s Deputy Speaker were reported to have increased by 32,985 and 4,435 percent respectively, based on bribes from national development agencies (Prothom Alo 2013). After the business interests of the central government are fulfilled, local governments and their associates in the local power structure exploit the project to fulfill their personal interests (e.g. Daily Star 2013a). Local elites are the rich farmers and many of them hold local government positions such as chair or member of the Union Council (UC) or Upazila Parishad (UP). They control these local power structures and political party organizations at the district, sub-district, union and village levels. Based on these power structures, they control the GRRP implementation activities. Laily termed these activities as “jor jar mulluk tar” (powerful people 85  control a constituency) which is helpful for protecting elite interests and for excluding marginalized community interests.  Elite domination causes major challenges for ensuring transparency and accountability. This domination creates an informal nexus between the UC and UP officials, bureaucrats and rich farmers. Focus group respondent Sabbir informed me the official government plan stipulates that dredging should occur at a specific place but local elites often divert the dredging activities to another site. They do so in order to protect their own properties including cropland, fallow land and household infrastructures from river bank erosion and embankment failures. Other focus group respondents, Harun, Rahim, Fajal and Jobbar, argued that local elites deposit the GRRP dredged sand in places that make it easy to sell for commercial purposes. Dredged sand is very valuable in the construction industry where it is used to construct new buildings, roads, houses and bridges. Furthermore, they develop commercial activities like fisheries projects in the dredging areas.   The GRRP illustrates a major gap between the government water management systems and local community needs and desires. According to the stated goals of the GRRP, the central government wants to develop flood control and facilitate drainage during the rainy season and secure enough water during the summer that freshwater can flow to more southern points in the GDA including Khulna city, Mongla port and the Sundarbans. On the other hand, community members like Billal and Joardar want natural resources back, like siltation, algae, earth-worms and water hyacinth, to promote cropland fertility. They also want to get back water-borne wild vegetables like marsh herb, water lily, ferns and hyacinth bean and natural fisheries like carp, barb and minnow in local water bodies. They want to be able to make agricultural materials like ploughshares, frames, ladders and sticks natural resources like bamboo, cane and wood which is 86  now in short supply. Natural resources like local wild vegetation are the major sources for nurturing domestic animals like cattle, chickens, ducks and goats. However, the GRRP has not restored any of these community resources. My focus group respondents, Kusum, Rahim, Bimal and Faruk, mentioned that the GRRP causes further challenges for existing usages. Chapra residents cannot take baths or satisfy their domestic water needs from the Gorai River. Kusum pointed out that it is difficult to predict the depth of the river and this leads to drowning deaths. During one visit to the region, on April 22, 2012, I and my PhD supervisor, Dr. John Wagner, saw a dead body floating in the Gorai River. Local people later informed me that this person had drowned when he was taking a bath. Two students of Kushtia Medical College died and three others went missing when they took baths in a dredged area of the river on 27 July 2012 (New Age 2012a).  Tofajjal and Joardar argued that the GRRP project planners did not properly assess the river flow patterns and did not listen to those in the community who could have helped them understand those patterns. Focus group respondents, Alamgir, Bakar, Borkat and Mofiz, stated that in one instance they explained to the contractor that the dredged sand should be deposited on the right bank of the river when he was depositing it on the left bank. Their logic was that the sand deposition on the right side could make the river bank stronger and reduce erosion and thereby protect buildings and cropland. On the other hand, the left bank is mostly charlands and does not require additional fortification. When the dredged sand is deposited on the left bank, it gets stronger and diverts the water more forcefully to the right side where erosion is already threatening buildings and fields.  Focus group respondents, Joardar, Billal, Faruk and Suman, expressed feelings of intense frustration with these sand deposition issues because of their increasing concerns about the 87  possible loss of household assets and cropland. They have shared these concerns with neighbors and submitted complaints to local government officials at the union and upazila levels. However, they are not hopeful about getting proper recognition of their voices from the government because they are not a part of the local power structures. Thus, they communicate with Tanvir, a wealthy landowner, who is a local leader of the ruling government. According to Joardar, Tanvir did not place enough importance on their concerns. Tanvir agreed with them about the negative effects of GRRP activities. However, he argued that he did not have enough power to change the GRRP activities. Joardar explained Tanvir’s intention differently. According to him, Tanvir is successful in gaining profits from the dredged sand. He did not want to lose these profits by raising local voices against the project activities.    Finding no other alternatives, Joardar, Alamgir, Harun and Kusum, demonstrated against the GRRP in front of the local government offices at the Kumarkhali Sub-district and Kushtia District levels. They composed different slogans against the ruling government due to their failures in protecting their local infrastructures and cropland. Because of this protest, the local governments face the central government’s dissatisfaction. The officials misinterpreted the protestors as opposition political party supporters who are unwilling to acknowledge the government’s success in local water resource development. Therefore, local governments in coordination with local elites, like Tanvir, find ways to resist the GRRP protestors. Local governments impose sanctions against demonstrations, protests and even meetings at outdoor locations. According to many of my informants, they also arrest many protestors and perform extra-judicial killings of protest leaders.  Local elites will also create financial hardships for protestors and threaten them physically. They raise demonstrators’ debt issues with local banks and NGOs and encourage 88  them to demand repayment of loans in full. Many of the local elites cancel sharecropping arrangements with people known to be among the protestors. They also have their supporters and hired musclemen to threaten protestors with killings, kidnappings and robberies. Rahim, a focus group respondent, informed me that the musclemen do not hesitate to kill the front line demonstrators almost every year at Chapra. The governments support these illegal activities of local elites with every possible mechanism. Consequently, the GRRP negative effects are increasing survival concerns among the marginalized households at Chapra.   Sedimentation and Charland Domination  Despite the GRRP implementation, sedimentation and charlands are increasing concerns for Chapra residents. The head of the Gorai River, where it branches off from the Ganges, tends to become blocked by sedimentation as a result of the reduced flow in the Ganges River. Joardar informed me that charland creation had not been a dominant issue before the Farakka Barrage was constructed. Previously sedimentation was mainly considered helpful for cropland fertility and agricultural production. However, sedimentation has now become a curse because of the effect of charlands on river bank erosion.  Focus group respondents, Jobbar, Morjina and Mofiz, informed me that the increasing charland creates major challenges for agricultural production and traditional employment practices. Jalil and Raju emphasized the greatly increased scale of charlands production since construction of the barrage.  Charlands now inhibit the river throughout its course and the river runs in different channels depending on the level of water. The Gorai River has been almost covered with these charlands, both in the middle of the river and along its banks, and the GRRP has failed to remove the majority of them. These charlands are not suitable for crop production because there is no 89  stability in soil structure and water flow due to the irregular seasonal flows, storms or flooding.  They are subject to encroachment from local elites and governments, however, which further deteriorates local ecosystems. Many local elites control the Gorai bank lands and build development infrastructures on them. They build fisheries projects, housing infrastructure, factories and business farms along the river banks. Local government agencies such as the Local Government and Engineering Department (LGED) worsen the problem by constructing roads and railways that block adjacent water bodies such as the Kaliganga River at Chapra. This river originates in the Gorai River and connects with local wetlands like the Chapra Beel. Construction of infrastructure for the GK project, which is described below, has also blocked many local water bodies including the Kaligangya, Sagarkhali and Dakua Rivers which creates major water stagnation problems at Chapra. Sixty-two percent of my household survey respondents stated that they have encountered major water stagnation problems because of these development infrastructures. The stagnant water contributes to agricultural production failures, environmental pollution and water borne diseases. Excessive rainfall or river flow increases the risk of these vulnerabilities. Joardar argued that this experience is different from the past. The past experiences were that the river flow and rainfall meant greater natural resources like siltation, fisheries and water borne wild vegetables.   River Bank Erosion and Embankment Failures  GRRP dredging has created more environmental vulnerabilities like extreme flooding or river bank erosion rather than resolving the existing problems. My case study respondent Joardar informed me that “bonna, khora, nodi bhangong, ebong jolaboddota holo Allahor gojob, manuj jokhon prokriteke niyontron korar chesta kore prokrite ei gojob dei” (when the outsiders 90  dominate nature with technologies, nature gets angry and takes revenge by creating multiple flooding, drought, river bank erosion and water stagnation). In addition to the in-depth case respondents, several of my focus group  respondents, Borkat, Fazlu, Sabbir and Kusum, also argued that the Farakka Barrage was the beginning of this natural punishment that is now made worse by the GRRP dredging, causing destruction of local infrastructures like houses, croplands and water bodies. The GRRP dredged sand is supposed to be deposited on the river bank so that the banks get stronger and help to reduce flood concerns. The embankments that have been built actually fail to control flooding; rather, they have increased the occurrence of sudden embankment failure and water stagnation.   According to Billal, “odhik baads maane odhik moron faad” (the increasing embankments mean increasing death traps). In the minds of many local farmers, the increasing technological domination over the Ganges Basin ecosystems causes nature to become exhausted and increases the amount and scale of vulnerabilities. My research clearly documents the fact that the Flood Action Plan and Integrated Water Resource Management in the GDA water management area have not restored the Gorai River flow and associated environmental characteristics and has not assisted agricultural productivity for most households. The central government established the GRRP to flush out sedimentation and charlands from the Gorai River but, according to my informants, the project increased flooding, drought, river bank erosion, water stagnation and embankment failures. The construction of embankments to reduce river bank erosion has simply displaced erosion effects to other locations. Focus group participant Tarun points out that, new embankments divide and block local water bodies like the Kaliganga River at Chapra, a distributary of the Gorai, and create water stagnation and environmental pollution.      91  My focus group respondents also argued that the river bank erosion due to the GRRP is creating major, additional livelihood vulnerabilities. Joardar argued that this project destabilizes the ecosystem relationships between the Gorai River flows, river bank, vegetation and cropland areas at Chapra. Focus group respondents Keramot, Jobbar, Suman and Laily argued that, previously, the Gorai River banks were covered by wild vegetation, forest land, bushes and trees and stable soil textures and this supported stability of local housing patterns, water bodies and cropland use. Keramot and Jobbar argued that the larger scales of dredging produce larger embankment failures and greater areas of river bank erosion. These erosion and embankment failures increase with seasonal flooding or rainfalls that wash away the GRRP dredged sand. My household survey data identified that 76 percent of people at Chapra encountered river bank erosion in their lifetime. During my participant observation in 2011-12, I saw many rural roads, houses, trees and gardens partially inside the Gorai River due to this erosion. Forty-four percent of the household survey respondents reported encountering this erosion more than three times in their lives. In Bangladesh as a whole, the total number of affected people because of this river bank erosion is twenty million who encountered losses of over $462 billion over the last 36 years. Among the affected people, in south-western Bangladesh more than ten million are now homeless (Prothom Alo 2014). The current top-down water management programs are a major reason for these disastrous consequences. The GRRP destabilizes the river bank due to the continuous flow reduction caused by the Farakka diversion. This project can only achieve success if the Gorai River gets the regular water flow from the Ganges River. According to my household survey data, 89 percent of people at Chapra believe that the Farakka diversion is the root cause of this Gorai River bank erosion. According to Joardar, the erosion began immediately after resuming the Farakka Barrage in 1975 92  and increased further with the other technological interventions, like the Lalon Shah Bridge over the Gorai River. Joardar argued that the Ganges River has been turned away from Pabna district toward Kushtia district for seven kilometers after eroding the villages of Mohanagartek and Belkaloya. Every year, the Ganges Basin flow reduction causes sedimentation in the Gorai River mouth which blocks the Gorai water flow. The level of sedimentation and charland is increasing every year due to the continuous Farakka diversion. In this context, the GRRP cannot effectively remove the larger amount of sedimentation, which creates river bank erosion and embankment failures at Chapra.   Due to the GRRP dredging activities, the area downstream of the Gorai railway bridge at Koya is currently encountering river bank erosion. Again, the Gorai River right bank is eroding at Chapra. Khoksa Upazila, downstream of Chapra in Kumarkhali, has also faced this erosion for several years. The Gorai bank erosion at Khoksa causes major risks to the five hundred year old Hindu Puja Mandir or prayer house and Hindu crematorium. The village of Kamalpur, downstream of the Khoksa, is also encountering this erosion. Two thirds of Hizlabot village has almost been destroyed due to erosion. Ganeshpur, downstream of Hizlabot village, is also encountering this erosion, which also destroyed areas further downstream, such as, Khoksa Upazila in Kushtia District, Shailokopa Upazila in Jhenaidah District and Sreepur Upazila in Magura District (Daily Ittefaq 2011).       My focus group respondents, Keramot, Harun, Tarun and Kusum, argued that the river bank erosion and displacements are major causes of losing bongshio shikor or the ‘ancestral roots’ of farming households at Chapra. In 2011, for example, Chapra communities encountered embankment failure nightmares. Because of their concerns about embankment failures they spent several nights sitting up without sleep and stockpiled emergency goods like dry food and clothes, 93  in case they had to abandon their residences. Billal bought fuel by reducing family food supplies to ensure available lights at a night to resolve potential emergency challenges like snake bites. Many of them visited the embankment site every thirty minutes to get updates on the embankment condition. Their fears were based on their past experiences of Gorai bank erosion.  River bank erosion did not originate with the GRRP. For instance, Kafil, a focus group respondent, told me that his family has encountered serious erosion problems five times over a period of three generations. Many of his grandfather’s croplands are now in the middle of Gorai River. He argued that the previous Chapra village is now charlands and the village has shifted two kilometers west in relation to the previous Gorai River channel. The current river location was their village during his father’s generation. Erosion and embankment failures have significantly intensified, however, with the GRRP. Again, Billal informed me that he experienced this erosion from the very beginning of his childhood, as did his father, grandfather and great grandfather. His grandfather was a rich farmer, although he is currently a landless day labourer. Many croplands of his previous generations are now charlands and inside the Gorai River. His current residence is twenty feet distant from the river and is on the verge of erosion. The erosion also destroyed many local structures like hospitals, mosques, gardens and residences. GK project infrastructure at Chapra, which I describe below, is now also at major risk of erosion and portions have already been destroyed twice.   Responses to my household survey data indicated that river bank erosion was the major cause of displacements for 86 percent of people at Chapra and, among them, 38 percent encountered this displacement more than three times. Every displacement disorganized their rhythm of livelihood, including household structures, cropland and agricultural production materials. Tanvir and Tofajjal faced major challenges for protecting their livelihood practices 94  like agriculture, fishing, education, housing and health care. Again, Joardar and Billal lost their last household assets due to this erosion and displacement. Many displaced people left Chapra forever and now live in urban slums. Joardar informed me that a victim of the Rana Plaza garment collapse in 2013 was displaced from Chapra in 2007. The major reasons for this displacement are river bank erosion and flooding.     Flood Vulnerabilities  Bangladesh is the third most flood vulnerable country of the world (UNDP 2004) and Chapra residents are among the most vulnerable in the country. My focus group and case study respondents express grievances against the GRRP based on the flood of 2011. Kafil, a focus group respondent, asked me a basic question because of his current livelihood sufferings: “aponi ki bolte paren amaderke ei durjog theke ke rokka korbe?” (could you please tell who will save us from disasters like flooding or drought?). He said that the Ganges River does not have enough water flow when they require it during the summer season. On the other hand, they encounter flooding when they need a regular flow. According to Joardar, the root cause of the 2011 flood was a sudden increase in flow due to combined effects of local rain and the river flow increase. In 2007, a second flood hit in October, when Joardar was still trying to recover from an earlier flood in August. The Gorai River lost its capability for transporting the sudden heavy water flow due to the major sedimentation and charland problems. When I was visiting their croplands at the Gorai River bank in 2011, Billal and Kafil showed me the aggressive flow of the Gorai River and they termed this aggressiveness a rakkhoshi nodi or ‘ravenous river.’ According to focus group respondents, Fajal, Keramot and Laily, the GRRP is responsible for increasing flooding due to the increasing number and scale of embankment failures and river bank erosion. This 95  flooding is even more dangerous than a regular flooding. In a normal flooding situation, local people can predict flooding characteristics and thus have time to save some household goods and assets.  Focus group respondents, Billal, Joardar, Jobbar, Suma and Morjina, encountered the different types of flooding and their negative effects on household assets and field crops. Billal informed me that flood water from sudden embankment failures reaches to a rooftop within a moment and offers no scope for saving household assets, domestic animals, reserved food, local seeds and crops. Joardar encountered losses of agricultural crops, damages to his cooking place and losses of next season’s seeds and production capital due to the flood in 2011. They lost domestic animals and shelter places and many of their household goods washed away in the flood water. Many of them encountered sickness, fevers and diarrhea. They lost household goods like utensils and structures like tube-wells. Community utilities like schools, electricity stations and mosques also encountered damage and loss. Focus group respondents, Billal, Joardar and Suman, passed the rainy season enduring these types of vulnerabilities and started the winter season with new challenges. They had no crop seeds, agricultural production capital, bullocks or plows for resuming agricultural production. Moreover, the marginalized people like Joardar and Billal did not have money to spend on food, education, health care and housing. Many of them are desperate to find out possible borrowing sources but face exploitation by local elites and NGOs.  Chapra neighbourhoods sought formal help from local government to recover from their damages and losses due to the 2011 flood. However, they failed to get any positive response from local government and, therefore, they made traditional arrangements to overcome the embankment failure effects. They collected harichaada (a traditional system for contributing 96  goods to a community event) from local people to raise money so that they can protect the embankment. Some community members contributed to this fund by cutting off daily expenditures on food and gas and selling domestic animals like chickens or ducks. They bought materials like bamboo, sand bags or poly bags with this money to make a stronger foundation for the embankment. Bilal argued that physical existence was the first priority (compared to starvation). Many community members provided manual labor, sacrificing time they could have spent at paid work. They put guards on the embankment site to get regular updates and ensure the embankment security. To make matters worse, after experiencing flood concerns during the rainy season, they encountered drought during the summer season.         Drought Vulnerabilities  The GRRP often fails to reach the minimum objectives of the summer seasonal water flow for the basin communities at Chapra due to the Farakka diversion. Because of this diversion, the Gorai River discharge was reduced from 171 cubic feet per second (cusecs) in 1971 to 2 cusecs in 2003 (Islam and Gnauck 2011). Currently, the Gorai River is itself encountering survival challenges due to the longer drought season (Islam and Karim 2005). Local people cross the river on foot or use mechanized transportation because the river becomes so shallow during the summer season. While conducting my fieldwork at Chapra on 2 March 2012, I played football in the middle of Gorai River with community people. The Sundarbans mangrove forest further south is also encountering survival challenges and mangrove trees and marsh crocodiles are declining significantly due to the failures of the Gorai River (Mirza and Sarkar 2004).     Focus group respondents, Fazlu, Bimal and Suman, told us that the GRRP fails in restoring local water bodies due to the continuous Ganges Basin flow reduction. Local water 97  bodies like canals, water depressions or wetlands are drying up due to continuous failures of the Gorai River. Tofajjal informed me that local rivers like the Hisna, Kaligangya, Kumar, Hamkumra, Harihar and Chitra are disappearing from local maps due to the basin flow reduction. The disappearance of these rivers deteriorates surface and ground water levels. The GRRP fails in restoring these rivers and water levels. The disappearance of local water bodies is a major cause of failing river water supplies, crop land fertility, water-borne wild vegetables, fisheries and domestic animals. These failures are responsible for major agricultural production and employment challenges.  