Negotiating Abundance and Scarcity

Permanent Link: http://ufdc.ufl.edu/UFE0042580/00001

Material Information

Title: Negotiating Abundance and Scarcity Health Sector Reform, Development Aid, and Biomedical Practice in a Tanzanian Hospital
Physical Description: 1 online resource (447 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011


Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: This study seeks to complicate understandings of the nature of hospitals and hospital work in aid-dependent countries by drawing on a case study in northern Tanzania. I ask: How are reconfigurations in health sector governance, donor-sponsored health programs, and local norms and values converging in the hospital, and how do they impact the possibilities and limitations of interactions on the ground? In what ways are hospital actors engaging or manipulating governance and biomedical regimes, and why do they do so? How do the encounters between these actors influence state and global initiatives? What might a hospital ethnography that grapples with these complex issues look like? Existing hospital ethnographies tend to focus their attentions on specific departments, maladies, or populations within hospitals, providing little understanding of how they link up to other forces or actors, and the effects of these connections and movements on the possibilities and limitations within therapeutic spaces. Conversely, tacking between transnational spaces where health policies are negotiated and a health facility where they are enacted, I suggest an ethnographic approach that traces how the hospital is unevenly entangled with multiple scales, simultaneously, and what this means for the lived experience of actors interacting within and beyond hospital spaces. I interrogate how global agendas, targeted health interventions, and government policies engage and are engaged by various hospital actors and the existing norms, values and practices of these facilities. In addition, this study highlights the areas and practices within the hospital that are decidedly beyond state/global governing apparatuses. I explore the implications for both health professionals and patients who must navigate hospital spaces marked simultaneously by abundance and scarcity of biomedical, technological, and bureaucratic resources. Highlighting the politics underlying engagement across multiple scales, this study suggests that what actors are trying to achieve?whether at local, national, or global scales?is critically important to the material configurations and daily interactions characteristic of bureaucratic and biomedical work in health facilities.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by NOELLE C SULLIVAN.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Chalfin, Brenda H.
Local: Co-adviser: Dilger, Hansjoerg.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042580:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042580/00001

Material Information

Title: Negotiating Abundance and Scarcity Health Sector Reform, Development Aid, and Biomedical Practice in a Tanzanian Hospital
Physical Description: 1 online resource (447 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011


Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: This study seeks to complicate understandings of the nature of hospitals and hospital work in aid-dependent countries by drawing on a case study in northern Tanzania. I ask: How are reconfigurations in health sector governance, donor-sponsored health programs, and local norms and values converging in the hospital, and how do they impact the possibilities and limitations of interactions on the ground? In what ways are hospital actors engaging or manipulating governance and biomedical regimes, and why do they do so? How do the encounters between these actors influence state and global initiatives? What might a hospital ethnography that grapples with these complex issues look like? Existing hospital ethnographies tend to focus their attentions on specific departments, maladies, or populations within hospitals, providing little understanding of how they link up to other forces or actors, and the effects of these connections and movements on the possibilities and limitations within therapeutic spaces. Conversely, tacking between transnational spaces where health policies are negotiated and a health facility where they are enacted, I suggest an ethnographic approach that traces how the hospital is unevenly entangled with multiple scales, simultaneously, and what this means for the lived experience of actors interacting within and beyond hospital spaces. I interrogate how global agendas, targeted health interventions, and government policies engage and are engaged by various hospital actors and the existing norms, values and practices of these facilities. In addition, this study highlights the areas and practices within the hospital that are decidedly beyond state/global governing apparatuses. I explore the implications for both health professionals and patients who must navigate hospital spaces marked simultaneously by abundance and scarcity of biomedical, technological, and bureaucratic resources. Highlighting the politics underlying engagement across multiple scales, this study suggests that what actors are trying to achieve?whether at local, national, or global scales?is critically important to the material configurations and daily interactions characteristic of bureaucratic and biomedical work in health facilities.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by NOELLE C SULLIVAN.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Chalfin, Brenda H.
Local: Co-adviser: Dilger, Hansjoerg.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042580:00001

This item has the following downloads:

Full Text




2 2011 Noelle Sullivan


3 To Tracy and Flynn


4 ACKNOWLEDGMENTS Dissertations are completed only after long journeys, with the support, ass istance, and encouragement of so many people. Many people in Tanzania, North America, and Europe have had a hand in helping me think through the ideas that emerge in this dissertation. I have benefitted from the gracious and unfaltering support of my advis ors. Hansjrg Dilger has been my Primary Advisor and CoChair. He has continually pushed me to expand and further clarify my ideas, and provided continual and invaluable feedback on this project from its first conceptualizations to the form that emerges on the pages below. Brenda Chalfin, my other CoChair, has been a constant source of stimulating ideas, enthusiasm, and encouragement. Having this special team of Co Chairs allowed me to imagine a project that brought together the complexities of healing wit h those of politics bureaucracy and transnational governance They have been unfaltering and invaluable mentors, continually inspiring me to explore interconnections that have not always been obvious. Luise White has pushed me to understand specific even ts in their wider and comparative context s and fostered in me a love for ethnographies past and present. H er enthusiasm, imagination and insights continue to inspire me Maria Stoilkova urged me to think more deeply about how globalization, neoliberalism, and politics impact my research and enticed me to assert my ideas and arguments by challenging me with important questions. I owe a special debt to Stacey Langwick, whose provocative courses and mentorship early in my graduate career significantly shaped my intellectual approach to biomedicine, science, Africa, epistemology, and notions of modernity. Although she is not on my committee, I hope she sees her mark in the pages that follow. I am also grateful to Kendal Broad, who gave me a foundation in feminist science and methodology, and was willing to share her professional and personal perspectives on very complex issues These mentors have been


5 fundamental to my theoretical and practical training, and any disservice I do to their careful guidance belongs to me alone. For Kiswahili training, I am thankful to Masangu Matondo, who introduced me to Kiswahili with a sense of humor, and to Rose Lugano and Charles Bwenge, who continued to nurture and strengthen my language skills while at the University of Flor ida (UF). Lioba Moshi first introduced me to Tanzania in 2004, and provided invaluable assistance during my return trip in 2005. Her unfailing commitment to students and dry wit made my first weeks in Tanzania incredibly rewarding. MS Training Centre for D evelopment Cooperation repeatedly offered me training, lodging, and support in Tanzania over the years. A special acknowledgement goes out to Yusta Mganga, Gaudentia Lwakatare, and Steven Ndosi. I would also like to thank the Chuo cha Uenye Ulimavu, whose staff and resident families continue to inspire with their tireless efforts to improve the lives of people with disabilities, and train them in skills that will allow them to be independent In Tanzania, my most profound gratitude lies with the staff pati ents and caretakers of the hospital where I worked, whose names cannot appear here for reasons of confidentiality. I wa s welcomed to the place I call Kiunga District Hospital as a researcher, but also as a member of a team The staff worked tirelessly u nder adverse conditions to build hospital capacity in the name of treating the ill and patients relatives were admirably involved in caring for not only their loved ones, but also other patients in need. These caretakers continually assisted hospital per sonnel in a way that was both remarkable and generous I am humbled and grateful that the staff, patients and relatives shared with me their joys, aspirations, frustrations, and sorrow only a fraction of which was able to spill out onto the pages that foll ow. The District Medical Officer (DMO) Dr. Saidi, was always a gracious host, and his commitment to building the hospital


6 was imaginative and moving. Being a researcher in the context of a hospital in Tanzania truly test ed my anthropological training. I mportantly, it also tested me as a person and the staffs pragmatism helped ease these dilemmas. Ninawakumbuka sana. I must also acknowledge the warm welcome I received from Kiunga District Council, and all of the councilors and staff that took time out of their busy schedules to talk to a researcher and allow me to attend their meetings so I could understand the way the health department intersects with other community concerns. There were many people in Tanzania who provided my family and I with frie ndship, support, and laughter. First among them must be Baba and Mama Simon and their children Simon and Catherine, who invited us into their lives and provided us with daily sustenance and companionship. Tunawashukuru sana, na tunawakumbuka kila siku Cla us and Trine Heim, and their children Camilla and August, as well as Stefan and Christiane V lk and their children Nielson and Tilmann, were constant friends and playmatesthey literally came to our rescue on numerous occasions, and so literally, this diss ertation would not have been possible without their timely and selfless assistance. I must also thank Jesper Mogensen and Lidet Wolde, who provided us with shelter, some times for long periods of time not to mention amazing meals and good company. Jesper Bosse Jnsson was always source of inspiration and humor. I am grateful I was in the right time and place to meet Charles Llewellyn, and his wife Deborah, who were constantly hospitable and entertaining hosts, and have continued to support me in my work. Ber nard I. Sefu, Mama Gertrude, Joho, and Sophia Moshi provided stimulating conversation and friendship. We owe a special debt to Diamond and Diana Carvalho, and their daughter Sophie, with whom we shared many great meals and even more laughter. I am quite ce rtain Diamond and


7 Tracy will drag each other to the top of Kili one of these days! I would also like to thank Irene Rwegalulira for her assistance in transcribing interviews. Tanzanias Ministry of Health and Social Welfare (MoH) deserves special thanks. D ue to their kind invitation, I was able to mine the MoH library as I attempted to reconstruct a history of the health sector, and many officers took time out of their very busy schedules to tell me their personal stories, and the stories of the departments within which they worked. Within the MoH I am particularly grateful to Dr. Faustin Njau, who not only agreed to an interview and provided additional information via email, but also graciously permitted me to attend the Joint Annual Review of the Health S ector meeting in 2008. Dr. Bjarne Jensen also took out so much time to share his experiences with me, both within and outside of Tanzania. I would also like to thank Dr. Z. A. Berege, for providing me with letters of introduction to the di strict hospital w here I worked. This research was supported by a Fulbright Hays Doctoral Dissertation Research Abroad fellowship for which I am very grateful. Tanzanias Commission for Science and Technology (COSTECH) provided research clearance for this study, and Dr. A. Mvungi of the Department of Sociology at the University of Dar es Salaam was my sponsor. I must also acknowledge the support of Drs Ruth Meena and Rose Mwaipopo, as well as Mrs. Mabel Kaaya. Special thanks must also go to the Am erican Embassy in Dar es Salaam I presented some of the ideas contained i n this study at two workshops: t he first, Institutions, Collaborations, Power: Workshop on Hospital Ethnography at the University of Sussex, UK, and the second, Urban Health in SubSaharan Africa in Bamak o, Mali. I would like to thank the organizers, as well as all of the participants for the fruitful conversations we shared, which have left important imprints on my thinking. I am also thankful to Claire Mercer,


8 Lisa Richey, and Stefano Ponte for their com ments and insights at the African Stud ies Association meeting in 2009. Although I had been following her work for several years, I finally was able to meet Claire Wendland in 2009, and our continued exchanges to date have leave a profound impact on my thin king. Her impact on the pages that follow is clear, and I look forward to future collaborations. The foundation in anthropology I received at the University of Victoria as an undergraduate has left its mark on me. I am especially grateful to Patricia Spitt al, whose work first inspired me to become an anthropologist, and to Eric Roth and Lisa M. Mitchell, who honed my interests in the intersections of health and technology. At the University at Albany, I am thankful to Iris Berger, who helped strengthen my t hinking and writing while a masters student. The University of Floridas Department of Anthropology and its Center for African Studies (CAS) have provided me with a very happy academic home. They have been constant sources of financial, academic, intelle ctual and moral support. From CAS, I benefited from three Fulbright Hays Foreign Language and Area Studies Grants (FLAS), as well as several travel grants. I am thankful for the continued support of Lonardo Villaln, Todd Leedy, Corinna Green, Ike Akinyem i, Renata Serra, and Julie Silva. The Department of Anthropology gave me numerous opportunities to teach, as well as a wonderful environment in which to learn. I would like to thank Allan Burns, Ken Sassaman, Mi chael Heckenberger, Clarence (Lance) Gravlee, Peter Schmidt, Peter Collings, and John Krigbaum for many stimulating exchanges, and for their interest in my development as a scholar and teacher Alyson Young has been a constant source of friendship, advice, and support. My progression through the program would have been far more tumultuous had it not been for the help of Karen Jones, Patricia King, Juanita Bagnall and Pamela Freeman, whose commitment to the department, its faculty and students is unrelenting


9 and remarkable Special thanks must also go to Dan Reboussin and Peter Malanchuk, our two African Studies librarians extraordinaires, whose passion for their work, undying support of scholars, and uncanny ability to predict what people need is such a valuable gift. I also owe many thanks to my fell ow graduate students (past and present) in anthropology and African studies Becky Blanchard, Anna Brodrecht, Scott Cady, Sarah Cervone, Jean Dennison, Lauren Fordyce, Jennifer Hale Gallardo, John Hames, Rachel Harvey, Angelina Howell, Cara Jones, Carmen A londra Laguer Diaz, Alison Montgomery, Joost Morsink, Tim Nevin, Meghann O Brien, Gypsy Price, Alan Schultz and Noah Sims and have all made my time at UF so rewarding I owe a major debt to Traci Yoder and Meredith Martin whose untiring support and friendship, not to mention editorial expertise, have been important in my development as a scholar and person. I have also benefitted from friends who take me outside of my academic world, including Karen Matthews, Kaitlin Earley, Sonny Assu, and Mireille van R aaij. I wish to extend my gratitude to Parker and Jenni van Hart, and their children Madison and Samantha for providing my family and I with love, support, humor, honesty, good friendship, and amazing meals As I reflect back on my lifes journey so far, I cannot help but be humbled by the grace and love that my family has shown me, even when they did not understand what I was doing. They have endured my absence from my Canadian birthplace for over a decade, but have always been pillars of strength for me even from afar. It was a great gift to grow up knowing that regardless of what I chose to do, I would have their support. In particular, my parents Guy and Shirley Gentner and Stu and Doris Colquhoun have been stalwart enthusiasts, and have motivated me wi th the passion and determination they demonstrate in their own lives My sister Maggie Russell, her husband Noah and their children Ethan, Kiera and Zack continually remind me of


10 the importance of love laughter and joie de vivre. My brother Adam moves me with his willingness to continually push towards his goals despite setbacks My grandparents Tony Wuitchik and Doreen Gentner remind me of the importance of getting the most out of life and finding respite in music My aunt, Shelley Smirfitt, inspires me with her inexhaustible perseverance in the face of considerable challenges. I regret that my grandparents, Bob and Lee Colquhoun, and Cliff Gentner, and my uncle, Rob Colquhoun, did not live to see me finish this dissertation. They have such an important impact on my life, and I carry their love with me always. Most of all, I want to thank my chosen partner of life, Tracy Sullivan. Tracy uprooted himself and our son from a known life in Florida to a country he had never before seen. For over a decade hes been a constant companion and friend, and upon our return to Florida gave up so much in order to be the physical, emotional, and moral support that I needed. He truly is what he claims to be: a spousal anthropologist. I am grateful for his presence in my life every day, and no one has scarified more than him to ensure that this dissertation was completed. I am also thankful for our son, Flynn, who so often of late has asked me, Why do you go out and do that crazy work? Crazy, indeed. But watching him ma rvel at the world around him, I remain hopeful that at some point the why becomes a little clearer to him. Ive been so blessed with his ability to make me laugh and be silly. Some people feel rooted in places. I am not rooted in a place, but rather in t he se two people. And so it is to Tracy and Flynn that this dissertation is dedicated. I hope what they find in its pages honors their daily sacrifices.


11 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF FIGURES .......................................................................................................................15 LIST OF OBJECTS .......................................................................................................................18 LIST OF ABBREVIATIONS ........................................................................................................19 ABS TRACT ...................................................................................................................................22 A NOTE ON THE TERMS DOCTOR AND NURSE IN TANZANIA ..................................24 A NOTE ON TRANSLATION .....................................................................................................26 CHAPTER 1 INTRODUCTION ..................................................................................................................28 First Impressions: Narrative of a Future Fieldsite ..................................................................29 A Second Arrival: 2008 ..........................................................................................................31 Cultures of Biomedicine and Hospital Ethnography ..............................................................33 Biomedicine, Bureaucracy and Governance across Scales ....................................................36 Medical Professionals as Whole Persons .............................................................................41 Medical Pluralism in Africa ....................................................................................................44 Multi Scalar Hospital Ethnography: Desire and Economies of Appearances ........................49 Orientations of Fieldwork .......................................................................................................52 A Researcher in the Hospital ..................................................................................................57 The Issue of Confidential ity ...................................................................................................61 Organization of this Study ......................................................................................................63 2 ESTABLISHING AND BUILDING A HEALTH SECTOR ................................................69 Biomedicine and the German Colonial Administration: 18881918 ......................................70 Biomedicine and the British Colonial Administration, 19181961 ........................................76 Postcolonial Biomedicine and Making a Living: From Independence to Ujamaa, 19611986 .....................................................................................................................................89 Structural Adjustment, Early Reforms and the Informal Economy: 19851990s ................103 Conclusion ............................................................................................................................107 3 THE HOSPITAL BEFORE HEALTH SECTOR REFORM ...............................................109 A Brief History of Arusha and its Biomedical Institutions ..................................................110 Methodological Considerations for a History of the Hospital ..............................................114 Kiunga District Hospital: 19771989 ....................................................................................117


12 Therapeutic Pluralism and Encounters with Patients ...........................................................124 Becoming a District Hospital: 1989 1999 ............................................................................129 Conclusion ............................................................................................................................136 4 ESTABLISHING TANZANIAS TRANSNATIONAL AID REGIME .............................143 Fashioning the Frameworks of Health Sector Reform: 19901995 ......................................147 Opening a Space for Collaboration ...............................................................................147 Proposals for Health Sector Reform and the Strategic Health Plan ..............................150 Hostilities in a Stagnant Sector .............................................................................................153 Donor Government Relations in the 1990s ..........................................................................156 Eroding Donor Government Relations: The Helleiner Report .....................................157 Planning Reforms ..........................................................................................................160 Financing a Mixed Health Sector ..................................................................................161 Developing a System of Aid Coordination ...........................................................................163 Tanzanias Current Aid Modalities ......................................................................................166 The Safety Net under the Basket .......................................................................................168 Conclusion ............................................................................................................................173 5 INFRASTRUCTURE AND PROFESSIONALISM: EXPANDING THE HOSPITAL .....176 Building the Hospital under Health Sector Reform: 19992008 ..........................................181 Institutional Reconfigurations .......................................................................................182 Speakable and unspeakable concerns in the wake of Health Sector Reform .........184 Public private partnerships: Reaching out to build up ...........................................187 Donors and the Infrastructure of Targeted Interventions ..............................................196 Donors by Happenstance: Building a Laundry Facility ................................................201 Reflections on Change: Staff Views of Donors and the Government ..................................207 Conclusion ............................................................................................................................209 6 OF PARALLEL AND HYBRID ORDERS .........................................................................225 Parallel Orders ......................................................................................................................229 Official Hierarchical Structure of th e Kiunga District Hospital ....................................230 The Unofficial Hierarchies of the Hospital ...................................................................235 From Parallel to Hybrid Orders ............................................................................................243 Official Duties and Membership in District and Hospital Management Teams ...........244 Merging Management ...................................................................................................246 Conclusion ............................................................................................................................253 7 CONCURRENT GOVERNANCE: SALARY, COMPENSATION, AND WORKERS RIGHTS ................................................................................................................................257 Professionalism, Compensation, and Concurrent Governance .............................................260 Inadequate Salaries ...............................................................................................................265 Delays in Salary and Promotion ...........................................................................................268 Extra Duty Allowances .........................................................................................................272 Working Between Abundance and Scarcity: From State to Donor Resources ....................278


13 The Ambiguous Benefits of HIV/AIDS Work ..............................................................279 The Desire for Donor Sponsored Workshops ...............................................................281 Negotiating Resources, Rumors, and Accusations: A case study .................................284 Neoliberal Desires: Negotiating a Commodified World ......................................................291 Conclusion ............................................................................................................................295 8 BIOMEDI CAL PLURALITY BETWEEN THE ENCLAVE AND ITS OTHERS ............298 The Hospital and its Enclaves ...............................................................................................302 Biomedical Subjunctivity Beyond State/Donor Interest .......................................................305 Acting upon Happys Malady .......................................................................................307 Th e Subjunctivity of Emergent Biomedicine ................................................................315 Punctuated Development: The Confluence of Enclaved and NonEnclaved Biomedical Practice ..............................................................................................................................317 From Pregnancy to Birth: The Case of Miriamu ...........................................................319 Biomedical Pluralism within and Beyond the Enclave: The Case of Tumaini .............327 Conclusion ............................................................................................................................337 9 APPLES WITH APPLES: THE AESTHETICS OF GOV ERNING WITH NUMBERS ...........................................................................................................................344 A Vocabulary for Thinking about Documents .....................................................................347 Behind the Indicators: Ideals of Governance and Aesthetics ...............................................349 Health Statistics, Quantitative Reality and Governance .............................................351 The Role of Aesthetics in the Economy of Appearances ..............................................355 Setting up the Ideal Health Management System .................................................................357 Paradoxes of Health Statistics in Action ..............................................................................359 Controversial Facts .....................................................................................................362 Purging Uncertainty for the Public ................................................................................365 Conclusion ............................................................................................................................368 10 THE SOCIAL LIVES OF DOCUMENTS AT KIUNGA DISTRICT HOSPITAL ............370 Engagements with Aligned Documents: Registers and Donor Reports ...............................372 Registers in the Reproductive and Child Health Clinic .................................................374 The Politics of Donor Reporting at the CTC .................................................................380 Working between Government and Donor Reporting Burdens ....................................385 Staging Partnership: An Honored Guest in the CTC .....................................................391 Non Aligned Documents ......................................................................................................395 The Social Life of the Adult Patient File .......................................................................396 If We Do Not Help Each Other What Will We Do? ..................................................398 Burdens of Work: NonCompliance with NonAligned Documents ............................401 I Do Not Ask the Nurse! I Ask the Patient ................................................................404 Conclusions ...........................................................................................................................407


14 11 CONCLUSION .....................................................................................................................415 Biomedical Pluralism ............................................................................................................416 Implications for Hospital Ethnography ................................................................................418 Looking for Resources Beyond the State .............................................................................419 The Importance of the Local .................................................................................................421 An Exit Story ........................................................................................................................426 REFERENCES CITED ................................................................................................................428 BIOGRAPHICAL SKETCH .......................................................................................................447


15 LIST OF FIGURES Figure page 31 Kiunga Health Centre, circa 1978 ....................................................................................138 32 Kiunga Health Centre, circa 1981 ....................................................................................139 33 Kiunga Health Centre, circa 1982 ....................................................................................140 34 Kiunga Health Centre, circa 1 988 ....................................................................................141 35 Kiunga District Hospital, circa 1989 ...............................................................................142 51 Kiunga District Hospital, circa 20012002 ......................................................................211 52 Kiunga District Hospital, circa 20022004 ......................................................................212 53 Kiunga District Hospital, circa 2005 ...............................................................................213 54 Construction of pediatric ward at Kiunga District Hospital. Photo from hospital archives, taken in 2005 ....................................................................................................214 55 Kiunga District Hospital, 2006. Note the addition of water tanks. ..................................215 56 District Medical Officer (DMO) of Kiunga showing the new pediatric ward to one of the missionaries involved with the organization that helped to fund it. Photo courtesy of Kiunga's hospita l archives, taken 2007. ......................................................................216 57 Kiunga District Hospital's new pediatric ward. Photo by N. Sullivan, June 2008. Female ward can be seen in backdrop. ............................................................................216 58 Kiunga District Hospital, 2007 ........................................................................................217 59 Kiunga District Hospital, circa 2008 ...............................................................................218 510 Kiunga Distric t Hospital looking from west to east. Photo by N. Sullivan, June 2008 ...219 511 Kiunga District Hospital looking from east to west. Photo by N. Sullivan, June 2008. ..219 512 Kiunga District Hospital laundry and kitchen building, April 2008. The laundry room itself only made up a quarter of the total building including the door and the window area. The remaining area was the hospital kitchen, which was accessed by a door on the back side of the building. Photo by N. Sullivan ...........................................220 513 Sheets drying during the rainy season at Kiunga District Hospital. August 2008. The rains meant that sheets often took several days to dry, sometimes creating shortages throughout the wards. Photo by N. Sullivan ....................................................................220


16 514 Interior of laundry room prior to installation of new machine. April 2008. Photo by N. Sullivan .......................................................................................................................221 515 New industrial washing machine, installed July 2008. Photo by N. Sullivan .................222 516 The door of the laundry room is expanded and a new gate installed to accommodate the new industrial washing machine. July 15th, 2008. Photo by N. Sullivan ..................222 517 Construction begins on the new laundry building extension. Taken October 15, 2008. Photo by N. Sullivan. .......................................................................................................223 518 Internal structure of new laundry building extension at Kiunga District Hospital, October 23, 2008. The extension was made very large in order to provide an interior drying area for sheets and a space for conducting training workshops. Photo by N. Sullivan. ...........................................................................................................................223 519 Centrifuge in the CTC Clinic at Kiunga District Hospital. To protect the actual identity of the NGO, its logo has been removed. The REFLECT marks where the original logo was placed. Note the tiling of countertops and walls in this laboratory where the CTC technology is housed. .............................................................................224 61 Official Hierarchical Structure of Kiunga District Hospital ............................................256 81 Pediatric overflow area in the female ward, Kiunga District Hospital. On the other side of the dividing wall lies the area dedicated exclusively to female patients. Photo by N. Sullivan. .................................................................................................................341 82 Neonatal section of the maternity ward at Kiunga Distric t Hospital. Note several beds have two women per bed. Photo by N. Sullivan. ....................................................342 8.3 A worn wheelchair without footrests sits beside a newer chair with a broken wheel. Kiunga District Hospital. Photo by N. Sullivan. ..............................................................343 101 A refrigeration unit at Kiunga for vaccines in the RCH. The white label has an image of the Japanese flag, and below is written "From the People of Japan". Phot o by N. Sullivan ............................................................................................................................410 102 Under five child development clinic card. A) Front cover of clinic card. Lower box provides space for recording immunizations and doses of vitamin A. B) Interior of c linic card. Bottom left has PMTCT (Prevention of Mother to Child Transmission) to indicate HIV status, as well as hati punguzo the Kiswahili term for the mosquito net vouchers. There is also a space for ngao, which is an insecticide provided at inte rvals to treat nets. Vouchers and insecticide are part of the Presidents Malaria Initiative (PMI) program. PMTCT sponsored by Presidents Emergency Plan for AIDS Relief (PEPFAR) funding. Growth chart is part of a World Health Organization (WHO) initiative ................................................................411


17 103 Resource, policy, and data circulations between Tanzanian health facilities, the Ministry of Health, donors, and multi lateral organizations such as the United Nations. ............................................................................................................................412 104 The Vice President of REFLECT (far left) is served a fresh Tanzanian lunch in one of the doctors offices of the Counseling, Testing and Care clinic. Photo by N. Sullivan ............................................................................................................................413 105 Clinical Officers share a small room, consulting with multiple patients simultaneously. Photo by N. Sullivan. .............................................................................414 111 N. Sullivan and acting M atron Laura participate in Nibebe dance. .................................427


18 LIST OF OBJECTS Object page 81 Women in the maternity ward of Kiunga District Hospital sing a hymn during the night shift. August 2008 (.wav file 784 KB). ..................................................................341


19 LIST OF ABBREVIATIONS AMO Assistant Medical Officer ART Anti retroviral therapy (to treat HIV/AIDS) CBOs Community Based Organizations CCM C hama Cha Mapinduzi the only Tanzanian political party from 1977 to 1993 CMS Church Missionary Society COs Clinical Officers CTC Counseling, Testing and Care [Clinic] acronym for the HIV/AIDS clinic at the hospital DANIDA Danish International Developme nt Agency DED District Executive Director DFID (United Kingdom) Department for International Development DHS Demographic and Health Surveys DMO District Medical Officer DNO District Nursing Officer DPs Development Partners (donors) DPG Development P artners Group FBOs Faith Based Organizations GoT Government of Tanzania GTZ Deutsche Gesellschaft fr Technische Zusammenarbeit (German Agency for Technical Cooperation) HSBF Health Sector Basket Fund HSSP Health Sector Strategic Plan ICTR (United Na tions) International Criminal Tribunal for Rwanda


20 JICA Japanese International Cooperation Agency MCHA Maternal and Child Health Attendant MDGs Millennium Development Goals M.ATT Medical Attendant MoH Tanzanian Ministry of Health and Social Welfare MO I Medical Officer In Charge MP Member of Parliament MTUHA Mfumo wa Taarifa za Uendeshaji wa Huduma (Tanzanias health information system) NGO Non Governmental Organization NORAD Norwegian Agency for Development Cooperation OECD Organization for Econom ic Cooperation and Development OPD Outpatient Department PEPFAR (U.S.) Presidents Emergency Plan for AIDS Relief PHC Primary Health Care PMO RALG Prime Ministers Office Regional Administration and Local Government (division) PMTCT Prevention of Mother to Child Transmission (of HIV) POA Plan of Action PO W Plan of Work PRSPs Poverty Reduction Strategy Papers RAF (British) Royal Air Force SAPs Structural Adjustment Programs SDC Swiss Agency for Development and Cooperation


21 SP Sufladoxine pyrimetha mine. A drug used as a malaria prophylaxis during pregnancy, as part of the PMI program STDs Sexually Transmitted Diseases STS Science and Technology Studies SWAp Sector Wide Approach TANU Tanganyika African National Union TB Tuberculosis USAID Uni ted States Agency for International Development VCT Volunteer (HIV) Counseling and Testing WHO World Health Organization


22 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy NEGOTIATING ABUNDANCE AND SCARCITY: HEALTH SECTOR REFORM, DEVELOPMENT AID, AND BIOMEDICAL PRACTICE IN A TANZANIAN HOSPITAL By Noelle Sullivan May 2011 Chair: Brenda Chalfin Co C hair: Hansjrg Dilger Major: Anthropology This study seeks to complicate understandings of the nature of hospitals and hospital work in aid dependent countries by drawing on a case study in northern Tanzania. I ask: How are reconfigurations in health sector governance, don or sponsored health programs, and local norms and values converging in the hospital, and how do they impact the possibilities and limitations of interactions on the ground? In what ways are hospital actors engaging or manipulating governance and biomedical regimes, and why do they do so? How do the encounters between these actors influence state and global initiatives? What might a hospital ethnography that grapples with these complex issues look like? Existing hospital ethnographies tend to focus their att entions on specific departments, maladies, or populations within hospitals, providing little understanding of how they link up to other forces or actors, and the effects of these connections and movements on the possibilities and limitations within therape utic spaces. Conversely, tacking between transnational spaces where health policies are negotiated and a health facility where they are enacted, I suggest an ethnographic approach that traces how the hospital is unevenly entangled with multiple scales, sim ultaneously, and what this means for the lived experience of actors interacting within and


23 beyond hospita l spaces. I interrogate how global agendas, targeted health interventions, and government policies engage and are engaged by various hospital actors an d the existing norms, values and practices of these facilities. In addition, this study highlights the areas and practices within the hospital that are decidedly beyond state/global governing apparatuses I explore the implications for both health professi onals and patients who must navigate hospital spaces marked simultaneously by abundance and scarcity of biomedical, technological, and bureaucratic resources. Highlighting the politics underlying engagement across multiple scales, this study suggests that what actors are trying to achieve whether at local, national, or global scales is critically important to the material configurations and daily interactions characteristic of bureaucratic and biomedical work in health facilities.