Kamal and Ibrahim, Focus group respondents, informed me that the GRRP fails in restoring the seasonal patterns of the basin flow. These failures increase challenges for agricultural production and employment opportunities. Joardar argued that sometimes the borsha or bonna ends early or later that alters the winter and summer seasonal patterns and this in turn changes the cropping patterns of kharif-2, robi and kharif-1 during the rainy, winter and summer seasons respectively. The khora season fails to get the regular water flow from the Ganges Basin due to borsha seasonal instabilities that creates major challenges for robi and kharif-1 seasonal crop production. Focus group respondents reported that they encounter trauma every year during the rainy and summer seasons because of these instabilities. The GRRP fails in restoring the ecosystems. Currently, they no longer expect water and ---ical services from the Gorai River for performing agricultural and livelihood practices.     The basin communities at Chapra are no longer hopeful about restoring the past agro-ecosystems on which they have relied for many generations. The reduced flow of water and the GRRP interventions disrupt these systems in the upper, mid-level and lower elevation croplands. The household survey data identified that 87 percent of respondents report facing major 98  livelihood challenges as a result of recent changes. The cropping practices based on maacha and vita or raised platforms with bamboo and mud respectively are no longer helpful to produce crops. The many wild, freely available natural resources described in chapter three are also much less abundant under current conditions. Drought, in addition to flood and river bank erosion, are major reasons for this resource depletion, which are deteriorated by the continuous sedimentation.     The Ganges-Kobodak Project     The Ganges-Kobodak (GK) project, like the GRRP, is a state owned and directed water modernization project. Since the 1970s its implementation has been based on the United Nations’ Flood Control, Drainage and Irrigation Program (Alexander et al. 1998; Talukder and Shamsuddin 2012) and in support of the goals of the Green Revolution (Cleaver 1972:177; Herring 2001:235). Unlike the GRRP which is focused on the flow of water in the Gorai River, the GK project is focused on the construction and maintenance of a series of canals that divert water from the Ganges River at a point several kilometers upstream from where the Gorai River branches off from the Ganges. Also, unlike the GRRP, the GK project was originally implemented before the construction of the Farakka Barrage in 1955 with the purpose of providing a secure supply of irrigation water to farmers throughout the year. It was, in fact, the first large scale irrigation project in Bangladesh. It was not completed until 1983, however, several years after the Farakka Barrage was constructed and, as a result, planners were forced to accommodate the changes brought about by the Farakka diversion and mitigate them to whatever extent possible. As in the case of the GRRP, the GK Project has interrupted the natural movement of water in the Gorai River Basin, undermining the ecosystems on which most 99  households depend and distributing the available water in ways that increase existing inequalities.     Figure 4.1: The GK Project Area. Source: Adapted from Banglapedia 2012.   The project encompasses an area bounded on the north and east by the Gorai and Madhumati Rivers, on the south by the Nabaganga River and on the west side by the 100  Mathabhanga River (Figure 4.1). Due to the decreasing flow of water in the Ganges and Gorai Rivers the GK canals are currently the main source of irrigation water for agricultural production in Chapra and throughout the GDA, despite the fact that the project had been established originally as a supplementary water supply.  The project has two main pumps with three subsidiary and twelve tertiary pump units (Table 4.1) capable of discharging 5,400 cusecs of water (Bangladesh Water Development Board n. d.). The project has an electricity requirement of 14 megawatts and has its own grid substation for providing this electricity. There is a 705-meter long intake channel that pulls the Ganges water to the main GK canal based on lift-cum-gravity flow for agricultural production. The main canal and sub-canals have 3,500 outlets to control irrigation water supplies to croplands. The project authority has established 2,184 hydraulic structures for the entire project area in addition to these outlets. These structures control water supplies at the main, secondary and tertiary canals. Project infrastructure also includes 228 kms of roads to promote project performance and resolve farmers’ water demand concerns (Bangladesh Water Development Board n.d.).          Table 4.1: GK Project Infrastructure Infrastructure Sub-section Quantity/Capacity  Pump Houses Main Pump 2 units  Subsidiary Pump 3 units Tertiary Pump 12 units Water Carrying Capability Main Pump House 3900 Cusec Subsidiary Pump 1500 Cusecs Total outlets 3500 Inspection roads 228 kilometers Irrigation Canals  main canals 193 kilometers  secondary canals 467 kilometers  tertiary canals 995 kilometers  Flood Control Embankments 39 kilometers  Drainage Canals 971 kilometers Source: Bangladesh Water Development Board n.d. 101    There are three main canals of the GK Project that cover 193 kilometers (Table 4.1). The secondary canal system covers an additional 467 kilometers and the tertiary system covers another 995 kilometers, a total of 1,655 kilometers. Most of the main canal is made with concrete and the secondary and tertiary canals are almost completely made of packed earth. All of the main, secondary and tertiary canals are interconnected and provide a lift-cum-gravity method of water supply from the source water point on the Ganges River to local cropland. The project also includes 39 kilometers of flood control embankment. Additionally, the project has a 971-kilometer long drainage canal for transporting extra water from the project area to overcome water stagnation (Table 4.1). The project has also constructed 2,184 fish projects to provide a supplementary fish supply for local communities.   The project encompasses a total area of 488,032 acres of which 286,642 acres of cropland receive irrigation water (Table 4.2). It covers thirteen upazilas within the four Districts of Kushtia, Chuadanga, Jhenaidah and Magura. The total population of the GK project area is 2.5 million. Among this population, 150,000 household heads own enough agricultural land to obtain project water service. These households represent 60 percent of all farming households in the area according to the Bangladesh Water Development Board (n.d.). Based on my fieldwork experiences, a person who has one-half acre or more of cropland is capable of affording the project water supply for agricultural production. The remaining 40 percent do not have enough cropland to afford these services. Among the 60 percent who do, two percent are rich households (3,000 households in total in the project area; 13 percent are intermediate (19,500 households) and 85 percent are small farmers (127,500 households) (Bangladesh Bureau of Statistics 2005).       102  Table 4.2: Cropland Areas and Population Served by the GK Project   Project Target Target Description Project Coverage Area 488,032 acres Irrigation Target Area 286,642 acres Project Area Population 2.5 million Owners of Land 150,000 Location Kushtia, Chuadanga, Jhenaidah and Magura Source: Bangladesh Water Development Board n.d.   The GK authority requires a massive organizational system for ensuring proper management of water supplies, infrastructural maintenance, accountability and transparency. The project has a complex chain of command from the central government to field supervisors. The central government established an office in Kushtia for operations and maintenance under the direct control of the BWDB head office in Dhaka. This local office operates under a Project Director who is at the rank of superintendent engineer overseeing several departments, for example, hydrology, morphology and agriculture. The departmental heads are also engineers.  They supervise various sub-sections within their departments and coordinate field level activities with field staff in the project area. The field staff are responsible for collecting infrastructural and water updates to develop better performance of the project. The project’s yearly operation and management costs are reported to be $24,881,1003 (Bangladesh Water Development Board n. d.).  In keeping with the emphasis on agricultural water use, the GK Project maintains a laboratory to test High Yielding Variety (HYV) seed quality to determine its suitability for crop production in the GK project area. The government also established a training center in Kushtia                                                           3 This figure represents the conversion value of Bangladesh taka to US dollars as of May 16, 2012. USD 1= BDT 80.3823.  103  city for local farmers’ skill development, so that they can farm the HYV crops based on the GK project water supplies. All of these activities are closely monitored by six major departments of the project office in Kushtia. The project also developed guidelines for community participation in local water management. For this purpose, there are 749 water management groups based on local farmers’ participation. These groups are divided into 49 farmers’ clubs which are again divided into seven water management associations. The water management federation is formed of these associations. However, as noted for the GRRP, this local water management system creates success for the rich farmers and failures for the marginalized farmers. The rich farmers control local water management associations. Many of them also chair or are a member of local Union Council or Upazila Parishad.        Outcomes of the GK Project The project has had some success in providing water supplies for agricultural production in Kushtia and Jessore Districts. For example, in 1968-69, it provided irrigation water for 67,933 and 14,384 acres of cropland to produce aman and aus rice paddy crops respectively in these two districts (Bangladesh Bureau of Statistics 1983). By 1978-79, in Kumarkhali Sub-district within Kushtia District, where Chapra is located, the GK project was successful in providing irrigation water to 14,287 and 4,451 acres of aman and aus cropland respectively (Table 4.3). In 1979-80, the water supply in Kumarkhali Sub-district was reduced to 10,685 acres of irrigated aman cropland during the rainy season but the water supply for aus during the summer season increased to 6,268 acres. The project faced serious challenges, however, when local water demand for robi crop production during the winter season increased when the source water was 104  decreasing due to the Farakka Barrage diversion. The GK Project thus had some successes but failed to provide a year around source of irrigation water (Table 4.3). The project also created a set of additional hardships for the majority of local farming households; however, (i) canal and road construction has caused significant damage to local ecosystems and associated natural resources; (ii) project infrastructure creates more vulnerability to both sudden floods and water stagnation; (iii) the project transforms water that was once freely available into a commodity that many cannot afford; and, (iv) the project provides elites more control over local water resources.   Table 4.3: GK Project Irrigated Acreage at Kumarkhali Sub-district  Year Aman (acres) Robi (acres) Aus (acres) 1978-79 14287 -- 4451 1979-80 10685 -- 6268 1980-81 12012 508 7929 1981-82 12594 122 8210 1982-83 13330 -- 9354 Source: Bangladesh Bureau of Statistics 1983      Focus group respondents, Kabir, Kusum and Laily, argued that the failures of the Gorai River management make them further dependent on the GK project for livelihood practices. Billal uses the project water for bathing domestic animals, e.g., cows and goats. Many of the respondents also take baths in this water. They raise domestic ducks and cattle based on this project water. Kusum and Laily, focus group respondents, collect water from the project to perform domestic activities like cleaning. Farm workers drink this project water when working in the fields. They also take rests under the GK project orchard. Many of them get seasonal fruit like mangos and jackfruits from areas irrigated by the project. They also cultivate seasonal vegetables on the project canal bank areas. Many of them harvest fish from the project water for household consumption. They are forced to use the GK project water for these purposes, due to 105  the Gorai River water failures. The canals thus provide a few of the ecosystem services formerly available from the river, but these services are now available in a much diminished capacity.    Construction Phase Conflict and Mismanagement The GK project construction further promoted domination of the ruling elites in local water management. Tofajjal, an in-depth case respondent, argued that the rich people who were inside the power circles supported the project’s construction and the marginalized people were very concerned about losing cropland, houses, stored foods, and production materials. As noted above, the GK project area has a population of 2.5 million and among them, only 150,000 households (Bangladesh Water Development Board n.d.) with a total population of about 700,000 are direct beneficiaries (based on average household size of 4-5 people). The remaining population of 1.8 million people includes urban and rural households, a large percentage of whom are marginalized, agriculture-dependent sharecroppers, day labourers, fishermen, boatmen, blacksmiths and other tradespeople, all with less than one-half an acre of land  Although exact figures are not available, it is clear that the majority of agriculture-dependent households are not direct recipients of the GK project services, although they have experienced damages and loss of household infrastructure and common property resources like fisheries as a result of canal construction. On the other hand, the rich farmers have been successful in gaining more benefits from the GK project construction. Tanvir’s house was a centre point of the project construction at Chapra. The project staff lived in his house. Tanvir and his father considered these activities as social services although this close contact with the project staff and control over project compensation money helped them to fulfill personal interests. They were successful 106  in saving assets like croplands, household structures and fisheries projects during the canal construction.  The GK project construction faced resistance from the local marginalized people at Chapra due to a number of concerns: the loss of cropland and household infrastructure, displacement, forced changes in cropping patterns, and water commodification. My household survey data confirms that 56 percent of my respondents tried to stop the GK project. They failed, however, because of the rich farmers’ non-cooperation and domination. Focus group respondents, Kamal, Kusum, Alamgir and Morjina, informed me that the government used the legal system to stop the protesters. Some local residents including Kamal submitted their complaints of eviction from residences and croplands to the court in Kushtia. However, their complaints were rejected because of the local elites’ domination of the courts. Forty-three percent of my survey respondents lost household assets and cropland. The project authority acquired these croplands and houses to build canals, flood control structures, roads and water bodies. Joardar lost his home and currently he lives alongside the project canal roadside at Chapra. Many displaced people lost their bongshio shikor or ‘ancestral roots’ at Chapra and now live in slums, on dams and embankments. Many of the displaced people moved to one of the 6, 000 slums in Dhaka (Daily Jugantar 2014).    The basin communities at Chapra who own cropland first experienced water commodification in agricultural production with the GK project. This commodification creates extreme economic burdens for the small farmers who are, as I mentioned, 85 percent of the total project farmers. Another 41percent people who fail to get the GK Project services encounter further livelihood difficulties like food consumption, due to price hike because of production cost increases. Their access to safe water for drinking and other household activities also deteriorates 107  as a result of this commodification. Rahim and Kafil, focus group respondents, argued that the Gorai River water and ecosystem services were free and readily available before construction of the Farakka diversion. However, the Gorai flow reduction makes them absolutely dependent on the GK project. Finding no other alternatives, they accept this water commodification and turn into water consumers, although many of their croplands and houses are within twenty meters of the Gorai River.  Water commodification displaces egalitarian water practices and creates an additional gap between the rich and marginalized people at Chapra. Their traditional water sources from the oxbow lakes, canals and wetlands that used to irrigate croplands, are no longer available due to failures of the Ganges flow and the Gorai River. As a result, their traditional irrigation practices like water pulling and drainage canal building with natural resources like bamboo, rope and mud are no longer useful. The rich farmers can overcome and adapt to these changes because of their surplus land ownership and access to capital. They have greater water demands and more influence within the GK management system. Their previous sense of community water ownership, responsibility towards laborers, local maintenance systems and traditional irrigation practices no longer apply. Local ecological knowledge about the seasonal timing of water supplies and close observations of crop and soil health are no longer applicable based on the current practices. Thus, the GK Project has had the effect of breaking down traditional systems of reciprocity and mutual dependence between different classes, removing the dependence of the rich on the poor and further marginalizing poor households.  Operational Phase Conflict and Mismanagement Resistance to the GK Project was not limited to the construction phase but continued during the operational phase as well. The source point of the GK Project at Bheramara is encountering 108  severe sedimentation, charland deposition and river bank erosion every year which reduces project performance. The GK authority performs maintenances regularly to keep the project source water point unblocked but the distance from the source point to the Ganges River is increasing due to continuous Ganges River flow reduction and sedimentation. Consequently, the majority of my survey respondents who are GK Project subscribers, failed to get a sufficient water supply from the project and 84 percent of my survey respondents believe that the Farakka diversion is the main reason for this failure.       Focus group respondents, Kamal, Ibrahim and Abul, reported the uncertainty of the GK water supply makes it very difficult for them to maintain the cropping patterns of kharif-1, kharif-2 and robi which describe aus, aman and winter seasonal crops. Kamal and Ibrahim explained that, the GK authority often provides water earlier or later than needed. They encounter surplus water supply problems during the rainy season when they require less water and fail to get enough water during the summer season when they desperately need water supplies. Both surplus water and drought are responsible for their agricultural production damage and loss. Among my survey respondents, 78 percent reported that they failed to get responses from the GK staff to resolve these problems and 63 percent of them reported agricultural production loss as a result.    Jalil and Raju, focus group respondents, reported that they do not receive any GK water. The project provides water on a plot-to-plot basis using gravity flow from the Ganges River. Croplands close to the main canal get surplus water, whereas distant croplands fail to receive minimum water supplies. Joardar informed me that the lower and higher croplands are subject to water stagnation and drought respectively.  109  Focus group respondents, Kafil, Joardar, Malek and Faruk, have taken desperate steps to bring GK water to their croplands. Kafil removed top soil from the higher croplands and deposited it on the lower croplands to raise the platforms to overcome water stagnation. Joardar built an informal earthen-canal across the land of other farmers in order to transport the GK canal water to his distant cropland. Many farmers create temporary blockades across project canals in order to move water to their land. Kafil sometimes uses temporary pipelines to get water and, after fulfilling his demands, he removes the pipelines.  Focus group respondents, Suman, Sabbir, Kusum and Fazlu, report major losses of HYV crop production due to the GK project’s irregular water supplies. This HYV crop production requires controlled water and therefore both water scarcity and extra water cause a loss of crop production. One major example of an over-supply of water occurred in 2011. Based on past water supply experiences, local farmers sowed IRRI (T-32) paddy, a HYV rice species, in mid-June and expected to harvest this crop at the beginning of August. However, they experienced a sudden rainfall on 3 July 2011 that damaged the IRRI (T-32) crops.  The project drainage channel and flood control system failed to remove excess water from croplands which caused water stagnation and crop damages.  The failure of the GK project forces farmers to adopt alternative cropping strategies at Chapra.  Rahim and Kafil, focus group respondents, report facing severe water crises almost every year during the kharif-1 season when they have their highest water demand every year. Kharif-1 occurs during the period from March to May based on the seasons of mid-spring and mid-summer. In this season, the project is only able to provide water supplies, on average, to 286,642 of 488,033 acres of croplands every year since its full operation (Bangladesh Water Development Board n.d.). Based on these past experiences, farmers produce alternative crops 110  like lentils, beans, wheat and onion that do not require the project water. However, the GK project authority does not want to recognize these alternative cropping practices because this recognition could be interpreted as evidence of the GK project failure.  In 2012, for example, the GK authority suddenly received surplus water from the Ganges flow during the robi cropping season although local farmers had sowed crops like pulses, mustards, wheat and onions, which cannot survive in wet soil. The robi cropping season is the period from November to February in late autumn, winter and early spring. My in-depth case respondents, Tanvir, Tofajjal, Joardar and Billal, requested the GK authority to stop the water supply because they were at the final stage of cultivating these crops. The authority could stop it at the source water point of the GK Project; however, the authority ignored their request because they wanted to demonstrate that the project was successful in supplying water during the robi season. Joardar, Alamgir and Billal termed the GK project authority’s act as “paka dhane moi dea” (destroy a success in agricultural production when it is almost ready to be enjoyed). Consequently, ten thousand acres of field crops were damaged and local farmers lost about $1,491,554 (Khulna News 2012). These crops were essential to their livelihoods as they attempted to recover from the flood and drought of 2011. As I discuss below, this destruction is a direct outcome of the lack of accountability of project managers to local people.    Lack of Accountability and Transparency Joardar argued that the project authority did not bother much about marginalized farmers’ water concerns. My own research confirms that the project operates independently of local water management institutions and is directly controlled by central government agencies. As a result the GK local authority does not attach enough importance to local water concerns like water 111  stagnation or drought. According to Jobbar and Fazlu, the GK field staff visit the rich farmers regularly but do not care about the water problems of the marginalized farmers.  Lack of accountability has also led to operational problems. Many canals have been out of order for years and many tertiary canals are disconnected from the main canal. Machinery of various kinds is stolen or inoperable. Consequently, rainwater spills into the canals and damages canal infrastructure. Kamal and Rahim informed me that many of the secondary and tertiary canals suffer blockages due to these maintenance problems. Portions of some drainage canals are covered with grass, bushes and rat holes that create water flow blockages and drainage congestion. These problems are further compounded by canal water leakage, misuse and wastage. Water leakage creates water scarcity for some croplands and surplus water for other croplands. Forty-four percent of my household survey respondents reported that they asked GK authorities to repair such problems but, nothing was done.  Due to lack of transparency, project funds are mismanaged and this is also a major cause of declining project performance. Kafil and Ibrahim informed me that many conductors of the GK project do not perform their assigned tasks, like repairing the project area, although they are successful in getting the full contract money. No government agency raises questions to the project authority because they share benefits from these malpractices. Joardar informed me that a contractor received a contract for a five kilometer long repair job for the tertiary canal at Lahineepara in 2010. The contractor performed the assigned tasks only for a half-kilometer although he took away the total money. Contractors have been doing these malpractices for a long time and people rarely complain to the local or central governments because of the lack of transparency and accountability in the government systems. Joardar argued that the project staff dismissed local complaints as politically biased activities against the government.     112  Kafil and Ibrahim, focus group respondents, explained to me the hardships they face when lodging complaints against project malpractices. The Kushtia District project office is responsible for more than 800 kilometers of canal infrastructure. Kamal needs to spend a whole day to reach this office, if he wishes to submit a complaint. He is often not able to meet with the Project Director directly due to the complex bureaucratic hierarchy. In addition to the Project Director, the project has several department heads, assistants to heads and section officers. Joardar needs to respond to several questions at every stage of this hierarchy before he reaches a department head and hopefully the Project Director. Joardar does not understand this office culture and feels that the bureaucrats treat him unfairly. He also reports that the authority does not fully recognize the importance of his water concerns. As a result, he is not comfortable visiting this office and often does not report water problems.       The GK authority also fails to resolve local water conflicts between the rich and marginalized farmers. Many rich farmers have built facilities that allow them to reserve project water for later use on their croplands. When the other farmers request the release of this water, the rich farmers get angry and they begin to treat one other like enemies. Fifty-eight percent of my survey respondents at Chapra report having executed illegal actions to obtain the GK Project water under these circumstances. Many marginalized farmers cut off aisles of the rich farmers’ croplands at night illegally. Many of them block a tertiary canal and divert the water to their land. Tanvir, a wealthy farmer, describes these actions as the outcome of hatred and jealousy against his or other wealthy farmers’ success and, consequently, he takes legal action against the marginalized farmers. On the other hand, Joardar describes these actions as revengeful activities based on Tanvir’s control over the local power structures. The GK authority and local governments fail to recognize the marginalized people’s frustration. Law enforcement agencies 113  execute legal actions such as law suits, jails and fines and also harass innocent farmers in order to protect the interests of the wealthy (Islam 2013).   Focus group respondents further informed me that the rich farmers initiate some informal actions. They withdraw sharecropping lands from the marginalized farmers who are not loyal to them. They also fire agricultural staff who do not support their water control over local resource management. Likewise, they exclude the marginalized people from the government safety net programs like employment, housing, health care or education. The local elites put pressure on local banks and NGO authorities to get their loans back from these marginalized farmers. In some cases, they are able to ostracize individuals within their community so that they cannot walk to neighboring lands or on public roads to perform everyday practices like schooling, marketing or cultivation.     Even though GK project staff and contractors ignore many of the infrastructural problems associated with canal maintenance, they do respond positively to requests from rich farmers. Joardar reported facing water shortages during the 2010 kharif-1 crop season and he made several requests to GK staff. However, the staff ignored his requests because of Tanvir’s water concerns. If the staff had responded positively to Joardar’s request, Tanvir’s croplands would have encountered a water shortage. Consequently, Joardar encountered production loss and Tanvir was successful in ensuring his production. In every village, a Union Council member and his associates, like Tanvir, own a major portion of local croplands. If something happens against their interests, they exploit the local power structures to secure their control over the GK project.    The central government has been forced to establish some alternative sources of irrigation water due to the GK project’s water supply failures. The government introduced shallow tube-well and deep tube-well construction programs for this purpose. More than one tube-well project 114  was established for every chalk or, cluster of croplands, in an agro-ecological area. The rich farmers were able to construct private tube-wells inside the GK project area and, as a result, Tanvir withdrew many of his croplands formally from the GK project. However, he left some croplands under the GK project in order to retain membership in the GK project water management association. He is, thus, still able to control GK project water while gaining extra security through the use of tube-wells. The GK project in combination with government supported tube-well construction has thus reinforced elite control over local water resources rather than provided more equitable access.  The pattern described here is consistent with what Das and Poole have referred to as the “colonization of margins” (2004: 3) and with the colonization of what Habermas (1987) calls the “lifeworld.” The intrusion of the state in this setting follows a typical pattern, in which scientific knowledge and techno-centric water management policies are developed, consistent with neoliberal approaches to agricultural modernization and water commodification. This approach fails to recognize community voices and supports elite control over local resources (Agrawal 1995; Leach et al. 1999:225; Peat and Watts 1996b:11-16). Elite interests—from the local level at Chapra to the international scale at the World Bank—can claim this approach is successful at the same time as local marginalized people are encountering continuous ecosystem service and livelihood failures.  As was the case with the GRRP, the GK project has failed to develop an effective or equitable water and agricultural management system. Both projects increase the inequality of rich and marginalized farmers. Both work in support of the central government’s agenda of introducing HYV crops, chemical fertilizers and pesticides to replace local crops, siltation and algae and traditional plowing. The rich farmers are successful in promoting commercial 115  agricultural production and agri-business based on these technologies. On the other hand, the marginalized farmers are encountering survival challenges and facing human rights violations.  In the next chapter, I focus in more detail on the transformations that have occurred to agricultural practices in the Chapra region as brought about by the three major government interventions I have described in the past two chapters: the Farakka diversion, the GRRP and the GK project.  116  Chapter Five AGRICULTURAL TRANSFORMATION AT CHAPRA  As a result of government policies and interventions the agricultural system at Chapra is being transformed from an natural to a technological system. I have indicated the nature of some of these changes in chapters three and four but, in this chapter, I describe them in greater detail. They include a wholesale shift to new crops, new cropping schedules, more reliance on export markets, increasing use of chemical fertilizers and greater reliance on capital-intensive technologies such as tube-wells. They also involve increasing levels of commodification of water, land and agricultural products including seeds, fertilizers and crops. Due to the differential socioeconomic capabilities of farming households at Chapra, the more wealthy households are generally able to benefit from these changes whereas the marginalized households face further marginalization.  The pattern of agricultural transformation I describe in this chapter is similar to that described for many other settings during the period of the green revolution (Blair 1978; Hossain 1988) and is continuous with colonial policies intended to transform indigenous agricultural systems in the developing world into ‘modernized’ systems. As many scholars have noted, this transformation leads to the loss of local ecological knowledge and its replacement by scientific knowledge and the commodification of ‘nature’ (Baviskar 2007; Cleveland and Murray 1997; Escobar 1996; Kottak 1999; Mehta 2001). James Scott (1998:241) points out that, according to colonial officials, “the practices of African cultivators and pastoralists were backward, unscientific, inefficient and ecologically irresponsible. Only close supervision, training and, if need be, coercion by specialists in scientific agriculture could bring them and their practices in 117  line with a modern Tanzania.” Science, in this context, is controlled by corporate actors, not by scientists (Latour 2004). States become political entrepreneurs promoting the production of scientific knowledge rather than indigenous knowledge consistent with neoliberal development agendas (Brandes 2005; Faber and McCarthy 2003:39; Guha 2000:4; Hanson 2007:599-600; Mascarenhas 2007:566; Robeyns 2005:94-5). The general outline of this process of agricultural transformation has been documented for Bangladesh by other scholars (Islam and Atkins 2007; Sillitoe 1998:204; Paul 1984; Zaman 1993). In this chapter I provide documentation of this process for Chapra specifically, on the basis of information gathered directly from local farming households.   Cropping Patterns As described in Chapter three, there are three major cropping patterns known as kharif-1, kharif-2 and robi that correspond to three distinct agricultural seasons which I describe here as the spring season, the summer or rainy season and the winter season. The kharif-1 crops are grown from March to May immediately after the winter season and before the rainy season. The kharif-2 crops are grown during the rainy season from June to October. The robi crops are grown during the winter season from November to February. As noted previously, the rainy season is called borsha and the dry months of both winter and spring season are called khora, terms that are best understood as agro-ecological in meaning, since they designate cropping patterns as well as seasonal variations. My focus group discussants and case study households pointed out that this seasonal cycle has changed, however, since the construction of the Farakka diversion.   Formerly, the kharif-1 season would begin with the arrival of the first spring rains that would bring the dry, hot weather of the late winter khora season to an end. Farmers would plant 118  and harvest crops during this time that were suitable for this particular season. Some of the traditional kharif-1 crops are muskmelon, watermelon, gourd, chili, spinach, okra and cucumber. Farmers have followed this seasonal cropping pattern for generations. Kharif-1 crops can neither survive in extreme drought nor flood conditions. However, the kharif-1 cropping pattern no longer occurs in the same way. Kamal and Ibrahim, focus group respondents, informed me that early floods inundated the kharif-1 seasonal crops—like oil seed, groundnut, chillie, vegetable, okra and bitter gourd—at Chapra in 2011. This occurred through a combination of heavier than normal rainfall and the fact that India stopped diverting water at the Farrakka Barrage, thus causing an abnormal and unexpected surge of water downstream in Bangladesh. In other years, rainfall amounts have been exceptionally low, causing drought conditions that can in turn create virus and insect problems for many kharif-1 crops. In these cases, India diverts a larger than normal percentage of Ganges water to Kolkata, creating an extreme drought situation in Bangladesh, which otherwise might have been manageable. Since the increasing incidence of both flood and drought are causing a decrease in the production of traditional local kharif-1 crops, many farmers are now growing the government supplied HYV crops during the kharif-1 season. The kharif-2 season begins immediately after kharif-1. During the kharif-1 crop harvesting period, farmers begin to prepare croplands and kharif-2 seedlings before the monsoon rain of the borsha season begins. Formerly most cropland was planted in two main kharif-2 crops, aman paddy and jute, which traditionally were the foundation of the national economy in Bangladesh. Aman paddy rice was a staple food that people in Bangladesh would eat three times a day. Jute was called the golden fiber because it was the main source of foreign currency. However, the kharif-2 season is facing irregularities today due to reductions of the Ganges River 119  flow. The disappearance of local water bodies and the associated lowering of water tables make it difficult to produce both aman paddy and jute. Kamal, a focus group respondent also informed me that bull frogs locally called kola bang no longer herald the rainy seasonal resumption, since they are disappearing from local water bodies due to ecosystem failures. Finding no other alternatives, local farmers plant the government promoted HYV rice paddy for the kharif-2 season just as they do for kharif-1.  Tanvir, Tofajjal, Joardar and Billal, in-depth case respondents, argue that rainy season failures in turn generate major challenges for robi crops during the khora season. Reduced water levels in the Ganges and Gorai Rivers lower local water tables, thus reduce the level of water in local water bodies at Chapra, like Lahineepara Beel. As a result, they often do not have the minimum water necessary for robi crop production. Under normal conditions, the level of water in Lahineepara Beel is such that it raises the water table and plants are able to capture some of this water without irrigation. Respondents state that the increasing dryness of the land creates higher temperatures, which cause a higher incidence of medical problems, such as heat stroke, as well as crop failures.  Farming families face major challenges when attempting to follow traditional seasonal cropping patterns. As the khora season length is increasing, due to different factors like the Farakka diversion, local wetlands face a longer shallow or dry time, which causes challenges for producing the kharif-1 and kharif-2 crops. To overcome these challenges, farmers produce one new rice paddy crop locally called boro at the end of kharif-1; boro production has been dominating cropping pattern since the 1990s. For example, boro was planted on 76,890 hectares out of a total of 236 thousand hectares in Kushtia in 2004-05 (Figure 5.1). When boro is planted in this way, there is not sufficient time left for producing kharif-2 crops like aman. Furthermore, 120  the rainy season failures, since they lower the water table, cause winter season failures that reduce production of robi crops like wheat, pulse, lentil, onion, chilly, vegetable, potato, cauliflower, radish, cabbage, tomato, mango, eggplant and okra. Reductions of wheat crop production also occur and are especially significant because this crop is second in importance to the paddy crop for household food self-sufficiency. In 2004-05, farming households in Kushtia District planted 31,160 hectares of wheat (Figure 5.1). The continuous ecosystem failures caused by reduced flows past the Farakka diversion are hidden by the fact that overall production figures remain high. By paying attention only to overall production figures, government analysts fail to recognize the ways in which their interventions are actually changing seasonal patterns, causing environmental degradation and increasing marginalization of smallholders. My in-depth case respondents termed these production challenges as “songkot” or “crisis” originating in the Ganges flow reduction.     Figure 5.1: Crop Types (in thousands of hectares) in 2004-05 in Kushtia District. Source: Bangladesh Bureau of Statistics 2005.   Focus group respondents explained that they are unable to maintain traditional land use patterns because of the Farakka diversion. Jalil argued that the diversion creates major challenges 02040608052.61 76.89 31.16 4.45 8.9 2.83 3.24 55.85 121  for the traditional agro-ecological system in all elevations within the Ganges River floodplain. Currently, the higher elevation croplands fail to receive river water, except during an extreme flood event and, therefore, lose ecosystem-based fertilizers like earthworms, water hyacinth and siltation, resulting in lost soil fertility. Mid-level croplands often encounter either drought or water stagnation because of river flow irregularities. Lower croplands encounter the worst problems due to flooding, drought, water stagnation and embankment failures. These environmental problems create agricultural production challenges for the robi, kharif-1 and kharif-2 cropping periods.   Focus group respondent, Tarun, pointed out that the Ganges flow reduction generates ecological knowledge failures in agricultural production. Chapra communities fail to practice the higher, intermediate and lower croplands in coordination with cropping and seasonal patterns. Tarun and Kabir’s deeper understanding about the kharif-2, robi and kharif-1 cropping patterns in coordination with the rainy, winter and spring seasons respectively is no longer applicable, due to the River flow reduction and disappearances of local water bodies. Alagmir and Gedi argued that cropland fertility practices and knowledge about siltation, algae and water hyacinths are no longer applicable in agricultural production and, as a result, croplands are losing productive capabilities. Intercropping practices are no longer applicable because of the changing and unpredictable characteristics of the seasons.  Kafil, a focus group respondent, argued that the Ganges flow reduction causes traditional agricultural production failure in vita systems. This system involves building a raised bed to grow crops like eggplant, chili, beans or okra. Because of the slightly higher elevation of vita crops, they remain safe from water inundation for a longer time than crops grown in the normal manner. Kafil and Rahim also use this vita system to produce seedlings so they can begin crop 122  production immediately after rainy season waters recede from croplands. However, use of these vita systems is diminishing day by day because of the Ganges River ecosystem failures.  The river bank communities at Chapra are also no longer able to practice the traditional maacha cropping systems. Maacha systems require building a platform with natural resources like bamboo, sticks and rope on a relatively higher level of a cropland to secure crop production during the rainy season. Joardar and Billal practice this maacha system for producing crops like bitter gourds, thus somewhat overcoming flooding or water stagnation problems. Moreover, they use the rooftops of residential houses to produce vegetables such as pumpkins. They plant seeds in a household gardening plot and connect the plants to rooftops using bamboo sticks. However, these maacha and rooftop systems are diminishing in use due to the current irregular climatic characteristics. Changes in the seasonal characteristics of the region thus lead to the failure of many traditional cropping practices, such as vita and maacha and other practices that traditionally allowed for the effective utilization of croplands at all elevations. In the short term, this results in lower crop productivity and in the long term it will also result in the permanent loss of the local ecological knowledge associated with traditional agro-ecological practices.   Crops and Seeds Indigenous Crop Varieties Many focus group respondents, notably Kusum, argued that the reduced flows of water in the Ganges River system are responsible for the reduced productivity and viability of local crop varieties. Joardar and Abul described how they would traditionally select the best portions of their croplands and scrutinize their elevation, temperature and soil quality so that they could produce the best possible quality of seed. Not all seeds for the next year would come from 123  specialized plantings, however; Abul and Faruk also reported preserving some seeds from regular crops when these crops pass their seed quality standards. Based on these traditional activities, they would retain seeds like paddy, onion, ginger and turmeric for the kharif-1, kharif-2, and robi seasons.  Suma, Joardar’s wife, produces local seeds like beans and watermelon. She has detailed traditional ecological knowledge about the specific temperature necessary to preserve these seeds. Suma maintains this temperature with traditional systems. Joardar and Suma preserve seeds like onion in an earthenware-jar locally called kolosh to ensure an environmentally friendly setting that allows no extreme hot or cold. Suma inspects the temperature and seed condition of this jar every fifteen days. However, the Ganges flow reduction creates major challenges for traditional seed production when it causes greater extremes of temperature or humidity. The failure of seed production in turn creates major challenges for traditional agricultural production. As focus group respondents, Joardar, Rahim, Harun and Laily explained, they need to plan their total seed demands one year ahead for each cropping season and preserve kharif-1, kharif-2 and robi seeds throughout a full year. This is difficult enough under normal circumstances but much more difficult today when technological interventions in the hydrological system create more unpredictability and turn manageable episodes of flooding and drought into more extreme, unmanageable events.  During focus group discussions, Tofajjal noted that traditional seed production practices involve the sharing and exchange of seeds and thereby the strengthening of community bonds. Keramot and Harun pointed out that the concept of selling seeds as a business was never part of their traditional agricultural practices and sometimes, they still exchange seeds like paddy or eggplant, based on good wishes and community sentiments. Some other seeds, like pumpkin, are 124  exchanged freely among neighboring women. Salma, Parvin and Suma, wives of Tofajjal, Billal and Joardar respectively, share these seeds without any cost so they can all produce seasonal vegetables. They also share the produced vegetables freely to generate good wishes and develop community bonds. However, the volume of sharing has also been reduced as a result of recent natural resource and farming practice changes. Reductions in vegetable production also reduce the food available for domestic animals, a topic discussed in more detail in the following section. My survey data indicate that 87 percent of the local population is facing major challenges to their practice of these traditional cropping patterns. The displacement of these cropping patterns have forced many farmers to accept the HYV crop production, finding no other alternatives.  Introduced Crops The Ganges Basin ecosystem failure is a main reason for seed commodification, which creates differential outcomes for rich and marginalized farmers. Kafil, focus group respondent, explained to me how local seed practice failures create external seed dependencies for agricultural production. The problem begins with the central government’s policy to promote HYV seeds and ignore the marginalized farmers’ voices about local seed species. This top-down seed policy promotes seed commodification, since it is now impossible for farmers to produce sufficient local seed. Better off farmers, like Tanvir and Tofajjal, are able to afford the new seeds and have the capital and resources necessary to grow them successfully; whereas, marginalized farmers, like Joardar and Kafil, cannot and thus face major socioeconomic challenges due to lowered agricultural production.   