24 A NOTE ON THE TERMS D OCTOR AND NURSE I N TANZANIA Tanzania has developed a health human resources structure that may be unfamiliar to some readers. Since my study describes relationships between these different levels of health professionals, I provide a general outline of the structure of the terms for the various levels of doctors and nurses employed within health facilities in Tanzania. Doctors : In Kiswahili conversation, there are multiple levels of biomedical practitioner that are referred to as daktari in general parl ance. There are two main terms used to denote doctor in Kiswahili: daktari and mganga. However, the Kiswahili word mganga is less commonly used to refer to a biomedical practitioner because the term may also be used to refer to a non biomedical healer. T he word daktari is more common, and can refer to three levels of biomedical practitioner. Clinical Officer (CO): The vast majority of biomedical care provided within clinics and hospitals in Tanzania is done by clinical officers, who receive approximatel y three years of medical training in clinical medicine, community health, and some training in sutures, bone setting, and minor surgical procedures. Assistant Medical Officer (AMO) : An Assistant Medical Officer receives an additional two years training a fter the CO training at a university or college of medicine, and can perform routine surgeries such as caesarian sections, appendectomies and laparotomies. They also have more specialized training in diagnosing more complex ailments compared to COs. Within the district hospital structure, they normally serve as the head of each ward or department, and complex cases are most often referred to them. Medical Officer (MO) : A physician, holding a Degree in Medicine. The majority of physicians in Tanzania do not practice medicine below the regional hospital level. Many work as administrators within the Tanzanian Ministry of Health and Social Welfare, within specialized hospitals, within regional hospitals, or in the private or NGO sector. Physicians present with in district hospitals in the country ordinarily work as the District Medical Officer (DMO), who is the chief medical administrator within a district. Many physicians also hold a Masters Degree in Public Health. In extremely short staffed district hospitals, DMOs may perform medical procedures. However, the majority of DMOs work primarily in administration at the district level. Nurses: Similar to doctors, the Kiswahili word for nurse (nesi or muuguzi ) can refer to several levels of practitioners. Nurse O fficer ( NO): Also called an A level Nurse, a nurse officer has either a bachelors degree (four years) or a diploma (three years) in nursing. Officers often serve as the in charge nurse of a ward or department, and are meant to provide the majority of spec ialized nursing care within the hospital. Officers are generally allocated to hospitals rather than to lower level health facilities such as health centers or dispensaries. NurseMidwife ( General Nurse) : Also called a B level Nurse, a nurse midwife recei ves two years of training in basic nursing and midwifery. Nurses of this level conduct the general


25 nursing duties and will also be asked to perform some cleaning duties within the hospital. They can also work within dispensaries and health centers. Nurse m idwives are the largest category of nurse at the district hospital Nurse Attendants: Nurse attendants receive one year of training in nursing and perform very basic nursing duties in the hospital, health center, or dispensary. They are also supposed to assist with cleaning duties. Maternal Child Health Attendants (MCHA) : Like nurse attendants, MCHAs receive one year of training, but it is specialized training on reproductive and child health. Many MCHAs work within dispensaries and health centers, or are responsible for reproductive and child health clinics and family planning clinics within hospitals. Ward Attendants (Medical Attendants) : These are the unskilled workers within the hospital. They have no medical training but are charged with general du ties in the medical facilities. These include cleaning, cooking, landscaping, moving patients, stocking wards, and other general duties not requiring specialized training.


26 A NOTE ON TRANSLATION On Code Switching Interviews with officers within the Ta nzanian Ministry of Health and Social Welfare, as well as those with donor representatives and expatriate volunteers in the hospital were conducted in English. Within the district where most of the research was conducted, however, all interviews with Tanza nians, whether hospital staff, administrators, or elected officials, were conducted in Kiswahili one of the official languages and the national language of Tanzania. I conducted these interviews myself, and translated them myself. However, it is common a mong health professionals in Tanzania to code switch between English and Kiswahili when speaking. Use of English words in the hospital is much more ubiquitous among those with advanced education; however even among nursing assistants and casual laborers who had received less formal schooling, code switching to English to convey particular ideas was common. There are two reasons for this. Several words in English have no translation in Kiswahili, and since advanced education (such as medical and nursing sc hool) is conducted in English, the hospital staff generally had a common vocabulary of English words and phrases that they would insert into their Kiswahili conversations. However, knowledge of English is often a status marker in Tanzania, demonstrating that a person has had advanced education (secondary school, college, professional training or the like) or that he or she attended private school. Code switching was common when staff spoke with each other, and would sometimes be employed with speaking with patients (see Chapter Seven), but more often in the latter case communication occurred in Kiswahili. Within the dissertation, for those interviews translated from Kiswahili, I have indicated where a word was spoken in English by italicizing that word or phrase within the quotation. Thus, italicized words within translated quotations do not indicate emphasis, but rather tha t the word spoken was in English. In cases where I have translated a statement from Kiswahili into English but provide the Kiswahili translation, it is found after the English expression, italicized in parenthesis. On translations of yes and no Furthermore, there are a few expressions in Kiswahili that translate into an equivalent of yes in Tanzania. Ndiyo indicates yes or it is so. It can also be conjugated with a particular noun class to indicate indeed. Eeh [pronounced as a long eh sound] is more often used, particularly in less formal settings, indicating agreement, or yeah. Eh hee [pronounced eh heh with t he heh also having a long eh sound] also indicates agreement, similar to the North American use of uh huh. Another indicator of yes or assent was bodily action in which the speaker would raise his or her eyebrows at the same time as taking a quick breath. In the translations of the interviews, I have translated the use of ndiyo as yes or indeed depending on context. However, I have maintained the word eeh in the transcriptions of dialogue rather than translating it into English, in order to stay closer to the ways that the participants expressed their thoughts.


27 Translating no is similarly somewhat complex. While there is an official word for no in Kiswahili [ hapana], it is not so commonly used in regular speech in Tanzania, and in certa in contexts could be considered too direct a negation. Instead, people commonly use eh eh sometimes while shaking their head, similarly to the North American use of uh uh. It was also common to use the word siyo meaning it is not or, depending on c ontext, is it not?, the latter used especially when someone is stating an opinion and looking for affirma tion from his or her audience. The word la can also be used, but this is usually an interjection that indicates a refusal of what someone has said or done, or that what a person has said is incorrect. On Gender in Kiswahili Language In Kiswahili, third person pronouns do not indicate gender. The word yeye can indicate either a male or a female, and it is very difficult to know whether the person being described is male or female unless the speaker specifies the persons sex within the context of the conversation. In the process of translating participant statements, when the sex of the person being spoken about was not indicated, I have tried to a lternate between using male and female pronouns in the English translation for the ease of reading. However, it is possible that someone I have referred to as one sex was actually of another.


28 CHAPTER 1 INTRODUCTION This study seeks to complicate unders tandings of the nature of hospitals and hospital work in aid dependent countries. In the pages that follow, I explore the ways that the convergence of health sector reform (HSR), donor sponsored health intervention projects, and the minutia of daily biomedical and bureaucratic practice impact the hospital what it is, and how its future is imagined. Tracing how policies and material practices interact within and beyond the hospital provides us with insight not only into the linkages between different levels of global and state health governance and how biomedicine mediates them, but also sheds light upon the gaps and dislocations within this system. Through following the actors, routines, practices, technologies, and linkages between different units that make up the hospital, I show what is created and what is at stake when the institutional and individual practices that make up a health facility converge with HSR and development aid. When I originally went to Tanzania in 2004 for advanced language training i n Kiswahili1, it was with a mind of developing a project about HIV/AIDS and how it was treated within government health institutions. However, when I arrived in Tanzania and first visited the grounds of a district hospital, it became clear HIV/AIDS is just one maladyalbeit a highly politicized and targeted one among the many that were treated within the hospital T he ways that health workers maneuvered and improvised the difficulties of their working environment deserved careful study. I returned in 2005 f or three months to do preliminary research within a small missionary clinic in order to learn the ways that biomedicine was practiced in a clinical 1 In Tanzania, Kiswahili and English are the two national languages. However, over 100 other languages are also spoken in the country. Kiswahili is effectively the lingua franca of the country, since it is the main language in which most children receive their primary education. English, while one of the national languages, is less spoken by the masses. Instruction in English does not start until secondary school, and it is mainly a language spoken by the highly educated, those individuals working with foreign companies or NGOs, or those working in the government. Thus, the ability to speak English is often an important marker of status.


29 setting. Upon returning to the United States, I developed a dissertation research proposal aiming to underst and how HSR and increased decentralization and privatization impacted both government and missionary health facilities in terms of bureaucracy and biomedical practice. I arrived in Tanzania in December 2007 to begin my research, planning to compare the di strict hospital I visited in 2004 with the missionary hospital where I conducted preliminary research in 2005. However, when I arrived I quickly realized that a comparative ethnographic project of the mission and the government hospital was no longer tenable. The changes that occurred at the government hospital during four years were so vast and unexpected that tracing these transformations and the hospital staffs experiences of them would be an almost unwieldy exercise in and of itself. Excerpts from my f ieldnotes trace these transformations. First I mpressions: N arrative of a F uture F ieldsite It is July, 2004, and as part of my language training program here in Arusha we are visiting a district hospital in Kiunga. We are all very excited as there are seve ral of us in the program who plan on doing future work or in healthcare. There are sixteen of us in total, crammed into two sport utility vehicles (SUVs ) We pull off the main, paved road but there is nothing surrounding us that reminds me of a hospital. H owever, just a few hundred yards off the main road we turn onto to a dirt path that is full of potholes. There are a variety of small buildings around, but I am confused and continue to look for the hospital, or a sign that will mark it. The program coordinator opens the back door of the SUV and we all file out. One of the students asks the coordinator what I am thinking: is this the hospital? She confirms that we have arrived, and I am shocked. There are a series of small buildings around but most of them are quite a distance from the dirt road where we have parked. We are led into a small office, which is quickly filled with our bodies. A woman in a dark blue dress is introduced as the Matron of the hospital, and she tells us she will show us around the grounds. As we are walking


30 on the tour, I see people milling about on the ground outside of each of the buildings. Several of them look very ill. The buildings are unlabeled and there appears to be little ability to determine which building has what funct ion. We are shocked at the conditions, and at how many of the patients are outside on the ground rather than inside the buildings. The matron walks us past most of the buildings, and takes us first to the HIV/AIDS clinic. It is tiny, with two small rooms w here counselors provide HIV testing, but have little else to be able to provide to their patients. The Matron takes us to the Maternity Ward. Its apparently the hospitals newest building, but when we go inside the conditions of the ward are quite diffic ult. There are many pregnant women outside walking, and when we go inside, we are escorted into the labor and delivery wing a long room containing a section with three cots, one next to the other, where women bring life into the world, sometimes simultaneously. The Matron is very proud of the new building, because now women who are in labor do not have to stay in the same bed as those who have given birth. Yet I am struck by how rudimentary many of the implements appear, and by the near lack of any electr onic equipment that I am so accustomed to seeing on media portrayals of maternity wards back home. As we are leaving from the tour to get back into our vehicles, I notice that there is a small boy laying in a ditch, grabbing his arm and moaning. He is alone there is no family member attending to him, and no one else nearby. I point him out to my program coordinator, who is Tanzanian but now lives in the United States. She immediately calls out to the people waiting outside of a nearby building, asking for the whereabouts of the boys relatives. A man runs over to the boy, and explains that they have been at the hospital since early this morning awaiting services. His son has hurt his arm, but they cannot seem to get the Clinical Officer to evaluate the boy, who is clearly in excruciating pain. The coordinator speaks to our driver about the


31 possibility of taking him to a private clinic, where she would pay for his services. Suddenly, the Matron comes walking briskly over and speaks sharply to the father. She speaks quickly and I do not understand what she says, but suddenly the boy is whisked up to his feet by the Matron and escorted, with the father trailing behind, into one of the unmarked buildings. The coordinator shrugs and says that the Matron denied tha t the boy is being neglected, and she has taken him to get services. When I arrive back at the classroom for a debriefing of the fieldtrip, my mind is consumed with worries was the boy simply removed from the scrutinizing gaze of the American visitors, or did he actually get treatment? A Second Arrival : 2008 It is January and I am excited: it has been four years since I visited Kiunga District Hospital and I am eager to begin my research there. I have all of the necessary documents clearing me as a researcher for the site, and I am hoping today that I can meet the District Medical Officer (DMO) and that he will give me the last approval I need to get started with my work: the approval of the hospital itself. I am on a daladala,2 which is to become my main mode of daily transportation from my newlyestablished home to my fieldsite. When I hear the conductor yell the name of my stop, I indicate that I will be getting off. The daladala stops, I hand the conductor my money and descend onto the road. When I get to the entrance to the hospital I notice that the gate is closed. I dont remember this gate having been there in 2004, nor that there had been any guard there, but perhaps our vehicles were simply waived through and I didnt notice. There are about ten people all at the closed gate, talking to the guard. They have food containers in their hands or balanced on their heads and I am surprised that they are not allowed in. The guard nods at me, indicating that he 2 Daladala is the Tanzanian term for the small minibuses that provide transportation shorte r distances within Tanzania.


32 wants to know why I am there. I tell the guar d in Kiswahili that I have a letter from Tanzanias Commission for Science and Technology ( COSTECH) and from the Ministry of Health (MoH) regarding doing research there, and that I need to speak to the DMO. He opens the gate a crack to let me by, and I fee l strange being permitted entry when all of those other people are left at the gate to continue trying to negotiate access to the hospital. The guard points me in the direction of the only part of the hospital I recognize from 2004: the administration buil ding. I head in its direction. I enter the DMOs office, where I give him my letter. He is not the same administrator I met here last time, and he calls someone on his cell phone once he learns why I am there. Five minutes later a woman enters. She is int roduced to me as the hospital Matronbut again, this is not the same person I met here four years ago. Obviously some of the administrators changed. The DMO tells the Matron in Kiswahili to show me around the grounds, as I will be doing research here until December. I follow the Matron out, and very quickly realize why I do not recognize the place as the same hospital I visited in 2004 so unlike the pothole filled and dust laden grounds that I obs erved just four short years ago. Between 2004 and 2008, the f acility doubled in size It now boasted a new male ward, pediatric ward, expanded HIV/AIDS treatment clinic, retooled reproductive and child health (RCH), and converting and expanded existing buildings to new purposes. Landscaping was carefully manicured, providing an external aesthetic that appeared carefully managed and meticulously clean. This dissertation traces the history of that hospital, and how it was perceived and imagined by the actors who converged there. It is a story that moves beyond the lev el of the hospital, however. In the pages that follow, I move across multiple scales to note the opportunities and constraints that came about at the confluence of HSR and development aid, and what


33 implications this had for the hospital as an institution, and for its workers and patients who interacted within the facilitys boundaries. Drawing on insights from medical anthropology, anthropology of the state, medical history, science and technology studies, and anthropology of policy, this dissertation asks: What is a hospital in an aid dependent state when both state and donors penetrate its boundaries? What does state or donor interest mean for the possibilities and constraints for providing biomedical care, and what politics underlies the distribution of t heir priorities? How do such transformations in the hospital working environment impact the ways that medical professionals think about their own futures, as well as their impressions of the state and various donors involved in healthcare provision? The im pacts of rapid transformations within Tanzanian healthcare, wrought at the interstices of global health policies, donor interests, and state reform, cannot be grasped without careful exploration on the ground, with important implications for how hospitals are approached as ethnographic fieldsites. Cultures of Biomedicine and Hospital Ethnography Anthropologists began studying biomedicine only recently. In a seminal article written in 1982, Allan Young argued that biomedicine required ethnographic scrutiny, in order to come to a fuller understanding of the specificities of how biomedical knowledge was produced and the underlying assumptions that made it up. In 1993, Shirley Lindenbaum and Margaret Lock published one of the first volumes addressing Youngs concerns, asserting that biomedicine continued to be undertheorized in anthropology. Inspired by these calls, numerous anthropologists began calling attention to the politics, cultures, and practices inherent in biomedical knowledges and therapeutic practice s. Attention to culture and practices have been central to these early works. Hospitals first came under the scrutiny of social scientists in the early 1960s, where they were seen as total institutions (Goffman 1961) or islands (Coser 1962), with orga nizational


34 and social rules and norms dictated by logical medical functions. Focusing largely on the movements and experiences of patients through the hospital, these functionalist approaches depicted hospitals as institutions isolated from the larger soci al context in which they were located, as a place of exception and exemption (van der Geest and Finkler 2004:1998). After several decades of latency, recently there has been a resurgence of social scientific interest in the hospital as the premire insti tution of biomedical practice. In a themed edition in 2004, Sjaak van der Geest and Kaja Finkler noted the undertheorization of hospitals within anthropological research. Several of the articles included in this edition maintain that hospitals offer an imp ortant lens on wider cultural and social processes of the societies in which they are located (Andersen 2004; Finkler 2004; Zaman 2004). Given that the majority of anthropological treatments of biomedicine concentrated on socalled Western contexts,3 and f ew examined hospitals in particular, the themed edition on hospital ethnography was one of the first collections that included hospital cultures, let alone the considering these cultures outside the socalled Western world in Ghana (Andersen 2004), South Africa (Gibson 2004), Mexico (Finkler 2004), Egypt (Inhorn 2004), and Bangladesh (Zaman 2004). The insights provided in this themed edition inspired a resurgence in hospital ethnography, making critical contributions to the theorization of the hospital as a lens through which a variety of global and local processes can be considered simultaneously (Horsley 2008; Mulemi 2010; Zaman 2008). One of the significant aspects of this recent work in hospital ethnography relevant to this study is the ways global biomedicine is reinterpreted based on cultural and social norms and values of a given locale (van der Geest and Finkler 2004; Finkler 2004). The acknowledgement that the communities in which hospitals are embedded matter for the contours of biomedical 3 See for instance Good and Delvecchio Good 1993; Good 994; Lock 2002; Martin 1994; Mol 2002; Rose 2007; Young 1993; Young 1995.


35 enco unters is a critical feature of the cases I present below. However, I do not trace the way that global biomedicine is reinterpreted as such, but how global and state forms are unevenly embedded in existing biomedical institutions, and to what effect. Rat her than seeing the global as reinterpreted, I aim to demonstrate how and in which specific contexts global, state, and local forces interact within the hospital, to what consequence for biomedical and bureaucratic encounters characteristic of hospital w ork. Notwithstanding the importance of decidedly local norms and values and their influence on daily encounters in the hospital, current scholarship, although significant, leaves unanswered some important questions about what the hospital is as an institu tion (see also Street and Coleman, forthcoming) Existing hospital ethnographies, and indeed critical studies of biomedicine as a whole, tend to focus nearly exclusively on one group of actors (patients, doctors, nurses), one malady (HIV/AIDS, cancer, infe rtility), or one health unit (outpatient clinic, cancer ward, HIV/AIDS clinic, orthopedic ward) in the hospital. We thus lack a hospital ethnography that traces the interconnections between different units within the hospital, and the linkages of specific departments to government and donor policies and programs that extend beyond the borders of the hospital itself. What is more, little is known about the politics underlying how health professionals, patients, family members, technologies, bureaucratic prac tices, biomedical knowledges, maladies, and other objects move or are constrained between and beyond different departments within the hospital. What are the politics of these movements or constraints? While existing hospital ethnography undermines the noti on that hospitals are islands substantially separate from local cultural and social processes, it remains to be seen how changes in state and global health governance impact biomedical encounters on the ground. While


36 hospitals are not separate from the com munities within which they are located, to understand them as devoid of state or global governance regimes risks a myopic treatment of the complexities of hospital entanglements. Hospitals, whether in aiddependent or wealthy countries, are not separate fr om apparatuses of the state (see for instance Street, forthcoming) In the cases of aid dependent states, they are also increasingly subject to multi and bi lateral development policies and initiatives. These issues are of critical importance in a world w here global health policies, governmental reforms, and a multitude of donor interventions punctuate daily encounters and practices within hospitals in aid dependent countries. How are reconfigurations in health sector governance, donor sponsored health pro grams, and local norms and values converging in the hospital, and how do they impact the possibilities and limitations of interactions on the ground? In what ways are hospital actors engaging or manipulating governance and biomedical regimes, and why do they do so? In short, how does this convergence impact the lived experience of the actors who come together within hospitals? Conversely, how do the encounters between these actors influence state and global initiatives? What might a hospital ethnography that grapples with these complex issues look like? With these concerns in mind, I draw upon scholarship within and beyond anthropology in order to suggest an ethnographic approach to the hospital that, rather than focusing on one scale (local, state, global) approaches the issue of how the hospital is unevenly entangled with multiple scales, simultaneously, and what this means for the lived experience of the actors interacting within and beyond these spaces. Biomedicine, Bureaucracy and Governance across S cales Biomedicine has a long history in Africa. That biomedicine was tied to colonization (Comaroff 1993; Iliffe 2002; Hunt 1999; Langwick 2005; Vaughan 1991; White 2000) and evangelization (Stirling 1947, 1977; Ranger 1992; White 1952) on the African cont inent is well


37 established. In a groundbreaking history of the medical profession in East Africa, John Iliffe argues that medical professionalism in Africa, and elsewhere, must be understood in relationship to the state (2002). Indeed, while prevailing defi nitions of professions argue that power and autonomy were the true essence of a profession, Iliffe notes that professions generally evolved under state control or within government owned universities thus, professions and the state are in large measure sy mbiotic (2002:3). At independence, health professionals both depended on the state to train and employ them, and also grappled with the state for recognition of their professional status. Today in Tanzania, the struggle between medical professionals and t he state shares significant parallels with the past. That biomedicine and governance are intertwined is not new, but the ways that they became entangled are important to how governance is performed through biomedical encounters today. Medical historians p rovide critical insights into how biomedicine became quantified and thereby linked up with systems of governance. George Weiszs (2005) and Theodore Porters (2000) work in particular sheds light on how, largely during the nineteenth century, and under considerable protest from physicians, biomedical therapeutic encounters became interconnected to universal notions of science and bureaucratic systems of both governments and insurance companies. Thus, Weisz argues, If the expanding role of quantification in medicine is part of the history of science, it is also one of the forms of knowledge that mesh most effectively with the prevailing contemporary forms of administration and government (2005:400). It is largely the quantification of biomedicine that allow s it to continue to appear universal and culturally neutral, despite considerable scientific and social scientific evidence to the contrary. Indeed, Claire Wendland argues that despite the lack of evidence supporting biomedicines neutrality, it remains a powerful narrative with real effects in the world (2010:246 n.3).


38 In terms of aid dependent countries in subSaharan Africa, the notion of an objective and universal biomedicine has important implications for states interactions with potential investor s and donors. In Africa, neoliberalism as a global form largely began to take root with the conditionalities of structural adjustment policies (SAPs). D ebt relief provided largely through the International Monetary Fund (IMF) and the World Bank favored sta te withdrawal from social services and the opening up of African states to the global market free of state interference. Similar to other regions of the world, this retooling has also brought a plethora of nongovernmental organizations (NGOs) into heavily indebted countries, providing opportunities for bi and multi lateral donor agencies to circumvent the state (Aretxaga 2003; Blundo and Le Meur 2008; Elyachar 2003; Ferguson and Gupta 2002; Mercer 1999; Trouillot 2001, 2003). Significantly for donor spons ored health interventions, which are largely funded by donor organizations from Europe or North America, the underlying guidelines structuring these programs are embedded in assumptions and values of the country within which the donor is headquartered (Dil ger 2001; Whyte et al 2010). The presumption of biomedical neutrality makes it a particularly compelling form through which to achieve development in the so called Third World. In addition, the fact that medical quantification allows for biomedicine to appear measurable and comparable across wide expanses of space and through time make this science a particularly compelling means of donor and state intervention into the daily lives of Tanzanians The result is a multitude of forms of transnational governmentality (Ferguson and Gupta 2002; drawing on Foucault 1991), where modes of governance extend beyond the sole purview of the state, to transnational networks made up of disciplinary bodies such as the World Trade Organization, as well as transnationa l and local NGOs and other organizations that perform state like functions. These configurations of transnational and state forms of governance call for


39 research that investigates the forms of power that are emerging at the interstices of the state, [inte rnational organizations] and NGOs (Elyachar 2003:598). If these interstices are examined, what comes to light is the politics underlying the ways that global forms travel. James Ferguson argues that rather than global forms flowing as they are oft descr ibed, they hop between strategic points or enclaves, skipping large expanses of space in between (2006). If one is to account for the particular ways that power is configured and engaged, it is criti cal to be attentive to how global forms are strategical ly placed, what i s included and left out of global interventions and what these externally funded and surveilled nodes of development mean for the contexts in which they operate. In Tanzania and elsewhere, the articulations of donor and state policies and programs within biomedicine are critically important to how hospitals function, and the lived experiences of actors converging within these health facilities. In addition, the engagements and practices of health workers, patients, and relatives within the hospital have crucial influence on donor and state programs and policies. Global and state priorities matter in terms of how resources and expertise are distributed throughout the hospital, and to the contingencies of biomedical encounters there. That sai d, local norms and values, and the local infrastructures and contexts into which global and state programs and policies are embedded, are equally important to how global and state policies and programs are planned, understood, and implemented. Lock and Nic hter argue: [W] e now need to reconsider medical cosmopolitanism from the vantage point of both local and global relationships. At this juncture, it is important to critically examine global public health agendas and how health and development agenci es influence national health policies everywhere in the name of health care reform, rationalizing choices through the use of evidence based criteria, homogenizing practice in th e name of efficiency, and so on. (2002: 910).


40 This dissertation is an eff ort to take up Lock and Nichters call to be attentive to multiple scales in order to interrogate how global agendas, targeted health interventions, and government policies engage and are engaged by various hospital actors and the existing norms, values and practices of these facilities. I wish to further extend Lock and Nichters this call by tracing the areas and practices within the hospital that are decidedly beyond state/global governing apparatuses, and what this means for the lived experience and therapeutic possibilities for both health professionals and patients who must navigate hospital spaces marked simultaneously by abundance and scarcity of biomedical, technological, and bureaucratic resources. Interrogating the interconnections between biomedi cine and the state through medical quantification offers a particularly potent means for grappling with the complexities of interactions across scales within the contemporary African hospital, and for tracing those governing apparatuses beyond the hospital to the macro spaces where policies and programs are planned, negotiated and contested. One of the ways that state and donor agents enact forms of governance is through bureaucratic practices. Much of the ethnographic literature on bureaucracy emerges not from medical anthropology, but from anthropologies of the state. According to these studies, bureaucracy must be understood not only being structured by policies, but also as structuring the meaning of policies through repetitive and often mundane everyday practices (Gupta and Sharma 2006). The emphasis on practices draws attention to the ways that people are situated and interact within the bureaucratic structure. Civil servants may, through their social relations and interactions, entrench governance stru ctures into the institutions where they work, but this process is always contingent and negotiated at the confluence of institutional and individual relations and enactments.


41 Importantly, while bureaucratic administration is characteristic of many biomedical encounters the world over, it remains notably absent from the majority of anthropological analyses of biomedicine and hospital pra ctice (but see Andersen 2004). Yet both states and donors govern, monitor, and evaluate health policy and programs through bureaucratic apparatuses that impinge upon the daily routines characteristic of biomedical encounters. If global forms create development enclaves, and global health agendas and development aid influence state health policies, then the hospital is not mer ely the premire institution of biomedicine, but it is also an important site for global and state governance modalities, largely achieved through specific engagement with administrat ive apparatuses on the ground. The conditions of those administrative sys tems are informed by particular historical trajectories, which often complicate global and state efforts at achieving their planned governance regimes. Many of these histories can be perceived by careful attention to the objects and infrastructure on which administrative apparatuses must depend. Medical Professionals as Whole Persons In Iliffes history of physicians in East Africa, he notes that a contemporary sociology (or, perhaps, anthropology) of the medical profession in the region was lacking, and highly desirable. Recently, there have been two recent ethnographies that explored the joys and contradictions of medical and nursing training and practice in Eastern Africa. Claire Wendland (2010) investigates medical training within Malawi, noting the a mbiguities that students encounter when the contours of their education do not mesh neatly with the conditions of medical work in the hospital. Helle Max Martin (2009) offers a comparable study regarding peoples movements through nursing education and the possibilities for nursing care in under resourced and overburdened health facilities. Both of these scholars note the discordances between the professional ideals that medical and nursing students learn in school and the immense difficulties


42 of translatin g these ideals into meaningful practice in the context of biomedical scarcity. Improvisation for both physicians and nurses became a key element in how they grappled with the contradictions between what they were trained to do, and the possibilities of ach ieving those standards in the hospital. Wendland asks, what happens to the moral order of medicine when much of the material technology required to manifest biomedical practice is unavailable, when scientific medicine is entwined with spiritual or politic al mission, when clinicians working lives are in every respect shaped by the same structural violence that produces patients suffering[?] (2010:24). Both Wendland and Martin (2009) assert that health workers are not only biomedical professionals in thei r own right, but, importantly, are part of the very communities they serve. Few ethnographies are so attentive to the social positioning of health professionals beyond the settings in which they work, with important consequences limitations for our unders tanding of how health workers understand their lives as what Martin calls a whole person. In short, ethnographies of medical professionals must consider not only health workers perspectives on their job, but also their personal and professional aspirati ons and obligations, and the ways that these multiple identities intermingle in the context of daily practices. In these works, the domestic and professional domains are blurred, revealing important insights into some of the improvisations and, at times, e thically murky practices characteristic of hospital work in aid dependent countries. What is not so prevalent in these works, and what I argue should be highlighted more specifically in hospital ethnography, and in anthropologies of health professions mor e widely, is health workers additional position as agents of the state.4 There exists some compelling anthropological work that grapples with the position of health professionals as state agents, 4 However, both Wendland (2010) and Martin (2009) do explore these health professionals views of their government. Yet hospital workers importance to governance is beyond the purview of these studies.


43 emerging primarily out of Europe. Yannick Jaffr and Jean P ierre Olivier de Sardan provide a comparative ethnography outlining the ethically murky engagements between health professionals and patients within government owned health facilities in five capital cities of West Africa (2003). To my knowledge, this is t he only existing ethnography that deals exclusively with health professionals as state agents. In addition, other anthropologists of the state interrogate health workers as state agents and members of local communities, but make little distinction between their professional status and the professional status of other civil servants who do not possess biomedical expertise (see Anders 2010; Blundo and Olivier de Sardan 2006). In an ethnography of daily governance in West Africa, Blundo and Le Meur call for studies that opt for a meso point of view, observing at intermediary levels how the different logics (of the central state, international and national NGOs,5 local bureaucracies, grassroots associations, etc.) interact in the daily delivery of public s ervices (2008:3, parenthesis original). Their approach is not only attentive to the ways that state agents navigate their situations, but also explores the discourses these actors develop in relationship to their daily practices and the strategic behavior s they adopt in the face of other state and nonstate actors. I suggest that health workers add to this intermediary level the possibility to additionally consider the ways that biomedicine intersects these various logics. Their perspectives therefore prov ide an important lens through which we might see the intertwining and contestations of multiple forms of expertise and experience as they play out within African settings. From day to day, African health personnel provide biomedical care, engage donor and state policies and programs, and interact with members of the same communities in which they live. Thus, their position at the 5 Non governmental organizations


44 interstices of state, global, local, and biomedical domains allows for the consideration of how these multiple scales intersect, in practice. The analytic strategy I adopt here explores the routines and perspectives of health professionals as socially and culturally embedded actors, simultaneously agents and subjects of the state, and wielders of biomedical and locally forged expert ise that allocates them particular status both within their communities, and within targeted donor/state interventions within the hospital. The ways that these actors navigate the multiple opportunities and constraints of a position at the interstices of t hese multiple forms and scales provides important insights into the ways global, state and local forms intersect, through biomedicine, bureaucracy, and social encounters, in practice. Their perspectives provide important insights into the contingencies of biomedical state institutions. Treating health professionals as whole persons requires attentiveness to not only the biomedical and domestic spheres of their lives, but also the uneven ways that state and global forms impact their experiences and practices. Medical Pluralism in Africa The convergence of local, state, and donor/global forms has important implications for the possibilities of biomedical practice within East African government health facilities. How these multiple scales articulate with eac h other through daily governance (Blundo and Le Meur 2008) and practices is contingent and emergent, informed largely by the history of the relationship between biomedicine and governance in the region. A s many scholars have demonstrated, neither historic ally nor currently has biomedicine been the only form of healing in Africa generally, nor in Tanzania in particular. Feierman and Janzen demonstrate that in precolonial and colonial Africa, a variety of therapeutic traditions existed, which overlapped bet ween groups of people, informed by interactions between different healers (1992). In essence, the therapies employed by any given healer had their own unique


45 histories, and the variety of practices among healers defies easy categorization (Feierm an and Jan zen 1992; Janzen 1978). Even today, the diversity of healing expertise can only be understood by exploring the personal histories of the healers who embody them (Langwick 2001). These practices, as Langwick argues, are more fluid than the categorization of traditional medicine implies. Indeed, even historically, several biomedical practices were indigenized (Vaughan 1991:24; White 1995), the most widespread of which was the injection (Malloy 2003). Indeed, this process of healers adapting and adopting biomedical therapies can also been shown to work in the reverse. Despite the boundaries between traditional and modern medicine being carefully maintained in colonial and current discourse (Langwick 2006), in practice, patients may use traditional treat ments in the clinic, and nurses engage these medicines, even at times suggesting nonbiomedical therapies when biomedical ones fail to provide relief from a malady (Langwick 2008). When the designation of medical pluralism or therapeutic pluralism is a pplied to subSaharan African contexts, the implication is that biomedicine as a unitary system, generally consistent and systematic, and that it should be seen as distinctive from the multitude of other, nonbiomedical forms of healing with a given region. What is specifically not articulated regarding biomedicine itself as a form of healing is itself potentially plural. While pluralism is discussed with reference to the unique expertise and trajectories of individual non biomedical healers in Tanzania, th is same characteristic of might well be applied to people engaging biomedical experts and treatments. Indeed, by treating biomedicine as a systematic healing regime rather than as contingent practice, much of the complexity within biomedical encounters rem ains obscured. Instead, biomedicine appears hegemonic, creating a biomedically


46 understandable body (2001:141) upon which biomedical practitioners can systematically intervene based on set diagnostic criteria. Some important works in medical anthropology and history suggest instabilities of biomedical interventions and practices. These contingencies are often expressed in two main ways. First, they reveal how individuals (often patients or non biomedical healers) creatively engage biomedical technologies and therapeutic obj ects (White 1995; Whyte 2002). Second, they note how the infrastructural and technological paucities of African health facilities intrude on health professionals abilities to deliver care based on biomedical standards and ideals (Anders en 2004; Booth 2004; Martin 2009; Wendland 2010; Zaman 2004, 2008). Turning to the ways that people attend to particular health problems in Uganda, Susan Reynolds Whyte argues that in their quest for improved health, people become actively involved in a process of trying out of undertaking particular interventions hoping that they will lead to better health but also understanding that a large degree of uncertainty and doubt punctuates these actions (2002:173). Whyte suggests that the ways that people de al with agents of misfortune in Eastern Uganda is similar to the ways that they deal with ill health. A malady, like an agent of misfortune, becomes known through acting upon it (see also Langwick 2007), and with the recent proliferation of pharmaceuticals in Ugandan shops and clinics, patients and relatives make use of drugs without necessarily caring whether or not the ways that they use them are in line with biomedical indications and guidelines for their use. Indeed, such attitudes towards biomedical interventions are hardly new even during colonial times, people would deliberately go to mission clinics desiring a particular treatment, often redefining the indications of those drugs or therapies through their own understandings of health and disease (Whi te 1995, 2000).


47 That uncertainty punctuates biomedical therapies due to local understandings of health and healing and the inadequacies of biomedical institutions is well established in literatures on biomedicine in Africa. However, I wish to articulate a third possibility for approaching biomedicine in Tanzania, and beyond. This understanding emerged as I thought about the work of three medical anthropologists and observed biomedicine in practice at the hospital that is the main site of this study. Margaret Locks work undermine s the assumption that all human bodies are universal a premise that underlies a variety of biomedical interventions. Instead, she argues that the biology of human bodies must be understood in context there are therefore local biologies, because humans are entangled with historical, environmental, social, cultural, and political contexts, thus dislodging an assumption that bodies are readily standardizable (Lock and Nguyen 2010:12; see also Lock 1998). Working on fertility in Egypt, Marcia Inhorn notes that biomedical practice differs from provider to provider, even within the same medical facility (2002). In a rather different treatment of therapeutic encounters in Tanzania, Stacey Langwicks work traces the ways that biomedicine and traditional medicine intersect and interfere with each other within the government hospital (2008). She shows that nurses in the hospital recommend patients see traditional healers for certain maladies that biomedical interventions fail to adequately address. Indeed, Pharmaceuticals move from being a curative technology to being a diagnostic technology (433) in that the failures of those pharmaceuticals tried out on a patient in the process of acting upon their maladies suggested that other therape utic modalities might achieve better success. Collectively, these works, as well as my daily observations within a Tanzanian hospital, suggest to me that there are instabilities within biomedicine itself that as yet remain unexplored. If bodies have local biologies, if biomedical expertise and practice differ from provider to


48 provider, if maladies are discovered through the processes of trying out pharmaceuticals, if people (health professionals, patients, relatives) engage biomedical therapies drawing o n their own understandings of health and healing, if state and donor priorities are unevenly distributed within the hospital, and if the conditions of hospital work necessitate considerable improvisation by health workers (Martin 2009; Wendland 2010), then might it be possible that biomedicine itself entails a degree of plurality that as yet has not been well explored? Indeed, biomedical pluralism has been suggested, although not fully explored, in select instances. In comparing biomedicine to Chinese medi cal pluralism, Volker Scheid suggests that all therapeutic practice, whether traditional or otherwise, is in fact pluralthat plurality is an intrinsic feature of all healing practices (2002). Kenneth F. Schaffner also hints at this possibility in a them ed edition on pluralism in alternative medicines (2002). He argues that contrary to what is often asserted within biomedicine, there is no unitary method underlying biomedical science. The possibility for exploring biomedical plurality within the Tanzanian government hospital is particularly salient for understanding the contingencies and discordances characteristic of biomedical practice as observed in Kiunga District Hospital. When we take seriously biomedical practices and their engagements with particul ar bodies, understandings of healing, forms of expertise and experience, bureaucratic apparatuses, material and spatial configurations, social norms and values, and specific histories, notions of a unitary but local biomedical culture appear inadequate i ndeed. As we shall see, multiple possibilities for biomedical understandings and practices exist in an unsettled relationship within the various departments of the government hospital. I thus assert that biomedicine, as other therapeutic modalities, is eme rgent and contingent, its intricacies only perceptible with careful attention to the practices and engagements that make it up.


49 Multi Scalar Hospital Ethnography: Desire and Economies of Appearances The approach to hospital ethnography that I advocate he re brings biomedicine and bureaucracy together in order to consider the ways that global, donor, and state entanglements impinge upon, and are also engaged and even circumvented by, social, structural, therapeutic, and technocratic relations within the hos pital. State and donor/global interventions must be embedded within existing (and often historically neglected and fragile) health facilities. My data suggest that while the hospital is as a site where state, global, and local forces collude, they come together in highly uneven ways throughout the hospital. Drawing on Fergusons notion of development enclaves (2006), I argue that anthropologists must be vigilantly attentive to where and how state and donor polices and programs are embedded in existing insti tutions, tracing the material and technocratic configurations through which these global and state configurations are actualized (or not). If hospital ethnography is to contend with the complexities and inequalities inherent in biomedical practice in aid dependent states, it must take seriously the ways that these global/state/local articulations are actualized through specific and often momentary engagements between actors, spaces, objects, and ways of knowing. The theoretical articulations and expansions I suggest in previous sections are important to how I approach the hospital. However, there are two additional conceptualizations that I have found important in the ways I have come to understand connections and practices across multiple scales within and beyond the hospital. The first is the notion of desire. Drawing on select aspects of Gilles Deleuzes scholarship, Joo Biehl and Peter Locke suggest that anthropologists take seriously the desires and imaginations of actors who move through the very s ystems that we study. They argue that Through close attention to people moving through broken institutions and infrastructures in the making and with careful observation always complicating the a priori assumptions of universalizing theory, ethnographic w ork can make public the constellations


50 through which life chances are foreclosed and highlight the ways desires can break open alternative pathways (2010:318, emphasis in original). Important to this theoretical orientation is not only describing the chal lenges and even disappointments with which actors grapple, but also, and most importantly, being attentive to the hopes and desires that they articulate, and the futures that they imagine. Desire and imagined future has an important impact on how people na vigate their worlds. Significantly, as Biehl and Locke suggest, bringing desire into anthropological analysis allows for institutional configurations to be explored, while simultaneously opening up the possibility for the people we study to propose alterna tives to the current state of things. As the authors note, People bear an understanding of their worlds, of the social problems they must circumvent or transcend, and of the kinds of politics that would actually serve their aspirations that is unaccounted for in policy discussions and decisions (336). I wish to link up this notion of desire to an aspect of Anna Tsings work. In what she calls an anthropology of the global, Anna Tsing (2005) argues that ethnography should highlight what is produced when global, state, and local forces interact by tracing their points of intersection. Tsing pays close attention to the role of appearances in the ways that global and local forms intersect in practice. She argues that companies attract financial investors th rough the self conscious making of a spectacle (57) through exaggerating what might be possible if adequate funds were made available. This marketing of the possibilities of a company is part of what Tsing calls an economy of appearances, and when brou ght together with Biehl and Lockes notion of desire, these concepts suggest an important trajectory for a hospital ethnography that is attentive to connections and disconnections across multiple scales. Being attentive to desire and the deliberate establishment of economies of appearances allows for a hospital ethnography that traces the subject positioning of a multitude of actors as


51 they navigate the intersections of global, state, and local forces. For instance, the ways that hospital workers engage their responsibilities hinges significantly upon what they want, professionally and personally, and whether or not their workplace opens up the possibility for achieving those desires. Thus, hospital administrators who are aware of those desires can create an economy of appearances directed at the workers, constructing the hospital as a place of institutional and professional opportunities. Yet in order to be able to convincingly perform such an economy of appearances, administrators would need to secure si gnificant financial investment (or at least, appear to be able to do so), requiring that they innovatively approach potential investors by appearing to be places where these donated funds could achieve results that are attractive to those benefactors. Tack ing between desire and self conscious performance, it is possible to perceive how hospitals create connections across multiple scales in the name of actualizing institutional possibilities. Similarly, in the states engagement with multiand bi lateral do nor entities, a government desiring to establish particular governance apparatuses may deliberately make itself appear attractive for development, largely by drawing on rhetorics recognizable and desirable to donors, thereby making itself appear to be a space of development possibilities. Thus, while governments, donors, and health facilities may collaborate across space, their motivations for doing so are not necessarily the same. Bringing the focus onto the aspirations of differently placed actors and ins titutions, and the performances they deliberately enact in order to achieve those desires, allows for a hospital ethnography that traces uneven engagement across multiple scales. It also brings attention to multiple layers of agency scattered throughout the connections in which the hospital is implicated. Importantly, such an approach also brings into view what these uneven collaborations mean for the possibilities of providing biomedical care in under resourced health facilities within aid dependent states.


52 Orientations of Fieldwork I arrived in Tanzania in December 2007 to begin my research, planning to compare the district hospital I visited in 2004 with a missionary hospital where I conducted preliminary research in 2005. However, when I arrived I quick ly realized that a comparative ethnographic project of the mission and the government hospital was no longer tenable. The changes that occurred at the government hospital during four years were so vast and unexpected that tracing these transformations and the hospital staffs experiences of them would be an almost unwieldy exercise in and of itself. I spent the next eleven months moving between the hospital departments and tracing linkages from the facility, to the District Council (DC),6 all the way to th e Ministry of Health and Social Welfare (MoH) and to the offices and informal meeting places of donor representatives with which the Ministry worked. I tried to pay attention to how health care provision influence s and is influenced by institutional histories and orders, and to understand how wider fields of state and donor governance were grounded and negotiated in the minutiae of daily practice. I also focused on the ways past and future ambitions of the hospital workers were built into the hospitals ar chitecture, and to see where health personnel maneuver ed the opportunities and limitations that came about at the intersections of HSR and targeted health interventions. To explore these broad themes, I followed hospital personnel and patients within the h ospital, tracing the ways that their movements and practices pointed to particular state and donor funding streams and governance procedures, and how these intersected with less prioritized areas on the hospital During the second half of my research, I tu rned to interviews and analyses of archival 6 The DC is the local level of government. Under structural adjustment programs in Tanzania, much of the decisionma king power moved from the central government to the DC, and the government ministries reoriented their priorities towards policy making. See Chapter Four for additional information about this process.


53 material, media, and reports and publications of a multitude of institutions, donors, ministries, consultants, and other organizations involved in the health sector in Tanzania My primary fieldsite in Tanzania was a government owned district hospital in Arusha regionan area teeming with expatriate businesses, nongovernmental organizations ( NGOs ) faith based organizations ( FBOs ), communitybased organizations (CBOs) safari companies, and a variety of other or ganizations and expatriate private businesses that made communities in the district highly familiar the presence of visitors, diplomats, development workers, vacationers, and any number of visiting volunteers aiming to help in some small way by working i n orphanages, schools, health facilities, churches, and any number of CBOs. Due to a long history of Arusha town being a favored site of colonial administration and potential employment opportunities for Tanzanians and others, the region i s multiethnic, a nd Kiswahili is the most common language people used to communicate with one another. Exchanges in other languages, when they occur, most often take place in villages that are far from main transportation roads, or with acquaintances known to be from the s ame tribe. Within the hospital as well as my other forays i n Tanzania, I was able to rely on Kiswahili primarily, although in some settings like the MoH and development organizations English was more appropriate. Fieldwork in a hospital is necessarily mul tisited. Although it is often presented as one entity, in practice hospitals encompass a variety of different places, each with its own workflows and routines. Kiunga District Hospital was home to four inpatient wards, two laboratories, two filing areas, numerous outpatient clinics, a dental unit, staff housing units, two surgical theatres, immunization rooms, nurses stations, an xray and ultrasound unit, and a plethora of offices and other d i visions. As I was interested in understanding what the hospital is as an institution, the multiple networks and movements that make it function, and the lived experiences of the people


54 that make it up, my research required a degree of strategic planning in order to be able to grasp these configurations. In the first s everal months of my research, I concentrated several weeks each on certain areas that I felt were of particular interest. This interest came about when, in the first days of my research, I realized that certain units were particularly well resourced, while others were struggling. Thus, during those first months, I spent consecutive weeks in the same ward or clinic, and once I felt that I had a grasp of the routines and quality of interactions of that unit and had established good rapport with the workers there, I would move to another unit. About eight months into my research, after considerable reflection on these routines, I bega n following connections between departments and out from these units to other organizations or bodies with which they were networked. I thus followed (and often assisted with) not only patients as they moved between units, but also laboratory samples, patient files, disease registers, budget summaries, NGO representatives, visiting groups, expatriate volunteers, administrators, dis trict trainers, donated supplies, and various other people and objects as they moved within, and often outside of, the hospital. As it turned out, I was able to attain some very rich and often unexpecteddata by utilizing this research strategy, and percei ved connections that would have remained wholly obscured had I stayed exclusively in particular units. Following these connections outside of the hospital took me to district council meetings, peoples homes, donor and government sponsored training works hops, and on three occasions, to Dar es Salaam. Dar es Salaam is home to the country offices of the majority of bi and multi lateral donors active within Tanzania, as well as to most of Tanzanias government ministries and larger libraries. On my trips to Dar es Salaam, I mined the National Library as well as the smaller library at the MoH, interviewed eight officers within the ministry, and sought out formal interviews and informal conversations with a variety of people who worked in bi lateral donor


55 offi ces and NGOs running vertical programs within the health sector I formally interviewed five representatives working for three separate bi lateral donors within Tanzania and three officers working with the transnational NGOs operating programs through the hospital. However, I had informal conversations and attended social gatherings with dozens more, culminating in numerous exchanges during the 2008 Joint Annual Health Sector Review meeting in Dar es Salaam Within the hospital itself, I did qualitative in terviews and life histories with 38 of the nearly 120 hospital employees, ranging from the top administrative levels with the District Medical Officer (DMO) and District Nursing Officer (DNO) all the way down to the ward attendants and nursing assistants t hat make up the bottom ranks of the official hospital hierarchy. In addition, the hospital recently began allowing expatriate volunteers affiliated with various organizations largely from the United Kingdom, United States, and Scandinavian countries to wor k within the hospital for periods ranging from a few days to four months. As these visitors were part of the strategies the hospital used to network with nongovernmental entities and increase its access to potential resources, I also interviewed nine expa triate volunteers Some I knew well from daily interactions, others I did not. I also made a trip to a regional hospital in another part of Tanzania, where I was able to interview the person who was the DMO in the hospital from the late 1990s up to 2004, w hich aided me considerably in building a history of the hospital prior to and after HSR was implemented. These interviews and life histories were critical to my ability to reconstruct not only the hospitals past, but also and just as importantly, an under standing of the ways that the challenges of working within the health sector have shifted through time. Interviews with all participants lasted between twenty minutes and two and a half hours.


56 While I had numerous casual conversations with patients at the hospital, I also conducted 27 brief exit interviews, to try to gauge patients impressions and experiences of care, and in the case of those who lived in the area over a long period of time, impressions of how the hospital changed. The exit interviews wit h patients lasted between five and fifteen minutes, but they had their own se t of complexities As a researcher working in a smaller community, and associating daily with the hospital staff, people often presumed I was a medical professional, despite my at tempts to identify myself as a researcher. Many patients and relatives assumed that through me they might be able to access additional resources or care, and on numerous occasions patients told me about their symptoms or struggles within and outside the hospital, assuming that I could intervene to assist them. Patients move in and out of the hospital. They must negotiate access to health services, and many suffer terribly particularly in hospitals where there are inadequate supplies of drugs, supplies, sta ff, and the health facility settings themselves are deteriorating. As someone who was visible on a daily basis at the hospital, the ethics of speaking to patients who were struggling to gain access to care were murky at best. In a system in which know who (Whyte 2002) is as or more important than know how, it became clear to me early on that any efforts to speak to patients about my research could give them the impression that the quality of their care might relate directly to the sorts of answers they p rovided My position as an ambiguous actor at the hospital made these exchanges very awkward. Worried that my questions as a researcher would increase the vulnerability of the patients being served there, I opted for informal conversations and exit intervi ews, rather than formal interviews at the bedside or during the course of their care.


57 However, the exit interviews provided minimal insights about the hospital or the challenges for patients and relatives in accessing care. W hen it came to shedding light on the challenges and opportunities available within the hospital for personnel, patients, and their relatives, informal conversations, daily interactions, and observations of practices were the richest source of data. As will be seen in subsequent chapter s, much can be gleaned from careful attention to how actors move within and outside of the hospital, and it was in following patients and observing therapeutic interactions that the experiences of patients most came into view. In this sense, patients have a very important presence in my study; however, how they felt about this movement or how they understood the system their voices are necessarily absent. Yet several other studies have patient focused perspectives providing important insights about the stru ggles to achieve health and access care (see for instance Fassin 2007; Obrist 2006; Whyte 1997, 2002; Zaman 2004, 2005). In the wake of HSR and donor funded health interventions, additional studies on these complexities are worthy of future study, and are just beginning to emerge (see for instance Kalofonos 2008, 2010). A Researcher in the Hospital Within the hospital itself, where shortages are abundant and staff in short supply, I spent most of my time not only observing the daily routines and interactions of various health workers in the different departments in the hospital, but also often fully participating. While often my participation was in the realm of the mundane assisting in transporting patients, mopping floors, taking laboratory results back to wards, changing bed sheets, counting pills on many occasions, my involvement with particular situations challenged the boundaries of what I understood participant observation to be. Prior to beginning my research at Kiunga, I thought that I would help out where I could, but I had a clear notion in my mind that I was easily going to remain in a researcher role, and that this participation would merely be about helping out.


58 However, as I followed the routines within the various departments of the hospita l, the assistance that was sometimes required challenged this very tidy idea about what exactly participation would entail. My lack of medical qualifications caused me a significant degree of stress at times. Back in Canada where I was born and raised, and the United States where I gave birth to my child, medical work was reserved for those having the training and certification to permit them to do procedures. From specialist physicians to massage therapists, the certifications, qualifications, and licenses of health care workers are commonly conspicuously displayed in American physicians offices for patients to view. In my mind, having no certification or training meant that I should not under any circumstances be involved in assisting in medical procedures. Yet the staff was often unsatisfied with this, and particularly during the night shifts I attended, I would be asked to perform some tasks that went beyond what I felt my role should be. In light of the staff shortages, particularly at night, my prot estations seemed very feeble. I learned some very rudimentary skills drawing medicines into syringes, changing intravenous (IV) bags, and even taking blood pressures but only after repeated observations and demonstrations by the staff. But I drew the line at others, such as providing injections. One particular evening when an elderly and hardworking nursing attendant asked me to provide an injection to a child and I declined, she looked straight at me from where she was helping a wheezing child and said, But Sista Noela,7 you have seen me give injections hundreds of times. Of course you know how to 7 At the hospital as well as more generally, I was oft en called Noela. In Kiswahili it is uncommon for a word to end in a consonant sound (although there are exceptions), and since many womens names end with an ah sound, this vowel sound was also regularly affixed to my name.


59 do it. If you do not give that patient the injection, how are you helping her? The words burned, but on that day I held my stance, and she just shrugged and s aid that patient would have to wait.8 There were times when my ethics beca me situational. There were times when, due most often to a lack of personnel, I was asked to directly assist with medical interventions on a patient, and in very select situations conceded because standing there to do nothing felt less ethical than taking instruction for those who needed the assistance and could direct me. There were particular situations that wrecked me emotionally, and I marveled at how the staff learned to cope. Ethical training provided little assistance in cases like this. I participated where I felt I could, and on a few unique emergency situations, in ways that I had never expected. Working as an ethnographer in a place teeming with well resourced expatriates (whether visitors, NGO workers, business owners, or otherwise) presents its challenges. Hospital personnel, patients, relatives, and people in the surrounding communities and even in Arusha town would assume that, despite my statements to the contrary, I must be affiliated with some kind of NGO or have access to people with resources. Yet the assumption that I was wealthy or had access to a wid e array of potential resources was correct in some respects. I did have access to people in powerful positions I was able to take advantage of my own social capital to an extent; unexpectedly, I encountered two individuals early in my research that were tied into very powerful donor organizations in Tanzania, and these contacts as well as contacts within the hospital administration allowed me to move vertically from the bottom levels of the system to powerful offices where decisions got negotiated and even finalized. Hospital workers, patients, and relatives did not have this kind of access. Even within Kiunga Distric t Hospital, the workers at the lower ranks knew that I was ab le to access the administrators, and I would often be asked 8 To clarify, I never did an y injections while at Kiunga, although each time I would be asked I would recall that nursing assistants words.


60 to tell these individuals higher up in the hierarchy about a particular problem. As often as I could, I tried to communicate these issues when asked and when I felt I could do so without revealing the source of that information or complaint. The unique ability I had to move between different levels of the system was not lost on the hospital personnel at Kiunga, nor to the highranked off icials I met in Dar es Salaam or in the District Council (DC). Knowing I would make a trip to Dar or to the DC in coming days, the staff (particularly in the middle and lower ranks, although sometimes administrators as well) often asked me to find out about particular issues of concern to them. Why were they not being paid their overtime allowances? Was it true that the donors pulled their funding from the health sector? What is the MoH planning to do about infrastructure at district hospitals? Could you te ll the NGO officers that we need a new vehicle for our mobile clinic? Whenever possible and appropriate I tried to find some kind of answers to these questions and incorporate them into my interviews and social conversations with people in powerful positions. When I returned from a trip to Dar es Salaam or the DC, I would often be asked to announce that information publicly as a kind of trip report in the daily morning meeting of all staff members. Similarly when I would arrive in Dar or to the DC of Kiu nga, when decisionmakers within donor officers or the ministry heard about my research, they often asked me questions about how things were functioning. This was one way I became aware of how little the policy makers and donor officers actually understood of the conditions of medical work at the ground level. When they asked me what I saw as particular challenges, I was able to articulate some aspects that could be improved, and back these up with examples of interactions or events I observed at the hospit al. Finally, there were a few other ways in which I was asked to assist in matters that where my expertise was deemed appropriate. The DMO requested on several occasions for me to help him prepar e


61 proposals to give to private donors for building projects a t the hospital, or for help in putting together PowerPoint presentations for speeches he would give in the future at national and even transnational workshops and conferences. I always obliged. When others came to me with computer issues, I used what limit ed knowledge I had to try to assist them. I set up several email accounts for hospital staff who had made contacts with others abroad and wanted to learn how to use computers, and I gave more than one lesson on computer use to staff members after working hours. T he Issue of Confidentiality With the exception of one official working within the MoH, whose identity would be impossible to obscure given his unique role in HSR, all names of people and in most cases, the names of the organizations with which they are affiliated, have been changed. In addition, I have created a pseudonym for the district hospital where I worked. There are aspects of the hospital where I workedwhich I have renamed Kiunga District Hospital that are not representative of the typic al district hospital in the country,9 and many of these details that might make the hospital readily identifiable have been left out of what follows. But I also realized in the process of writing that by identifying the hospital or even those individuals who approved the use of their name, there was a risk that the hospital itself or particular individuals working within it might become the subjects of intense criticism or disciplinary actions, while drawing attention away from the particularly problematic configurations that inform how biomedical bureaucracies and health care provision unfold. 9 At present, I am not convinced that there is such a thing as a typical district hospital in Tanzania, particularly when one delves into the details of each institutions past and the intricacies of how these facilities intersect with the communities within which they are located. However, as will become clear within this study, there are aspects of Kiunga District Hospital that provide it wi th significant advantages compared to other district hospitals in the country.