The central government policies promote the substitution of HYV seeds for indigenous seeds for most of the important crops. In the case of rice, the government introduced HYV aus, 125  HYV t-aman and HYV boro (see also Nightingale 2003:528) after increasing their seed development budget 510 percent from $8,154,140 in 1972 to $497,403,000 in 20064 (Ministry of Agriculture 2007). Figure 5.2 displays the government budget5 to improve just five crop types (rice, wheat, maize, pulse and oil seeds) from 2010 to 2013 (Ministry of Finance 2012). These budget allocations are dominated by different factors like government policy priorities. This greater budget allocation for HYV seed development led to increases in the planting of new seed varieties of aus, aman, boro by 1085, 541, and 2983 percent respectively between 1990 and 2003 (Department of Agricultural Extension 2007). These HYV seeds dominate local agricultural production. On the other hand, production of indigenous local seeds, like aus, t-aman and b-aman fail to receive government support and local people fail to produce them due to ecosystem failures.     Figure 5.2: Two Seed Production Program Budgets in Bangladesh. Source: Ministry of Finance 2012.                                                               4 Conversion to US dollars was calculated using the 2006 conversion rate. 1 US dollar = 67.4504 Bangladesh Taka (BDT) on July 14, 2006. 5 US$ 1= BDT 73.1568, July 14, 2011. 02000000400000060000008000000100000001200000014000000160000002009-10 2010-11 2011-12 2012-130 14,999,300 6,793,630 4,841,660 1,792,040 2,735,220 2,781,700 2,733,850 Seed Production of Rice, Wheat, and Maize Pulse and Oil Seed Project126   The government seed policy promotes private sector involvement in seed development and marketing and this worsens the differential outcomes between the rich and marginalized. Currently, local seed businesses are a new profit-making opportunity for the rich farmers. However, the marginalized farmers are excluded from this business; consequently, they turn into seed consumers of the local seed business farms and the national and international seed industries. Since 1998, the central government has permitted private companies to produce HYV seeds (Table 5.1). Many of these companies use their own marketing systems to sell their own new varieties of seeds. For example, BRAC, a major NGO in Bangladesh and one of the largest in the world, encourages their clients to buy its HYV seeds. Additionally, the central government permits local and international companies to import the HYV seeds from abroad and sell them in local markets.    Table 5.1: Privatization of Seed Production in Bangladesh  Year Organization Hybrid Rice Seed names  1998 BINA (BADC) BINA 4, BINA 5, BINA 6 1998 BRRI (BADC)  BRRI dhan34, BRRI dhan35, BRRI dhan38,  1998 GDC Amar Siri 1 1998  McDonald Bangladesh (Private) Limited  Loknath-503 1998  ACI Limited AILOK-6201, Allok-93024,  1998  Mollika Seed Company CNSGC6(Sonar Bangla 2), HTM-4(Sonar Bangla-6) 2003 Supreme Seed Company  HS273 2003 BRAC GP4 2006 National Seed Company Limited Taj 1(GRA2), Taj 2(GRA3) 2006 North South Seed Limited HTM-606, HTM-707 2006 Sea Trade Fertilizer Limited LP 108 2006 East West Seed Bangladesh Limited HTM 202, HTM 203 2006 Sungenta Bangladesh Limited LU YOU-3(Surma 2), LU YOU 2(Surma 1) 2006 National Seed Company Limited Taj 1 (GRA 2), Taj 2 (GRA-3) Source: Ministry of Agriculture 2007 127  The focus group respondents pointed out how the seed development programs have increased the cost of local seeds. In 2005-06, one acre of cropland required spending an average of $9 a year for HYV t-aman, HYV boro, HYV aus rice seed (Figure 5.3). The cost of the equivalent local seeds was higher than the HYV seeds because of government seed policy. The central government provides major subsidies for the HYV seeds, while no subsidies are given to the local seeds that are now produced in lesser amounts and are harder to come by. Therefore, marginalized farmers like Joardar and Billal cannot afford to buy local seeds when they are unable to produce their own and also find it difficult to afford the subsidized seeds. However, the rich farmers like Tanvir and Tofajjal benefit both as commercial agricultural producers and through involvement in the seed business.      Figure 5.3: Subsidized Seed Costs (US$) for One Acre of Cropland in 2005-06. Source: Department of Agricultural Extension n.d.   My in-depth case respondents, Tanvir, Tofajjal, Joardar and Billal are thus all increasingly depending on the HYV crops. My household survey data indicates that 60 percent of the farmers were using these HYV seeds in 2010. In 1969-70, the HYV aus, aman and boro 3 5 3 5 3 4 Aus HYVAus LocalT. Aman HYVT. Aman LocalBoro HYVBoro Local128  crop production was 1,397, 36, and 1,910 metric tonnes respectively in Kushtia but this increased to 40,570, 253,080 and 242,970 metric tonness respectively in 1999-2000. On the other hand, the indigenous rainy season dependent broadcast aman production fell from 51,885 metric tonnes in 1969-70 to 5,490 metric tonnes in 1999-2000 (Table 5.2). This local crop production decrease and the HYV crop production increase promotes profit for the rich farmers but produces major agricultural and survival challenges for the marginalized farmers.      Table 5.2: Comparison of HYV and Indigenous Crop Production (metric tonnes) between 1967-70 and 1999-2000 in Kushtia District Year HYV Aus  Broadcast Aman HYV Aman  HYV  Boro  1969-70 1,397 51,885 36 1,910 1999-00 40,570 5,490 253,080 242,970 Source: Adapted from Chowdhury and Zulfikar 2001: 88, 90, 109, 111, 121, 123, 151, 152  Environmental vulnerabilities like flooding and drought create differential outcomes for rich and marginalized farmers when growing HYV crops. These crops require strictly controlled watering schedules and cannot survive with water extremes of flooding and drought. The Ganges Basin farmers encounter sudden, unpredictable flooding, earlier or later flood periods, and more periods of drought, all of which are not suitable for these HYV crops. Focus group respondents, Joardar and Kafil, cannot cope with these environmental vulnerabilities because of their limited land ownership and economic resources. However, the rich farmers are successful in overcoming these vulnerabilities based on their larger and varied cropland holdings and capital resources that allow them to purchase the water which is now available as a managed commodity rather than a free resource. 129   Water Commodification The reduced volume of water in the Ganges and Gorai Rivers is responsible for many of the hardships now being faced by poor farming households but the commodification of water, made possible by the Ganges-Kobadak Project and the proliferation of tube-wells, is also the cause of new forms of hardship. The central government has introduced these technologies following the advice of the World Bank, the International Monetary Fund and other international agencies, in the hope of achieving the UN Millennium Development Goals that address issues of hunger and environmental sustainability (Water Resources Planning Organization 2009:9). The benefits of these technological innovations are almost entirely captured by small elite, well-off farmers.  As outlined in chapter four, those who want to obtain water from the government operated GK Project must purchase it and this creates unequal access for rich and poor. Much more inequality has been created, however, through the privatization of water supplies by construction of tube-wells. This form of privatization is facilitated by Article 4.7 of the 1999 National Water Policy Act, which specifically addresses irrigation privatization for agricultural production. This is accomplished mainly through the construction of shallow and deep tube-wells by farmers themselves, on their own properties, so they can withdraw water from local aquifers. The number of shallow tube-wells increased by 5307 percent between 1979 and 2005 and deep tube-wells increased by 296 percent during that time period (Bangladesh Bureau of Statistics 2005). Based on these irrigation projects, rich farmers are able to gain greater control over local water management. Tanvir has two shallow tube-wells and one deep tube-well in Kumarkhali sub-district.  130  The government provides subsidies and tax exemptions for the purchase and installation of tube-wells on the basis of various development programs, including structural adjustment plans, the National Poverty Reduction Strategy Paper and the Millennium Development Goals as adopted by Bangladesh. Local rich farmers, like Tanvir, own the wells as private property and operate them without restrictions or regulation. According to Tanvir, he was initially required to obtain a license to install and use the wells but he is not required to renew them periodically. He can withdraw as much groundwater as needed for irrigating his own lands and he can charge as high a water fee as he can get when selling water to others. Tube-well owners can also sell the tube-wells themselves without restriction.   Joardar and Billal do not have the financial resources to purchase this technology and often have to purchase their water from the rich farmers. Joardar and Suman stated that Tanvir controls the price of water supplied in this way and this control over water has become an important component of how local power structures are maintained.   The cost of irrigation water constrains agricultural production among poor households. According to the Department of Agricultural Extension, the average annual irrigation cost per acre for HYV and boro rice crops was $ 87 in 2005-06 (Figure 5.4). This includes the cost of water delivered from all sources, including GK water and private tube-wells. This total cost includes irrigation water for aus HYV, t-aman, and boro HYV but does not include irrigation costs for robi crops like wheat. Since most of marginalized farmers, who make up 93 percent of the population of Kushtia, have been forced by circumstances to grow HYV crops, irrigation water expenses have become a significant new economic burden for this group. Joardar often fails to receive water supplies he is entitled to from the GK project and becomes vulnerable to exploitation by Tanvir since Tanvir controls the local GK project water management association 131  and extends his control of the local water supply through his ownership of shallow and deep tube-wells. These tube-well projects work independently of the GK Project system. They have their own canal infrastructure for water transportation and drainage, separate from the GK canal infrastructure. Tube-well owners like Tanvir control water supplies, service charges, canal construction and repair works for their own projects. In this context, community egalitarian water practices are replaced by monopolized market economic systems. The basin communities at Chapra are failing to resist these systems due to the Ganges flow reduction.        Figure 5.4: Irrigation Costs (US$) for One Acre of Cropland in 2005-06 in Kushtia. Source: Department of Agricultural Extension n.d.    Focus group respondents reported spending a significant amount for domestic water and irrigation water. Earlier, Kamal and Ibrahim used the Gorai River and local water bodies for domestic activities and dug ring-wells or kuya for drinking water, supplies that were readily available without any expenditure. However, due to lower water tables, these sources are no longer available. Tanvir has two tube-wells, one inside and one outside of his residence. On the 24 0 10 0 33 20 Aus HYVAus LocalT. Aman HYVT. Aman LocalBoro HYVBoro Local132  other hand, Billal and Joardar do not have any tube-wells and need to seek permission from Tanvir for tube-well water for drinking and domestic activities.  In addition to tube-well water, Tanvir buys bottled water when his son Shaon and grandchildren, who now live in Australia, come for a visit. There are numerous private companies in Bangladesh producing bottled drinking water at local markets. Tanvir’s grandchildren always drink this bottled water when soft-drinks like Coca-Cola are not available at Chapra. Parvin, Billal’s wife, sees these practices during her work as a domestic helper in Tanvir’s house and feels frustrated for her own children and grandchildren. Billal fails to get tube-well water and sometimes drinks pond or wetland water that has been contaminated by chemical fertilizers, which can cause diarrhea, dysentery and skin diseases. Clearly, this differential water access is a major outcome of water commodification and an important cause of reinforcing social inequality at Chapra.   Fisheries Commodification  Water commodification has led in turn to fisheries commodification. Wild fisheries are failing due to reduced river flows and lowered water tables and in response to this the local elite develop commercial fisheries projects, thus creating a new business out of this environmental crisis. As the Gorai River and its dependent local water bodies disappear, many fish species are becoming extinct. The central government’s water dredging operations create further challenges for fish species like eels (kuichya) and snakehead fish (gajar). Furthermore, chemical fertilizers and pesticide contamination in local water bodies destroy local fish species like the minnow (puti) and also crocodiles (kakla). Khan (2013) informed us that more than twenty-eight indigenous fish species have became extinct due to interventions such as the Farakka diversion and 133  environmental pollution. Currently, ten percent of 260 local fish species are reported to be encountering survival challenges (Khan 2013). Ninety-three percent of the households who completed my household survey reported significant reductions in their ability to obtain fish for personal use or for sale in local markets.    Focus group respondent Sabina informed me that she can only afford to buy lower priced fish like tilapias and Thai pangasius. She and her family encounter major challenges even to buy lower priced fish since one kilogram of tilapia costs $1.50 in the market at Chapra. Sabina informed me that a household with five members needs 157 kilograms of fish for a year, which requires spending $204. This amount is beyond their economic capabilities and so they reduce their fish consumption. Islam et al. (2004a) report that a landless household in this region consumed on average 157 kilograms fish in 1990 but reduced to as low as 58 kilograms in 1999. Local fish species extinctions have been followed by the introduction of new fish species suitable for commercial production. For these purposes, the government allocated $179 million in 2013 (Ministry of Finance 2013:74). The Bangladesh Fishery Research Institute has developed new fish varieties, like Thai climbing perch (gugi) and catfish (nona tengra) (Ministry of Finance 2013:74). The central government has also supported the introduction of foreign fish species like carp, tilapia and Thai Pangasius which in turn serve to create new fisheries business opportunities for the rich farmers at Chapra.  Joardar and Billal argued that rich people like Tanvir control local fisheries projects and that these projects ensure that local fish populations will not recover and that fewer fish will be available for consumption by poor families or as a source of traditional fishing employment. Local political leaders of the ruling government in coordination with rich farmers, often use local water bodies to develop their fisheries projects. Political leaders even blocked the Ganges River 134  for two kilometers downstream from Kushtia to develop a major commercial fisheries project (Prothom Alo 2013b). At Chapra, Tanvir has blocked the Kaliganga River to establish one of his fish projects. In addition, he obtained leases for two other fisheries projects from the GK project by exploiting the local power structure. As I discuss in more detail below, commercial fish farms also provide a means of transforming local employment practices from ones that were relatively autonomous, such as catching wild fish on one’s own schedule, to practices that involve a greater dependency on elite families of poor households and thus more opportunity for their exploitation.  Land and Equipment Costs  Households without sufficient land to support the family are forced to rent land or work as day laborers. According to my survey data, 49 percent of the farming households at Chapra depend on sharecropping for a portion of their livelihoods and are forced to pay high rents for this land to the wealthy land owners (who are the only people who have excess land). Land rentals were less common in the past when land scarcity was less prevalent and rich people often shared croplands with poorer households based on principles of reciprocity, kindness and social justice. The commercialization of all aspects of the agricultural economy makes such arrangements highly unlikely today, however. In 2005-06, land rental charges for boro HYV, HYV t-aman and HYV aus crop production, for one cropping period, were $34, $29, and $25 respectively for one acre of land (Figure 5.5).    135   Figure 5.5: Land Rental Costs (US$) for One Acre of Cropland in 2005-06. Source: Department of Agricultural Extension n.d.     Sometimes, as an alternative to sharecropping, land poor households obtain kot or loans from an employer or other wealthy individual, a practice that creates further dependence. However, this practice is currently following the opposition direction and the rich farmers are exploiting it for further benefits. In 2001, Joardar received $1000 from Tanvir in exchange for 0.33 acres of cropland. Trust and tradition between them were the two major criteria for this kot and Joardar made a verbal commitment to return the money within one year. However, Joardar failed to return the money due to a combination of flooding and then drought that occurred during that year. Later on, he borrowed an additional $500 but failed to recover from his economic losses. He eventually had to sell this land to Tanvir in 2007 for $3,000 although its market value was $6,000.   All households in Chapra today also face additional costs to purchase farm equipment that can no longer be made locally because of a lack of materials like wood. There is insufficient wood, for instance, for making boats, bullock carts, winnow fans and husking pedals. To 25 23 29 25 34 27 Aus HYV Aus Local T. Aman HYV T. Aman Local Boro HYV Boro Local136  overcome these shortages, the national government subsidizes tractors, transportation vans, boats and paddy husking machines that are imported with assistance from the World Bank (International Monetary Fund 2013). The government permits private companies to import these technologies and to sell them directly to local farmers. The central and local governments are officially responsible for ensuring accountability throughout this process. The shortage of locally manufactured farm equipment means that many poor farming households have to rent equipment. Joardar, for instance, rents a tractor and plow from Tanvir once Tanvir has finished using it to prepare his own fields (Figure 5.6). Many poorer households cannot afford to rent tractors and have to do the work manually. Billal and his family, for instance, prepare their field with hoes; his son Jamal and wife Parvin will then shoulder the yoke and try their best to pull the plow that would normally be pulled by a bullock. Ninety-three percent of my survey participants report practicing this type of cultivation system. Tanvir, on the other hand, is able to earn $1,000 by renting his tractor two seasons in a year.      Figure 5.6: Land Preparation Costs (US$) with Tractor and Machine for one Acre of Cropland in 2005-06. Source: Department of Agricultural Extension n.d.    0510152025HYV Local HYV Local HYV LocalAus T. Aman Boro25 19 25 19 24 20 137  Fertilizers and Pesticides As described in chapter three, Chapra farming households would traditionally benefit from several forms of freely available ‘natural’ fertilizers such as siltation, algae, earth-worms, water hyacinths and animal manure. However, most of these natural fertilizers are available today only in much reduced quantities or are not available at all, due to changes in the environmental characteristicsof the region. As a result, farmers need to purchase chemical fertilizers or suffer significant losses in crop productivity.   Natural Fertilizers Focus group participants, Hanif and Naser, described the ways in which they would obtain natural fertilizers from the Gorai River and Chapra beel. During a regular borsha season, the Ganges River flow carries silt that would cover all croplands connected to the Ganges River and its tributaries by local wetlands, canals and oxbow lakes. This siltation is a major source of soil nutrients and no other fertilization is needed. However, since the Gorai River is itself encountering survival challenges due to the Farakka diversion, many local wetlands and ponds no longer exist at Chapra. As pointed out in Chapter four, the Gorai River is filled with sedimentation and charland that prevent the replenishment of local bodies of water during the rainy season.   Focus group respondent Kamal demonstrated a particularly detailed understanding of natural resources like algae, which are declining due to the river flow reduction. He informed me that local wetlands, lakes, canals and croplands themselves will produce algae as long as Gorai River water is available in sufficient quantity during the rainy season. This algae grows naturally in croplands and water bodies during this season, which keep croplands fertile and increases crop 138  production. All farmers like Joardar and Morjina receive these ecological services equally, irrespective of socioeconomic hierarchies. However, this algae dependent cropland fertility has diminished significantly due to the reduction of flow in the Ganges and Gorai Rivers.        Water hyacinth is another source of cropland fertility that is reducing day-by-day because of the Ganges River ecosystem failures. This hyacinth used to be readily available in local water bodies and was easily movable from one place to another during the rainy season. Joardar would pile up water hyacinths in croplands at the end of the rainy season and make them into green fertilizer through traditional methods. The hyacinths were allowed to compost for about two months before they were applied to the fields. Sometimes, local farmers would place water hyacinth compost on their croplands to retain soil humidity for a longer time during the winter season for the robi crop production. They would also use them as cooking fuel. In the past, the plant was traditionally abundant as well as free; however, this water hyacinth is no longer available in most of its previous habitat due to the reduced flow of water throughout the system. Much local cropland at Chapra is also disconnected from local streams, ponds and wetlands due to interference by canals and other water irrigation infrastructure failures.    Earthworms are another traditional source of fertilizer but are much diminished in numbers due to the river flow reduction. One particular species of earthworm, locally called kecho, used to be abundantly present in croplands as a natural consequence of the local ecosystem. My focus group respondent Barek reported seeing a greater number of kecho every year at the beginning of the rainy season. This kecho burrows into the topsoil, eating organic matter and excreting castings which enhance soil fertility. This fertility source is freely available for every farmer at Chapra without discrimination. River fish like snakehead also eat kecho and local fish production also therefore depends on its availability. However, according to my 139  informants, the Ganges River flow reduction has significantly decreased the number of kecho in croplands as this requires a specific environment connected with river water and croplands. Moreover, contemporary excessive usage of chemical fertilizers and pesticides further reduces the kecho population in croplands.  Harun, a focus group respondent, reported that he is now unable to obtain animal manure fertilizers due to a reduction in domestic animal populations. Earlier, the Ganges flow maintained natural resources like grasslands and forests that were helpful to raise these animals. Harun and Kabir described how they used to gather dung from these areas and deposit it in pits on their homesteads for eleven months until it had broken down sufficiently to be used as manure that they would deposit on their croplands at the beginning of the winter season. This would ensure cropland fertility throughout the robi, kharif-1and kharif-2 cropping seasons. No expenditure was involved in making this manure and everybody could make it equally without any discrimination. However, these manure fertilizers, in addition to water hyacinth, siltation and earth-worms are diminishing day-by-day due to the Ganges River flow reduction.   Chemical Fertilizers To overcome the loss of natural fertilizers, the central government supported the creation of more than ten major fertilizer industries and established an import program for fertilizers from abroad. As in the case of the HYV seed industry, well-off farmers can afford to purchase these fertilizers and are also able to create new businesses based on their sale. However, marginalized farmers face extra economic burdens and survival challenges because of this commodification.   