62 That said, while I changed or omitted many details in order to obscure the location of the hospital and the individual identities of those individuals who interacted with me, I remain sensitive to the fact that a researcher can only provide so much protection to her fieldsite and informants. It is entirely possible that people in positions of power may be able to easily identify the fieldsite or particular individuals statements despite my best efforts. Nonetheless, what is described in the pages that follow was written with every effort to protect the individuals and facility involved, while simultaneously pointing to the ambiguities of the configurations that come about when particular policies and programs are mapped onto chronically under resourced and understaffed health institutions. As an ethnography of an entire hospital undergoing important transformations, my study is necessarily partial, and is a depiction of a particular historical moment within an institutional process that remains ongoing. Further, while my analysis at times problematizes some of the effects of HSR and of donor funded programs that target particular kinds of individuals or ailments, it is important to note that many of the health personnel with whom I interacted would not necessarily agree with my analysis. Several of the issues I point to as problematic or suspect are the very arrangements that the staff look on hopefully or with a sense of pride (see also Whyte et al. 2010). I never encountered an individual who was critical of the projectification of hospital services; on the contrary, the vast majority of staff members articulated formally in their interviews or informally in conversations that they very much hoped that more projects and donors would become involved in the hospital (see also Booth 2004). Below, I attempt to explore, and at times celebrate, the inventiveness of the health personnel working at Kiunga both collectively and i ndividually. However, I also point to cases that are ethically murky, since these often highlight wider structural and political issues that inform the daily practices and interactions in the


63 hospital The fragmentation of the hospital and the uneven distr ibution of resources within its boundaries create significant inequalities that are, in my view, highly ambivalent, and which have not received adequate attention. These structures are embodied, enacted, and negotiated. Organization of this Study This stu dy tacks between the spaces where health policies and programs are enacted and those where they are to be implemented. Chapter Two, Establishing and Building a Health Sector traces historical trajectories through which biomedicine and medical professions became entrenched in Tanzania. The chapter calls attention to the ideals of policies and state initiatives and their incomplete actualization on the ground. Drawing on descriptions from health workers themselves, Chapter Two describes historical linkages between governance and biomedical practices and the challenges of providing care and clamoring for recognition of professional status within a state that was consistently unable to live up to its promises. In Chapter Three, The Hospital Before Health Sect or Reform, provides a brief history of the establishment of biomedical institutions in Arusha region generally, and in Kiunga District specifically. Drawing primarily on the perspectives of health workers present during the 1970s through 1990s at Kiunga, I trace the establishment, growth and difficult working conditions within the hospital prior to HSR. Pointing to the ambiguities of hospital work in a chronically understaffed and under resourced health facility, I note how personal and professional aspira tions met with continual disappointment as the health facility failed to actualize the kind of working environment that would make biomedical training meaningful. The hospital looked to the local community to provide the resources necessary to expand the hospital, and largely succeeded in building up some of the facilitys infrastructure. Yet despite these transformations, hospital staff morale remained low, as health professionals found little within central government policies and programs that would help them achieve their own goals.


64 Subsequent chapters move to consider how HSR was planned through specific donor state collaborations, and the opportunities and limitations that came about in the hospital as a result of important transformations in aid and governance modalities within the country. Chapter Four, Establishing Tanzanias Transnational Aid Regime, provides an important counterpoint to Chapter Two. It reveals how the Tanzanian government negotiated sectoral reforms among a donor community that l ost faith in the states ability to enact meaningful transformations in its governing systems. At the same time, the chapter points to the establishment of a common language and set of priorities among donors, and the ways this impacted the MoHs relations hips with donors. During the 1990s, the MoH enacted an economy of appearances directed at donors, in order to make the country appear to be an attractive locale for donor investment. These efforts were largely successful because the MoH adopted policy fram eworks and discourses that were in line with donor values prevalent at that moment. Further, the chapter points out discordances within these discourses, as individual donor officers articulated concerns about the existing transnational aid regime and fina ncing modalities underlying the health sector in Tanzania established during the planning stages of HSR In Chapter Five, Infrastructure and Professionalism: Expanding the Hospital, I turn to the ways that HSR and donor sponsored targeted interventions helped the hospital expand its infrastructure, and the impacts of this on medical professionalism among the workers at the hospital. Under HSR, health facilities were expected to be entrepreneurial by seeking out their own partnerships with private individuals and organizations that could help them procure resources that the government was unable to provide a significant departure from the orientations of the previous system that only allowed health facilities to garner financial support from local communities and government coffers. Detailing how the hospital doubled in size in


65 the span of eight years, and which donor entities sponsored these efforts, I explore the effects of infrastructural expansion on the ways that health professionals perceived their workplace, and their own possibilities and desires within it. I argue that infrastructure is intricately linked to health personnels notion of professionalism, and that the hospital administration was largely able to garner staff cooperation and investment in the institution by establishing an economy of appearances that made the site appear to be one of institutional possibilities. Chapter Six, Of Parallel and Hybrid Orders, outlines how formal health sector policies and guidelines, rather than replacing existing norms and values within the hospital and its wider community, actually intermingled with existing tacit codes and unofficial norms of the past. The chapter highlights how the official hierarchies meant to dictate roles and relational positions of hospital workers were actually often circumvented by parallel hierarchical orders that favored the social ties that specific health workers had to the larger community regardless of their rank. This parallel order allowed lowerranking and higher ranking workers with influential ties beyond the hospital to thwart official policies to their own advantage, largely compromising the ability of hospital workers with the most training to provide high standards of patient care. The chapter also demonstrates how t he existing and newlyinstituted structuring orders congealed in hospital and district management teams, which served to silence highly trained health professionals concerns in favor of existing political and religious authority structures within the community. Importantly, this chapter argues that although these parallel and hybrid orders of authority thwarted official policies and guidelines, they were in fact completely critical to the hospital administration to achieve the kind of local and state inves tment necessary to achieve the institutional possibilities envisioned for Kiunga District Hospital. In short, those unofficial orders based on pre existing patronage ties within the community were critical to the facilitys ability to


66 appear compliant with state and global governance regimes, and crucial to the hospitals success in procuring funds In contrast to the parallel orders allowing particular individuals to use their social capital in the wider community to negotiate their responsibilities withi n the hospital, Chapter Seven, Concurrent Governance: Salary, Compensation, and Workers Rights interrogates the role of state and donor compensation policies and the ways these (dis)articulate with the professional and personal desires and obligations of hospital workers. Compensation through the government was an issue that was non negotiable to any health professional, regardless of rank. This challenge became further highlighted with the infusion of targeted donor programs within the hospital, which c reated abundance in select units of the hospital, increasing workers dissatisfactions with MoH compensation policies as select few workers were able to access benefits from donors while others were not. Because hospital administrators were poised to distr ibute both MoH and donor resources among hospital staff, they became increasingly vulnerable to accusations of corruption and favoritism as health workers struggled to gain access to insufficient resources in fulfilling their basic domestic and professiona l needs. Thus, the chapter argues that donor programs increased overall dissatisfaction of health workers, because the failures of the government to recognize their personal and professional aspirations were made all the more perceptible when development enclaves were established within the hospital. Donor sponsored resources reminded workers on a daily basis of the states failure to live up to its promises. Chapter Eight, Biomedical Plurality between the Enclave and its Others considers how the uneven distribution of resources and expertise within the hospital impacts biomedical encounters. Arguing that biomedicine as a therapeutic modality is emergent and knowable only


67 through practice, Tanzania is presented as a case that is particularly helpful for u nderstanding the plurality of biomedicine because infrastructure, technologies, expertise, and maladies characteristic of biomedical encounters are never stable, and never taken for granted. Noting the marked differences between departments within and beyond the purview of state and donor interests, this chapter traces which maladies and actors are able to move across the boundaries between enclaves and their others, which are constrained, and the punctuated politics of these movements. Drawing on a vocabulary developed within Science and Technology Studies (STS), Chapters Nine and Ten take documents and their representations seriously as important ethnographic objects. Chapter Nine, Apples with Apples: The Aesthetics of Governing with Numbers, returns to the spaces where health policies and programs are planned and monitored. The chapter outlines the ideals and assumptions underlying health statistics and their significance for governance. Tracing the disjunctures between idealized understandings of the capacities of medical statistics and their complexities in practice, I show that the very facts that donors and governments hotly contest are those that are erased in public documents that circulate widely to depict the complexities of the Tanzanian con text on a global scale, allowing for the country to be compared to other places elsewhere. By turning to numerical representations and the documents through which they circulate across multiple scales, this chapter argues that there is considerable instabi lity within the universal languages characteristic of the transnational aid regime, and that these contestations are largely effaced in the production of global health statistics. Departing from STS approaches that maintain that facts are able to travel because the debates underlying them have subsided, I demonstrate that even hotly contested facts have considerable power to move, and to depict carefully crafted images of the world. Facts are shown to represent across


68 multiple scales, but what they repres ent to differently positioned actors and governance entities differs significantly Although donors and states are able to govern using numerical representations that are known to be inadequate to the ideals underlying health statistics, I argue that they govern only partially, and with continual anxiety. Chapter Ten moves to different engagements with documents within Kiunga District Hospital, and the various motivations underlying these interactions. Moving from documents that are beyond the purview of s tate and donor interests to those that fit neatly within governance priorities, the chapter investigates documentation practices and understandings, and what these mean for the possibilities for biomedical practice within the hospital. Documents not aligne d with state and donor priorities circulate widely within the hospital, and are variously taken up or contested as health workers vie for professional recognition or improvise strategies to deal with the paucities of their workplace or in their own experti se. For practices surrounding documents tied to state and donor surveillance, I draw a distinction between the qualities of statistical reports and representations, and the improvisations that are possible and not possible with different kinds of reports I further reveal how health professionals think about the burdens of documents and the ways that they directly and indirectly impact the possibilities for providing he alth care within the hospital. In the concluding chapter, I review the significance of my findings for Tanzania, and for biomedical institutions in other aiddependent countries.


69 CHAPTER 2 ESTABLISHING AND BUI LDING A HEALTH SECTOR Writing about colonial biomedical discourses and practices in Africa, Megan Vaughan argues, whilst medical discourses must themselves be seen as constitutive of the problems they describe, they may also reflect, albeit very indirectly, material and political circumstances outside the immediate realm of the medical (1991:7). Taking seriously Vaughans call to be attentive to long term political economic and historical processes, this chapter brings together intertwined histories of governance, biomedical institution building, professionbuilding, and daily life in Tanzania from the period when biomedicine first arrived in the area to the 1990s, just before Health Sector Reform ( HSR) was implemented. Inspired by Frederick Coopers call for work that specifies how large scale, longterm processes can be analyzed with due attention to their power, their limitations, and the mechanisms that shape them (2005:111), the discussion below outlines the processes that underlay the establishment not only of a biomedical health sector in Tanzania, but also the creation of medical professionals to work in the sector T he introduction of biomedicine is not politically neutral in Africa or other economically marginalized countries in the world. Rather, the health statistics of such countries populace have continually been the objects of surveillance, as international lenders and bi lateral donor agencies use health as a measure of a states success in its progress towards development. Thus, the development of the health sector in Tanganyika, and later in Tanzania, and the development of an array of classes of medical pr ofessionals cannot be separated from the state and its political goals, let alone from the wider bi lateral and global interests that put pressure from on government authorities in the name of development and health for all. The histories presented below are in line with recent work on civil servants and governance in Africa, which argue that attention should be paid to the historical and social processes and


70 practices that underlie governance (Chalfin 2010; Martin 2009; Blundo and Olivier de Sardan 2006). Yet medical professionals are more than just civil servants. They are also professionals, tied to particular kinds of authority and scientific expertise that have, over the past century, become recognizable nearly everywhere ( Martin 2009; Wendland 2010). The specificity of being a medical professional within resourcepoor biomedical settings located in heavily indebted African states cannot be ignored (see Martin 2009; Wendland 2004, 2010). Many of the difficulties and moral ambiguities experienced by workers in the health sector, both under colonial administration s and subsequent governing regimes, were related to peoples ability to carve out a life for themselves despite inadequate salaries and working conditions. These challenges continue into the present. For many health workers, the changing contexts of their profession had to be balanced against familial and social expectations and obligations of support in contexts of hardship (Martin 2009). While the various colonial and post colonial governments planned to build and improve the biomedical health sector, at no point were objectives fully achieved. In this sense, one might argue that an overarching continuity in the history of biomedical institutions and professions in Tanzania has been the consi derable gap between ideals outlined in policy and the realities of daily practice between the ideals of professional practice and the realities of biomedical working environments Amid chronically understaffed and under resourced conditions, medical profe ssionals grappled to find satisfaction in their ability to provide adequate care, and often struggled to convince their patients of the superiority of their healing modality when the tools and infrastructure demanded by biomedical healing were only sporadi cally available. Biomedicine and the German Colonial Administration: 1888 1918 While in earlier years there are some accounts of biomedical physicians and medical missionaries traveling and even establishing small hospitals in what is today the United Rep ublic


71 of Tanzania,10 the beginnings of Tanzanias government health sector can be traced back to 1888, when two German physicians and four orderlies came to the region with the German army (Clyde 1962) At the time, the region was called German East Africa, and contained the mainland of Tanganyika as well as the regions that later became Burundi and Rwanda. While German missionaries initially brought biomedicine to the region, it soon became part of the broader colo nial endeavor. German biomedical practition ers came to the region primarily to provide health services to German colonialists and military personnel, and to those Asian and African laborers who were part of the civil service (Clyde 1962; Nsekela and Nhonoli 1976). Overall, throughout the period of German colonial rule, there were efforts to develop preventive and curative services sidebyside, with the colonial government engaging several of its military surgeons in campaigns against communicable diseases such as sleeping sickness and malaria S ma ll military hospital facilities were opened in select areas of the region, but early in the German colonial period the majority of Tanganyika was largely untouched by biomedical institutions. In the 1890s two hospitals were built in Dar es Salaam: Sewa Ha dji and Ocean Road. Sewa Hadji Hospital was named after the Ismaili benefactor who donated funds for its construction. He was a prominent trader in the area, with considerable political influence The hospital opened in 1893, but continued to be expanded until 190405. In 189495 the German government financed a government hospital, the construction of which was completed in 1897. The Ocean Road Hospital, as it was named, became the main medical facility serving Europeans in Dar es Salaam (Clyde 1962). 10 One of the most well known of medical missionaries to travel in the region during the precolonial period was Dr. David Livingstone, of the London Missionary Society, who tra veled to Zanzibar in 1866 to attempt to find the source of the Nile River. See Livingstone 1874


72 In subsequent years other facilities were built. In 1893, there were two small hospitals outside of Dar es Salaam, one in Kilosa and another in what is now Moshi, and a military surgeon was posted to each. The following year, the colonial government established hospitals in Tabora, Ulanga (now Ifakara), and Bukoba, followed by three others in Mwanza, Lindi and Iringa in 1895, and in 1896 a hospital was opened in Ujiji. In 1910, hospitals were erected in Arusha and Mwanza. By the beginning of the First World Wa r, there were dispensaries or hospitals in most of the main towns in Tanganyika (Nsekela and Nhonoli 1976). Yet in practice, very few Africans sought out biomedical treatments in the early colonial period, with the exception of those communities near estab lished missions (Clyde 1962). Missions cropped up throughout the region from the middle of the nineteenth century, and this development accelerated under the German colonial administration. Up to the 1920s, although the German colonial service erected a fe w biomedical establishments were erected, missionaries provided most biomedical care to Africans. The francophone White Fathers had been in the area the longest, and their missions were also among the quickest to spread. They established early missions at Karema, Tabora, Bukumbi during the 1890s, and extended from Bagamoyo to Kilimanjaro, Ugogo and Usambara, and later through Ufipa, Mbulu, Buha and near Lake Nyasa. The Universities M ission to Central Africa (UMCA) established itself between Bonda and Uzigua, while the Church Missionary Society (CMS) had some stations at Ukaguru and Ugogo. In the south, there were several German Benedictines posts. Missions were thus not only competing against Islam in Tanganyika they were also competing with each other. The missions often used education and medicine as a means to attract converts, and many of those converted to Christianity were people of low status within their communities (Iliffe 1979). A


73 missionary doctors memoir illustrates the way that biomedicine and missionary work might be combined: [T]o heal the sick not only to make them well, but also to make the works of God manifest. It is this witness of Gods nature and this witness to the gospel of Jesus Christ that fulfills the mission of the hospital. Let no one argue that it is not good merely to place the healing profession in the midst of a people that has not produced its own healers; this is good in itself. (Dibble, 1965:26. Emphasis in original)11 Incorporating biomedicine into missionary work was seen as a means of attracting converts, combating African superstition, and competing with the spread of Islam. However, rather than eradicating African therapeutic modalities or Islam, these missionary endeavors often increased therapeutic pluralism in the re gion (Ranger 1992). An African biomedical practitioner produced one of the earliest accounts of missionary medicine in Tanganyika: Adrian Atiman. During his childhood, Atiman was captured somewhere in West Africa, and became part of a slave caravan travel ing across the Sahara to the coast. During his transit, White Fathers missionaries rescued him. He received an education in Algeria, and was later sent to Malta for additional studies. Atiman accompanied some of the first missionaries to Tanganyika, and fr om 1888 worked as a medical missionary at Karema. His is one of the few existing autobiographies of an African biomedical practitioner (Iliffe 2002) When he arrived in Karema, he describes the tools and conditions with which he worked: My medicines compr ised 20 kilos of camphor, a little carbolic acid, and ipecacuanha. For instruments, a thermometer, an irrigator, and nothing else. There were very many ticks, which we swept up each morning, before they had finished their meals of blood for the day. Each morning I visited the orphans and treated those who were ill. Persistent fevers were prevalent owing to lack of quinine. Since this cost 1000 francs a kilo, it is 11 Dr. Dibbled worked in Tanganyika. It is ironic that he states that no other healers existed, since nearly every other missionary and colonial physicians acco unt at the period makes specific mention of other healers existing in the areas where they worked. Some of these accounts are provided below.


74 obvious that it could not be given to everyone, and it took a year to obtain fresh supplies of medicines. Bandages were made from clothes used by the orphans. (Atiman 1946:61). Atiman further recounts a number of deaths among his colleagues from the diseases prevalent in Karema at the time: In 1890, Father Auguste Carmoi, died from haematuria (bla ckwater fever); in July 1891, Father Jean Marie Josset died from the same cause; and in November 1891, Father Emile Pruvost died after a fever lasting nearly a month, and was buried by Monseigneur Lechaptois. By these deaths the staff of Tanganyika Vicaria te was reduced too. F our missionaries only! (Ibid.:64). Despite the presence of medical missionaries, Africans did not abandon their therapeutic practices in favor of biomedicine and Christianity. B iomedicine became one among many other healing moda lities available. According to a German military doctor who worked in Uhehe for two years at the turn of the twentieth century, The natives state that in Uhehe for as long as they can remember there have been doctors, that is people who not only know how to perform magic but also attempt to heal the illnesses of their fellow men. Indeed, their influence as doctors, which is still considerable, was even greater previously; competition from European colleagues and missions is now making life more difficult for them (Weck 1969[1908]:31. Punctuation in original ). M issions and biomedical practitioners worked to convince Africans of the utility of their healing modality While Europeans were convinced of the superiority of their medicine the Africans among w hom they worked were not necessarily interested in utilizing one modality exclusively. From the time when Tanganyika became part of a German colony to the period when German rule ended, there were significant increases in the number of biomedical practiti oners posted to the region. In 1896, of the military medical personnel, there were sixteen surgeons, twenty one medical orderlies, and approximately ten German nurses. In addition, there were a few missionary doctors and nursing sisters in hospitals in mor e remote areas. By 1910, the numbers increased significantly, with 53 physicians in the country (43 affiliated with the


75 military), 100 German orderlies, and 11 German nursing sisters (Nsekela and Nhonoli 1976). By 1914, some German civilian doctors came to work in German East Africa, but overall the government did not prioritize biomedical services for African populations. Hospital work was one of the main areas in which Africans could gain employ within the colonial medical service. The majority of Africa ns working in the medical system at this time were appointed menial positions and provided on the job training, By 1914, European nurses were providing training to African attendants, and despite the fact that many of the latter were illiterate, they becam e highly skilled (Iliffe 2002). During World War One ( WWI ) the majority of German medical professionals were reassigned from work at military bases and colonial establishments to war related duties, and thus the majority of organized biomedical services came to a halt. In 1917, Africans were recruited as combatants rather than simply laborers as part of the African Native Medical Corps (ANMC). The ANMC arrived in Dar es Salaam in July 1917, staffing the Sewa Haji hospital and several other medical facil ities inland and on the coast. Africans within the ANMC did specialized work and were to be treated respectfully. According to Iliffe (2002) the war elicited a sense of comradeship between the Africans and the Europeans. It was within the ANMC that the Eu ropeans learned how to train East Africans in medical skills, and they continued this training after the war. When Germany was defeated in WWI, its colonies were redistributed through the League of Nations, with Rwanda and Burundi coming under Belgian auth ority, and mainland Tanganyika becoming a mandate of Britain, renamed Tanganyika Territory. The British took over Dar es Salaam in 1916, but even several years after the war ended, the biomedical services available in Tanganyika were minimal at best (Nsekela and Nhonoli 1976).


76 Biomedicine and the British Colonial Administration, 1918 1961 As the British took over Tanganyika territory from the Germans, medical officers from other posts in Africa were brought into the colony However, particularly in the years prior to official takeover of the Tanganyikan mandate in 1923, staffing levels were entirely inadequate. Nurses were in particularly short supply. In 1919, the Ocean Road Hospital had only one nurse, Elizabeth Kemsley, who had to care for sixtyfive p atients without any assistance (Clyde 1962). By 1920, the British consolidated its control over the major administrative centers of Tanganyika. They inherited twelve hospitals from the Germans, as well as a sanitarium and a mental hospital. According to the Handbook of Tanganyika: During 1920, several Medical Officers, most of whom continued to serve on contract after their release from the forces, expressed their unwillingness to accept permanent appointments or to continue indefinitely under terms which w ere provisional, while others who were long overdue for leave wished to return to England. As a consequence, only nine Medical Officers were left to provide for th e medical wants of the country. It was not, however, until the beginning of the year 1921 that the furnishing of regular monthly and other routine medical reports was effected. (Handbook of Tanganyika 1930:362 363) As a therapeutic practice linked up with colonial rule, documents such as medical reports were important to governance in Tanganyika. From early in the establishment of health facilities in the region, biomedicine and bureaucracy were intertwined.12 In 1921 the British colonial government employed additional staff: 26 medical officers, 17 European nurses, and one dentist. The government also imported 19 Indian subassistant surgeons. The Territory housed two mission doctors and three private doctors (Nsekela and Nhonoli 1976). In addition to the government and Christian medical facilities, the Muslim Ismailis set up a private d ispensary in 1929 in Dar es Salaam, which would later be expanded to 12 The role of documents in governance and in biomedical practice are discussed in Chapters Nine and Ten.


77 become the Aga Khan Hospital, which today provides a variety of preventive, curative, and specialized services to the city to Africans, Asians, and Europeans alike (Kaiser 1996). One of the British doctors who joined the Tanganyikan Medical Service after the change in colonial administration was Dr. Donald Latham, who, along with his wife Gwynneth, was stationed in Lushoto in 1925.13 Mrs. Latham, who was not biomedically trained, quickly l earned how to work as a surgical assistant to Dr. Latham. Recalling life in Lushoto in 1925, Mrs. Latham writes, There was no other doctor or dentist within hundreds of miles, so poor Don did everything and had to bear all the responsibility of diagnosis and treatment, medical or surgical (Latham and Latham 1995:34). Describing the typical day of providing health services in Lushoto, Mrs. Latham states: I would sit at a table with a register and a great box of small squares of blank white paper, while Don, stethoscope round his neck, would stand beside his array of bottles. The askari14 would herd the crowd into an orderly queue and the days work would begin. Each patient would be examinedchest, eyes, heart, mouth and spleen and even to me, a lay person, certain symptoms would be discernible at a glance. It did not take us long to realize that a great number of the attendants were there merely out of curiosity. For example, on enquiry a patient would say he had a cough and after confirmation he would be gi ven a bottle of cough mixture. I would write his name in a register and if it were necessary for him to return the next day I would write his number on one of my slips of paper and tell him to bring it with him. Very often, when the next day came an entire ly different person would present the numbered chit. Usually a whole string of coughs would follow, then a genuine patient would produce a pain the tummy, then a string of tummy aches would follow until the next patient arrived, and so it went on. (Latha m and Latham 1995:92) 13 The m emoirs of Gwynneth Latham, which are said to have been written in the early 1940s about her and her husbands life in Tanganyika during the 1920s and 1930s was preserved and finally published by their son, Michael Latham, who himself during starting in 195 5 worked as a District Medical Officer in Tanganyika for nine years. Mrs. Latham passed away in 1972. All quotes taken from the Latham and Latham 1995 source are the portions of Gwynneths diary that was published there. 14 Askari is the Kiswahili term for guard or police officer. In this context, Mrs. Latham is likely referring to a guard.


78 As this excerpt suggests, medical professionals were short staffed and grappled with limited supplies. Meanwhile, people were deliberate in how they engaged with the missionaries, not necessarily submitting to biomedical and bureaucr atic regimens in the ways the medical practitioners expected Under German colonial rule, and in the early years of British rule, very few of the biomedical treatments developed and tested within the colonies were effective. Quinine for malaria and bismut h for yaws were two of the major successes (Malloy 2003). In 1921, the Europeans discovered that bismuth oxide was effective against yaws, which was prevalent and highly endemic in Tanganyika. The salt was administered via injection widely throughout the t erritory, with 75,000 cases being treated by 1925 (Nsekela and Nhonoli 1976). This remedy spawned a reputation for particular biomedical treatments, with the African populace beginning to favor injections over other biomedical therapeutics.15 Indeed, working in Lindi in the late 1920s, Mrs. Latham recalled: In recent years much had been done of the disease called yaws using injections. The sindano or needle, had become a sort of miracle cure. On any of these medical excursions hundreds of natives would sim ply demand an injection without describing their ailments, even saying they were sure it would do them good, give them strength and make them physically and sexually virile. They would almost weep with disappointment if a sindano was refused and an alternative treatment provided. (Latham and Latham 1995:94) However, many of treatments attempted during this period were toxic, and there were no effective biomedic al treatments for parasites at the time (Malloy 2003). After 1923, the British began replacing te mporary biomedical structures left over from the German administration and bringing more British doctors to the territory. At the time, more 15 Injections as a favored treatment among Africa populations are discussed below for the 1970s through the 1990s in the hospital where this research was conducted. T here are several good histories of the emic understandings of the power of injections in colonial Africa. See Malloy 2003, White 1995, Vaughan 1991.


79 African patients were being treated in biomedical facilities, due in large part to the popularity of injections. Thus, the British began to expand the biomedical workforce by training African paramedics. In 1924, the first course for training African dispensers began in the territory.16 African dispensers were mainly trained in curative medicine, with the aim of having them assume the responsibilities of outpatient clinics, laboratory investigations and prescribing in the government dispensaries. Initially the course was 18 months long, and later it was expanded to three years. There were 81 dispensers trained by 1929 (N sekela and Nhonoli 1976). African tribal dressers became another cadre of medical training developed under the British administration. The training consisted of three months, where dressers were given instruction in diagnosis of common ailments, sanitation, and first aid. They were to become the grassroots, first line biomedical practitioners in the colony. By the end of 1929, 247 tribal dressers worked in the colony, and by the 1930s they were treating the majority of African patients coming for biomedica l services in the Territory. A physician and researcher working in Tanzania during the 1950s described the difficulties dressers faced when they went on their placements: [S]ince there were never enough doctors to fill the needs of the medical centers, dre ssers often found themselves having to take responsibility far above the level they had been trained for. There they were, these young African men, working alone in small, badly equipped dispensaries out in the bush. Always in dire want of essential equipm ent, with only limited drug supplies and without a hospital within reach to send patients to, they had to treat even the most gravely ill patients and therefore worked under severe stress. (Jilek Aall 1979:50). With biomedical expertise but little of the s upplies necessary to utilize their professional skills, African tribal dressers grappled with working environments that undermined their abilities to 16 This cadre of worker would later be known as hospital assistant and ultimately became the cadre that is today known as medical assistant.


80 fulfill their own or their patients expectations of healing. From the beginning, they learned to grapple with scarcity. By 1930 the colonial government trained laboratory technicians in an eighteen month course, later to become a three year curriculum (Iliffe 2002). However, prior to this some Europeans were informally training uneducated Africans to use a m icroscope. Mrs. Latham writes, Don had soon realized that a reasonably trained but relatively unschooled African was excellent with a microscope. They have keen eyes and, having been taught the rudiments only of certain diseases (Latham and Latham 1995:9 3). Even in more remote mission hospitals, missionary doctors were known to have trained Africans in biomedical treatments. Dr. Leader Stirling,17 who was a missionary doctor in southern Tanganyika beginning in 1935, began training Africans as medical pract itioners within a few years of arriving in the colony, stating It was obvious that for any promising future, African doctors and nurses were needed who could eventually replace us completely (1977:49). However, in the 1930s, Dr. Stirling felt that training Africans as physicians would be impossible due to their overall lack of education, But nurses were as much needed, and here we could make a start, although suitably educated girls were few. The idea of trained female nurses was received on most sides w ith amazement and even disapproval. It had never been done and was an impossible project (Stirling 1977:49). Notwithstanding initiatives to scale up training African medical professionals in the late 1930s, there was inadequate staff available to bring s ervices up to desired levels. During the 1930s, the only available medical school for East Africans was Makerere College, in Uganda. However, although the program started there in 1923, very few Tanganyikans underwent the 17 Leader Stirling was a major figure in Tanzania. Having begun work as a missionary doctor in 1935 in Masasi, southern Tanzania, he stayed in the area for many years, became a Tanzanian citizen in 1961, and was Minister of Health of Tanzania from 1975 1980. He was also one of the first physicians to become a Minister of Health in Tanzaniain prior years the post had usually been held by politicians (Stirling 1987). Dr. Stirling died in Tanzania in 2003, at age 97. For f urther information, see Sunday Times 2003.


81 training to become assistant medical officers (AMOs) .18 In 1938, three secondary school boys were sent to Makerere College in Uganda for training, and the first Tanganyikan to complete the program, Mr. J. R. Mutahangarwa, graduated in 1940 (Iliffe 2002; Nsekela and Nhonoli 1976). By 1948, S ewa Haji Hospital had a training program for AMOs as well as hospital assistants, the duties of which were described by an unpublished pamphlet of the Tanganyikan Medical Department as assisting: the medical officer in a hospital. He may be asked to treat African outpatients, assist the medical officer at operations, mix medicines, look after the work of the African nursing staff, or take charge of the sick people in the hospital when the medical officer is on safari. Sometimes he has charge of a small d ispensary where he does everything himself. (quoted in Titmuss 1964:15) Training of African personnel in the lower cadres was somewhat more successful. Formal nurse training began in 1940. Both men and women were educated in nursing, and several young wome n were also trained in midwifery. In 1950, a nursing school was established in Moshi, and in 1953 the colonial government took over an existing nursing school run by the Overseas Food Corporation, which allowed them to double the output of male nurses (Nse kela and Nhonoli 1976). Dr. Stirling established the first nursing school in Tanganyika at the Lulindi hospital in Masasi. A student of the Lulindi school, Thecla Grace Mchauru was the first Africa woman to qualify for the government nursing certificate in Tanganyika in 1943 (Nkya and Anduru 1993; Stirling 1977). Working as a nurse in the 1940s, Ms. Mchauru states, During those days I would be worried about a patient who was beyond recovery. Nevertheless I would do all in my power to 18 Originally, graduates of the Makerere Medical School were called Senior Native Medical Assistants. The title was renamed Senior African Medical Assistants in 1931, and again renamed African Assistant Medical Officers in 1937 (Iliffe 2002).


82 help until I convinced myself that the patients death was simply inevitable and that there is nothing left that I could do (Nkya and Anduru 1993:16).19 At that time, it was not common to give a patient drugs to take at home. This practise is just not good. Some people do not complete taking the prescribed doze [ sic] when left on their own. One might abandon the drugs and forget about them as soon as one felt one was getting better. So during those days, we made sure that the patient completed the dosage and even those who found it difficult to take their medicine orally received our help. The main thing was to make the patient realise that he or she had someone who cared, someone to whom he or she could turn to for reassurance. This encouragement from a nurse, this demonstratio n of affection from one who is ill, is sometimes better and more effective than taking medicine. (Nkya and Anduru 1993:1617)20 Clearly, patients were supposed to submit to the biomedical expertise of nurses and doctors in the course of their healing regime ns an idealized notion of biomedical interactions that rarely, historically or currently, depicts the realities of patients and relatives agency in therapeutic encounters (see also Martin 2009). Although several Africans received biomedical training, Eur opean attitudes towards their African colleagues and patients could be extremely negative. For instance, while Africans often referred to the African AMOs as doctors, Europeans refused to do so. Europeans often exhibited suspicion about the ability of Afri cans to practice biomedicinea suspicion that continued to fester throughout the remainder of the colonial pe riod and after i ndependence (Iliffe 2002). Dr. Stirling recalled his experiences in the 1930s and 1940s with the negative attitudes of his European colleagues to the Africans with whom they worked and whom they treated: 19 Nkya and Anduru 1993 contains excerpts of the authors interview with Ms. Mchauru conducted during the early1990s. 20 While she was the first female African nurse in Tanganyika, Ms. Mchauru did not practice nursing f or long. She was selected to undergo training in Community Development in Kenya, and upon returning to what was then Tanzania in 1957, she became a Community Development Worker. In 1968 Julius Nyerere appointed her as the Secretary General of the womens u nion, Umoja wa Wanawake Tanzania (UWT), which was affiliated with the nationalist party that eventually helped secure the colonys independence. She later went on to several other political leadership roles in Tanzania, and never returned to the nursing pr ofession (Nkya and Anduru 1993).


83 Once when I was visiting the hospital of another mission, the doctor showed me a rather primitive outpatient building, contrasting badly with the pleasant rooms provided for sick miss ionaries. Good enough for the native s, she remarked quite casually. I began to realize that I had to contend not only with the devil and disease, but with the shocking prejudices of some of my senior colleagues. Another said flatly All Africans ha ve bugs, even the cleanest of them, a preposterous statement that I found totally false (Stirling 1977:13). African AMOs were not well compensated, and earned considerably less than Indian Sub Assistant Surgeons and European Medical Officers working in Ta nganyika. Initially, the AMOs made little protest about the differentials in pay, but by 1937 they began to express their displeasure when a European doctor remarked publicly that they were better trained than their Indian counterparts (Iliffe 2002). Nor w ere nurses salaries much better. Recalling the conditions under which she worked during the mid1930s through retirement, an African nurse wrote to the Daily News newspaper in 1973 that When I started training 35 years ago we were so underpaid that it wa s almost impossible to buy a pair of stockings or tooth paste. We were grossly over worked compared to other students and naturally enough we resented it. However, I think it has been, and always will be, true that nurses are exploited ( The Daily News Jan. 20 1973:5. Punctuation in original ). This is a telling example of how biomedical professionals began clamoring for recognition of their professional status. From the late 1930s on, there have been numerous altercations between health professionals and the Tanzanian government over inadequate compensation and poor working conditions elements that undermine their professional status as biomedical practitioners. The development of biomedical services in Tanganyika slowed considerably during the Great Dep ression of the 1930s, and it was not until 1943 that the level of financing of the biomedical health sector reached same level as 1930. Vacant posts remained empty, and hospital maintenance funds were severely cut. Training came to a standstill. With the S econd World War (WWII) following the depression, development of biomedical services and personnel in the


84 colony dropped sharply at a time when demand for the services was on the rise ( Iliffe 2002 ). Resources were scarce, and most were directed towards war efforts. Dr. Stirling reflected back on the conditions he encountered and what was required to cope: Improvisation was therefore forced on us, and I had to use my ingenuity to supply what was needed from time to time. The first step of course was to go thr ough the cupboard to see what existing instruments could be converted (perverted, if you wish to be hypercritical) to some essential use for which they had never been intended. (1995:106107, parenthesis in original). The remaining staff at government dis pensaries and hospitals was overburdened by demands for services, and medical personnel willing to move to Africa during the war and serve in the colonies were unable to do so unless they joined the Royal Air Force (RAF) Nursing Service (see Robertson 1993). Those medical personnel left to do the work in Tanganyika were entirely overburdened. In one of the letters Dr. Stirling wrote during the late 1930s, he entreated, There is still no news of any more doctors for this diocese, yet the volume and complexi ty of work keeps on growing. If you are a doctor, please come; if you know a doctor, please send him. If you dont even know one, please pray for some to come, and keep on praying (Stirling 1947:128). At the end of WWII, the government resumed its expans ion of medical services, doubling the budget for medical expenditures between 1940 and 1945. However, the services still did not meet demand. In a Tanganyikan tenyear development plan published in 1946, the Director of Medical Services stated, it should be noted that there is no existing provincial establishment and that the ad hoc nature of the staff now employed is based largely on factors of expediency and the availability of personnel and funds (Tanganyika Development Commission 1946:49). Further, [O ]ur medical service at present cannot be considered to be more than a token service. Within the framework has been included a health service even more diminutive than that qualification implies. If the modern concept of public health is


85 to be accepted as comprising the guiding and advisory purposes of preventive medicine concerned with urban, rural and labour health problems, we must face the issue of providing an adequately manned health service considerably in excess of our present absurdly small branch of the department. (Tanganyika Development Commission 1946:49). T he war caused considerable disruptions to colonial services, which were slow to regain their momentum. When expatriate medical professionals were recruited for colonial service after 1945, the territories were ill equipped to receive them (Holden 1984a). Through the 1950s, demand for curative services was rising, exceeding the limited capacity of the existing health system. Even the increases in trained staff were unable to meet the demand. A ccording to Nsekela and Nhonoli (1976), between 1955 and 1960, the number of patients at dispensaries doubled from 2.7 million to 5.6 million. Missions provided many of the services in the colony, particularly for child and maternal health. They also estab lished training facilities, and worked in government hospitals in the colony (Nsekela and Nhonoli 1976). However, the British colonial government was biased in favor of Christian missionaries when it came to providing social services (Dilger 2011). While m issionary and government health facilities were set up to cater to both European and African populations, there were no official provisions for healthcare or education for the established Asian communities living in the colony. Thus, Ismaili communities be gan erecting their own facilities under the British c olonial administration; however, the government only provided grants to medical missions established in rural areas, and because most Ismaili communities were established in cities and towns, their facil ities were ineligible. I n the late 1950s, Ismaili clinics, often supported by Aga Khan IV, began allowing Africans to receive services in their clinics, and at independence, there were 12 Aga Khan supported medical facilities in Tanganyika (Kaiser 1996:63) Dr. Louise Jilek Aall, a Swiss physician and medical researcher, went to Tanzania to work at Ifakara during the 1950s. She describes the crowded conditions of the hospital in her memoirs:


86 The hospital complex, consisting of one store y brick buildings, w as spread out over a large area, each wing connected by a roofed walkway. The wards were always over crowded and sometimes patients shared their beds with family members. Others slept on mats on the floor between the beds. I could not make out who were the patients and who were the relatives or visitors. In the evenings people were sitting around small fires everywhere on the hospital grounds, cooking food for themselves and for their sick. (1979:1718). E ven in the 1950s, although African biomedical pract itioners were better trained than prior to WWII and had more responsibilities, they were still considered to be of lower status than their European counterparts, which bred resentment among the African AMOs (Iliffe 2002). Dr. JilekAall describes that Onl y African personnel worked on the wards during the night. One European nurse was on duty and stayed in a house close enough to the hospital to be quickly at hand if called (1979:19). Indeed, Africans often shouldered the most menial and cumbersome duties (Iliffe 2002). However, negative attitudes towards Africans were not universal. Bridget Robertson, a nurse in the colonial service worked in multiple British colonies in Africa from 1947 to 1963. Robertson served as a nurse in the colonial medical servic es, transferring from post to post within Kenya, Seychelles Islands, Zanzibar, and Nigeria. Her posting to Zanzibar began in 1954. In her memoirs, Robertson describes the caliber and commitment of British and African nurses with whom she worked: At the tim e of my arrival, many of the medical and nursing members of the staff had long records of service. The matron and the sister tutor had been in Zanzibar for more than ten years, while some of the African staff nurses, whom I had the privilege of presenting to Princess Margaret some months later, could claim periods of up to thirty years service (Robertson 1993:71) Dr. Jilek Aall was also impressed with the talents of the African dressers with whom she worked as a traveling physician in Ulanga district. She explains: I was amazed how much the two young dressers were able to do in their small laboratory. With a microscope, a centrifuge and a few reagents they were able to diagnose the different types of malaria and most parasitic diseases. My two


87 assist ants were easy going, goodhumored young men, thankful for every opportunity to learn and eager to discuss the problems we met with when working together (1979:39). Meanwhile, prior to during, and after the colonial period, the majority of local people in Tanganyika relied on healers to procure treatments for their maladies, and African therapeutics were an important part of social practice and daily life not only for individuals, but for entire communities (Malloy 2003; Langwick 2007). Healers intervened not only upon the body, but also upon communal and community issues relating to family relationships, infertility, misfortune, epidemics, warfare, witchcraft, and other disasters.21 These therapeutic practices were not uniform, and were not bound to one eth nic group or another. Instead, each healer or diviner had his or her own unique history (see Feierman and Janzen 1992). The therapeutic modalities of individual healers might be learned also through apprenticeship or by divine inspiration from ancestors or spirits (see Langwick 2007; Malloy 2003). It was not uncommon for patients to try out (Whyte 2002) multiple kinds of medicine and healers in order to treat their maladies, and this practice continues today in Tanzania (see Chapter Eight ; Langwick 2008, 2010; Whyte 2002). For instance, working in Tanganyika in the 1930s, Dr. Stirling recalled only being solicited by Africans for very severe cases, stating Modern scientific medicine was still considerably suspect and was only to be tried as a last resort when the traditional local remedies had failed (1995:64). Another missionary doctor working in Tanganyika, Dr. Pascal Imperato, describes his experiences with patients and their health seeking behaviors thus: Many of the Africans looked upon us as just one other place among many, where a cure or an amelioration of symptoms could be obtained. If we failed, the witch doctor would be consulted. If he failed, then still another witch doctor would be 21 For further information on colonial and postcolonial therapeutic practices in eastern Africa, see White 1995, 2000; Whyte 1997; Feierman 1990; Feierman and Janzen 1992.


88 sought out. There was no set rule to determine we would pre cede the witch doctor or follow him in the treatment an individual African (1964:38). Dr. Dibble made a similar observation, noting a complex intermingling of the old superstitions with the new concepts which are being introduced. The ancient framework is still there and oftentimes the new ideas do not even supplant the old ones, but merely take their place in the thought structure of the individual (1965:98). Meanwhile, Dr. Imperato lauded the benefits of germ theory while simultaneously acknowledging th e challenge of convincing people in Tanganyika of their superiority: It is to be hoped that modern medicine, which is changing cultural concepts of disease causation and treatment, will nevertheless engender in Luo society that confidence which the tradit ional practitioners have enjoyed for centuries (Imperato 1966:201). P atients were deliberate in their interactions with doctors, arriving at the mission hospital in search of a particular treatment, and refusing treatments for other conditions (see White 1995) Many biomedical therapies were reinterpreted based on local etiologies of disease: Africans who swallowed three days supply of tablets at once may not have misunderstood a nurses instructions but may instead have considered those instructions to be a misguided and inappropriate way to deal with disease and pain (White 1995:1391). Thus the power and meaning of particular European treatments were redefined in terms of local understandings of healing. Doctors working in Tanzania during the colonial period and just after independence describe similar instances. Dr. Imperato reflected that he always had to allow for the possibility that the well meaning relatives and friends, who served as interpreters, often exaggerated that patients symptoms, with the hope that he would receive an extra large dose of medicine (Ibid.:33). Dr. Stirling experienced many difficulties getting patients to continue treatments once


89 they felt better: After one or two injections they would see such a dramatic improvement that they were quite satisfied, and no words would persuade them to continue with treatment until they were fully cured (Stirling, 1995:78). In a letter he wrote in the late 1930s, Dr. Stirling describes how he was somewhat troubled with one group of pati ents who had each brought 50 cents but had no need of injections, and others who wanted injections but had not brought 50 cents (1947:71). Dr. Dibble experienced similar challenges, when his patients would feel better and thus save the remaining drugs given them by the doctors for future use, or would sell them to others who had what they believed was a similar illness (1965). In another example, Dr. Imperato was somewhat surprised one day when administering vaccinations to children and recognized some as having been immunized on a previous occasion. H e confronted the mothers, who deduced that two immunizations must certainly be better for their children than one (Imperato, 1964). Thus, decades after biomedicine was first introduced to the area, it continue d to be challenged by other healing modalities relying on very different understandings of health and healing, and this environment of medical pluralism and multiplicity of understandings of health and illness continue into the present (see also Langwick 2 006, 2007; Malloy 2003) Postcolonial Biomedicine and Making a Living: From Independence to Ujamaa, 19611986 Tanganyika Territory gained independence at the end of 1961, and in 1964, the mainland merged with Zanzibar to form the United Republic of Tanzani a. Parallel to challenges of the past the country faced a critical shortage of medical staff and supplies, and struggled to train the medical personnel needed (Beck 1981; Iliffe 2002; Turshen 1984). The Development Plan of 1961 to 1964 emphasized curative medicines, and was largely a continuation of colonial health policies. However, in July of 1961, the Tanganyikan Minister of Health, Mr. Bryceson, requested that a team of scholars review the health sector and determine how to better integrate the health services available in the country. A team of British scholars visited the country on several


90 occasions between 1961 and 1962. Reflecting on the previous twenty years of development in Ta nganyikan medical services, their report states: [T]here has been no l ack of careful thought about the present and the future; no shortage of plans indeed, three major programmes were formulated within ten post war yearsand ample evidence of enthusiasm for a better health service. But none of these programmes has proved to be within the economic capacity of the country. Progress, where it has been achieved, has quickly been caught up by population growth and rising demands ( Titmuss 1964:31). Reflecting back on the health system, the Titmuss Report found that economic problem s were a main impediment to building up the health sector to meet the countrys needs. Based on the recommendations of the Titmuss Report, the Tanganyikan government formulated a new plan for 1964 to 1969 aimed to improve the socioeconomic status and living conditions of the Tanzanian people T he central government assumed authority over hospital health services aimed at general public health, while rural services remain e d the responsibility of the local authorities. The plan entailed a major build up of rural health centers ( Nsekela and Nhonoli 1976). At the time, m ost medical facilities in the country were located in urban areas, where the majority of expatriates lived (Beck 1981; Clyde 1962). While r ural health centers were supposed to provide both preve ntative and curative medicine, in 1961 only twenty of these facilities existed, despite colonial government plans to increase their number throughout the country. Most local authorities did not have the resources to pay medical personnel in the rural healt h centers, and the district and provincial medical staff who were supposed to supervise them were largely unable to do so due to their own difficulties with staff shortages and the lack of communication infrastructure. Instead, missionary groups operated s everal of the rural medical facilities (Beck 1981; Nsekela and Nhonoli 1976). Conditions in existing biomedical health facilities after independence remained difficult. In her study of medical pluralism and health seeking behaviors among the Sukuma in nor thwestern


91 Tanzania in 1966, Reid found that biomedical practitioners felt their healing practices were superior to local healing traditions, but that they were consistently constrained by the lack of proper medicines (1969:132). This situation required t hat hospital personnel adopt a variety of criteria by which they would distribute the limited resources to the abundance of patients they served, particularly when it came to drugs or supplies that were scarce or particularly expensive: In a case such as bilharzia, if the diagnoser prescriber feels that the patient understands about his disease, he will give the patient a course of stibophen injections. If, on the other hand, he does not think the patient understands well enough he gives the patient pallia tive treatment which probably consists of aspirin and perhaps potassium iodide solution. Another criterion is age. If an old patient has a disease which a severely limited supply of penicillin or chloroquine phosphate could cure, he may not be given the tr eatment. The drugs may be saved for the possibility that a younger person will require the drugs before the supplies are replenished. A third criterion is severity of the symptoms. If the patient does not seem to be in acute distress, the cheaper and more plentiful medicines will be tried first. (Reid 1969:132133). However, the ethical dilemmas posed by restricted resources could cause considerable strain on health workers. Working in the early 1960s as a traveling doctor, Dr. Louise Jilek Aall describes coming upon a small mission station in Iragua, where the mission Father received a note that an outbreak of measles hit a remote village several hours away. Hearing of the problem, she and the mission Father took all of the penicillin they could muster fro m the mission station to the village, and began to treat the children. The doctor realized that some o f the children were beyond help. She faced the dilemma: preserve an ever depleting supply of penicillin for those children who could be helped, or continue to provide the treatment to everyone: I began to mix distilled water with the penicillin and to give sufficient doses to only some of the children. I felt like a criminal, and my tension mounted as the penicillin supply diminished. The mothers sensed my uneasiness and anxiously pushed forward. At last the Father stood up and declared we had to leave. A storm of protest arose. Why should some of the children get help and others not? Could we not see that the little ones would die if they did not receive the injection? The women closed the circle around us, sobbed and begged, clutching their crying, coughing children, holding them up to our faces. I looked around in amazement. Where had all the people come from? The mothers must have run for miles through


92 th e bush with their sick children. There was no way that I could treat them all. (Jilek Aall 1979:85) As these cases illustrate, although biomedical health services were supposedly free under the postcolonial government, they also had to be carefully negotia ted and biomedical practitioners were forced to adopt unofficial norms to prioritize how services were distributed in a context of scarce resources. Yet determining who was worthy of treatment took an emotional toll, and Dr. Jilek Aall describes her feeli ngs of hopelessness, and eventually frustration and anger, with the incessant demands of mothers with sick children amid scarce resources and the overburdens of work. At independence, there were four hundred (mostly expatriate) doctors registered, with pr ivate doctors making up nearly half of the total number of physicians in the country. The Ministry of Health allowed government physicians to practice private medicine in their spare time, but this privilege was revoked in July 1962. Of the 180 private pra ctitioners, most were located in major towns: Moshi, Arusha, Tanga, Dar es Salaam, and Mwanza. Others were located in smaller towns where government services were also available. At independence, the Ministry of Health (MoH) suggested that private practice be curtailed by requiring private physicians to pay a monthly fee to practice, as well as potentially restricting licensing to those physicians who provided a minimum number of years of service in the under staffed rural areas. However, this idea was aban doned after protests from the Medical Association and physicians in the country (Nsekela and Nhonoli 1976). Training nurses, medical assistants22 and rural medical aids was a major priority for the government in the 1960s ; however, it also posed significan t challenges (Iliffe 2002). Entrance standards for the cadres of nurse A, health inspector, and medical assistants were increased. The 22 The level of medical assistants would later b e re named Clinical Officers


93 government proposed reforming nursing training from the existing three cadres (nurse, hospital midwife and health nurse) to create three new cadres: nurse A, nurse B, and midwife. Nurse B students would receive three years of general nursing training, including midwifery, disease prevention, and infant welfare. Those who were most promising would be admitted to the nurse A c ourses, receive an additional year of training, and become eligible to act as a Matron or nurse I nC harge. Candidates for the nurse A courses had to complete a certain level of secondary education, whereas trainees for nurse B and midwife could begin training straight out of primary school.23 Medical assistants and rural medical aids were in high demand at a time when entrance requirements reduced the number of people eligible for biomedical training. Meanwhile, the Ministry of Health decided to establish its own medical school at Muhimbili Hospital, and the school began to t ake on medical students in 1963 (Nsekela and Nhonoli 1976). In 1967, the Tanzanian government signed the Arusha Declaration, whereby the country adopted an African form of socialism, cal led Ujamaa (meaning family in Kiswahili). Tanzania became a oneparty state, under the leadership of the Tanganyika African National Union (TANU), which in 1977 became the Chama Cha Mapinduzi (CCM) party when TANU merged with the ruling party in Zanziba r. The main priorities under Ujamaa were education, industry and agriculture. Although the health sector was not a main priority under the Arusha Declaration, the government looked to socialist countries elsewhere for inspiration on how to build up the hea lth sector (Iliffe 2002). Several private physicians, most of Asian decent, left in the first years after independence at the governments proposal to impose fees for private practice. After the signing of the Arusha 23 The cadres of Nurse A, Nurse B and midwife continue to the present.


94 Declaration, and its Acquisition of Bu ildings Act,24 many remaining private doctors and other professionals departed to Britain or Canada. Others remained, and received some financial support from Aga Khan IV through the East African Muslim Welfare Society (EAMWS), established in the mid 1940s. EAMWS supported social services to the Ismaili community in the country. Soon after Ujamaa was adopted, Aga Khan IV financ ed clinics, dispensaries, village wells, and other materials to support Ujamaa villages (Kaiser 1996). This stopped soon thereafter, however, when the Tanzanian government dissolved the EAMWS in 1968, and replaced it with the government sponsored Muslim Council of Tanzania (BAKWATA), which many Muslims felt was a state initiative meant to pacify and control Muslims in Tanzania (Dilger 2 011). One initiative of the Tanzanian government under Ujamaa was to organize massive radio studygroup campaigns to spread educational materials throughout the country from those in the highest offices all the way down to the household level, and from those with highest levels of education to those with little education at all. These were successful, with attendance rates in study groups ranging from three people to more than two hundred, and with 42 of the countrys 65 administrative districts participating in the radio campaign. Among these initiatives was a campaign called Mtu ni Afya (Man is Health), which promoted community health through prevention. P revent ion was the main thrust of the campaign, so people could avoid putting pressure on fragile biom edical h ealth services. The campaign advocated control of the environment to promote health to build latrines or boil water, for example. Two million Tanzanians participated in the radio study groups, where they would listen to radio broadcasts on ways to maintain health in daily practice (Hall 1978). However, the lack of infrastructure or 24 This act allowed the government to nationalize private businesses and buildings.