In 2006, a total of 3,551,000 metric tonnes of fertilizers such as Triple Super Phosphate (TSP), Single Super Phosphate (SSP), Diammonium Phosphate (DAP), Muriate of Phosphate (MP), 140  Ammonium Sulfate Phosphate (ASP) and Nitrogen-Phosphorous-Potassium-Sulfur (NPKS) were purchased for crop production in Bangladesh. This represents an increase of 306 percent between 1980 and 2006. In Kushtia, 69,155 metric tonnes of fertilizers of urea, TSP, DAP and MP were used in 2011 on 348,000 hectares of cropland (Table 5.3). The rich farmers are the major beneficiaries from these fertilizer usages and sales. Kofil, Tanvir’s son, has a fertilizer business and Tanvir promotes commercial crop production based on these chemical fertilizers. On the other hand, marginalized farmers encounter major economic challenges for affording these fertilizers in agricultural production and they do not have scope for establishing fertilizer business.      Table 5.3: Fertilizer Usage in Kushtia   Kushtia   2008-09 (metric tonnes) 2010 11 (metric tonnes) Increase (percentage) Urea  46,608 50,898 9 TSP 1,561 6,971 347 DAP  524 7,196 1,273 MOP/MP 923 4,090 343 Total  69,155  Source: Ministry of Agriculture 2012    Chemical fertilizer use represents a highly significant extension of agricultural commodification. Currently, one acre of cropland requires an average of $90 a cropping year for fertilizer expenditures to produce HYV aman, HYV t-aman, HYV boro or irrigated wheat crops (Figure 5.7). In this context, one acre of cropland requires $41, $31 and $35 a year for fertilizer expenditures in HYV boro, HYV aus and HYV t-aman crop production respectively. Rich farmers like Tanvir and Tofajjal can afford these fertilizers for commercial crop production. 141  They are the major beneficiaries of the central government’s fertilizer subsidies and they have the capital and government connections necessary to start a fertilizer business. On the other hand, marginalized farmers like Joardar cannot afford these fertilizer expenditures because of their economic crises.            Figure 5.7: Fertilizer Costs (US$) for One Acre of Cropland in 2005-06. Source: Department of Agricultural Extension n.d.   Kamal argued that chemical fertilizer usage is a main cause of loss of soil fertility. He also encounters increasing crop production failures because of new insects, fungi and weeds that appear as soon as he begins to use chemical fertilizers. To overcome these production failures, the central government introduced insecticides, fungicides and herbicides, thus further increasing the cost of crop production. Finding no other alternatives, Kamal uses these pesticides to overcome insects and fungi problems. In Bangladesh, farmers used 17,393 metric tonnes of insecticide, fungicide, weedicide and rodenticide in 2012 for these purposes. Pesticide use increased 262 percent between 1989 and 2002 in Bangladesh (Bangladesh Bureau of Statistics 2005). Currently, farmers need to spend more than $22 per cropping year for pesticides on one 31 25 35 24 41 26 Aus HYVAus LocalT. Aman HYVT. Aman LocalBoro HYVBoro Local142  acre of cropland (Figure 5.8). These monetary costs were not necessary when the Ganges Basin ecosystems were active.                   Figure 5.8: Pesticide Costs (US$) for One Acre of Cropland in 2005-06. Source: Department of Agricultural Extension n.d.   Despite these pesticides, farmers have not been entirely successful in overcoming insect and fungi problems.  According to Joardar, insects develop resistance to many insecticides within two years but new insecticides are not made available to them within this time period. Once available, these insecticides are often more expensive and he cannot afford these increasing costs given the crop production losses he experiences. On the other hand, rich farmers like Tanvir, face fewer insect and fungus problems because they are better able to access the scientific knowledge necessary to these new systems of agricultural production. Pesticides have thus become one more cause for major social inequality at Chapra.  Domestic Animals  Kamal explained that domestic animals are essential to traditional agricultural production but that populations are diminishing day-by-day, due to natural resource failures. Wetlands, grasslands and forests are needed for nurturing animals like bullocks, oxen, cows, chickens, 02468HYV Local HYV Local HYV LocalAus T. Aman Boro143  ducks and goats. Bullocks cultivate croplands and cows provide milk for household nutrition. Traditionally, Billal and Kafil would get bullock calves from domestic cows without cost and use them to reduce their agricultural production costs. They raise these animals domestically and use wild grasslands and plants as their fodders. The bullocks were used to cultivate croplands based on their traditional agricultural practices. Focus group respondent Kusum explained how local natural resources can be used to nurture domestic animals. Domestic ducks feed on small fish and snails from local water bodies such as the Chapra Beel. Cattle receive food from local grasslands and forest resources and vegetation like water hyacinth and banana trees during the borsha season. However, all of these free natural resources are now in much diminished supply due to the reduced flow of water in the system (cf. Lansing 2006).  Therefore, focus group respondent Bakar explained how natural vegetation has been contaminated by chemical fertilizers and pesticides and how new insects and fungi destroy many forms of vegetation and diminish local biodiversity. As a consequence, Billal and Joardar reported no longer having any domestic animals like hens, ducks, goats and cattle. Many of these animals encounter health concerns and die because they eat pest affected grasses and leaves.  To recover from this domestic animal reduction, the central government in Bangladesh introduced poultry and animal farms to ensure meat supplies. For example, in 1997, the government supported the establishment of 29,649 dairy farms, 20,833 goat farms, 10,289 sheep farms, 30,760 duck farms and 60,670 poultry farms throughout the country (Rahman 2012). The government received assistance from the USA, Germany, France, Australia and Japan to develop hybrid species of poultry and cattle. In every village at Chapra, rich people established more than one poultry farm. Tanvir’s son-in-law has a poultry business at Chapra and the whole village 144  depends on this farm for household meat and egg supplies. In 2013, one kilogram of poultry meat cost $1.50. Based on this cost and on a daily consumption of 25 grams per person, Billal needs to spend $123 a year to feed his nine member family. This does not include the cost of eggs and milk. Billal and Joardar cannot afford these expenditures and, thus, go without meat for many of their meals. On the other hand, Tofajjal and Tanvir have generous surplus supplies of meat, eggs and milk. Livestock commodification compounded by vegetation loss is thus another cause of differential class outcomes at Chapra. Rich farmers benefit from government supported opportunities for livestock production, while marginalized farmers face extra economic burdens as they become consumers of the rich farmers’ livestock operations.  Kamal, Raju and Billal all reported having no domestic bullocks for cultivating croplands and thus no opportunity for getting calves from their own cows. The lack of bullocks not only limits the amount of cultivation they can do but also reduces their ability to manufacture agricultural production materials like hoes, sickles, paddy husking pedals, ploughshares and cleavers that are produced traditionally from the slaughtered animals’ bones, horns and hooves. They fail to get milk from domestic cows and eggs from chickens and ducks. These domestic animal and resource failures generate major livelihood challenges and thus human rights violations for the marginalized communities at Chapra.  Domestic animal reductions also damage traditional exchange relationships within the community. In 2010, Joardar borrowed a bullock from a widow and bought her groceries from local markets in exchange. In 2011, he loaned a tractor and plow to Billal for agricultural production during the summer season and received boat rides from Billal during the rainy season. These practices reduce level the differences among farmers and promote community 145  cohesion and egalitarian practices. However, these exchange practices are decreasing due to continuous increase of market dependencies.  Employment Traditional patterns of employment at Chapra have also changed dramatically as a result of the commodification of agriculture and the inability of marginalized households to produce sufficient food through traditional methods. Among my survey respondents, 56 percent reported significant periods of unemployment in 2011 in the agricultural sector or in the other occupations such as boatmen, potters and fishermen who depend directly or indirectly on agricultural productivity. Women, the elderly and the disabled face the worst employment challenges in Chapra.    Local agricultural employment practices have different socioeconomic dynamics. The intermediate farmers like Barek perform supervisory tasks for more wealthy farmers like Tanvir, while landless day laborers like Billal work for Tanvir in agricultural production activities such as cultivation and irrigation. The day laborer is locally called kamla and this individual generally works for the rich farmer from dawn to dusk and is fed three times a day as part of an informal contract. The small farmers like Joardar and his wife Suma mainly work on their own land but also perform day labor for more wealthy landowners on occasion. Joardar’s daughters and sons also perform some gender based agricultural activities like rice winnowing or seed preserving as a part of the family-farming process. Joardar and Kafil, another small farmer, help one another through informal labor exchanges.  There are fewer employment opportunities for kamlas today than there once were and many are facing increasing periods of unemployment. Many kamlas want job security and seek a 146  yearly contract locally called rakhal. The rakhal works for 365 days a year and performs traditional agricultural tasks from early morning to late night. A rakhal stays in an employer’s residence after completing his assigned tasks. During their employment contracts, employees like Billal, who work as a rakhal for Tanvir, are allowed to eat seasonal fruits like mangos from the employers’ fruit gardens and eat onions or pumpkins from the fields. Parvin, Billal’s’ wife, reports receiving vegetables, fish and lentils in addition to some cooked food from Tanvir’s wife, Sohali.   Focus group respondent Keramot informed me that local natural resource failures have led to a reduction in traditional employment opportunities such as the manufacture of traditional tools. Wooden plows, for instance, were traditionally made by day laborers from a local wood known as langol. Day laborers like Keramot would also build irrigation canals with local natural materials such as wood, bamboo, stick and rope. During the rainy season, boats are an important means of transportation and are made with local resources. When field paddy is brought home, it must be husked and winnowed and the winnowing is done with a fan, locally called kula, that is made with bamboo and cane. Billal performs this winnowing task as a day laborer of Tanvir. Parvin works in Tanvir’s home under his wife Sohali’s supervision to winnow paddy, boil it and make rice. The deki or husking pedal is also made with two pieces of wood and manufactured locally. These agricultural production materials are less used today, however, partly because the local materials are no longer available. Labor opportunities are thus lost both in their construction and use.  Focus group respondents Kafil, Harun and Suman pointed out a number of other ways in which reduction of the Ganges flow traditional employment opportunities in the areas of seed production, plowing, traditional irrigation, care of livestock and the preparation of fertilizers. 147  Traditional plowing systems are no longer a major source of local employment. Traditional irrigation water supplies have been replaced with the GK project water and deep and shallow tube-wells so construction and maintenance of traditional systems is no longer a source of employment. Bullock carts are being replaced with mechanized vans. Traditional paddy husking is increasingly mechanized, reducing gender-based job opportunities for Kusum, Morjina and Parvin. Women laborers are not hired to collect drinking water from local water bodies, since that water is no longer available; agricultural laborers are also not asked to prepare fertilizers from traditional sources, since they are no longer available and well off farmers now purchase chemical fertilizers.  The Ganges flow reduction generates challenges for other occupational groups, like boatmen and fishermen, as well. Currently, many of them are losing their ancestral occupations and are having to leave Chapra. The boatmen cannot transport field crops due to the river flow reduction. Boat making materials like wood, bamboo and cane are not available in sufficient quantities. Thatchers and basket-makers fail to get employment opportunities due to reductions of necessary local materials. Fishermen cannot catch sufficient fish due to the condition of the Gorai River and local water bodies. Blacksmiths have lost job opportunities due to the imported technology that now dominates agricultural production. Potters are unable to gather sufficient local materials to carry on their occupation and compete successfully with alternative products. Traditional ecological knowledge is diminishing in importance as a result of changes to traditional employment and occupational patterns. Previously, the younger people began to learn this knowledge in their early childhood, when they observed their elders’ ecological knowledge practices. Grandparents were passionate teachers of their grandchildren. Many children carried breakfast or lunch to their fathers or grandfathers in the fields and observed agricultural 148  practices. Many of them worked as assistants while learning this knowledge. Through this process, Billal developed agricultural production skills by the age of fifteen. However, these learning practices are no longer important due to the Ganges flow uncertainty and ecosystem failures. Local ecological knowledge in this setting is not being lost simply through its displacement by alternative scientific and technological knowledge; it is being lost as part of the wholesale transformation of the environment itself and the commodification of common property resources.    New Employment Patterns Eighty-five percent of those who completed my household survey reported changes to their traditional employment practices, such as agricultural laborers, thatcher, boatmen and potters. The government failed to recognize any of these employment groups as a specialized group. Consequently, these traditional employment systems are in decline because of a government supported “strategy to withdraw laborers from low productivity to higher productivity activities in manufacturing and modern services” (International Monetary Fund 2013:60). Employment sectors are not categorized as “higher productivity” on the basis of wages or working and living conditions, however, so workers who make this transition are not necessarily any better off than before. Many of those who can no longer find employment in traditional rural occupations now work as wage laborers on poultry farms and fisheries projects owned by their more affluent neighbors. Those who work in these new employment sectors do not receive the same benefits as in the traditional system, when they would be fed by the employer or be allowed to pick fruits or vegetables for personal use from the employer’s land. Neither do they receive the benefits that should be available to them in a modernized economy, such as sick benefits and weekends off. 149  Focus group respondents, Jobbar, Fajal and Kusum, reported that farm owners control wage and working hours and employees fail to protest unfair working conditions out of fear of losing their jobs. The ‘new’ rural worker thus faces a triple vulnerability: loss of the traditional ecological knowledge; limited skill training for new jobs; and, new and more intense forms of workplace uncertainty and exploitation.  Billal stated that the new contract based employment opportunities are not sufficient for him to maintain a decent livelihood. During one conversation, he recounted a glorious past when a day laborer earned sufficient money to maintain a decent livelihood. A day laborer in the 1970s earned fifty paisas6 a day and could save ten paisas after spending enough for daily necessities like rice, oil and cloth. Other items like fish and wild vegetables were freely available. As with all the other changes described in this chapter, changes in employment patterns favor the rich over the poor since the rich have ready access to new agricultural technologies and to government subsidies to assist in their purchase. Labor costs for indigenous and HYV crops are much the same (Figure 5.9) but government subsidies for HYV seeds, chemical fertilizers and pesticides, tube-wells and other technology provide powerful incentives for well-off farmers to switch to HYV crop production. The HYV cropping pattern comes together with a reduced need for traditional agricultural labor and the ecological knowledge associated with it.                                                              6 One Bangladeshi Taka is composed of 100 paisas. 150   Figure 5.9: Labor Costs (US$) for One Acre of Cropland Cultivation in 2005-06. Source: Department of Agricultural Extension n.d.    During my focus group meeting in 2011, Tanvir informed me that he has two tractors and two plows, two vehicles, one boat and nine paddy husking machines. This equipment helps Tanvir reduce his labor costs and he can earn extra money by renting his equipment to neighbors. Rahim, Fajal and Laily pointed out a tractor can do the work of twenty laborers. One small van can do the crop transportation tasks of ten day laborers. One shallow tube-well can provide as much water as thirty day laborers’ could provide using older technologies.  Differential access to training and skill development by different classes also worsens the divide between rich and marginalized farmers. The rich farmers are better prepared to benefit from training programs because of their educational backgrounds. Tanvir, a rich farmer, completed grade twelve and Tofajjal, an intermediate farmer, completed grade ten. However, Joardar and Billal do not have sufficient reading and writing skills to participate in the training programs that are offered. Joardar dropped out in the ninth grade and has rarely used his literacy skills since then. Billal never enrolled in a school. Elite domination at the local level also ensures that the programs are not designed to accommodate those with less education. Most poor farmers cannot read English and are also disadvantaged as laborers since they cannot read the labels of 39 39 43 43 43 38 Aus HYVAus LocalT. Aman HYVT. Aman LocalBoro HYVBoro Local151  agricultural materials and technologies. Billal writes Bengali names like sada sar (white fertilizer) on containers of urea so that he can perform his job correctly. Sometimes, he makes mistakes and faces scolding and a wage reduction from his employer.  The new agricultural economy can create benefits for those employed at the managerial level.  Tanvir, for instance, employs Kafil, an intermediate farmer, as his manager. Kafil’s role has expanded in the new economy; he maintains farm machinery and operating schedules, manages irrigation systems, hires and supervises laborers, keeps track of labor costs and sales records and collects crop price updates to assist the marketing process. Day laborers like Billal occupy the bottom layer in this new division of labor and find the gap between them and the rich farmers has widened significantly.    Transformation through Commodification As this chapter demonstrates, the process of transformation now underway at Chapra extends into every domain of the community’s life. Water commodification under conditions of increasing scarcity is central to the transformation but it is also supported by a broad set of agricultural modernization programs initiated by the national government with the support of and the advice of powerful international agencies, upon which the country has grown increasingly dependent over the past several decades. The rich farmers are getting richer and connecting more with globalized economic and concomitant production systems. The vast majority of the population is becoming increasingly marginalized, facing ever more severe livelihood challenges and many are being displaced from their ancestral homes. Many formerly middle class farmers are either becoming rich or falling into the marginalized class. The central government is not entirely unaware of these problems and has introduced safety net programs in an effort to reduce 152  displacements and assist marginalized households. Local governments are major implementing agencies of these programs which will be discussed in the next chapter.   153  Chapter Six LOCAL GOVERNMENTS’ ROLES TO PROTECT MARGINALIZED HOUSEHOLDS  Local-level government in rural Bangladesh administers a number of agricultural subsidy and safety net programs established by the national government on the basis of guidelines provided by the Poverty Reduction Strategy Paper and the Millennium Development Goals for Bangladesh. In theory, these programs could alleviate some of the worst hardships caused by the ecosystem and crop production failures described in previous chapters. Since many components of the local water management system, such as the Farakka diversion in India and the GK Project in Bangladesh, are entirely outside the control of communities like Chapra, these communities have become increasingly dependent on local governments to manage the scarce and unpredictable water supply that remains. Communities expect local-level government to ensure agricultural production and livelihoods and, in order to do so, it needs to address seed and fertilizer management, as well as water management and provision of social service benefits pertaining to food, employment, education, health care and housing programs. Such programs have, in fact, been implemented but, as is so often the case in this setting, they fail to incorporate the voices of the marginalized communities and have fallen well short of meeting their goals. In this chapter, I critically describe the performance of these programs in the context of local power structures. My analyses suggest that local elites in Bangladesh, based on their connections to the national government and external agencies, are able to manipulate subsidy and social programs and thereby extend their resource control over rural communities. Agrawal and Ostrom (2001:488) reached a similar conclusion in their study of India and Nepal.             154  Power Structure of Local Governments  Local elites are major decision-makers in local government institutions that implement central government decisions about agricultural policies and safety net programs. As outlined in chapter three, the political system in Bangladesh is highly centralized and local-level government institutions at the Upazila Pasrishad (UP) and Union Council (UC) levels are controlled by local elites acting in concert with national elites, including elected MPs, high-level bureaucrats and high-ranking officials of the political party in power. All UPs have similar administrative departments, for example, the Department of Agricultural Extension (DAE) and the Local Government Engineering Department (LGED), which constitute the “street level bureaucracy” within this system (Heyman 2004:493). The DAE and LGED are local offices of national departments run partly by the UP. Innovation occurs almost entirely in the form of what Hayami (1981:169) calls “induced innovation,” that is, as a consequence of the actions of external elites in coordination with local elites, rather than as local grass-roots responses to local or regional issues.   UPs and UCs also manage local social service agencies that deal with issues like water management, seed and fertilizer distribution and subsidies and food, employment, education and health care programs. The Upazila Nirbahi Officer (UNO) is a UP staff administrator who serves as a coordinator among UP and UC departments and committees, serving as the chair or sub-chair of the committees that administer the above services. Every elected member of a UC is a rich farmer from a local village. These local elites are, thus, the major power brokers between the marginalized people and the ruling elites who promote the central government’s control over local resources. Similar situations have been described by Agrawal and Ostrom (2001) in the context of their study of the World Bank’s 155  decentralization approach in India. The UP is also dominated by local elites but, as noted in chapter three, many UP members are appointed rather than elected. Local elites are able to exploit local power structures to ensure their control of UCs and UPs. Tanvir, Tofajjal, Joardar and Billal provided me with details about how this exploitation works at Chapra. Joardar, a small farmer with minimal landholdings, works for Tanvir, a wealthy farmer, needs access to safety net programs like health care or housing and government support to obtain seeds and fertilizer. Joardar also needs to rent land from Tanvir on occasion, for sharecropping. As a sharecropper, he needs to vote in local elections according to Tanvir’s preferences and failure to do so can lead to the loss of sharecropping land or government benefits. Eighty-three percent of my household survey respondents report similar forms of exploitation by local elites. Elites also do not keep election promises to poor farmers, such as the inclusion of Joardar on local resource management committees.  Local Members of Parliament (MPs) also exercise extreme levels of control over power structures at the local government level in ways that are not conducive to democratic practices or inclusion of marginalized people. The MP controls local resource management like irrigation water, seed, food, education, employment, health care and housing by controlling the nominations of chairmen and member candidates for UP and UC elections. Many local elites (all wealthy landowners) practice nepotism or bribery to get these nominations (New Age 2014a). They consider these bribes as investments that make profits for them later through local resource management decisions (Khan 2013). This system effectively excludes marginalized people from receiving a nomination to run for UC or UP positions. This top-down elite domination corresponds to what Scott (1998:6) has termed the implementation of a“high-modernist” agenda. Ferguson similarly (1994:15) argues that “a rural development project is a part of the expansion 156  of the capitalist mode of production, which is often not so good at all for the poor peasants.” Local government involvement in this type of capitalist expansion leads to what many researchers have termed the destruction of “local commons” (Bardhan 1993) and “sovereign selves” (Agrawal 2003).   Water Management   The central government controls the Gorai River Restoration Project and Ganges-Kobodak Project described in chapter three, while local government manages smaller scale water projects, like the Shallow Tube-Well and Deep Tube-Well Projects, created as part of the country’s Poverty Reduction Strategy (International Monetary Fund 2013:283). The large projects are thus controlled by national bureaucracies, while the local projects are controlled by local elites. The average community member has no opportunity to raise his voice in the decision-making process for any of these projects.  Local projects are managed by water management committees at the district and sub-district (upazila) levels (Table 6.1) created on the basis of guidelines provided by the central government. At the upazila level, the UP and the UNO are the chair and vice-chair respectively of the irrigation management committee. Ten of the remaining 11 positions are filled by government staff from various other agencies and only one position is filled by a representative of the more than 200,000 farmers in the Kumarkhali Sub-district. This representative farmer is selected from among the local elites and works as a middleman between local governments and the other farmers in the sub-district. Joardar termed this representative as sorkarer dalal (an agent of the government).   157   Table 6.1: Irrigation Management Committees at District and Upazila Levels District Level Members Deputy Commissioner (Chair) Executive Engineer, Bangladesh Agriculture Development Corporation or Senior Agriculture Engineer, Department of Agriculture Extension (Secretary)  Superintendent of Police Deputy Director, Department of Agriculture Extension Deputy Director, Bangladesh Rural Development Board Representative of the Department of Environment Executive Engineer, Local Government Engineering Department Representative, Health Department Executive Engineer, Bangladesh Water Development Board Executive Engineer, Power Development Board Executive Engineer, Rural Electrification Board Representative, Fisheries Department Sub-District or Upazila Level Members Upazila Parishad chairman (Chair) Assistant Engineer, Bangladesh Agricultural Development Corporation or Department of    Agricultural Extension (Secretary)  Upazila Nirbahi Officer (Vice-Chair) Upazila Agriculture officer, Department of Agricultural Extension Upazila Engineer, Local Government Engineering Department  Officer, Upazila Rural Development Representative, Bangladesh Water Development Board Representative,  Health Department Officer, Upazila Fisheries Department  Representative, Power Development Board/Rural Electrification Board Officer in Charge (OC), Police Representative Farmer Source: Ministry of Agriculture 2011   Table 6.1 clearly demonstrates that there is virtually no involvement from community members in local government irrigation management committees. The upazila committee simply implements the central government’s irrigation policy guidelines. This committee is officially responsible for coordinating with NGOs based on these guidelines. There are no irrigation 158  committees at the union council level, even though unions and villages are the major focusing points for these irrigation management activities of the committees.  None of the water management systems in the Chapra region are being effectively operated. As documented in chapter three, the GRRP is producing further environmental vulnerabilities like flooding and river bank erosion that the local upazila irrigation management committee does not have the authority to correct even if they did have the political will. The GK authority is unable to provide sufficient water supplies due to reduced flows in the Ganges River and has not coordinated their supply system with the agro-ecological system that includes high, middle and low elevation croplands. As focus group respondents Kamal and Kafil informed me, the higher croplands face drought when the project authority provides proper water supplies to lower croplands; whereas, lower croplands face water pooling and stagnation when they provide adequate water supplies to the higher croplands. The central government hoped to overcome these problems by establishing tube-well projects which, though under the control of the richer farmers, have the benefit of localizing the water delivery system. However, these projects are currently facing major challenges as well. Focus group respondents Sorkar, Taher and Kamal informed me the deep tube-wells withdraw arsenic contaminated groundwater and that water levels are dropping. The groundwater depth was 7 meters in 1981 but that had decreased to 3.5 meters by 2010 (Ministry of Finance 2013:61). The upazila irrigation management committee is fully aware of these challenges but the centralized approach of the national government does not provide them with the means to find remedies.  The central government has recognized and attempted to resolve water drainage problems throughout the country, including Chapra where portions of the Kaligangya River were cut off by GK Project canals and portions of other local water bodies were cut off by roads and 159  highways. The central government provided $79,994,500, $170,685,000 and $344,386,000 in 2009, 2010 and 2011 respectively, to overcome water drainage problems throughout Bangladesh (Ministry of Finance 2012). Local government officials and the irrigation management committees control whatever funds come to their region and, once again, project managers fail to consult with community members about these problems. If marginalized people resist the actions of the local committee they are reported to the police as has having carried out illegal activities against government rules and regulations. It is for this purpose that the local irrigation management committee always includes a representative of the local Police Office. This committee structure is a major reflection of the local power structure that benefits the rich people and deprives the marginalized people.   Seed Management  Local elite control of resource management also extends to seed management. As described in chapter five, most farming households find it difficult to produce their own seeds due to lowered crop productivity and the government is actively discouraging the use of local seeds in favor of HYV seeds. Committees are formed at district, upazila and union council levels on the basis of central government guidelines and these committees organize the distribution of HYV seeds and chemical fertilizers and, through safety net programs, give a small amount of seed to farmers who cannot afford to buy it. The UP committees determine who will be licensed to sell seeds and they must submit progress reports periodically to the central government. As in the case of the water management committees, very few local farmers are able to serve on these committees which are dominated by appointed civil servants (Table 6.2). Local elites compete for membership on these committees and use them to promote their own interests.  160  Table 6.2: Seed and Fertilizer Management Committees in Local Governments Sub-District or Upazila Level Members  Member of Parliament  (Advisor)  Upazila Parishad Chairman (Advisor)  Upazila Parishad Vice-Chairmen (Advisor)  Upazila Nirbahi Officer (Chair) Upazila Agricultural Officer (Secretary) Officer, Upazila Animal Resource Officer, Upazila Fisheries Department Officer, Rural Development  Officer, Cooperative Department  Officer in Charge, Police Union Council chairmen  Representative, Bangladesh Agricultural Development Corporation Representative, Bangladesh Fertilizer Association Farmer, nominated by Upazila Parishad   President, Upazila Press Club   Union Level Members Union Council Chairman (Chair) Union Council Secretary (Secretary) Union Council Members  Two respectable local persons, Member of Parliament nominated  Local Imam or spiritual leader, Member of Parliament nominated Representative, Teacher, Member of Parliament nominated Officer, Deputy Assistant Agriculture Officer Upazila Nirbahi Officer nominated  Source: Ministry of Agriculture 2009a  Joardar and Billal argued that local elites control the seed business by acquiring the right to sell the seeds. Local governments collect the HYV seeds from the central government and provide them to local dealers. There are 54 seed dealers in Kushtia and three of them are in Kumarkhali Upazila where Chapra is located (Ministry of Agriculture 2009b). The seed dealers control seed prices and the poorer farmers like Joardar face major challenges for getting seeds of proper quality and price in time for planting. According to Joardar, the committee chairman and 161  its members dismiss those who complain as supporters of the opposition political party who want to tarnish the ruling government’s reputation.  Focus group respondents, Fajal, Faruk, Mofiz and Kusum informed me that the poorest farmers cannot afford the commercial HYV seeds and that the central government has created safety net programs to assist them. The central government provides guidelines for implementing these programs through agricultural rehabilitation committees created at the district, upazila and union levels (Table 6.3). The district committee finalizes the total amount of seed grants for each upazila and distributes them to all upazila within the district. The upazila committees in turn distribute them to all union committees within the upazila. The union committee finalizes a list of the total vulnerable people in a village, then distributes the free or subsidized seeds to the listed people. During the period of my fieldwork, the government was distributing HYV seeds to poor households under two major programs-one for rice, wheat and maize and the other for pulse and oil seed. In 2012, the records of the Ministry of Agriculture reported they provided free Kharif-1 season rice seeds of Bona Aus (Nerica) and Ufsi Aus for 1500 acres of cropland in Kushtia (Ministry of Agriculture 2012). The government gives this seed to local farmers free as a major initiative to popularize this HYV seed. This allocation is only enough for 3000 of the 215,768 marginalized farmers in Kushtia District who own less than 2.49 acres of croplands (Bangladesh Bureau of Statistics 2005). Furthermore, the government only provides this seed for the kharif-1 cropping period and not for two other longer periods of kharif-2 and robi.   All farming households that own less than 2.5 acres of cropland are eligible to get a maximum of one third of an acre of cropland seed. The government has specific guidelines for local decision-making processes of distributing this free seed. The district agriculture  162  Table 6.3: Agriculture Rehabilitation Committees in Local Governments  Sub-district or Upazila Level Members Upazila Nirbahi Officer (Chair) Officer, Upazila Agricultural Officer (Secretary) Officer, Upazila Animal Resources Officer, Fisheries Department Officer, Rural Development  Officer, Project Development Representative, Joint Task Force Representative, Non-Government Officer Union Level Members Union Council Chairman (Chair) Senior Deputy Assistant Agriculture Officers (Secretary) Union Council Members Representative, Non-Government Organization Deputy Assistant Agriculture Officers Source: Ministry of Agriculture 2009    rehabilitation committee finalizes a priority list of the marginalized farmers submitted by the upazila committees. After receiving the approved list from the district, the upazila agricultural rehabilitation committee directly distributes the seeds to the listed farmers. Joardar and Billal insisted that the Union Council committee chairman and members include their supporters on the list of recipients and exclude those who support their political opponents.    Focus group participants Raju, Sabrina and Bakar informed me that many students of Kushtia Government College, who are the ruling government supporters, received the free rice seeds in 2011 although they do not have any involvement in agricultural production. According to them, nobody gets these agricultural benefits who fail to ‘satisfy’ the officials of UC or UP committees. Local elites also exploit the marginalized farmers by other mechanisms, for instance, by extracting free labor. Tanvir exploited Joardar’s labor on his own farm in exchange 163  for the government’s free seeds. This elite exploitation of local seed programs thus increases marginalized farmers’ seed deprivation and agricultural production challenges.    Fertilizer Management   As noted in chapter five, farming households traditionally rely on siltation, algae, earth-worms, cow-dung and water hyacinth as natural fertilizers but are unable to obtain these materials in sufficient quantities today, due to ecosystem failures caused by government interventions and the reduced flow of water in the Ganges Basin. They therefore need chemical fertilizers to maintain crop productivity and the national government has implemented fertilizer distribution programs for this purpose. The fertilizer distribution system is managed by the same committees that oversee seed distribution and with all the same outcomes (Table 6.1). Members of the local elite monopolize the sale of fertilizers at prices that poorer farmers cannot afford and manipulate the distribution of free seeds so as to consolidate their authority over workers and political supporters. Kofil, Tanvir’s son, is a fertilizer dealer in the Kumarkhali upazila. He augments the amount of fertilizer he can sell by obtaining a retail license both for himself and for an appointed salaried worker, Bokul.  Most smallholder and all marginalized farmers cannot afford to buy chemical fertilizers at full market price. The cost of urea fertilizer, for instance, an essential source of nitrogen, was 12 Bangladesh Taka per kilogram in 2009, increased to 20 Taka ($0.26) in 2011 (Ministry of Finance 2013:68). This forty percent increase means that one farmer needs to spend an extra $40 for one acre of cropland. Poor farmers therefore desperately need access to the small amount of free fertilizer that the government distributes. In 2012, the Ministry of Agriculture provided free fertilizer in Kushtia for 1,500 acres of cropland with every farmer owning less than 2.5 acres of 164  cropland being eligible to receive 16 kilograms of urea, 10 kilograms of Diammonium Phosphate (DAP) and 10 kilograms of Muriate of Potash (MOP) to produce the Ufsi aus crops and, 16 kilograms of urea, 10 kilograms of DAP and 10 kilograms of MOP to produce the bona aus crop (Ministry of Agriculture 2012). The total market value of this free fertilizer is $19 but one acre of cropland, on average, requires $130 of fertilizer per year. Farmers desperately need more than the fertilizer the government provides for free. Smallholders who do have some cash assets will thus buy as much additional fertilizer as they can, even at the inflated prices sometimes created by local price-fixing. According to them, local governments ignore their complaints regarding price manipulation due to the nexus between the local government officials and fertilizer dealers. According to Jobbar and Harun, as with complaints about seed distribution, government officials argue that only opposition political party supporters lodge these complaints to tarnish the ruling government image. Consequently, the marginalized people face double vulnerabilities: political repression and agricultural production challenges. In summary, this fertilizer management system promotes elite interests and deprives marginalized people of key agricultural resources.    Food Security Program Management  Local governments’ water, seed and fertilizer management failures inevitably result in major food insecurity for marginalized people. In order to address this problem, the central government created the food safety net programs to secure minimum food rights for marginalized people, based on the guidelines of Poverty Reduction Strategy Paper (Ministry of Food and Disaster Management 2010) and the Millennium Development Goals (Ministry of Food 2013:19). As in the cases described above, the central government places local governments in charge of implementing these programs.  165  As with the water and seed and fertilizer committees, the local government food management committees operate within a local power structure that excludes marginalized people (Table 6.4). Committees at the district, upazila and union levels operate several food programs such as the Vulnerable Group Feeding and Food For Work programs. Some major activities of the district food management committee are to review previous project performances, allocate money to upazila committees for the current program and review project activities. The upazila food committee approves the priority lists of the target people that are submitted by the union council food committees. Based on this approval process, the union council food committee distributes food benefits to eligible people.  Focus group respondents Fazlu, Abonti and Bakar explained to me some of the ways that food programs are used to promote elite interests. Local MPs control the district and upazila food management committees based on their advisory positions. As a result of their political control, the district food committee members are local political leaders who violate the central government’s own guidelines (Raihan 2013). Local elites at Chapra, Tanvir, Mokles and Khalek, are therefore key decision-makers when it comes to developing a list of eligible food benefit recipients. According to my informants, these lists are drawn up on the basis of personal interests, nepotism, bribery and political bias, rather than actual need.   One of the most important and most funded food relief programs is the Vulnerable Group Feeding (VGF) program. This program had a budget of $700,000,000 in 2012-13 and provided food to 400,000 people. People who own less than 0.15 acres of cropland are eligible to receive 10-20 kilograms of rice or wheat a month.   166  Table 6.4: Food For Work Committees in Local Government Sub-district Level Members Member of Parliament (Advisor) Upazila Parishad Chairman (Chair) Officer, Project Implementation Management (Secretary) Upazila Nirbahi Officer (Vice-chair) Upazila Parishad Vice-Chairmen Executive Engineer, Local Government Engineering Department Officer, Upazila Agricultural Department Officer, Upazila Rural Development  Officer, Upazila Social Welfare  Officer, Upazila Account Department Officer, Upazila Fisheries Department Comptroller, Upazila Food Department Officer, Upazila Assistant Engineer, Local Government Engineering Department  Officer, Upazila Youth Development  Union Council Chairmen Two respectable persons, Upazila Nirbahi Officer nominated  One teacher, Upazila Nirbahi Officer nominated   One woman, Upazila Nirbahi Officer nominated Union Level Members Union Council Chairman (Chair) Union Council Secretary (Secretary) Union Council Members Deputy Assistant Agriculture Officer Member, Family Planning Visitor Field Assistant, Bangladesh Rural Development Board Three respectable persons, Upazila Nirbahi Officer nominated One teacher, Upazila Nirbahi Officer nominated One woman, Upazila Nirbahi Officer nominated   Source: Ministry of Food and Disaster Management 2012a   Local governments provide specific benefits to poor households during special occasions like Eid-ul-Azha, an important Islamic holiday. In 2010, the VGF program provided a special food ration to 3.5 million people in 482 upazila for the Eid-ul-Azha celebration. Every VGF card holder received 10 kilograms of rice (Ministry of Disaster Management and Relief 2010). The  167    Figure 6.1: The Central Government’s Food Safety Program Budgets (US$7). Source: Ministry of Finance 2012.  VGF program also provides free food during Eid-ul-Fitr, another religious holiday that occurs at the end of Ramadan. These VGF allocations ensure the participation of marginalized households in important Muslim religious celebrations, as well as provide them with a basic level of nutrition throughout the year. Food is not provided to non-Muslims during religious holidays and the program is discriminatory on that basis. The political motivations and economic self-interest of those who control the food committees also means that many eligible recipients of food do not receive it. Billal, for instance, failed to get the VGF benefits in 2010 because of local elite domination over the VGF program and he sent his wife and children to his father-in-law’s house to celebrate the Eid-ul-Azha holiday. Abonti took an NGO loan to buy special food for his children during the Eid-ul-Fitr. This exclusion and multiple socioeconomic challenges are an every year experience for marginalized people at Chapra.                                                            7 US$ 1=BDT 80.3028, June 6, 2012  0100,000,000200,000,000300,000,000400,000,000500,000,000600,000,000700,000,000800,000,000General ReliefActivitiesVulnerableGroupDevelopmentVulnerableGroup FeedingTest Relief  Food Gratuity ReliefFoodFood For Work2012-13 2011-12 2010-11 2009-10168  The central government has also funded a Food for Work (FFW) program, a sixty day program that generates employment opportunities for marginalized people. One employed person, based on this program, is eligible to receive eight kilograms of rice or wheat a day in payment for providing labor for the construction or maintenance of local infrastructure, such as GK project canals, embankments and roads (Ministry of Finance 2012). However, the FFW program goals are thwarted by the local power structures and FFW program implementation lacks transparency and accountability. Based on Bangladesh Bureau of Statistics (2005) data, forty-one percent of people in Kushtia who own less than half an acre of cropland report they did not receive FFW program benefits in 2010. Although the safety net programs have the capacity to mitigate food shortages, they are mismanaged and do not operate on a scale sufficient to alleviate more than a small percentage of the food shortages being experienced in Chapra and throughout Bangladesh. As I will argue in the next chapter, since these food shortages are entirely the outcome of human intervention, they constitute an abuse of the human rights of smallholder farmers and agricultural laborers in this setting.  Employment Programs Management As documented in chapter five, employment patterns at Chapra have changed radically over the past several decades. New jobs have been created in sectors like fish farming, tube-well construction and maintenance, and operation of the new farm machineries used for paddy husking and plowing. These new jobs are far fewer in number than the jobs lost in traditional fisheries, conventional irrigation system maintenance and the practice of traditional farming techniques that include the collection of many freely available natural resources for use as fertilizer, for the construction of tools or for sale in local markets. To mitigate these losses local 169  governments have created various employment management committees at the district, upazila and union levels based on guidelines provided by the central government.  