95 provisions such as access to water eroded peoples ability to achieve the standards of health and hygiene depicted in the Mtu ni Afya campaign. The 1970s ushered in a ma jor influx of foreign aid to Tanzania. Donors, mostly from Scandinavia, provided over 70 percent of the budget for health during this decade. Foreign assistance to the country increased fr om US$51 million in 1970 to US $650 million in 1980 (Bagachwa et al., 1997). Much of this aid was distributed in the form of vertical projects rather than to the sectors as a whole (Havnevik et al. 1988). Although it was favorable in the period for donors to sponsor large construction ventures, in the health sector, foreign assistance went largely to smaller projects such as building rural dispensaries and health centers, and establishing training institutions for the lower cadres of health professionals (Young 1986). Writing in 1977, thenMinister of Health Dr. Leader Stirling discussed some of the ambiguities of foreign assistance in the health sector: It is true that we have received, and are still receiving, a lot [of aid] from many different countries, for which we are indeed grateful, and without which we should be una ble to develop our services, especially our hospitals, health centres and training schools (which have now in training a total of over 2,000 health workers doctors and all sorts). The trouble is that every piece of development, be it a new building, a new apparatus, a new technique or a new intake of students, automatically involves an increase in running costs, and the hard fact is that almost no donors are willing to commit themselves to recurrent expenditure, except on a very temporary basis. The only notable exception is the paying or subsidising of the salaries of certain foreign experts, but as we have no shortage of manpower and are training our own experts as fast as we can, this piece of aid is self limiting and will soon cease to count. (1977:13738. Parenthesis in original ) Foreign assistance to Tanzania was not merely monetary, but also included volunteer activity by expatriates. According to T. Kue Young, a Canadian volunteer who trained Assistant Medical Officers (AMOs) in the country during 197 9 and 1980: As a volunteer working in Tanzania, I was struck by the multiplicity of foreign nationals from both the East and West jostling for an opportunity to help. The dependence on international aid seems to make dubious Tanzanias claim to be self rel iant and truly socialist. However, in a country where the paucity of


96 resources is overwhelming but the commitment to extending coverage of basic services firm and genuine, all available resources including external aid, whether capitalist or socialist, have to be utilized. (1986:132, emphasis in original) Volunteer p hysicians came from China, Cuba, Canada, and the Soviet Union, to work not only in large specialist hospitals, but also in training facilities for the lower cadres of health professionals (ib id.). Young was struck by how egalitarian collaborations between Tanzanians and foreign nationals could be, although some of the foreigners were not pleased with the degree to which their position was perceived as commensurate to their African counterparts : From my observation, Tanzania was one of the few countries where foreign nationals, be they volunteers, advisors or consultants, did work firmly under the control of Tanzanians. The practice in some developing countries (particularly Francophone) of hir ing foreign nationals (especially those from the former metropolitan colonial power) as line managers in the civil service was not adopted by Tanzania. I recall how foreign professionals in Tanzania, in moments of frustration and despair, often commiserate d with one another and expressed envy of their counterparts in neighboring countries where they were assigned to positions with real clout! (Young 1986:132. Parenthesis in original ) In 1969, the central government adopted a new health plan to govern the health sector until 1974. It entailed a significant shift in policy, making rural health services a top priority. Between 1970 and 1974, the government increased its health expenditures from 5.2 percent to 8.9 percent (Iliffe 2002). Finding rural health de velopment too slow, the Ministry took over rural health centers from local government. Only fifty rural health centers were operating during this time, and the plan called for an additional eighty facilities to be built (Nsekela and Nhonoli 1976; Iliffe 2002). There were no plans for building additional dispensaries in the country, of which there were 1200 at the time; however, as Nsekela and Nhonoli note, these were unevenly distributed and several were in serious disrepair and extremely short staffed (197 6). An additional major government initiative was a shift of financing from hospitals to rural dispensaries, village health posts, and rural health centers. However, in 1974 this plan had to be temporarily abandoned due to rising oil prices, drought, popul ar resistance, and financial drains of villagization. The


97 implementation resumed in 1975, as the government adopted a third five year plan for the development of the health sector. Emphasis remain ed on developing rural districts and preventive medicine, and training personnel such as rural medical aids, medical assistants, and maternal and child health aids who would be able to staff the rural health facilities. However, rural emphasis in health policy did not necessarily translate into practice. Dutch Soc iologist Geert M. van Etten conducted a long term study of Tanzanias rural health system from mid 1969 to mid1973 in which he explored impediments to developing healthcare in rural areas. Van Etten undertook this project primarily in Mwanza region, where he and others conducted research and trained rural medical aids in public health, and preventative medicine in particular. One of the major findings of his study challenged predominant global health discourses arguing that rural health workers were the be st equipped to provide preventive health services in rural areas because they were part of the culture where they would work and would understand and value life at the village level. By contrast, van Etten found that the attitudes of people being trained a s rural health workers were negative and highly patronizing towards the villagers, largely because those being trained in the profession were more educated and wealthier than the populations they were meant to serve, and many came from urban backgrounds. I n essays these workers composed regarding their ideas about rural people, van Etten quotes students as stating that rural populations lack knowledge, follow tribal customs, and are uneducated and uncivilized (van Etten 1976:120). Further, van Etten quotes one students patronizing attitude directly from one of the essays he received: in order to overcome all these problems we can start educating all the people in rural areas to know what is good for their future and what not; telling them how to live together and decide what to do for their life and their families (ibid. 121). Van Etten argues that this negative attitude towards rural populations and


98 preventative medicine reflects the overall health systems history favoring clinical hospital services rather than rural, preventative services: this significant change in the pattern of resource allocation, from the hospital sector to the primary care institutions, did not coincide with a major change in the underlying system of values which people stil l attached to hospital medicine (ibid. 123). This was, according to van Etten, largely because of existing hierarchies in the system, whereby clinical workers such as physicians and nurses enjoyed more status and higher salaries than rural public health w orkers, thus informing attitudes among trainees that diagnostic and curative services were of higher value than preventative ones. Nearly a decade later, a Canadian physician in Tanzania observed similar dispositions among the students he trained to becom e Assistant Medical Officers (AMOs). Comparing his experience to van Ettens findings, Young notes, Many of these attitudes were present to varying degrees among the students I taught in Mbeya Region in the late 1970s. During their village health projects many students turned up in their polyester shirts, bellbottoms, platform shoes and reflector sunglasses, all set to do labor with the peasants! (1986:133 n. 2). To Young, this display of urban style to a rural area reinforced the conclusions of van Etten regarding the problematic attitudes of rural health workers. Van Etten concludes that it is not only the poverty and ignorance of rural populations that encumbers the development of rural health services, but also the negative attitudes and the selective approach of health workers themselves which contribute to the limited communication with the rural population and to the delay in health development (1976:129). This is a telling example of the ways that health professionals worked to mark their professi onal status as separate from the people they served. Conditions for health workers during the 1970s were difficult. Salaries were inadequate and the government was unsympathetic to workers complaints ( Il iffe 2002). Several medical


99 professionals voiced their concerns in the newspapers during the 1970s. In the Nationalist newspaper in 1970, a medical professional expressed that he felt health workers were reducing the amount of work they did because their salaries were not commensurate with their training a nd the difficulties of their work. According to a quote by Mr. A. M. Ntauka, who worked in Rufiji, Doctors take seven years at university colleges while Arts students take only three, yet when both join the government their salaries are either equal or Ar ts graduates earn more than doctors ( The Nationalist July 10 1970:1).25 In response to a workers public complaint about salaries in the news, someone responded, Salary scales are fixed according to qualifications and profession. work is done on shifts with two or more persons working for 24 hours. In fact this requirement is necessary for building our young nation. What is required is more output and not necessarily more working hours ( The Daily News May 31 1972:9). Staff shortages continued to plague the sector throughout the 1970s. As Dr. Stirling noted (above), larger infusions of funds and trained staff into the sector was complicated by increased demands for services and the overall costs of maintaining and supplying more health facilities an d providing salaries to additional staff. The shortages were further exacerbated by the fact that many medical professionals were leaving Tanzania for opportunities in neighboring countries or abroad (Iliffe 2002). In an interview for a news segment on Wo men at Work, the Matron of Aga Khan Hospital states, Many qualified nurses have left the country suddenly. As a result we could not make replacements immediately. We have however, been advertising vacancies in the local newspapers. But there has been lit tle response ( Sunday News April 2 1972:11). In addition, patients and family members were not satisfied with the quality of facilities, or more often, that of health workers at government facilities. There were several complaints about 25 These quotations from healthrelated news articles in the1970s are ta ken from van Amelsvoorts compilation, Medical Anthropology in African Newspapers (1976).

PAGE 100

100 corruption in the system. In an anonymous letter to the editor, a person referring to him or herself as Constructive Citizen states, Some social centres like Government hospitals are increasingly becoming anti socialistic in their services, describing how doctors atte nd primarily to those patients with money. Constructive Citizen argues: this kind of procedure especially when it is carried out during official hours is very insulting indeed and is undoubtedly jeopardizing the services of the centre to the public as the doctor becomes tempted to attend to the cash patients thereby ignoring the common men, the very people for whom and by whom the doctor was sent to school to gain knowhow supposed for putting into practice. ( The Daily News May 6 1972:9). Patients could also express dissatisfaction with how their maladies were diagnosed. In a Your Point of View column in the Sunday News a person describes how he became ill and went to a dispensary, where he did not get better, and finally went to Muhimbili Hospital. O n arrival, he met the medical officer, and Without a glance at my distorted body or even a reading of my temperature, the officer simply prescribed injections and half a dozen tablets ( Sunday News June 18 1972:6). Such complaints were not limited to government health workers either: A person with limited time who cannot afford to wait in the long queue at the Government dispensaries would normally go to a private doctor, but these people have nowadays become so moneyconscious that they do not examine t he patient at all. Once you get there, they ask you what is wrong; if you reply fever the next question is have you got 5/ ?26 You say yes, you get a jab and two tablets of A.P.C,27 and very unconcerned, they shout Next. Whether you have headache, stomach paints or ear trouble, it is a jab and two A.P.C., and 5/ . I very much doubt if they are not using the same injection for all ailments. (The Standard, April 5 1971:2) Overall, the 1970s were a decade of important advancements in the heal th sector, but also some major setbacks. There were improvements to health indicators and immunization 26 The symbol / denotes Tanzanian shillings. 27 A.P.C. is a combination drug that was often available over the counter containing aspirin (acetylsalicylic acid), p henacetin and caffeinea painkiller and fever reducer.

PAGE 101

101 distribution among children, particularly after 1975. The first thirty years after independence marked a major increase in life expectancy of Tanzaniansup fifty percent from the colonial period (Heggenhougen 1986; Iliffe 2002). The sector also made strides in controlling Tuberculosis (TB). F rom the point of view of an Australian physician who had worked in Uganda as a missionary doctor from 19291931, and later worked in Kenya, Ethiopia, and Somaliland between 1932 and 1962, what Tanzania had achieved by 1978 was nothing short of remarkable. Invited to the Silver Jubilee Conference of the Association of Physicians of East and Central Africa at Muhimbili Ho spital in 1978, Dr. A.D. Charters commented: I was amazed at the advances which the Africans had achieved since my departure from East Africa 16 years previously, for, whereas there was at that time not even a medical school in Tanzania, there are now professors in every discipline and about 360 medical students, some 60 qualifying as doctors each year. The Conference was most efficiently organised and the papers were of a high standard. (Charters 1985:140) However, other initiatives, such as nutrition and clean water provision, did not fare well. Several aspects of the health sector were in decline during the 1970s. Expenditures not only dropped during this period, their distribution also changed significantly. Financial support of rural health grew betwee n 1971 and 1981, while it dropped for hospital services. The emphasis on rural facilities had an adverse effect on distr ict and referral hospitals, which deteriorated as a result. Furthermore, demands for biomedical services in towns and urban areas continued to rise, as people critical of villagization abandoned the rural areas and moved to cities. By the mid1980s, the vast majority of Tanzanians lived within ten kilometers of a health facility; however, many of these were poorly supplied and staffed (Ili ffe 2002). Many Tanzanians were critical of the government in the late 1970s and early 1980s, in part in response to the overall deterioration of the health sector. According to Iliffe, corruption was not a major issue in health facilities in Tanzania unt il the 1980s, when public critiques and

PAGE 102

102 economic decline were having major impacts on the morale of health professionals. Although health professionals received small pay increases in the early 1980s, inflation rates at the time were nearly thirty percent per year (Iliffe 2002). Salary increases were larger among the lower cadres of the health profession, which led to doctors protesting their inadequate salaries and arguing for the right to practice privately. S ome began informally offering health services from their homes or the homes of patients. Other doctors left Tanzania in the late 1970s and early 1980s (Iliffe 2002). According to a consultant report to donors: In theory, private medical practice is illegal in Tanzania. In practice, many doctors working for the state sector do take private patients, and others are simply operating behind official or legally admissible fronts, e.g. being a school or company doctor or working for a charitable foundation or church. Given the low salaries of Tanzanian medical practitioners in relation to their traditional social status and their earning power in other countries, it is not surprising that many (most?) are involved in private practice, and the more unscrupulous use their official premises and medicines for per sonal profit. The temptation to leave the country is very real, especially for senior medical personnel, who could earn many times their Tanzanian salaries in neighbouring counties. It is not known how many trained medical workers succumb to this temptatio n, however. (Cooksey 1986:17) As economic hardships continued in the 1980s, it was not possible for the Tanzanian government to achieve its policy goals Three quarters of the MoH spending for 198586 was spent on curative services, still with an urban bia s. According to a consultant report on the sector in 1986, 38% of the curative budget for the country was allocated to Muhimbili National Hospital alone. The consultants note that what the MoH was doing in terms of allocating funds to the sector is diamet rically opposed to what official policy says it has done or is trying to do. The discrepancy should be a matter of public concern and debate, but to date this has not been the case (Cooksey 1986:12). Yet, the consultants note that while the budgeting allocations were problematic, more health services of some kind are available to the rural population in Tanzania than in any comparable African country (Cooksey 1986:13).

PAGE 103

103 Structural Adjustment, Early Reforms and the Informal Economy: 19851990s The late 19 80s and early 1990s was a period of planning, and in some cases, implementing, major reforms in the structure of the Tanzanian state. It was further a period of significant shifts in donors development modalities, which would have profound effects on the w ays reforms were implemented and subsequently managed (see Chapter Four) However, for many Tanzanians outside of the government, as well as those working within local government institutions, few of the se macro level transformations were felt on the groun d. Thus, this section outlines some of the major events leading up to the planned reforms, and gives a brief summary of the challenges of making a living in Tanzania during a time of planned, but as yet unimplemented, reform. The economic decline in Tanzania from the late 1970s into the 1980s was a result of multiple factors. The 1979 oil crisis, the break up of the East African Community, and a war with Uganda in late 1978 to mid1979 precipitated serious drains on the countrys finances. The Tanzanian go vernment received no financial assistance from the international community for the war with Uganda, despite the widely expressed support for bringing down Idi Amins regime (Biermann and Wagao 1986). Meanwhile, by 1980 export prices were dropping and the t erms of trade were extremely unfavorable, causing further decline in Tanzanias economic situation. The increase in urban density from the 1970s into the 1980s was met by decline in living standards and the quality of basic services. Food was increasingly scarce, with essential staple items such as cooking oil and soap becoming ever more difficult to come by. Unable to procure basic necessities or living wages from the formal economy, many Tanzanians turned to informal selling and buying practices in order to survive, despite the official ban on the informal sector (Lugalla 1997; Obrist 2006; Tripp 1997). In 1982, the economic situation of the country was dire, and the government went to the International Monetary Fund (IMF) to attempt to get a loan.

PAGE 104

104 However unable to negotiate an agreement with the IMF, the government created its own structural adjustment program (SAP). Economic policy changes cut expenditures and investments, enacted a limited devaluation of the currency, and increased liberalization of trade; however, bi and multi lateral donor institutions considered the measures inadequate (Biermann and Wagao 1986; Kiwara 2003). The countrys main foreign donors rescinded their support (Bagachwa et al 1997). Thus, the Tanzanian government announced that people should expect economic hardships in the coming years. Unlike other countries where political and social unrest ensued, Tanzanians remained largely supportive of their government (Biermann and Wagao 1986; Tripp 1997). At the same time it was attemp ting to restructure the country, the Tanzanian government continued to negotiate for loans However, the World Bank and other donors were not satisfied with the Tanzanian SAP, or with the government more generally, and it was not until the countrys first president, Julius Nyerere, stepped down in 1985 that the IMF agreed to negotiate An agreement was reached in 1986 between the World Bank, the IMF and the Tanzanian government (Rusimbi 2003; Biermann & Wagao 1986). The Tanzanian government put together an Economic Recovery Plan (ERP), which ultimately convinced the IMF and the World Bank to provide loans. The program was formally launched in 1986. The IMF/World Bank SAP required that the healthcare system privatize and decentralize. Due to the governments agreements with the IMF and the World Bank, donor support to the country returned, and increased substantially from the time the agreement was in place into the early 1990s. In 1986, Tanzania received US$670 million in donor support, which increased to US$1,345 million by 1992. The main donors at the time were the UN, Norway, Sweden, Canada, Denmark, Finland, Germany, Italy and Japan. (Bagachwa et al. 1997).

PAGE 105

105 Overall, the 1980s was a decade of extreme hardship in Tanzania. Government sectors experienced massive shortages of resources, and health was arguably one of the sectors worst hit (Havnevik et al. 1988). Health was not a priority in the ERP, and conditions within health facilities worsened. While the Essential Drugs Program, funded by the Danish International Development Agency ( DANIDA) helped to lessen the dearth of drugs in the rural areas, hospitals were practically ignored. By 1988, the majority of hospitals in Tanzania reported that they lacked the vast majority of necessary drugs, and many buildi ngs deteriorated. Meanwhile, although salaries comprised the majority of government expenditure on health, inflation rates significantly reduced the value of wages, resulting in further erosion of staff morale ( Havnevik et al 1988). In the 1990s, prices continued to increase as wages began to drop. As a result, both the wealthy and the poor turned to informal economic activities as a means to supplement their income (Lugalla 1997). Corruption ran rife within government sectors. In the health sector, many workers resorted to selling prescription drugs on the informal market, and charging their own fees for medical services that were supposed to be free. Other medical professionals began engaging in informal activities to supplement their incomes, and these activities often absented them from their workplaces. Some opened private practices, while others provided health care informally from their own homes (Lugalla 1995). As the government health sector was eroding, the MoH worked to maintain the cooperative foundation they had established with Aga Khan medical facilities, in particular due to the inability of the government to meet the health services demands of the country. Many people were forced to go to the private sector for services that were no longer available in government facilities, and the Ismaili clinics were able to fill some of the gaps in the existing public sector. In

PAGE 106

106 the late 1980s and early 1990s, as the government conceptualiz ed new health sector policies it looked to Ismaili facilities fo r ideas on how to improve the quality of services (Kaiser 1996). Signifying a major shift in governing philosophies, the Government of Tanzania signed the Zanzibar Declaration in 1991, which officially distanced the new government from the Arusha Declarati on and the system of Ujamaa, in favor of structural adjustment and liberalization of the economy (Tripp 1997). As the conditionalities of the World Bank and the IMF were implemented in the late 1980s and early 1990s, the country experienced some improvements in its macroeconomic performance, particularly in terms of the mining and agricultural sectors. However, inflation remained unchecked throughout the 1990s, and the pace of the reforms did not impress donors (Bagachwa et al. 1997). In the wake of the governments neglect of the health sector, the deterioration of medical facilities, and the inadequacies of worker salaries, the relationship between government health workers and the state grew exceedingly bitter. Several doctors and nurses left Tanzania. Th ere were a series of stoppages and strikes at hospitals in Tanzania between 1990 and 1992. The most significant of these took place at Muhimbili, where junior doctors went on strike to protest inadequate allowances and the transfer of three of their leader s to other parts of the country. As the hospital administration attempted to crack down on the participants in the strike, the latter got support of university students and attempted to spread the strike out to other health facilities. The administration dismissed junior doctors involved in the strikes, and had the police evict them from staff housing. In some cases, police harassed or even physically attacked doctors and their families (Lugalla 1995). As a result of their treatment by the hospital administration, the police, and the government, the public became sympathetic to the junior doctors cause, and hospital workers, trade unions, and students went on a march to the Prime Ministers Office. The fired

PAGE 107

107 doctors were reinstated and the situation at Muhi mbili was investigated (Iliffe 2002). By 1998, the country still did not have a new system in place, and conditions were precarious at best: The country was littered with the debris of past systems, with the remains of socialism lying thick and the surfac e patches of private practice still spreading (Iliffe 2002:219). Procuring biomedical treatment in the 1990s was exceedingly difficult for the majority of poor Tanzanians. While they avoided most private and missionary facilities due to user fees, with t he deterioration in quality of public services in the 1980s and 1990s, and the obligation to pay bribes in order to get services, even government health facilities became out of reach for the poor (Lugalla 1995). Meanwhile, the overall health of Tanzanians also deteriorated as income dropped, reducing peoples capacity to feed and care for themselves and their families. The quick pace of urbanization also had a deleterious effect on health, as most Tanzanians living in the cities occupied informal settlemen ts lacking sewerage and drainage systems. These living conditions meant that more people were suffering from illnesses relating to food and water contamination and from unsanitary living conditions increasing the public need for health services at a time w hen conditions in the sector were deteriorating (Lugalla 1995). Conclusion From its first establishment to present, the health sector in Tanzania has continued to be tied to the goals of the government whether under colonial administrations or under the postcolonial central government. Much of the helplessness that health professionals felt in response to their situation from the colonial period into the 1990s was a result of a longer history of their workplace being linked to particular modes of governance, and the inability of the government to actualize the kind of working environment that would make providing biomedical health services meaningful. Having spent decades establishing health professions in Tanzania, and fighting for recognition of their ha rd earned professional status, the conditions in Tanzanian

PAGE 108

108 health facilities belied the aspirations that brought people into the health workforce to begin with. Spattered throughout the memoirs of medical professionals working in Tanganyika and later Tanzania are stories of the difficulties these workers faced having to watch people suffer or die for want of certain supplies, drugs, or sufficient personnel to assist. This history of chronic understaffing and underfunding of the sector continues to seep into the present, where health workers desire the compensation, technologies, upgraded training, and infrastructures demanded by their profession. Having outlined the challenges faced by the health sector overall in Tanzania, I now turn to the specific history of Kiunga District Hospital, which until Health Sector Reform was marked by many of the paucities, challenges, and morally ambiguous practices characteristic of government health facilities more generally in the country

PAGE 109

109 CHAPTER 3 THE HOSPITAL BEFORE H EALTH SECTOR REFORM Throughout the history of the establishment of biomedical institutions and the distribution of professionals to work within them, location always mattered. As discussed in Chapter Two, health facilities were initially constructed near m ilitary bases, expatriate communities, and mission basses in Tanganyika. It was under the British colonial administration that the health sector was significantly expanded into smaller towns and even villages an initiative continued after independence and with Ujamaa However, finding health personnel willing to work in the more remote areas of the country was a constant challenge, and many workers abandoned posts that were located within the more isolated and under resourced areas of the country (Langwick 2008). Yet location matters in a different way as well. Those health workers stationed close to larger urban areas had to evaluate their salaries and conditions of work against the costs of living near the city, which as will be seen below, became increasi ngly difficult during the 1980s and 1990s with the economic crisis and structural adjustment programs (SAPs). John Iliffe argues that in East Africa, medical professions and the state are in large measure symbiotic (2002:3). Indeed, much of the scarcity and institutional inadequacy that medical professionals face in their daily practices shortages of drugs, diagnostic tests, equipment, and training are tied to the governments inability to financially support or adequately coordinate health facilities and activities. The history of health policies in Tanganyika and later Tanzania may be retraced through the myriad governmental policy documents and health status reports that have emerged since colonial rule was first established in the area. What is not so easy to see in these reports, and what is often unclear in other studies of African medical systems and professionals, is how health policies and medical professional ideals are actually implemented on the ground. How do medical professionals and the inst itutions within which they

PAGE 110

110 operate interact with, negotiate, and translate donor and government policies and programs? How does the location in which a health facility operates influence the ways that policies and opportunities are implemented and negotiat ed? Kiunga Districts proximity to Arusha town has been crucial to the hospitals history, and in particular, to the nature of transformations experienced at the facility under health sector reform (HSR) What follows is a brief history of Arusha, followed by an account of the establishment, growth, and working conditions of the health facility that would later become Kiunga District Hospital. A Brief History of Arusha and its Biomedical Institutions During the German colonial administration, nearly half of the hospitals built were in coastal cities, while the remainders were established on major rail lines (Tabora, Ujiji) or in areas with major cash crops (Kilosa, Marangu) Mt. Meru Hospital, currently the regional hospital and the largest health facility in Arusha region, was opened in 1910one of the some fifteen hospitals the Germans built during their colonial reign, and one of only two buildings remaining in the town that were erected during the German colonial era While mission schools were quickly er ected in the town, m ission clinics were not a major mission priority in the area. While the first mission school was established just outside of Arusha town in 1904, the first mission dispensary was not opened until 1942. Major expansion of mission health facilities in Arusha and surrounding districts occurred during the 1950s, as the Lutheran Church constructed the Selian and Leguruki clinics in 1953, and the Catholics established facilities in Poli/Singisi in 1957 and Sombetini in 1958 (Benson 1996). Ove rall, Arusha and the northern province more generally had far fewer biomedical health facilities than other comparable regions in the colony. In the years after 1942, other government and private clinics were established in the region, but Mt. Meru hospita ls outpatient services were consistently overloaded, echoing the overcrowding in the town itself : The towns cramped and unhealthful living conditions, including the lack of

PAGE 111

111 sanitary facilities and clean water, probably contributed to disease and epidemi cs, and these, no doubt, burdened the limited health services for Africans (Peligal 1999:244). Among the African populations living in Arusha during the early 1950s, the main ailments being treated at the hospital were malaria, worms, venereal disease, respiratory infections, injuries, ulcers, smallpox, polio, and tuberculosis. M any Africans were resorting to private doctors instead of utilizing the government hospital an action meant to demonstrate their dissatisfaction with hospital services. However, pa ying for private clinic services also increased pressure on scarce financial resources ( Peligal 1999 ). As the state adopted Ujamaa and privileged rural development, the infrastructure of Arusha town deteriorated at a time when urban population was on the rise. From the late 1970s to the early 1980s, Arusha, like other areas of the country, was hit heavily by chronic shortages of necessary goods and poverty. According to Benson (1996), ten new government medical facilities began operating in Arusha and the villages in surrounding areas in the late 1960s to early 1970s. Further, since Arusha was a town targeted for industrial growth, various private clinics were established to service the industries there. Meanwhile, many villages built dispensaries due to the governments emphasis on self help, and by the late 1970s the proportion of people living near a health facility in the region was one of the highest in the country. However, in Arusha town and its surrounding villages, as elsewhere in Tanzania, many of these facilities were not completed or functioning due to the lack of funds to pay personnel or complete construction (Benson 1996). While in the 1970s the government health sector in Arusha and neighboring districts expanded and private clinics were limi ted, by the 1980s the trend was reversed. The government health sector was stagna n t, while private clinics open ed in increasing numbers. Under Nyereres

PAGE 112

112 presidency from 1962 to1985, 52% of the new health facilities in Arusha and Arumeru Distric ts were gove rnment owned and 27% were private. I n the period from 1985 to 1993, 63% of new facilities were private and only 18 % were government owned (Benson 1996). While some private clinics were legitimately affiliated with religious institutions or charitable organ izations, several in the Arusha and Arumeru Districts were privately operated clinics as practitioners circumvented government policies on private clinics by paying charitable organizations for the right to use their name on the clinic sign. However, thes e private clinics had no other formal affiliation with the charities. The distribution of the private clinics was also notable they were most often set up in places with high population density, nea r to a main road, and also near an existing government fac ility. This meant more urban or denselypopulated areas had a variety of clinics to choose from, while many villages farther from the main roads remained under served (Benson 1996). Medical work in Catholic missions increased particularly quickly in the e arly 1980s, despite Catholics being a minority religious group in Arusha. Catholic congregations saw medical services as an important role of the Church, and income from the dispensaries were seen as a means to support the congregation. In contrast, the Lu therans the largest Christian denomination in Arusha and surrounding areas only opened one dispensary during the 1980s. There were some Islamic dispensaries set up in the mid 1980s, but they were few and far between ( Benson 1996). In the 1990s, when Benson was conducting his field research in Arusha and Arumeru Districts, he asked health professionals within all of the health facilities in the two districts about their attitudes to their work. He inquired about what aspects of their job made them happy and what frustrated them. Four of the government workers stated that they found nothing in their jobs

PAGE 113

113 that brought them satisfaction. One District Medical Officer (DMO)the head of government health activities in the district stated that he was unable to reall y practice medicine and the job was too over burdened with administrative tasks. Most of the people Benson interviewed, however, did find some satisfaction in their work, and particularly valued their relationships with their patients. Of the frustrations they outlined, the vast majority of respondents highlighted the lack of equipment to facilitate their work (82.3%). In the urban districts, respondents interviewed said that the biggest problems in their districts were deficiencies in staff morale, transpo rtation, supervision, training and education, and drugs (Benson 1996).28 These challenges echo those that the staff at Kiunga District Hospital would articulate in 2008. I now turn to the establishment, growth, and conditions of Kiunga District Hospital from the late 1970s until just prior to the implementation of HSR. While a variety of public dispensaries and private hospitals and clinics were nearby, Kiunga District Hospital, the main site for this study, was the largest health facility in its district. I n 2008, the hospital compound was long, stretching from west to east, with road running through it, cutting the grounds in half. Within the hospital compound, there were a number of departments that each had its own atmosphere. Similar to many district hos pitals that were outside urban areas in Tanzania, the buildings were separate from one another, built several feet above ground level so that they did not flood during rainy seasons. The majority of buildings were within a few meters of the hospitals road but there were few paved walkways to ease movement between the various edifices. W ard buildings were long and rectangular, accessed by steep ramps. Structures that provided other services, such as the clinics or outpatient departments, were most often open, awningcovered outdoor areas with smaller rooms built into 28 Benson employs very few quotes in his study, so it was not possible to draw directly on workers specific statements.

PAGE 114

114 the edges of the foundation. Patients awaiting services sat or stood in undercover waiting bays, and services were provided in the maze of small rooms or areas within.29 The layout of the hospit al in 2008 was a function of history, space, and aesthetic imaginings. B elow, I trace the history of the hospital from the point of view of those who worked there between 1978 and 1999, when HSR was first implemented.30 Methodological Considerations for a History of the Hospital Constructing a history of the conditions of Kiunga District Hospital prior to health sector reforms presents a variety of methodological challenges. T here were several potential sources available: Kiunga Hospital files, Ministry of Health reports, other scholarly work, newspaper articles, and oral histories from health personnel who worked at the site during the years in question. Unfortunately, these sources posed some problems. At Kiunga, archives of hospital files in 2008 were housed in two very small sheds with dirt floors. While the hospital administration allowed me access to these files, many of the boxes and the papers contained within them had all but disintegrated from the moisture of countless rainy seasons past, or had bee n damaged by rodents. Of the few minimally damaged boxes that remained, the earliest documents went back to the 1970s. Of these documents, there were reports of disease burdens in the district, charts of the number of health facilities, correspondence from medical workers in the dispensaries to the DMO asking for particular supplies or personnel, and several documents from local chapter of the Chama Cha Mapinduzi ( CCM ) party correspondence during the 1970s and 29 Based on my visits to other district hospitals in Tanzania, there did not appe ar to be a standardized plan for layouts. The histories of infrastructural upgrades on individual buildings were discernable by differences in roofing materials, visible areas in the floors and walls where it was possible to see the building had been exten ded, and hospital archives. However, the overall aesthetic that I observed in several district hospitals allowed for a lot of air to move through the various wards and departments, and many were built similarly well above ground to avoid flooding during ra iny seasons. At Kiunga, the exterior design of the wards was remarkably consistent from the original ward built in the 1970s to the newest wards built in 2004 and 2005. It was not until one entered the ward that a difference in design was detectable. 30 Ho spital developments and conditions after Health Sector Reforms (HSR) are discussed in Chapter Five.

PAGE 115

115 1980s about various partyrelated activities. U nfortunately, I only found two reports that provided details about Kiunga District Hospital in the past, and the details from these reports have been integrated below as appropriate, although neither provides much of the specificity I hoped to find to support the other available sources .31 Archival research in the Ministry of Health and Social Welfare library posed similar problems any reports I found from the regional office in Arusha or from the district were from the years after HSR was implemented. Newspapers were similarly problematic In post independence Tanganyika, the Tanganyika African National Union ( TANU) party favored a centralized and Tanzanianized mass media system, out of Dar es Salaam. The government nationalized the medi a in 1969. In 1972, the government merged the two main daily newspapers of Tanzania The Nationalist and The Standardat which point it was named the Daily News owned by the government. The other two newspapers available at the time, Uhuru and Mzalendo were sponsored by TA NU. The only private paper left during the 1970s was Ngurumo, which lost readership and eventually ceased publication in 1976. Thus, from 1976 to 1993, only two daily newspapers were available in Tanzania, both affiliated with the government ( Sturmer 1998) Overall, searches through these newspapers from 1968 through the 1980s turned up no articles about Kiunga Health Centre, or its establishment as a district h ospital in the late 1980s As for scholarly work, Bensons 1996 study was the only source available providing details about government health facilities in Arusha region in the years prior to HSR. The remaining sources of information about conditions in the hospital prior to HSR were those personnel who worked there in prior years. In 2008, approximat ely twenty of the existing staff of nearly 150 worked at the hospital during the late 1970s or 1980s. None started working 31 The only scholarly reports I was able to locate regarding the health sector in Arusha region were Alger 1981 and Yoder 1980.