Local governments manage several employment programs some of which are specialized in the economic sectors they wish to promote. They include: (1) the Employment Generation Program for the Poorest, (2) Fund for Assistance to the Small Farmer and Poultry Farms, (3) Swanirvar or Independence Training Program, (4) Employment Generation for the Ultra Poor, (5) Rural Employment for Public Asset, (6) Rural Employment and Rural Maintenance Program, (7) Micro-credit for Women Self-employment and (8) Jatka (Fish) Protection and Alternative Employment for Fishermen. The central government provided a total of $498,115,000 over four years from 2009-10 to 2012-13 to promote employment opportunities for 400,000 marginalized people in the poultry industry through the Fund for Assistance to the Small Farmer and Poultry Farms. This program pays for the importation of poultry species that are used to displace domestic species considered not suitable for commercial poultry operations. A second program, the Swanirvar program, trains unemployed people in how to set up and operate these poultry businesses.   In this section, I focus mainly on a less specialized form of assistance, the Employment Generation Program for the Poorest (EGPP) (Table 6.5). In addition to the employment management committees at district, upazila and union levels, the EGPP program requires a Project Implementation Committee (PIC) at the bottom level of the committee hierarchy. The UC chairman nominates five to seven UC members to serve on PIC excluding all marginalized people even though they are the programs’ target group. The union committee is responsible for EGPP project implementation while PIC looks after everyday activities such as taking attendance and paying the employed people. 170   Table 6.5: The EGPP Committees at Upazila and Union Levels Upazila Level Members Member of Parliament (Head Advisor) Upazila Parishad Chairman (Advisor)  Upazila Nirbahi Officer (Chair) Officer, Upazila Project Implementation (Secretary) Upazila Vice-Chairmen Union Council Chairmen Officer, Upazila Health and Family Planning  Officer, Upazila Agricultural Department  Officer, Upazila Fisheries Department  Officer, Upazila Engineer, Local Government Engineering Department Comptroller, Upazila Food Department  Office, Upazila Education Department Officer, Upazila Youth Development  Officer, Upazila Women Affairs  Officer, Upazila Cooperative Department  Officer, Upazila Animal Resources  Officer, Upazila Social Welfare Officer, Upazila Ansar & Village Defence Party Officer, Upazila Rural Development Member, Upazila Field Supervisor, EGPP Teacher, Upazila Nirbahi Officer nominated Volunteer representative, Upazila Nirbahi Officer nominated Manager, Upazila level Banks                One respectable person, Upazila Nirbahi Officer nominated   Union Level Members Union Council Chairman (Chair) Union Council Secretary (Secretary) Union Council Members Union Deputy-Assistant Agriculture Officer Manager, Local Bank  Field Assistant, Bangladesh Rural Development Board  Women Member of Union Council reserved seats, Upazila Nirbahi Officer nominated One teacher, Upazila Nirbahi Officer nominated Source: Ministry of Food and Disaster Management 2012a   171  Focus group respondents Suman, Jobbar, Fazlu and Kamal reported that the ruling elites control the EGPP for their own interests just as they do all the other local committees. Many eligible participants, including Suman and Jobbar, report not even receiving news about their eligibility for EGPP employment. Fazlu and Kamal also pointed out that there is a huge gap between the total number of unemployed people and the number employed by government programs. In 2010-11, for example, the central government offered EGPP opportunities to 2,380 out of 98,450 eligible people in the Kamarkhali Upazila in Kushtia (Ministry of Food and Disaster Management 2012b). Suman and Jobbar argued that the ruling elites in local governments exploit the gap between government employment opportunities and the total number of unemployed in order, once again, to consolidate their positions within the local power structure.  Focus group respondents Jobbar, Fazlu and Mofiz reported that local elites will sometimes pay them from EGPP project funds, rather than own personal funds, for work they perform for the elite farmers on their farms, work that is not part of the EGPP program. Jobbar and Fazlu, for instance, will be given work at Tanvir’s deep tube-well project when they are supposed to work on GK project canal maintenance. They do not protest this malpractice out of fear of losing the work. They also reported that, many people whose names are on the EGPP benefits list are local political leaders and their family members including brothers, brothers’ wives or domestic servants. None of them work on approved EGPP projects but they all receive full payment from EGPP funds. Local elites are also reported to put false names on the EGPP list in order to gain access to additional benefits. Local government authorities know about many of these malpractices but they fail to execute legal action, due to their own vulnerability within the local power structure.   172  Alamgir, Bakkar and Faruk argued that the employment programs actually increase their employment challenges. This occurs because the central government prefers to fund projects that displace local practices and further erode their self-sufficiency. The central government provides micro-credit loans to help women establish poultry farms using imported species but focus group respondents Abonti, Gedi, Kusum and Morjina argued that they would like support for their contribution to domestic species production. The Swanirvar program provides skill development for raising foreign poultry species but this displaces local ecological knowledge and creates dependencies on distant markets and vulnerability to market conditions they cannot control. The rural infrastructure maintenance programs help maintain government infrastructures like the GK project but are not directed towards the restoration of local water bodies that are blocked by GK project infrastructure. My informants also pointed out that the central government’s employment programs do not recognize traditional occupational groups like blacksmiths and potters. These top-down employment programs promote elite interests and exclude marginalized people and subsistence livelihoods.  The bureaucratic system of payment for those who do gain access to these programs also creates problems. Suman, Alamgir and Kamal, for instance, report having to open bank accounts in order to be paid by the EGPP program. Many of them have never had bank accounts and do not understand the procedures and requirements for opening an account. Many do not have property documents, for instance, or other forms of documentation necessary to confirm their identity. Suman reported that bank officers treat him as a low caste person whenever he has bank business, a process that reinforces his sense of marginalization and discourages his participation in government employment programs.  173  Education Program Management   Survey and focus group respondents agreed that their ability to provide education to their children is compromised when agricultural production is low or employment opportunities are not available. My informants disagreed, however, on the type of support they wanted from government in the area of education. Barek and Morjina stated that they would prefer government support for agricultural production and employment, so they could pay education costs themselves, rather than have the government create special support programs for the poor. Parents from different socio-economic backgrounds also place very different demands on the public education system. Tanvir, a wealthy landowner, is interested in having his children obtain enough education to maintain the family status in the local power structure. Tofajjal, an intermediate landowner, argued that a good education is a better investment for his children than money spent in the agricultural sector. Joardar, on the other hand, a smallholder, desperately wants education for his children in the hope that they can reverse the pattern of marginalization he has experienced when it comes to participating in local resource management processes. Ninety-four percent of household survey respondents hold the belief that educated children could help them get better access to government programs and services.  Billal, a landless laborer, dreams of his second son, Lablu, becoming a government magistrate in order to protect their rights. However, 67 percent of the marginalized household survey respondents reported not being able to afford the costs necessary to educate one or more of their children.    Local governments administer education support programs at the district, upazila and union levels based, as always, on the guidelines of the central government. There is also a School Management Committee for every school that is entrusted with the task of implementing the actual programs.  174  Parents do not need to pay tuition or school fees for their children in order to attend primary and secondary institutions up to grade 10. However, they need to pay fees for grades 11 and 12 and university. Despite this free education to grade 10, poor households cannot afford other educational expenses such as books and school uniforms. To assist with these costs, the central government has created eight scholarship and grant programs: (1) Stipend for Disabled students, (2) Grants for Schools of the Disabled, (3) Stipend for Primary Students, (4) School Feeding Program, (5) Stipend For Dropout Students, (6) Stipend For Secondary and Higher Secondary Level Students, (7) Preliminary Education for Development of Children and (8) Post Literacy Education Project (Figure 6.2). These education programs are developed based on policy guidelines of Poverty Reduction Strategy Paper (Ministry of Food and Disaster Management 2010). The government is also decreasing government education budget and encouraging private sector involvement in establishing new schools. In 2012, there were three government colleges and 10 schools in Kushtia District whereas there were 30 private colleges and 173 private schools (Kushtia Zila Samity 2012).       Figure 6.2: The Eight Scholarship and Grant Program Budgets. Source: Ministry of Finance 2012.    01000000002000000003000000004000000005000000001 2 3 4 5 6 7 82012-13 2011-12 2010-11 2009-10175  In addition to the fact that school costs are more than some households can manage, some of my focus group respondents indicated that they need to keep children home to help them with basic subsistence tasks. When I asked Billal about access to formal education for his children, he expressed a deep concern with the comment “bhaat paina, shikhya dibo kibabe?” “How can we afford education for our children when we do not have enough food for the whole family?” Many families are thus forced to make choices between food and education. Suman and Sabbir reported making different choices for their male and female children. Based on their limited economic capability, they prefer to educate a son than a daughter with their logic influenced by bongsher baati (line of inheritance8). The governments operate school food programs but one focus group respondent, Tarun, informed me that his children were not able to benefit from this much valued program because his children do not have suitable conditions at home to study and he therefore cannot send them to school. He cannot afford a light for his children to study by at night and his whole household resides in a single cottage, locally called kureau ghor, where his children do not have even a minimum study space. Many children with these economic challenges, fail to attend school on a regular basis, do not complete their grades and are, therefore, not eligible for scholarship benefits.  Many focus group respondents, including Kamal, Rahim and Harun and Borkot, stated that scholarship and grant programs are currently under political control. As in the case of other social services, local MPs and their supporters on district, upazila and union councils control the school management committees for their personal interests. Head teachers are dependent on the upazila and district administration for their promotion, salaries and job postings. Many staff members in schools use their time doing private consultancies and are dependent for their                                                           8 Inheritance of land is usually through the male line. Females acquire rights to land only though marriage and upon marriage normally go to live on the land of their husband’s family. 176  positions on nepotism and political support for local elites. The school management committees are unable to take legal action against these malpractices even when some of their members would prefer to do so. Consequently, many of the program benefits end up going to the children of rich farmers and their relatives, while children from poor households receive nothing and are forced to drop out from school in high numbers. Consequently, 84 percent of my marginalized survey respondents reported not receiving any benefits from the government’s Stipend For Secondary and Higher Secondary Level Students. Among these respondents, fifty percent dropped out from school before completion of grade five.   Health Care Program Management  Local governments perform important roles in implementing the central government’s health care programs. Unfortunately, these health programs are also controlled by local elites similar to the education programs. Illness is part of the vicious cycle of marginalization for poor households, the result of too little food and poor nutrition, which then, in turn, causes food production and household income to drop still more. When a marginalized household head encounters sickness, the whole household faces economic vulnerabilities and multiple sicknesses. Fifty-six percent of household survey respondents report encountering sickness in 2010 and needed support from local government to regain their health and wellbeing. Seventy-two percent of my survey respondents answered yes to the question: “did you encounter agricultural production failures due to sickness? Fifty-two percent reported sickness associated with crop failures during all three cropping periods.  According to focus group respondents Bakar, Borkat, Suman and Sabbir, many of the illnesses they experienced are preventable but preventive health care is not funded by the 177  central government and they cannot afford to pay for curative health services. Currently, local people suffer from high incidences of malnutrition as a result of food shortages and from diarrhea caused by unclean drinking water and water contaminated by chemical fertilizers, insecticides, fungicides and herbicides. Sixty-one percent of survey respondents report health problems in 2010, caused by unsafe drinking sources. Some of these problems could be resolved by government health care programs that provide benefits such as free tube-wells and sanitary latrines but local elites control these resources. Eighty-four percent of my survey respondents report having no access to tube-wells or sanitary services. Some foods, notably fish caught in local water bodies subject to agricultural run-off, are also a source of illness. Because of the overuse of TSP fertilizer on rice crops, one kilogram of rice in Kushtia, on average, has been found to contain 0.099 milligrams of cadmium (Bhuyan 2013). Cadmium, like other heavy metals such as mercury, can create diseases like kidney failure and cancer. As stated earlier in this chapter, deep tube-wells often withdraw arsenic contaminated water and this is also a source of health problems. The Farakka diversion in India can be considered a significant contributing factor for arsenic contamination since the deep tube-wells have been built mainly as a way to make up for the lack of natural flow in the Ganges and Gorai Rivers (Rahman et al. 2010). Currently, about 85 million people in Bangladesh are facing arsenic contamination in drinking water and field crops. Forty-four percent of the total population in Kushtia is facing arsenic contamination (Hossain 2006). Despite the pervasiveness of illnesses caused by environmental factors of this type, the central government health care services do not include any preventative measures and do not  monitor or enforce water quality regulations. The failure to provide preventive health services dramatically increases the disease burden for poor households at Chapra. Many diseases like cadmium poisoning are new to them 178  and have no remedy within traditional medical systems. To make matters worse, the central government is reducing their expenditures on doctors, nurses and new medical technologies based on the terms of the current structural adjustment program. Community clinics and NGO-operated first aid service centers are being introduced based on the guidelines of the Poverty Reduction Strategy Paper and Millennium Development Goals. This further contributes to the commodification of local health care services. Thirty-five community clinics have now been established in Kushtia (Ministry of Health & Family Welfare 2013) mainly in urban areas. The privatization of health services benefits those who can afford the services but this is not the case for most marginalized people. Eighty percent of my survey respondents report not being eligible able to receive needed health services due to lack of ability to pay. Since the central government understands that many people cannot afford privatized services, they have encouraged more NGO involvement in providing health services. Figure 6.3 depicts the pattern of health service provision in rural Bangladesh as a whole.               Figure 6.3: Percentage of Health Care Sources in 2010 in Rural Bangladesh. Source: Bangladesh Bureau of Statistics 2010.   2.57 0.31 3.49 1.13 0.31 8.52 13.11 0.22 25.04 41.2 0.9 0.58 2.62 051015202530354045179   Standard health services in Bangladesh are generally organized in four tiers. Division level hospitals and medical colleges service the several districts located within them;9 district level primary hospitals service the district population, an upazila health complex services the sub-district population and union level health and family welfare centers service a village or group of villages. The Kumarkhali Upazila Health Complex Bulletin (2012) indicates there are twelve doctors at this facility, serving a population 352,210 people in the Kumarkhali Upazila. Chapra residents can also access health services at two Chapra Union Sub-Centers (Ministry of Health and Family Welfare 2014). These public health services are not enough to ensure the health and wellbeing of marginalized people at Chapra. Abonti and Keramot informed me that one health officer with rudimentary training, operates one center and they are not able to provide treatment for serious illnesses like arsenic poisoning. The center is less than one standard apartment bedroom in size or about two square meters.  In addition to the health services provided by the above institutions, the central government has introduced the following seven health care safety net programs: (1) Maternal Health Voucher Scheme, (2) National Nutrition Program, (3) Micro-nutrient Supplementation, (4) Revitalization of Community Health Care Initiative in Bangladesh, (5) National Sanitation Project, (6) Special Program for Irrigation and Water Irrigation and (7) Maternity Allowance Program for Poor Lactating Mothers (Figure 6.4). These programs were developed based on the Poverty Reduction Strategy Paper and Millennium Development Goals guidelines (Ministry of Planning 2013; International Monetary Fund 2013: 143).                                                            9 Bangladesh is divided into seven divisions and each division is further sub-divided into several districts. There are a total of 64 districts in the country.  180  Local governments implement these programs for different target groups. The maternal health voucher scheme and maternity allowance programs seek to assist mothers who do not have the economic capability to maintain good health during pregnancy. In 2012-13, $933,965,000 was spent nationally on this program. Nutrition and micro-nutrient programs (numbers 2 and 3 above) represent a significant effort to overcome malnutrition in poor children. However, 90 percent of focus group respondents report not receiving any safety net program benefits.    Figure 6.4: The central government’s Health Care Programs from 2009-10 to 2012-13. Source: Ministry of Finance 2012.  Focus group respondents Borkat, Fazlu, Kabir and Mofiz argued that, in this case, too, local elites control the central government’s safety net programs and exclude most marginalized people at Chapra from receiving benefits. Tanvir did not have any problem obtaining health services in the Kumarkhali Upazila Health Complex. Medical staff are always available for him and other local elites. But many of my focus group respondents stated that doctors sell hospital medicines and use government medical technologies illegally for personal gain. They also do private consultancies even during hospital office hours. Consequently, marginalized people like 0500000001000000001500000001 2 3 4 5 6 72012-13 2011-12 2010-11 2009-10181  Billal, Fazlu, Kabir and Mofiz receive inadequate or no care. Sometimes they are forced to buy medicine from local pharmacies, even when these medicines are supposed to be available for free to hospital patients.   Housing Program Management  Housing is an issue for many poor households in Chapra due to the limited availability of building materials like wood, which could formerly be obtained for free in the local environment, and due to the fact that, many people have lost their homes due to river bank erosion or displacement by GK canal infrastructure. Low wages and limited employment opportunities also make it difficult to buy the corrugated iron they generally use for roofs.  The central government has implemented six housing safety net programs for poor people throughout Bangladesh: (1) Housing Support, (2) Ashrayan Project, (3) Gucchagram, (4) One Household One Farm, (5) Comprehensive Village Development and (6) Char Development and Settlement Project.  As indicated in figure 6.5, significant amounts of money are spent on these programs. Ministry of Disaster Management and Relief (MoDMR) provides housing grants to the Director General of Disaster Risk Reduction, which then allocates housing credits to the district councils throughout the country. The Deputy Commissioner of each district then distributes the housing credits among Upazila Councils and the Upazila Nirbahi Officer (UNO) is then directly responsible for distributing these credits to the eligible people.   182    Figure 6.5: The Governments’ Housing Safety Net Budgets (in $US). Source: Ministry of Finance 2012.   Different housing safety net programs focus on different services. The ashrayan project provides assistance to displaced people. This project builds houses in a specific area where local governments rehabilitate displaced people. The central government increased this project budget from $4,069,600 in 2009-10 to $22,415,200 in 2012-13. The Gucchagram assists people who are considered to be climate victims. The budget for this program was $10,646,000 in 2009-10 but was reduced to $7,755,640 in 2012-13. The one household, one farm program provides eligible people with a small piece of land and money for housing materials. The budget for this program increased from $996,229 in 2009-10 to $58,528,500 in 2012-13. The government has also introduced a char development program to settle marginalized people on charlands but the Gorai River charlands at Chapra are not included in this budget. Although overall annual expenditures total several hundred million dollars, these programs assist only a small number of all those who are in need. Also, as is the case with the other programs described above, local elites are able to capture many of these benefits for themselves, thus reducing the effectiveness of the programs still further.  Joardar reports having visited UC officials several times about his housing situation but received no help.  He lost his ancestral home in 1960, when the government expropriated some 050,000,000100,000,000150,000,0001 2 3 4 5 62012-13 2011-12 2010-11 2009-10183  of his land for construction of the Ganges-Kobodak Project. He built another house on the edge of his cropland that was destroyed by river bank erosion in 1987. He then applied to the GK authority for permission to erect a dwelling on their land and was allowed to build a small shack on a flood protection right-of-way, where he now lives along with his wife and children. Tanvir, meanwhile, is one of the major beneficiaries of local governments’ housing programs. In addition to his house at Chapra, Tanvir has a house in Kushtia City and one in Dhaka that are used by his children. According to Joardar, Tanvir does not follow government procedures regarding housing safety net benefits. He puts fake names on the beneficiary list and sells the housing benefits for personal profit. As a result of these illegal appropriations, Billal is also unable to receive housing assistance. He and his family currently reside within six meters of the Gorai River and face imminent displacement and loss of housing due to flooding, embankment failure and river-bank erosion.  