PAGE 116

116 there in the 1960s, and when I asked whether they were still in contact with anyone who retired but was present when the facility ope ned, the only people mentioned had passed away or were no longer in contact with the participants. Of the nearly twenty people I identified as having a longterm perspective on the facilitys growth and change through time,32 I formally interviewed fifteen. Major improvements to the hospitals infrastructure since HSR made garnering specific details about changes in the working environment prior to 1999 difficult. Knowing that the facility was a health center from 1968 to 1989, I divided analysis of working environments into two periods: prior to 1989, and the 1990s. However, when asked about the working environment and the hospital infrastructure of the past several participants made very brief statements about the past and then moved immediately to the ch anges they experienced since reform, as a statement from one maternal health nurse present at the time illustrates: There were few staff, and no buildings. It is much easier to do work now, since cooperation is there, we have motivation now. In light of the rapid infrastructural changes that occurred after HSR, it appeared that the period prior to reform seemed stagnant by comparison for many of the people I interviewed. However, I was able to get enough details from some participants in order to garner a general picture of the hospitals infrastructure, supply availability, and the general working atmosphere. Wherever possible, I include approximate dates for the infrastructural developments of the hospital, and indicate areas of divergence between differ ent oral histories when appropriate. 32 For the purposes of this h istory, I considered any staff having substantial experience at Kiunga District Hospital from before 1999 to have a long term perspective. Thus, this group included staff that worked at the hospital between ten and thirty years. Of those interviewed rega rding the history of the hospital, only eight were employed there since before the late1980s. However, there were enough consistencies in their statements to build at least a tentative picture of the conditions of the hospital during the late 1970s and ea rly to mid 1980s.

PAGE 117

117 Kiunga District Hospital: 1977 1989 According to hospital files, Kiunga District Hospital opened its doors as a small health centre in 1968. In 1977, when some of the interviewees first began work there, the health center occupied a large compound, but few buildings The buildings were constructed at quite a distance from one another. According to one report, Kiunga was one of only a small number of health centers in Arusha region that had electricity, although achieving a constant supply of electrical current was a problem (Alger 1981). The small building closest to the road leading to the villages was an administration block, containing two offices: one for the DMO, and one for the District Nursing Officer (DNO). Across from the administrative block was an inpatient building containing two small rooms, one for male inpatients and one for female inpatients, with the capacity for six beds in each area.33 Beside the inpatient ward was an approximately 200 square foot building that served as a storage unit (commonly termed store ). Adjacent to the administration building was a small outpatient department (OPD), with two rooms for consultations with clinical officers (COs). The OPD also had a very small laboratory, an injectio n room, a nurses room, a dispensing room, and a small room for wound dressing. About 500 feet down the road from the OPD was the labor and delivery ward, which had two rooms one for pre and post natal patients, and another for deliveries. According to st aff accounts, because the maternity ward only had six beds, women in labor often share d beds with women who recently gave birth. The compound also had a larger building, across from the maternity ward, which at the time was used as a hostel to house perso nnel who were either training at the facility or new to the area and had not found permanent housing. There were two rooms attached to the hostel shared 33 This small building would be renovated in the early 2000s to become the pharmacy.

PAGE 118

118 by the dental officer and the doctor on call during night shifts. Close to the hostel was another buil ding that served as a kitchen and a small laundry room, since at the time, the Center provided food to patients.34 Meanwhile, the compound also had an isolation ward, at a significant distance from the other buildings of the health center. This ward was for those patients suffering from communicable diseases such as Tetanus, Meningitis, or Tuberculosis (TB). Finally, to the north of the main health center were nine small staff houses, and one larger house allocated to the DMO (Figure 3 1) According the a survey of Arusha Regions medical facilities produced in 1981, while several health centers in the region required some minor repairs, the r epairs required at Kiunga were major, although construction had begun on a new female and pediatric ward that was an ticipated to ease some of the congestion (Figure 3 2) (Alger 1981). Staff who worked at Kiunga Health Center during the late 1970s and into the late 1980s characterized the working environment as very difficult. The facility lacked electricity35 and running water, not to mention many of the basic necessities needed to do their work. Mary, a nurse who began working at Kiunga Health Center in 1977, remembered the shortages and lack of capacity of the health center during the late 1970s and early 1980s: Mary: The working environment, that is, for example, the equipment for work, some of the equipment was scarce. It was there, but not a lot. That is, we were able to get maybe gloves Except for here, even water it was not like it was coming [out of the pipes]. For example, we had to go all the way to the river to get water to clean the wards! Other equipment for work like gauze we might get but things like antiseptic we were without. So you just cut a piece of gauze, you did procedures that you completely knew it was supposed to be a sterile procedure but because you did not have a way to sterilize you just did it like that. Otherwise, we had no way to get transport, so it was not very good. So even referrals of patients, it was not good because you found there was only one vehicle and it was being used for other things. But that, for that period of time, it was a health center and as a health 34 The hospital stopped providing food to patients after HSR, around 2004. 35 This lack of electricity reported in the interviews was in contrast to the Alger report, which states that Kiunga Health Center had a connection to city electricity, but the Alger report (1981) also acknowledged that maintaining consistent electrical current to the facility was an issue.

PAGE 119

119 center even those referrals were not so many because the only real hospital was the one at Mount Meru [in Arusha town], a nd of course the amount of people we had was not many. There were not many customers, so it was not like we were failing to send referrals because there were too many, no! Because also other people were using local healers.36 (Interview, 1120 08, Kiswahili ) In 1982, a nurse midwife named Elda transferred to Kiunga from a remote district in the western part of the country. She described the environment at Kiunga as better than that of her previous placement: Elda: but still there were difficulties. Because e ven here, there was no electricity. It was a problem, we had to use oil lamps to be able to assist in deliveries, many times you even ran out of oil. There were even times when we would have to burn paper for light. A mother would push and at that point we would light the paper. Uh! (Interview 10 208, Kiswahili).37 The conditions described were consistent with the resource poor working environments prevalent throughout the region. According to the 1981 report on health facil ities in the area : With few exc eptions, health facilities in the Region are experiencing shortages of basic supplies, medicine, and equipment. When such goods can be obtained, it is often with great difficulty and expense. In the past year, petrol, diesel fuel, and kerosene have been co stly and sometimes impossible to procure, although the problem is now less severe in most areas. Government institutions must order equipment and supplies from the Central Medical Stores in Dar es Salaam, where many items are out of stock and orders have been delayed for as long as eight months. Administrators indicated that as little as 25% of the goods ordered may arrive, although the norm is 5075%. Equipment is even less likely to be received, with some orders still outstanding after a years time [ sic] (Alger 1981:33) A ward attendant at the time, Rehema, began working at Kiunga Health Centre in 1986. She left a year later to do the one year course to become a medical attendant, but when she returned in 1988 she found the working conditions to be just as they were when she left. She spoke about the 36 The context of medical pluralism in the district is discussed below. 37 This statement contrasts with what was written about Kiunga in the Alger report regarding the facility having electricity. However, none of the people interviewed about conditions at Kiunga mentioned having electricity, and there were several people who talked about working without it. It was not clear from any of the interviews when electricity was made available at the facility, and even in 2008 it was not uncommon for the electrical supply to cut out, particularly during t he day.

PAGE 120

120 lack of medicines, and the need to send patients to private pharmacies to buy their drugs, So the patient who had been admitted, he was just admitted but that bed was the only thing you could give him. You didnt have services and treatments. If it was penicillin or chrystapen,38 whatever, everything you had to tell him he must go to buy (interview, Kiswahili, 112708). Dr. Joseph, who was a ward attendant at the center during this time described the diffic ulties of working within a context of scarcity: Dr. Joseph: The salaries were not enough for us. Things were not good, and the equipment for doing work, you couldnt do much of your work, and there was more work than you could do. Living conditions here we re very bad. Even medicines, even gauze, we didnt have it. You had to use your brain to think, you had to learn to cope and improvise with what was there. (Interview 11 308, Kiswahili) Due to the inadequate buildings and shortages of supplies necessary t o perform biomedical treatments, staff reported that the patients and their relatives compl ained a lot, and those who had the financial means went to private clinics or practitioners. According to a medical attendant, Rehema, the patients and relatives we re complaining because they were saying that we have our own hospital and it is nice, that is, for example here the hospital has doctors and it has nurses but now, they have no medicines! (Interview, 1127 08, Kiswahili). Amid the shortage of supplies, t he workers also grappled with chronic understaffing of the facility, which increased their burdens and adversely affected staff morale. There was one clinical officer39 (CO) at the facility until 1983, when another was allocated. According to two of the interviewees, there were only five nurse midwives40 available, although several nurse attendants were there to assist. The nurses might have two nurse attendants in their department, 38 Chrystapen is also called benzylpenicillin, another type of penicillin that kills bacteria. Crystapen is administered by injection or intravenous drip. 39 A clinical officer is the lowest level of doctor working in Tanzania. A full outline of the different levels of doctors and nurses at the hospital outlined in a note at the beginning of my study. 40 Nurse Grade B

PAGE 121

121 and there were no nurse officers41 at the facility Once the female and pediat ric ward was completed in 1982 (Figure 33) ,42 more nurses were brought to the facility, but the health center remained short staffed, causing a burden on those who worked there. When I asked her about the number of staff working at the health center during the 1980s, E lda responded: Elda: I cannot remember the amount, but we were not big like we are now. If you came for night shift you were only one person for the entire place. NS: So you assisted deliveries, you did injections? Elda: U h uh [no]. One person in each department. Maternity one person, female ward person looked on in the male ward and the TB isolation [ward]. One doctor. (Interview, 10208, Kiswahili) Despite government policy to improve rural health provision at the time, Kiunga, like other ru ral health facilities in the region, was not as well resourced as its urban counterparts in Arusha town, but had considerably more staff than rural dispensaries in the area (Alger 1981) Transportation of patients in serious condition was a constant struggle. The health center only had one car, which according to two nurses often underwent repairs or lack ed petrol, or it would be in use at the time when the staff needed to transfer patients. In instances when the staff was faced with emergency cases requiring transfer to the regional hospital, they relied on others. If it was daylight, nurses would walk to the main road and attempt to flag down cars to ask them to take the patient to Arusha. They also relied on a local college, the police department, and the post office nearby in order to negotiate a vehicle for patient transport. Nurse Elda described the difficu lties involved in securing a vehicle for patient transfer during the night shift: 41 Nurse Grade A 42 There was some discrepancy about the date construction of the female and pediatric ward was completed among the part icipants interviewed. However, the Alger report of 1981 on health facilities in Arusha region confirmed that the wards construction was underway at the time the report was published, so it appears highly probable that the construction was completed in 198 2.

PAGE 122

122 Elda: If you went to night shift you arrived by foot and you had t o ask a security guard to escort you way up there to [a nearby] college when it was raining, and beg for the college to give you a car to pick up a sick person and take her to the maternity ward Or you went with the guard all the way to the post office there, you went to ask the postal office to call the police so you could get a police car. And at night when it was raining, the road was really bad. You went and you would be rained on until you were completely wet. Because there was no phone, nowhere to ma ke a phone call. The postal office would help you to use the phone, call the police and say that the hospital requests [ inaomba] for the use of a car. Or we would go by foot to the college because there was no communication otherwise. NS: Did they agree t o help? Elda: Many times they would agree to help. They would say that the sick person could pay for the gas. (Interview 102 08, Kiswahili) Such a working environment, rife with chronic shortages, instilled little cooperation and morale among the workers. Lydia, a ward attendant at Kiunga who began working there in 1978, stated that during the late 1970s through the 1980s, each indi vidual governed him or herself, ( ulikuwa unajitawala zaidi mwenyewe ) and the administration was not very involved in the dai ly routines of the facility. She described how we forgot ourselves ( tumejisahau), and people came and left work whenever they wanted, without providing good service to patients. Lydia explained, They [the workers] could not arrive on time in the morni ng, maybe they were supposed to be at work at seven oclock in the morning, but workers would arrive past then. Like in the maternity they might not see patients until nine oclock! Nine oclock, eleven oclock because of that character [ tabia] of the wor kers made them late (Interview, 11 2608, Kiswahili). When I asked her how she felt about this working environment at the time, she responded: It was really bad, the person who was hurt was the patient, the person who suffered was the patient! So then it was that even many of the patients were not coming, and the reputation of the hospital was bad. Many went to private [clinics] because our services were not good.

PAGE 123

123 While staff often referred to corruption among their peers in other health facilities in th e country, and at the regional hospital in Arusha in particular, rarely were allusions made to corruption at Kiunga either in the past, or in the present. There were two exceptions, and only one participant was totally explicit regarding what occurred. Lydia, provided significant details about the informal payment schemes operating at the hospital at the time Lydia was an adolescent when she began working at Kiunga Health Centre. S he stated outright that staff was taking bribes: Lydia: Everything was free in the past, eeh, but people were just taking also. Eeh, people were taking so that that issue of taking money [bribes] was there except now it is not here anymore. That [bribery] was there in the past much more because first, it was not investigated [ hai fuatiliwi] Eeh, in the past yes it was really bad and this is why I have told you that now, services are good because in the past it was that every person charged whatever they saw fit [ kila mmoja anachaji anavyoona] you see? (Interview, 112608, Kiswah ili) When I asked Lydia how this worked in practice, and whether it was each person individually who participated, she responded: Lydia: Lets give an example, like here we will draw a picture. We each took o n our own here. They [the patients] come t o the male ward and they [the workers] take their own. In the re ception they take their own. Eeh, because many things went into peoples own pockets Two of the patients that I interviewed upon their exit from the hospital co nfirmed that they witnes sed this kind of conduct among the hospital staff in the 1980s. A woman in her fifties who lived in the area since she was a child commented even in the 1980s, I always got my services here, so I have always been satisfied and have never had anything what soever bad to say (Interview, 102908, Kiswahili). However, this woman remarked that nonetheless in the 1980s when she came to the hospital, you may have had to wait for a long time or pay someone off to be ab le to get services.

PAGE 124

124 As noted previously, t hese comments about the bad conditions of the past must be understood in light of the vast changes that occurred at Kiunga District Hospital not only in terms of infrastructure, but also in terms of the ways work was organized. Amid Lydias statements was one that was reiterated by several of her fellow workers who had been at the hospital for over a decade: Now we have come to understand what the meaning of work is, ehe! It was common during the interviews as well as in casual conversations to talk about how their work in 2008 had meaning ( kazi ina maana sasa) compared to the daily practices prior to HSRlargely because they now worked with some of the infrastructure and resources that allowed them to put their medical training to meaningful use (see Chapters Seven and Eight ). Therapeutic Pluralism and Encounters with Patients As several scholars have outlined, when biomedicine was introduced to Tanganyika, it joined the many existing African therapeutic resources available to people in their quest for he alth (see Langwick 2001, 2006; Malloy 2003; Nseke la and Nhonoli 1976; Ranger 1992). Maintaining the crucial separation between so called modern and traditional medicine was an important aspect of the colonial endeavor, as a way to establish certain forms of expertise and authority that were critical to the disciplinary civilizing of colonial modernization (Langwick, 2006:143; see also Malloy 2003). From the time when biomedicine was first introduced to Tanganyika up to the present, people have regula rly rejected, innovated, and combined biomedical technologies and practices with other therapeutic resources in accordance with their understandings of health and the body (White 2000, 1995; Langwick 2007; Whyte 1997, 2002). While Africans considered some biomedical treatments, such as the injection, as extremely potent, others were considered ineffective or inappropriate for dealing with their ailments (Ranger 1992). In a recent example, Susan Reynolds Whyte, working in Uganda with the Bunyole from the 1970s to the 1990s, traces

PAGE 125

125 the ways that pursuits of healing were negotiated and renegotiated, by exploring the continuities and changes in understandings about healing and misfortune through time (1997). With the deterioration of services through the 1970s a nd 1980s and the increasing availability of biomedical drugs and treatments, people in the 1990s became disenchanted and distrusting of biomedical workers. However, Whyte found that people were highly experimental in their uses of medicines, both biomedica l and otherwise Further, health workers operating in Uganda in the 1990s did not have access to many of the resources they would have liked in order to determine what disease they were treating. In practice, rather than determine a diagnosis, they have simply adopted the symptomatic idiom of their patients to the extent that they take the patients brief statement of complaint as the basis for medicating. They do not ask further questions to elicit an illness history or exclude possibilities (Whyte:213). As Stacey Langwick has noted, in a place where government health facilities are chronically under resourced, understaffed, and overburdened, the fact that the government encourages the use of biomedical facilities rather than local healers is somewhat cu rious ( 2007, 2008). In Tanzania as in other areas of Africa, the pragmatics of therapeutic choice have historically been influenced by social and familial relationships, as Leader Stirling describes: Besides the patients natural fear of an operation there was the complication that any injury to an individual is in African eyes an injury to the family, and so before an operation could be done not only was the consent of the patient essential, but also that of all his relatives. Even if the patient were will ing, the obstinate refusal of some old man, or even more often some old woman, could often make it impossible to operate, and one had to watch the patient dying slowly of a condition that could easily have been relieved. (Stirling, 1995: 92) The social context of healing in Africa has important impacts on where and when and how people seek treatments for their maladies (see for instance EvansPritchard 1976; Janzen 1978; Turner 1969; Whyte 1997). For maladies that are chronic, the costs of long term care can be highly expensive, and t his burden often falls to the family. Thus, as Whyte et al. demonstrate in a

PAGE 126

126 more recent context, the financial burdens to the family of chronic illness may impact a persons willingness to seek treatment at all as well as what treatment is most preferable (2006). While the intersection of these multiple therapeutic resources were important aspects of the realities of daily practice at Kiunga District Hospital. Indeed, non biomedical and informal biomedical therapeutic behaviors and practices were prevalent in staff interviews both about the hospitals history and the issue of health seeking behaviors of patients in 2008. Nonbiomedical and even informal biomedical43 therapeutic practices were consistently spoken of in a negative light, and many blamed local healers for the poor state of health of some of the patients that came to the hospital. According to the report on health facilities in Arusha in 1981, Still another segment of the population elects to use the services of the approximately 248 traditional healers found in the Arusha Region. Practicing even in urban centers, these doctors offer alternative treatment that is often more accessible and more familiar than that received at modern health care facilities (Alger 1981:31 punctuation in original ). Several among the longterm staff at Kiunga spoke about their encounters with patients, and at these moments it was common for the issue of nonbiomedical therapies to be brought up. When asked about what the environment was like in the late 1970s and early 1980s, a nurse administrator, Mary, began speaking immediately about the kinds of struggles that the staff encountered with patients: Mary: A person from you know there are a lot of cattle owners in this area, eeh? So some one would come from collecting grass [for the cows], she would drop the grass there [at the outpatient department], and went to the doctor. She said, my chest hurts! and the doctor understood. Or a person was suffering serious from something else, now she was very interested in getting an injection If you 43 By informal biomedical therapeutic practices I refer to those practitioners who provide biomedical treatments outside of a registered private or government clinic such as from within their homes. While these kinds of informal biomedical practices e merged largely in the late 1970s and early 1980s (see Benson 1996), the staff at Kiunga often noted that some people continued to operate in these informal therapeutic practices up to present. See discussion below.

PAGE 127

127 prescribed her a tablet she would just take it, throw it away just there under the tree and leave taking her load [with her] because she knew that what she wanted was an injection and not tablets. So at that time there was no cost sharing so they did not feel what, that pinch that these tablets are money. So that was what I saw, that people were, that is, that they were given medicine but they did not accept medicine without an injection. (Interview 112008, Kiswahili). At a time remembered for chronic shortages of medicines and supplies, Mary was not alone in describing her frustration with patients apparent lack of appreciation for their prescriptions. Three other nurses who were present at the time described similar examples of patients discarding their prescriptions because they were not the treatments that the patients came to the hospital hoping to receive Nurse Elda discussed village healers of the past at length, and their role in providing th erapies to people living in the area: Elda: Even here, the village healers gained a lot because people really depended on them. Local medicines, like medicines for stopping diarrhea, medicines for vomiting, for malaria, people would eat herbs for malaria. But also, there were many medicines that were not measured You were surprised because somehow they would work but others were not effective. Eldas description of the nonbiomedical therapies points to the precedence she puts on biomedical measurements of medical efficacy, as well as the inability of biomedicine to adequately explain why such therapies actually worked. While it was clear that she privileged biomedical explanations of therapeutic interventions, there was also an underlying acknowledgement t hat not all nonbiomedical therapies should be dismissed out of hand (see also Langwick 2007, 2008). Elda felt that local healers were taking advantage of the poor conditions under which the hospital was operating in order to profit, at the expense of the health of the villagers they treated. To Elda, being treated by local healers was hazardous, and could be potentially fatal. She described one healer in particular with whom she had several dealings during the 1980s:

PAGE 128

128 Elda: many people liked village healers because they did not know the dangers that could happen. Even witchdoctors there were many people who went to witchdoctors There was even one who was very tricky. He44 cured, but he knew that certain people had really big problems. He knew the symptoms of problems with the blood, he knew the symptoms of dehydration. But he still deceived them that he was giving them medicine. If he noticed that people were really having problems, he told them to run to the government hospital. Because he feared being cha rged for not helping. But he had already seen that person had a problem and still did not send them. He could see that the person was serious [condition] but he deceived them and said that the person had been bewitched He got his money there, afterwards h e would bring them [to the hospital]. There was one day when a village healer came to the hospital and she said, this person is vomiting a lot, and has a fever. I think it is malaria. I ask that you help this person and give a drip, be cause I see that it is malaria. But the person had already been staying at her house for at least a week She saw that the situation was getting worse and so sh e ran to the hospital and said I see that it is malaria. And at that time malaria was very high Every person who came i t was malaria, malaria, malaria. P eople would come to the hospital and throw away their medicines and did not use any prophylaxis so every person knew malaria, even the village healers. Even cerebral malaria they knew it. When they saw cerebra l malaria they were happy and they would say that the person had been bewitched and you see this is why he is confused. But he confused and deceived people for two or three days and eventually sent them to the hospital. (Interview 10 208, Kiswahili) Similarly, in a report about health care in Arusha Region produced in 1981, the author interviewed several health professionals, and she described their attitudes as follows: Health professionals interviewed tended to disdain the use of traditional medicine, c iting incidences of poisoning from herbal mixtures and infection from circumcision and other surgical procedures performed under septic conditions. Other procedures were termed harmless but of no medical use. Several, however, admitted that certain practic es could be beneficial to the patient. Often, it was observed that patients combined traditional and modern treatments or reverted to the other method when one form of treatment produced no results. (Alger 1981:31). Stacey Langwick has described similar p olitics of medical pluralism in southern Tanzania (2007), where staff commonly blamed local healers for patients in serious condition delaying coming to the facility. At Kiunga, this was particularly the case when it came to pregnant 44 The sex of the healer being referred to was not made clear in the context of the conversation. Since Kiswahili has no gender in its third person pronouns, it is also possible that the healer being referred to in this quotation was female.

PAGE 129

129 women, many of whom pr eferred the use of family members or traditional birth attendants to traveling the long distance to the hospital. While the issue of the health facilitys encounters with nonand informal biomedical therapies are not a central theme of this dissertation, they are nonetheless important to acknowledge. H ealers provide a wide variety of therapeutic services in Tanzania, but at Kiunga District Hospital, they were also highly vilified as precipitating medical complications with which the staff later had to deal and as causing many unnecessary deaths (see also Langwick 2007; Richey 2008) Becoming a District Hospital: 1989 1999 In the mid 1980s, the Ministry of Health (MoH) announced that it wanted to create a district hospital in Kiunga district, which up to th at point had none. At the time, there were only two facilities large enough to make the transition to district hospitalKiunga Health Center and another Center in the western part of the district. In order to become a district hospital, a facility neede d a minimum of sixty inpatient beds. The two health centers competed with each other to expand their facilities A motivating factor for the staff at both centers was that the government would allocate more staff and equipment to a district hospital than a health center; thus, the designation as a district hospital might ease the burdens of work, increase work satisfaction, and allow more specialized services to be provided. In short, it would bring in the kinds of resources that might appeal to the workers sense of professionalism, because their working environment would potentially resemble what many termed as a real hospital.45 According to those staff members present at the time, there was a massive impetus to become the new district hospital, and workers attempted to mobilize community members and local politicians in their efforts to secure the funds they would need to build up the hospital. Dr. 45 For a history of medical professionalism among East A frican physicians, see Iliffe 2002. For a recent treatment of medical professionalism among nurses, see Martin 2009.

PAGE 130

130 Joseph, who was a medical attendant at the time, discussed the many plans that the hospital administration put forward for building up the hospital: The workers were happy to be building! First, they wanted to expand the wards to reduce the overcrowding so that there were not three people in the beds. And they wanted to build a [surgical] theatre, and so that people were given knowledge also, and they brought in four more workers to increase capacity. It promoted [Kiunga] to be lifted up. With the erection of a surgical theater, the staff knew that the MoH would provide surgicallytrained AMOs to the facility, who would likely pass their expertise on to other staff at the facility. Lifting up the hospital was about more than expanding its infrastructure. It provided the potential for the staff to expand their professional skills, thus potentially earning them promotions or additional training. The official plan was to renovate the existing matrons office and hostel to become a new male ward (Figure 3 4) The staff looked to the local community to provide the funds they needed to expand. People in the s urrounding community contributed what they could and, according to the staff, were very motivated to help. Local politicians also donated funds, and the facility received support from the regional government. In approximately 1989,46 renovations of the new male ward were complete (Figure 3 5) As a result, Kiunga Health Centre defeated the other health center and became the new district hospital. The government provided the hospital with beds, mattresses, more medicines and staff. According to Sister Gertrud e, the hospital gained approximately fifteen more nurse midwives. She did not recall how many more clinical officers were brought to the facility, but she did remember that they were finally able to do round 46 There are discrepancies on the date at which the building was completed. One respondent said it was completed in 1983, but five other interviewees stated that the completion of the female/pediatric ward was part of the facilitys efforts to become upgraded to the status of a district hospital. Thus, given the critical role of building the ward in upgrading to the status of a district hospi tal, I have adopted the later date.

PAGE 131

131 rotations, so she estimated that there were betw een seven and eight new clinical officers present at the upgraded facility.47 Under the administration of the 1990s, there were efforts to continue expanding the infrastructure of the hospital, but these initiatives were slow going at best. From 1990, its f irst year functioning as a district hospital, until 1999 when HSR began to be implemented, there were only a few additional changes to the hospital. In 1994, the United States Agency for International Development ( USAID) donated funds to build what is toda y the major theatre. 48 At the time, the new space was for voluntary surgical contraception such as tubal ligation and vasectomy. Furthermore, the MoH promised to donate an x ray machine to the hospital if it was able to generate funds to construct a building within which to house it. However, according to the DMO at the time, Dr. Kalete, there were several political and bureaucratic constraints to securing the funds to construct the building. While the construction began in 1990, it was stalled for years as the DMO and the hospital administration attempted to finance the construction. In 1997, after the foundation for the new building sat without further construction for several years, Dr. Kalete went to the D istrict C ouncil (DC) to petition for the expansi on of the hospital, and the xray building in particular. She felt that because the C hairman of the C ouncil was from the other side of the district, where the health center with which Kiunga had competed was located, the politicians delayed the expansion. The project thus stalled, t he DMO went away to complete a m asters degree in Public Health. When she returned to Kiunga in 1999, she began the initiative 47 Sister Gertrude was the only staff member interviewed who was able to provide approximate numbers in the staffing increases after the facility was designated a district hospital. However, five other interviewees refer red to more staff being allocated. When asked about specific numbers, none of the others had an estimate. It is also not clear from the interviews with people present in the 1990s at what time the facility gained assistant medical officers (AMOs). However, based on the life histories I took, by 2000 there were at least four AMOs available at Kiunga. 48 As described in chapter 4 the major theatre is the place where in 2008, more complex surgical procedures such as caesarian section took place. The minor theatre, which was part of the x ray building described below, was used for less complex surgical procedures such as removal of superficial growths. See chapter 4 for further discussion.

PAGE 132

132 anew. Dr. Kalete managed to get one politician to support her, and then she took the project to the hospital administration directly: Dr. Kalete: I told [the administration] if we want to succeed we have to show a way first before [ pauses ] before somebody else. So we decided we should start to build on our own first. And as leaders, we should be the first one to [give] the money. So we proposed that [donation] to be paid according to salary scale. So we decided on certain amount of money. When we went to staff meeting they agreed with our idea, and they said the amount you propose is too little we should add more. So they [the staff] proposed another rate higher than what we proposed. (Interview, 61108, English) Each staff member agreed to contribute a portion of his or her salary to the building project. The staff that was present at the time spoke with pride about their contributions to the x ray building. Elda, a nurse at the hospital, stated, We felt good because we found that it was not good to stand alone. It was not that we were forced. We sat in our own committee, we said that we wanted to st art to contribute. We said we knew that the administration knew our salaries, and some would give 6,000 [Tsh]49 and others 3,000, willingly. Dr. Kalete and the administration then went to the DC and told them about the initiative, and the Council formed a mobilizing committee to go into the communities in the district and solicit donations However, by that time what little had been constructed of the xray building had deteriorated. The erosion of the existing construction became another bureaucratic hur dle for the administration to maneuver. According to Dr. Kalete, Because there was no more money and there were cracks because that they built the buildi ng of the sewerage system. When with the discussion with [District] Council they wanted us to abolish the building and build a new one. Then it came one activist, he said You people you are joking! This building is okay! Just block the sewerage system and you know, correct the defaults which you know had been occurring then complete this building, it is okay! So what we did, we completed that building, we got an X ray and a theatre. (Interview, 61108, English) 49 Amounts are in Tanzanian shillings. In 2000, 6,000 Tsh equaled approxi mately US$9.80.