Focus group respondents Borkat, Jobbar, Joardar and Billal also stated that local elites exploit displaced people by offering them informal housing opportunities on government lands, such as those belonging to the GK project, and then collect tolls from them or require them to work for no wages. Sometimes a member of an elite family will force a displaced person to commit criminal activities, such as smuggling or trafficking, in return for a place to live. Bakar, Fajal and Joardar stated that vulnerable people will also be coerced into political activities, including the intimidation or beating of supporters of other parties. After a few years, however, they are generally evicted from their informal housing and sometimes their houses will be burned down.     184  The Consistent Failure of Safety Net Programs In this chapter I have documented the consistent failure of local government institutions to alleviate the chronic problems faced by the majority of rural households living in Chapra. These hardships begin with water shortages, ecosystem service failures and reduced crop productivity. They are made worse by agricultural modernization programs that are controlled by local elites working in concert with national political leaders and national and foreign corporations. The modernization programs benefit wealthy and intermediate farmers but reduce employment opportunities for the majority of households. These economic crises in turn lead to food shortages, health problems, low levels of educational achievement and inadequate housing for large numbers of people. Even though hundreds of millions of dollars are spent annually on safety net and agricultural subsidy programs, the scale of the problem is well beyond the capacity of these programs to remedy and the little good they might do is undermined by corruption and manipulation by local elites. Government leaders, planners, aid agencies and international development banks represent the chronic problems faced by Chapra residents as the result of insufficient development but, in the next chapter, I propose an alternative theory, one that focuses on these problems as human rights abuses caused by too much development in the service of ecocracy.     185  Chapter Seven HUMAN RIGHTS VIOLATIONS AT CHAPRA  In Chapra, access to a sufficient and predictable supply of water is the prerequisite for realizing other human rights to food, employment, health care, education and housing. The ability to realize these human rights is undermined by the effects of Farakka diversion constructed by India and the highly centralized, neoliberal approach to water management and agricultural development in Bangladesh. Political and technological interventions in the Ganges dependent area have resulted in the economic marginalization of the vast majority of households living in Chapra and in many other similar rural GDA communities. My goal in this chapter is to evaluate the degree of hardship experienced by marginalized households in relationship to international human rights agreements and international customary law in respect to water. If it can be demonstrated that these hardships do constitute human rights failures then the governments of India and Bangladesh, as the main agents of marginalization, can legitimately be held responsible for these failures.   I rely on several United Nations agreements and conventions in my analysis and on the Berlin Rules created by the International Law Association. Most important among the many applicable UN agreements are the 1948 Universal Declaration of Human Rights, the 1966 International Covenant on Economic, Social and Cultural Rights, a 2002 amendment to the 1966 Covenant referred to as General Comment 15, the 1997 United Nations Convention on the Law of the Non-Navigational Uses of International Watercourses and the 2010 General Assembly Resolution on the Right to Water and Sanitation. The 1948 UN Declaration of Human Rights recognizes basic necessities like food, employment, education and housing as human rights. The 186  1966 Covenant, which came into force in 1976 and was subsequently ratified by the governments of both Bangladesh and India, provides additional details about rights to education, health and an adequate standard of living. In 2002, General Comment 15, which deals explicitly with water and the relationship of water to other human rights, was added to the Covenant. The 1997 UN Convention on International Watercourses specifically addresses the responsibilities of states in respect to international rivers. The General Assembly Resolution of 2010 reaffirms the human right to water, citing all previous relevant agreements and covenants approved by the General Assembly. The Berlin Rules on International Water Resources, which was finalized in 2004 by the International Law Commission, is the most comprehensive statement globally in respect to customary international water law. The 2004 Berlin Rules were developed as a revision of the 1966 Helsinki Rules in light of changes to international law including those brought about by 1997 United Nations conventions on International Watercourses. Unlike the 1966 UN Covenant, neither the 1997 UN Convention or the 2010 General Assembly Resolution have been ratified by either India or Bangladesh; however, international customary law does have legal standing in all countries and in international courts even when it is not formally recognized by a given country and no local enforcement mechanisms are in place.  Human rights violations in Chapra are inter-generational when families are displaced permanently from their ancestral homes with no chance of recovering them. Moreover, these violations are mutually reinforcing since lack of food, for instance, causes malnutrition and health problems which in turn further increase the causes of unemployment and a deepening cycle of food insecurity.   In this chapter, I describe in detail the nature and extent of the hardships being experienced by Chapra villagers in respect to the six human rights identified in Table 7.1 (water, 187  employment, food, education, health and housing). I conclude with a consideration of how a better understanding of the relationship of river water to human rights might contribute to the restoration of human rights through community empowerment initiatives and improved policies for environmental governance. Table 7.1 demonstrates the percentage of households in my survey that I consider to fall below basic human rights standards in the six areas included in this study. These figures are based on responses to several survey questions about the status of these rights in Chapra. The survey was not designed primarily to facilitate the quantiative analysis of human rights violations but by summarizing response rates to several questions in each area of concern, it is possible to provide rough estimates of the percentage of households experiencing violations. The factors used to assess violations are provided in a series of tables below, one table for each human right category that is being assessed. Each factor within a table corresponds to one survey question. The response figures for each factor are averaged within each table in order to provide an estimate of the percentage of households experiencing human rights violations in each of the six human rights categories under study. Table 7.1 provides the summary results of these calculations.   Table 7.1: Human Rights Violations at Chapra  Human Rights Percentage of Households Falling Below Minium Standards Sufficient water for basic needs  79 Adequate level of employment 77 Adequate level of food consumption 85 Reasonable access to formal education 79 Reasonable access to health care 66 Secure housing 73  188  I begin with a discussion of water as a human right and then examine, in turn, employment, food, education, health care and housing.    Water Rights Violations My evidence, as summarized in Table 7.2, indicates that the water rights of the majority Chapra households, as defined by several international agreements and conventions, are being violated. The most important provisions clearly being violated are: 1) those pertaining to access by all citizens to “sufficient” and “affordable” water for basic needs, including agricultural subsistence; 2) the obligation of one state to not harm another by unilateral withdrawals of water from a shared river; 3) the special responsibilities of all states towards vulnerable communities; 4) the rights of affected people to participate in decision-making processes; and, 5) the protection of “the ecological integrity” of a watershed. I discuss these factors in this order below, presenting quantitative information from my household survey when available and otherwise relying on qualitative information from focus group and case study informants.   Table 7.2: Water Rights Violations at Chapra Factors Preventing Access to an Adequate Supply of Water Percentage of respondents reporting this problem Insufficient local water supply due to reduced flow in the Ganges River 100 Water Stagnation problems  62 Gorai River channel change problems  90 River bank erosion problems  84 Embankment failure problems in 2011 63 GK Project water supply problems  68 GK Project’s source water point affected by the Ganges water reduction   84 Water rights violations (average) 79  189  Water Access for Basic Needs Article 17 of the Berlin Rules (International Law Commission 2004:17) provides one of the clearest legal statements on “the right of access to water.” Article 17.1 states that “every individual has a right of access to sufficient, safe, acceptable, physically accessible, and affordable water to meet that individual’s vital human needs.”  Article 17.2 states that right of access must be available on a “non-discriminatory” basis. Article 17.3.a specifies the need for the state to not interfere in an individual’s exercise of their right to water. Similar water rights definitions are included in articles 11 and 12 of the United Nations General Comment 15 (2002). Data presented in chapters four, five and six demonstrate that a majority of poor households in Chapra experience chronic shortages of the water required to grow subsistence crops; the water used for drinking, cooking and bathing is regularly contaminated by agricultural chemicals and pathogens associated with standing water; water access is highly discriminatory; the states of both India and Bangladesh have interfered with people’s abilities to access the water they need for  basic human needs. Ninety-five percent of my survey respondents informed me that proper water access failures are responsible for agricultural failures. Their primary source of water was formerly the Gorai River and the failure of this river forced them to depend on GK Project or tube-wells. However, eighty four percent of my survey respondents believe that the Ganges Basin flow reduction due to the Farakka diversion causes the GK Project water supply to be insufficient. Furthermore, seventy eight percent of survey respondents fail to get sufficient support from GK Project staff to resolve their water problems. Neither did they get any cooperation from the local water management association. Local elites control the water management association in such a way that seventy four percent of marginalized households report not receiving minimum information about their activities. When they fail to get affordable 190  water access from rivers, streams or GK canals they are forced to use purchase water from tube wells. Despite this alternative mechanism, sixty three percent of my survey respondents report experiencing agricultural production damages and losses due to water supply problems. Their suffering increases further with environmental vulnerabilities like water stagnation. Among my survey respondents, sixty two percent report water stagnation problems in their neighborhoods, which cause multiple problems like health and crop production concerns. These problems are escalated with the river bank erosion, which is encountered by 76 percent of my survey respondents. Among them, 44 percent encountered this erosion three to four times in their lives. Sixty-one percent of my survey respondents do not have access to safe drinking water sources such as uncontaminated tube wells or hand-dug wells known as kuya. Instead, they use sources like stagnant water that is often contaminated with agricultural run-off. Ninety-one percent of my survey respondents report no longer practicing traditional sports and festivals due to the Gorai River water failures. Some of these celebrations, like boat races, occur less often because of reduced water flow, while the others, like pitha (sweetmeat) festivals, occur less often because of crop production failures and a lack of cash to buy the requisite ingredients.    State Obligations in Regard to Shared Watercourses As documented in chapter three, the flow of Ganges River water to Bangladesh during the dry season is now less than one-third of its volume prior to the construction of the Farakka Barrage and about two-thirds of its volume during the wet season. This is a major cause, though not the sole cause, of water shortages in Chapra. This represents a clear infringement by India of Article 7.1 of the UN Convention (United Nations 1997) and Articles 12.1 and 16 of the Berlin Rules (International Law Association 2004:20,22). Article 7.1 states that “watercourse States shall, in 191  utilizing an international watercourse in their territories, take all appropriate measures to prevent the causing of significant harm to other watercourse States.”  The government in India violates these articles not only through the operation of the Farakka Barrage but also through the operation of a whole series of hydro-infrastructures throughout the Ganges Basin in India (Khan 1996). Article 12.1 of the Berlin Rules specifies that, “basin States shall in their respective territories manage the waters of an international drainage basin in an equitable and reasonable manner having due regard for the obligation not to cause significant harm to other basin States.” Article 16 of the Berlin Rules states that, “basin states, in managing the waters of an international drainage basin, shall refrain from and prevent acts or omissions within their territory that cause significant harm to another basin State having due regard for the right of each basin State to make equitable and reasonable use of the waters.” Due to violation of these articles, the flow of water in the Gorai River, a distributary of the Ganges, is reduced well below the one-third level of the Ganges itself due to sedimentation that is particularly intense around the mouth of the river. Chapra villagers during the dry season do not receive the minimum amount of river water they need for agricultural production. Only four percent of my household survey respondents report being able to access Gorai River water for agricultural production. However, one hundred percent of local people used this river and other people distant from the river, report using local ponds, wetlands and lakes before the Farakka diversion.  Since the government of Bangladesh has not been able to prevent India from withdrawing this amount of water from the Ganges River water, due to the nature of hydro-politics in the region (Bhattarai 2009; Brichier-Colombi and Bradenock 2003; Elhance 1999; Hossen 2012; Iyer 2008; Nakayama 1997; Paisley 2002), they have failed in their obligations to “respect,” “protect” and “fulfil” the water rights of their citizens as outlined by the Berlin Rules 192  (International Law Association 2004). Additionally, the government of Bangladesh is responsible for worsening water shortages through its own top-down water management programs like the Ganges-Kobodak and Gorai River Restoration Projects.   Vulnerable Communities and the Right to Participate in Decision-Making Processes Data presented in the preceding chapters also makes it abundantly clear that the process of marginalization of smallholder farmers and farm laborers has occurred as a result of their exclusion from decision-making processes and the ability of local elites to control local water management institutions and decision making. Class relations have a long history in this region and some degree of marginalization and poverty was certainly in place before the 1970s. Marginalized households, however, did not previously exist in the numbers they do today and the scale of previous hydrological interventions was not such that people could be displaced or impoverished in such large numbers. With the encouragement of international donor organizations, the government has introduced a number of safety net programs to address the needs of vulnerable communities but have done nothing to reverse the underlying cause of their vulnerability. This clearly contravenes Article 20 of the Berlin Rules which reads: “States shall take all appropriate steps to protect the rights, interests, and special needs of communities and of indigenous peoples or other particularly vulnerable groups likely to be affected by the management of waters, even while developing the waters for the benefit of the entire State or group of States” (International Law Association 2004:26). Chapra residents are also entitled to compensation based on article 21 of Berlin Rules that specifies “duty to compensate persons or communities displaced by water projects or programs” (International Law Association 2004:27). This compensation has international and national 193  dimensions. Local people in the Ganges Dependent Area should receive the compensation from the Government of India due to the harms created by the Farakka diversion. For example, India paid this compensation to Pakistan (Parua 2010:252). In addition to this international aspect, the government in Bangladesh creates much harm to local communities with water and agricultural development programs. My data documents the fact that many households have lost land due to project interventions and have been entirely displaced on some occasions with little or no compensation. Forty-three percent of my survey respondents report losing lands to the GK Project construction and thirty-eight percent report losing more than one acre of land. The government in Bangladesh provided compensation for GK Project expropriations; however, this compensation was controlled by local elites and often did not find its way to marginalized people. The government did not provide any compensation to the victims of the Gorai River bank erosion. This erosion caused the loss of more than three acres of land for 46 percent of my survey respondents. This includes lands used for crops storage.    Ecological Integrity Water rights violations at Chapra create ecological resource rights violations. As noted throughout this dissertation, the natural resources formerly available to Chapra residents as common property resources are vital to their sustainability. Several articles in the Berlin Rules and the 1997 UN Convention use the concepts of ecological integrity and ecological flows to define the nature of the customary water rights in this domain. Article 3.6 of the Berlin Rules defines ecological integrity as “the natural condition of waters and other resources sufficient to assure the biological, chemical, and physical integrity of the aquatic environment” (International Law Association 2004:9). Article 3.1 of the Berlin Rules describes the aquatic environment as “all surface and groundwater, the lands and subsurface geological formations connected to those 194  waters, and the atmosphere related to those waters and lands” (International Law Association 2004:9). Article 20 of the 1997 UN Convention provides guidelines for “the protection and preservation of ecosystems” and states that “watercourse States shall, individually and, where appropriate, jointly, protect and preserve the ecosystems of international watercourses” (United Nations 1997).  Article 15 of the Berlin Rules emphasizes that water rights include protection of “water necessary to assure ecosystem services or otherwise to maintain ecological integrity or to minimize environmental harms” (International Law Association 2004:22). Reduced river flows at Chapra reduce local ecosystem services like cropland siltation, wild fisheries, wild vegetables and vegetation. All of my survey respondents unanimously argued that local water bodies like Chapaigachi oxbow lake, Lahineepara, Shindha, and Shaota canals, and Lahineepara and Chapra Fakirapara wetlands are facing survival challenges due to the Ganges-Gorai River ecosystem failures. The demise of these water bodies violates article 22 of the Berlin Rules that reads: “States shall take all appropriate measures to protect the ecological integrity necessary to sustain ecosystems dependent on particular waters” (International Law Association 2004:28).  Furthermore, article 24 of the Berlin Rules specifies that “States shall take all appropriate measures to ensure flows adequate to protect the ecological integrity shall take all appropriate measures to ensure flows adequate to protect the ecological integrity of the waters of a drainage basin, including estuarine waters” (International Law Association 2004:29). The current management system also contributes to the creation of “harmful conditions” such as more extreme and unpredictable flooding, prolonged drought and river bank erosion (Table 7.3). Article 27 of the 1997 UN Convention describes the obligation of states for the “prevention and mitigation of harmful conditions” that include flood, erosion, sedimentation, 195  salinity and drought (United Nations 1997). Article 32 of the Berlin Rules describes the need to protect citizens against “extreme conditions...that pose a significant risk to human life or health, of harm to property, or of environmental harm” (International Law Association 2004:33). Article 21(2) of the UN Convention notes; “watercourse States shall take steps to harmonize their policies in this connection” so that no “harm to human health or safety” will occur (United Nations 1997). Article 28 of the 1997 UN Convention further describes the responsibility of states to “immediately take all practicable measures necessitated by the circumstances to prevent, mitigate and eliminate harmful effects of the emergency” (United Nations 1997). Due to the Ganges Basin management failures, sixty-two percent of my survey respondents report facing water stagnation problems that create disease pathogens, mosquitoes and loss of production. The unregulated use of chemical fertilizers promoted by the state also violates laws that require states to minimize environmental harm by controlling pollution.  My analysis of water as a human right also incorporates principles regarding the importance of water rights to many other human rights. General Comment 15 emphasizes that priority in water allocation should be for “personal and domestic uses” but article 1.6 also states that: “Water is required for a range of different purposes, besides personal and domestic uses, to realize many of the Covenant rights. For instance, water is necessary to produce food (right to adequate food) and ensure environmental hygiene (right to health). Water is essential for securing livelihoods (right to gain a living by work) and enjoying certain cultural practices (right to take part in cultural life)” (United Nations 2002). Article 2.11 further states that, “the adequacy of water should not be interpreted narrowly, by mere reference to volumetric quantities and technologies. Water should be treated as a social and cultural good, and not primarily as an economic good” (United Nations 2002). In the sections that follow, I assess the extent to which 196  human right failures in respect to water create a cascading effect on several other human rights and on their cultural practices in general.   Table 7.3: Harmful Ecological Effects as Reported by Case Study Participants        Harmful Conditions Smallholders and Landless Laborers Wealthy and Intermediate Landowners Billal Joardar Tofajjal Tanvir Loss of land/crop  X (no) √ (yes) √ √ Water access failure  √  √ X X Soil depletion & nutrient loss  X √ √ X Embankment failure √  √  √  √ River bank erosion √ √ √ √ Flood damage & loss  √ √  √ √ Drought damage and loss X √ √ √ Water stagnation problems √  √  X X   The Right to Employment Article 23.1of the UN Human Rights Declaration states the following: “everyone has the right to work, to free choice of employment, to just and favorable conditions of work and to protection against unemployment” (United Nations 1948). Article 1.6 of General Comment 15, as noted above, links water rights to livelihood rights (United Nations 2002). Article 13.2.b of the Berlin Rules explicitly links water rights to “economic needs” which include the right to employment International Law Association 2004:21). Ninety percent of my household survey respondents informed me that obtaining sufficient local employment is dependent on success in traditional agricultural production, whether on one’s own land or through employment on another’s land. Successful agricultural production generates employment opportunities for farme