PAGE 133

133 Thus, the efforts to build the facility after it was designated a District Hospital were slow, but in the end, successful. The xray building was complete by 2001over a decade from when the foundations were first dug. This initiative was completed by soliciting funds within the community, from local politicians, and from hospital personnel themselves. The search for funding for health facilit y construction and expansion from among community members was characteristic of the period prior to reforms; however, the ways that hospital personnel understood who could provide them with the funding needed for expansion would change considerably in the wake of HSR (see Chapter Five). Despite the increase in staff and the extension of the facility the people working at the hospital continued to speak of chronic shortages of necessary personnel. Indeed, prior to the implementation of HSR, staff morale rem ained extremely low at the hospital because of the lack of supplies, infrastructure, and human resources. The hospital expanded, but so did demand, providing little relief from the difficulties they encountered. Mary, who was an administrator prior to the facility becoming a district hospital, described the period prior and post 1989. According to her, It was 1989 to be a district hospital, but it was only in name. It was not that we had a lot of changes. So it really continued like a health center until, I can say until the end of the 1990s, when indeed things started to change (Interview, 112008, Kiswahili). Frida, who began working at Kiunga in 1992 as a nurse in charge of a ward, and who was later upgraded to become an administrator at the hospital, characterized the difficulties of leadership and the working environment thus: Frida: You remember, this kind of hospital, it was upgraded to the status of a hospital but it was even without buildings, so you can imagine to work in an environment that had no buildings nor work equipment, it was difficult. Noela, you cannot imagine. It was truly difficult because you came to work, that is, a nursing administrator came to the staff and she tells you, please work, do this and this and [the worker] tells yo u I have nothing to do. It was truly difficult even to

PAGE 134

134 assess that performance of a person. Because [the hospital] has not enabled her [ haijamwezesha] [The administrator] has not been able to get those things she needed so that [the worker] could do her work well. So it was really difficult, and then there was a big shortage of staff. By that time there was no employment [of new staff], employme nt had been completely stopped. So the staff that we had were those that had to be given all three shifts and when [a worker] came to work he did not have even an envelope in which to put the patients medicine. And even medic ine, firstly there was none. But as an administrator of course it was to try to encourage people, my goodness, let us fulfill our responsibilities [ jamani fanyeni timizeni wajibu ], we must help patients! But how could we help them? You had no medicine, no equipment. So it was really difficult. (interview 71608, Kiswahili) For the workers, the poor morale at the time was r elated to not only the shortage of the necessities of their work, but also the ways such shortages affected the quality of patient care. It meant little to be a health professional without an environment that allowed them to put their expertise to work. Re hema felt hopeless in her inability to help patients: Rehema: It was difficult because you could get a serious patient and he could die on you. So you feel that you are a bad person, eeh, because that is, you know for example that here we have an eme rgency Now, if a patient came who was asthmatic case, that is, that asthma or maybe the case was a person who had fallen and has suddenly lost consciousness [ akazimia ebu], think about it! You, until you prescribe medicine, the prescription is taken to th e far away [pharmacies], and a person brings those medicines, that patient, will you meet him again? Do you know if you will be able to find him? Is it not that that patient can die!? So also you feel bad. (Interview 112708, Kiswahili). Rehemas descript ion of the hardships faced in being unable to provide good services to patients due to a lack of supplies is reminiscent of the comments other health workers, such as Dr. Jilek Aall, made during the colonial period (see Chapter Two) As stated above, several of the workers present from the late 1970s and early 1980s through the 1990s characterized working conditions similarly between both periods. An exception was Dr. Joseph, who spoke of changes that he felt were significant: Staff was increased. And som e of the equipment was increased in the 1990s. He stated later, it was better because little things had improved. We were talking more, we had a few more things

PAGE 135

135 (Interview, Kiswahili, 11 1308). When I asked what was significant about the changes, he ex plained, The reason was because already we had a few of the buildings, and several things that were made better, like the reception, they continued to do more activities. And the TB unit, there were a few things that were made better. They were little but they were improvement Christine, a nursing officer who started working there in 1986, expressed a similar sentiment, acknowledging that in the 1990s they experienced some changes because the donors and the government increased salaries slightly (Intervi ew 11 2408, Kiswahili). Overall, although several of the interviewees spoke of the hardships of their work and difficulties practicing medicine in a context of scarcity, many saw this as just part of the routine. Sister Gertrude, who began working at the facility in 1981, des cribed how they had no supplies, and were often compelled to ask patients to bring their own bags in which to put their medicines, but they were used to these conditions, we saw it as normal (Interview, 81708, Kiswahili). Similarly, Christine said, Equipment [during the period] was still a problem. But if you are already used to something, you just do your work. I wanted supplies but I knew I could not get them. Overall, when it came to how she felt about the conditions under whi ch she worked, she said, It was our government, we were already used to it. What can we do ? (Interview 11 2408, Kiswahili). In the 1990s, some of the workers heard that health sector reforms were being planned, but many were unaware. Of those informed o f the pending changes, waiting for the kinds of resources needed to provide adequate health services had adverse effects on fragile morale. Frida described the process of waiting for the government to implement the promised changes: Frida: Of course the wo rkers thought that [reform] was a requirement. That is, to change the system Of course, it came to a place where people were saying, now, if the government has failed, fine, maybe even all of the hospitals will privatize or something like that. But by t hat time even privatization was not there. So it was just

PAGE 136

136 people talking, and that complaining. But they did not have the ability to compel [ kuishinikiza ] the government to do this or this or this. But because people believed that maybe it [reform] would c ome later and the condition would change, basically they were encouraged to think this year, may changes come so at least the conditions improve and people can do their work in an environment that is a bit better. So they were really in a condition of w aiting [ walikuwa katika hali ya kusubiri kwa kweli ]. (Interview, 7 16 08, Kiswahili) Having experienced the deterioration of conditions at Kiunga over the previous years, several of the interviewees expressed a sense of ambivalence about working conditions in the 1980s through the 1990s. This feeling of helplessness with regards to effecting changes in the governance of their workplace would transform in significant ways in the years following HSR. Conclusion From the time when the health center at Kiunga first ope ned to the late 1990s when it was a full fledged district hospital, the staff experienced a variety of challenges. Morale was low, the equipment they required was unavailable, and there was a general sense that staff was unable to influence the government in order to procure the necessary resources to adequately do their job. Several struggled with the moral ambiguities of their work: requiring patients to purchase their own supplies and drugs, opening up informal clinics from their homes, or requi ring patients to pay for services that were officially supposed to be free in order to supplement their inadequate salaries. With the economic crisis of the 1980s and the effects of SAPs during the 1990s, eking out a living was made more difficult for the majority of Tanzanians. Corruption was not a major problem in the health sector until the 1980s (Iliffe 2002), when salaries were too meager to allow for health personnel to subsist and support their families. Working in the hospital was morally ambiguous because one could not rely on this kind of work alone to support ones family, but also because it was not uncommon to see patients suffer or in some cases die because the supplies that were supposed to be at the hospital to help patients were unavailable. Trained in health professions that promised to use technologies, pharmaceuticals, and specialized expertise

PAGE 137

137 to heal the sick, health workers felt largely cheated when they worked within an environment that undermined their ability to use their professiona l training to its desired use. They desired reform to realize the kinds of infrastructural possibilities that would make their professional skills meaningful, and access to salaries commensurate with their specialized expertise. Next, I explore how such re forms were conceptualized and later implemented at the level of the state.

PAGE 138

138 Figure 31. Kiunga Health Centre, c irca 1978

PAGE 139

139 Figure 32. Kiunga Health Centre, c irca 1981

PAGE 140

140 Figure 33. Kiunga Health Centre, circa 1982

PAGE 141

141 Figure 34. Kiunga Heal th Centre, circa 1988

PAGE 142

142 Figure 35. Kiunga District Hospital, circa 1989

PAGE 143

143 CHAPTER 4 ESTABLISHING TANZANIAS TRANSNATIONAL AID REGIME In the 1990s, as the Government of Tanzania (GoT) worked towards dec entralization and privatization, nongovernmental organizations (NGOs), faith based organizations (FBOs), communitybased organizations (CBOs), and private for profit entrepreneurs proliferated in Tanzania. They became involved in health provision and other social services previously offered primarily by the government (Mercer 1999; Di lger 2009). While NGOs and FBOs were present in Tanzania during Ujamaa their roles were limited by a centralized state inhospitable to competition except in remote areas beyond its reach. Thus, in the 1990s, as donors and t he Tanzanian government planned and operationaliz ed sector reforms, the surge of NGOs, FBOs and CBOs in the country allowed more stakeholders to be involved in the health care provision than ever before. After 1995, the GoT planned and implemented a wide r ange of reforms in its governance structures, at both central and local levels. The new systems put in place included programs to improve transparency, private sector participation, public sector management, and particular procedures such as anti corruptio n policies meant to move the country towards good governance. Prominent stakeholders within the internat ional donor communitynow to include several NGOsprovided input and financial backing to for these governance reforms (see Harrison 2008; Gould 2005).50 Indeed, these reforms can be understood as taking place in a context of a transnational aid domain, within which aid recipient governments interact with a select group of both public and private aid agencies, all employing a remarkably similar set of objectives, procedures, policies and frameworks despite being very differently positioned within 50 Currently, stakeholders in Tanzania include the MoH, a group of bi and multilateral donors (Development Partners or DPs), civil society, and Tanzanian communities. Each of these actors is expected to become involved in the initiativ es and projects aimed at improving the health and livelihoods of all Tanzanians.

PAGE 144

144 the system (Gould 2005:63; see also Green 2003). According to Gould, this remarkable convergence on vocabulary among the key actors in the aid domain wa s a major divergence from the aid domain of the 1980s and early 1990s (2005:62). During the 1990s, the structural adjustment programs (SAPs) to which many aiddependent countries signed in the 1980s were not achieving the expected results in economic perf ormance in those countries. The World Bank and the International Monetary Fund (IMF), collectively referred to as the Bretton Woods Institutions (BWIs) formulated the SAPs, and therefore their credibility was being undermined in the donor community. Tanzania and Uganda became important test sites for a different approach to development assistance. The new approach highlighted by the BWIs enacted a different approach to debt reductionone that highlighted poverty reduction, and thus allowed the BWIs to refor mulate their reputation in the eyes of donors. Thus, Tanzania played an important role in establishing the transnational aid domain characteristic of the 2000s (Gould 2005). T he relatively recent shift in Tanzanian governance regimes and the character of donor support echoes processes occurring in several other aid dependent countries. The transnational aid domain involved in governmental reform emphasizes values such as r ecipient country capacity building, and even ownership of policy and programs oc cur ring in partnership with particular stakeholders supporting the government in its progress towards good governance (see Anders 2005, 2010; Chalfin 2010; Gould 2005; Mosse 2005a; Zacher 2007). These processes are part of what David Mosse refers to as a new architecture of aid (2005a:3), characterized by economic liberalization and privatization, wherein various public and private donors (stakeholders) collaborate through particular frameworks to support governments

PAGE 145

145 However, t he forging of deve lopment relationships from the 1990s until recently should not be seen as a case of donors dictating agendas to a vulnerable and aiddependent government. Instead, as Gould argues: [T]he Tanzanian evidence conjures up a vision of a constructed community of interests that transects the borders of agencies, bureaux and nationsThis community of interest is the main site of the social construction of partnership, and is the source of a specific logic of engagement within which specific technologies of subjec tification are at work. Tanzaniais a prime candidate for such an analysis by virtue of the extent to which responsibility for donor designed social agendas have been taken over by various local agents. (2005:67, emphasis in original) Given the difficult economic conditions of the 1980s that led to the countrys adopting SAPs, the GoT was only too well aware of the risks of alienating the stakeholders on which it depended. The logic of engagement to which Gould refers was established because the MoH realized that it needed to make itself attractive to foreign investorswhether NGOs, private organizations, or bi and multi lateral donors. Gould defines the logic of engagement as one that largely (though not with any great strategic consistency) serves to in still self governmental disciplines in the client subjects of the aid relationship: state and non state actors alike (Gould 2005:70, parenthesis in original). In his analysis of the remarkable changes in Tanzanias aid relationships to its donors, techno logies of subjectification are the precise means by which these disciplines are imbued onto the various actors involved in the aid domain. In large part, these technologies operate through documents. Drawing on Anna Tsings (2005) notion of the economy o f appearances, I argue that the specificities of Tanzanias transnational aid regime came about due in no small part the MoHs self conscious performance of particular discoursesand thereby, possibilities within the countrys health sector that appealed to donor values prevalent at the time. The major transformations observed in Tanzanias health sector are tied largely to the MoHs ability to

PAGE 146

146 foster and maintain a collaborative space across differences. Tanzanias health sector is made a site of possibil ity through the various strategic frameworks and contracts to which the MoH and its stakeholders became signatories and over which they continue to collaborate. These documents are what Annelise Riles calls good specimens of a particular genre (2000:72 see also Stirrat 2000), attentive to particular languages of management, accountability, transparency and good governance attractive to donors, while producing variations applicable to the Tanzanian context. In line with the work of both Gould (2005) and Riles (2000), I find that these exercises in creating a particular aesthetic of documents were not about substance of the policies, but rather on making them appear in a particular way in order for them to be recognizable to, and appealing to, donors. Rat her than market its accomplishmentswhich in the 1990s were few the MoH was found a means of selling its potential by drawing on imaginaries of what could be accomplished in the sector if the correct governance regimes and routines were established. The Mo H was largely able to accomplish this through an economy of appearances that required the production of policy and procedural documents of a particular style. Overall, this chapter provides critical background to policy and procedural re arrangements that opened up certain opportunities at Kiunga District Hospital to transform the practices, routines, and infrastructural arrangements that characterized it prior to Health Sector Reform (HSR). Below, I outline particular moments during the 1990s and 2000s that had important impacts on how aid and governance in the health sector were oriented. Accordingly, I trace how HSR was initially planned in the context of eroding relations between the MoH, the donor community, and people working in the health sector. I t hen introduce the factors that brought about a major shift in donor government relations, how this led to a significantly different aid regime in Tanzania marked by coordination and the values of partnership ownership good

PAGE 147

147 governance, and capacitybuilding. Finally, I outline remaining conflicts and contradictions underlying a transnational aid regime that, on paper, appears highly collaborative Underneath the published and oft cited values of partnership and coordination lay considerable dis cordances among donor organizations. Certain donors only partially agree with the overall frameworks to which they became signatories, and the differing values between particular donors remain the subject of considerable contestation in practice. Fashioni ng the Frameworks of Health Sector Reform: 1990 1995 From 1990 to 1995, the MoH set itself to devising a series of policy documents that would move the countrys health sector towards the decentralization and liberalization required by the dictates of the SAPs. The various policy and framework documents the MoH produced during this period provided a foundation conducive to increasing the number and types potential stakeholders that could participate in health care provision in the country or at least, ha ve some kind of voice in the priorities that would come to characterize the logic of engagement that became the foundation for donor government cooperation in the country. However, during this early period, as NGOs, CBOs and FBOs multiplied throughout the country, and while the GoTs ability to truly conform to the dictates of its SAPs was still in doubt, these initial frameworks and reports served to demonstrate a failing governments commitment to reform. It was these documents, along with a particular in tervention by a joint Tanzanian expatriate consultation team, which fostered the kind of environment that made the relationship between Tanzania and its current and potential donors a possibility. Opening a Space for Collaboration Early in the 1990s, the MoH was receiving pressure from both donors and health professionals within its facilities to create a system to govern the decentralized health sector. At the district level, there was no over arching guideline directing health facilities on

PAGE 148

148 administratio n, management, and evaluation of their facilities. In 1995, the Principal Secretary of the MoH wrote: Various attempts by the MOH to encourage districts and regions to develop plans of action have, hitherto, not been very successful. Without standards to go by, district attempts led to plans which were not always successful in qualifying for funding. Frequently, such plans were seen as mere shopping lists and often found to be quite unrealistic. (MoH 1995:ii, punctuation in the original). Thus, the Ministry of Health created a series of guidelines to enable districts to plan for their health needs. The World Bank provided a loan for the purposes of paying external consultants to help in planning reforms and creating a set of strategies for carrying out the decentralization process. The Danish International Development Agency (DANIDA) donated money to pay for the local costs of the process, and the German Ag ency for Technical Cooperation ( GTZ ) provided technical assistance. The MoH subsequently formed a national team of Tanzanian government officials and nongovernmental organizations. According to Dr. Faustin Njau, who in 2008 was the Head of the Health Sector Reform Secretariat at the MoH and who was an original member of the national team the decision to decentralize the health sector was independent of the overall decision in Tanzania to decentralize the government: Dr. Njau: Reforming of the health sector was a necessity, because first of all we had a very big [ pause ] infrastructure network, and then we wanted to strengthen the paramedical part of it because we had only one medical school by then [in the early 1990s] and [it] was difficult to produce enough doctors to be in the forefront in the health system. So we had to strengthen the main areas but also we had [ pause ] to focus on improving the management of the health system, im proving the maximization of the [ pause ] little amount of funds that we had. So, and also to decentralize. Whether there was decentralization of the government itself or not, the Ministry of Health during that time had decided that at the district level, let it go. (Interview, 070208, English) T he teams major concerns in developing proposals for reforming the health sector included improving overall management of the sector, human resources, capacity building at the district level, and ensuring referral hospitals were providing adequate support to the district level. This

PAGE 149

149 notion of capacity building was, according to Gould, embedded in interventions with specific goals tha t [were] almost always about conforming to formal demands or expectations of aidrelated actors and processes, and was therefore linked to specific policy agendas of the aid community (2005:70). As he aptly notes, there is no end to the extent to which one can acquire new capacities, and thus the process towards building capacity can be infinite and may always be incomplete (Gould 2005:71) According to Dr. Njau, i n terms of financing, we wanted now to introduce cost sharing in the system because the amount of fund in the budget we get was not sufficient. Being efficient with limited funds, managing them through certain monitoring and evaluation routines, was paramount. The team also worked to devise suggestions of ways to ensure greater community part icipation in the health sector, as well as better coordination of donor operations. This commitment toward both donors and public participation marks one of the earliest instances where the MoH worked to construct a shared community of intere sts (Gould 2005), necessarily enjoining local and global actors in the endeavor to forge a new, decentralized health sector that performed based on the capacity building measures to be forged through HSR. In 1990, a new National Health Policy was released to the pub lic, to provide an underlying foundation for the subsequent reform process. The stated goals within the Policy were to reduce maternal and infant mortality, increase life expectancy, make health services accessible and available throughout Tanzania, improve human resources through upgraded training, educate communities about preventable diseases and appropriate treatments, and emphasize the necessity of multi sectoral cooperation,51 and promote good health within families (MoH 1992, 1994a). 51 This multi sectoral cooperation was to include the education, water and sanitation, and agriculture sectors, as well as other entities such as community development and wome ns organizations, and NGOs.

PAGE 150

150 Here again, we see the MoH attempting to forge a community of interests across multiple scales and layers within and beyond its national health sector. A revision of the National Health Policy was undertaken in 1993, and released in 1994. The National Health Policy of 1990 originally stated that health care would remain free to the public, and that the GoT would continue to be the main health care provider. However, in 1994, the Policy was amended to suggest avenues for decreasing the governments financial burdens in the health sector. The document pointed to the establishment of user fees and a public/private mixture of health care and encouraged research to determine o ther financing methods. Importantly, practitioners in the private sector were to be considered partners instead of competitors (MoH 1994b). Due to t his provision, a plethora of research emerged suggesting strategies for transfer r ing some of the costs of health services to citizens. The move was controversial, and several studies grappled with what user fees would mean for Tanzanias poor (see for instance Abel Smith and Rawal 1992b; Kiwara 2003; Lugalla 1995). The liberalization and semi privatization of Tanzanias health sector was thus made official policy through this policy document a technology of subjec tification that moved the country into line with the a new architecture of aid that would allow multiple public and private donors to have an increasingly important role in Tanzanias health sector. Proposals for Health Sector Reform and the Strategic Hea lth Plan Collectively, the MoHs team devised a plan for HSR, and circulated a draft to the senior management. These plans were subsequently published as Proposals for Health Sector Reform in December 1994. The Proposals stated that there were several fatal flaws that challenged reform implementation. The lack of effective participation between stakeholders in the health sector the various ministries, donors, technicians and other parties involved in the sector stalled efforts. Further, the Proposals crit iqued donor funding approaches that circumvented the

PAGE 151

151 MoHs governance structure. Focusing particularly on vertical, targeted donor initiatives52 such as those relating to HIV/AIDS, MCH, TB and Leprosy, and the Essential Drugs Programme, the Proposals argued that vertical programs caused: duplication of generic functions, training and supplies as well as a serious lack of coordination. Some of the donors prefer to operate without the coordination from the Ministry of Health and deal directly with the Regions /Districts. The final outcome is an absence of integrated planning, monitoring and evaluation. (MoH 1994b:21) Indeed, within several sections of the Proposals donor sponsored vertical programs were criticized for undermining and fragmenting health sector administration more generally. Among the suggested improvements was better integration of health related programs in the country, as a matter of both policy and implementation at all levels. Here we see the Proposals pointing to contradictions in donor val ues and donor practices if donors wanted to see the government perform according to their notions of good governance, they would have to truly commit to allowing the state to become a meaningful partner. This is a telling example of ways that the state attempted to bring its donors into a particular logic of engagement based on a coordinated partnership emphasizing values through which these different actors could collaborate meaningfully. In addition, the team argued that the re was no framework for effec tively monitoring reform implementation. Ultimately, the team wanted to ensure that all actors had a common understanding of the way forward, from the donors and ministries to communities and families. The Proposals argued that health planning and policy m aking required creating and effectively utilizing new tools and strategies for measuring inputs and outcomes within the sector, to ensure 52 Vertical projects are those initiatives donors sponsor directly, bypassing government structures to operate programs directly on the ground, within various governmental and non governmental facilities throughout aid depende nt countries. Because these activities circumvent the state and operate largely independently within these countries, they are often perceived as threatening to values underlying the SWAp and JAST.

PAGE 152

152 that cost effectiveness remained a main priority of future reforms (MoH 1994b). Drawing on James Scotts notion of le gibility practices (Scott 1998) Michel Rolph Trouillot points to a legibility ef fect, by which numerous state and nonstate entities negotiate the tools (policies, procedures, plans, vocabularies) of governing through classifying and regulating in parti cular ways (2003) As a legibility effect (Trouillot 2003) the MoHs emphasis on continual monitoring and evaluation carried the promise of maintaining state power in a decentralized system by embedding structures that would allow the central government t o maintain a degree of oversight at the local level. Ultimately, the Proposals suggested a variety of avenues by which the health sector could be reformed. In terms of managing the sector, district health services would be decentralized, and come under the management of the local government. District Medical Officers (DMOs) would be the officers in charge of district health services, and those physicians with training in public health would be given preferential treatment for the post. For managing human r esources, the Proposals recommended introducing monetary incentives such as extra duty allowances and improved salaries to entice workers to remain at their assigned facilities, since post abandonment (particularly in more remote facilities ) was a recurren t problem in the sector. Instead of coordinating promotions, salary increases, and disciplinary procedures through the MoH, the reform proposed transferring this responsibility to the local government level, where health sector employees would fall under t he District authorities. Further, the Proposals suggested that remuneration in the public sector be commensurate with the private sector, to stem the gravitation of public workers toward the private sector. Members of the District Health Management Teams53 were to receive training in management and planning for the health sector, 53 The DHMT would later change title to become the Community Health Management Team (CHMT) see Chapter 7.

PAGE 153

153 and the courses were to be distributed equitably within the districts. At Kiunga District Hospital, this move of compensationrelated issues to the local level would create significant challenges (see Chapter Seven). Once the Proposals were created, the team drafted broad plans for reform implementation. In 1995, the team published the Strategic Health Plan 19951998 (Tanzanian Ministry of Health 1995a), which outlined ways that the Ministry proposed to address the issues outlined in the Proposals. Both the Proposals and the Strategic Health Plan were circulated to the Tanzanian Ministerial Cabinet, where they were endorsed in early 1995 (Semali 2003). I asked Dr. Njau whether the Wor ld Bank was also invited to endorse the Proposals He clarified that while the World Bank and some other donors had financed the development of the document, and it was offered to those donors (such as the Scandinavian foreign assistance organizations) who had a vested interest in HSR, we were not asked to submit [the Proposals ] by the World Bank. Because it was our document, our designs. However, despite Dr. Njaus statement it was clear that the Tanzanian government could not go ve ry far without donor endorsement and that donors had considerable influence on the reform agenda (see Gould 2005; Harrison 2008). Given that the health sector was largely dependent on external financing, and in light of the economic decline of previous years, it was clear th at donors influence was also needed in the process of planning and implementing the health sector reforms (Semali 2003:73). In short, the GoT had to play into a complex and systematized economy of appearances in order to ensure its investors would financ e the reforms. Hostilities in a Stagnant Sector Though several reform plans were in place during the mid1990s, actually turning those plans into practice was a highly complicated affair. Due to a long history of neglect of the sector, there were a variety of problems from within the health sector workforce that caused general

PAGE 154

154 complaints within and outside of Tanzania about the quality of healthcare provision in the country. Few health facilities in the country had been upgraded since they were first erecte d. While there were minor improvements in the 1990s in drug and supply chains, many health facilities lacked the basic necessities and human resource requirements to make providing meaningful biomedical care a possibility. Many of the staff with whom I spoke who worked in the public or private sector during the 1990s described difficult conditions within health facilities, and many workers were frustrated that the rights the MoH guaranteed them were not upheld. In response to the difficulties of work, many health sector employees turned to private business endeavors, increasing absenteeism or tardiness within health facilities. Others responded by informal (and ethically questionable) income generating activities within the health facilities. Still others we nt on strike. The MoH lacked support from the very workers critical to actualizing its plans for implementing reforms. Few workers felt that the government was doing enough. C orruption within the sector was increasingly a source of complaint. Drawing on r esearch in Ulanga, southern Tanzania during the 1990s, Maia Green found that people were deterred from seeking care at the district hospital because Access to care [at the facility] is said to be conditional on the payment of gifts to hospital staff, even in the event of a medical emergency (2000:410). Some Ministry officials were critical of the unofficial income generating practices of medical professionals. In 1995, while visiting Muhimbili Medical Centre,54 the Minister of Health, Mrs. Zakia Meghji, st ated, We shall not put up with those who receive bribes as an excuse for low salariesand those who think that they cannot do without bribery better leave the offices before it is too late ( The Guardian, Dec. 13 1995:1). The Minister pointed to the 54 Muhimbili is Tanzanias national hospital, and the largest government owned hospital in the country.

PAGE 155

155 probl ems at Muhimbili with worker absenteeism during business hours She heard reports that people were showing up to work briefly, then departing to work in their own business esoften their own for profit clinics and private hospitals. Mrs. Meghji publicly war ned, I really wish that the first person to be netted is one of these bigshots so that the rest can learn, and she urged hospital workers to anonymously report bad conduct. Further, the difficult conditions in the health sector sparked a series of medi cal worker strikes during the 1990s (Iliffe 2002). The Minister of Health responded by banning these protests, but this prohibition received mixed reviews with Tanzanian politicians. According to a report in The Guardian in 1997, during the Minister of Hea lths budget speech, a Member of Parliament (MP), Mr. Ndimara Tegambwage argued, I disagree completely with the ministrys statement on page 25 of the ministers speech saying that it was a big foul for doctors to strike as it was contrary to their profes sional ethics (July 22:5). He continued, strikes are legitimate the world over, not only for other professionals, but also for medical professionals. The MP argued that instead, the Ministry should try to address the issues that doctors were bringing to their attention, improve the conditions in which they lived and worked, and thereby avoid future strikes. At the same Parliamentary meeting, another MP from Iringa, Mr. Mfwalamagoha Kibaasa, held that given their inadequate salaries, doctors should be provided with free medical care when they themselves were sick (Ibid.). Moreover, while the MoH made on call all owances an official policy in 1992, few districts were in a position to be able to compensate health sector employees for their overtime and oncall work. I n The Daily News on August 27, 1997, the Minister of Health, Dr. Aaron Chiduo, announced to the National Assembly that only select health workers were receiving their allowances. The Minister argued that it was the right of those medical practitioners to receive

PAGE 156

156 their arrears, and all persons owed arrears should send documentation to the Treasury, since the districts were unable to pay for their allowances. However, since the government was short of money, it will pay the allowances on phases whe n funds are available ( Daily News Aug. 27 1997:3) Thus although the districts were unable to procure the finances to pay doctors their oncall allowances, it was the right of those medical practitioners to receive their arrears, and the MoH was going to find a means of paying them. The Ministers commitment to its policy favoring worker compensation was a potential means by which the MoH could regain its employees trust, and thereby potentially curb some of the problems with absenteeism and ethically mu rky practices occurring within health facilities. By increasing investment in the status of health workers, the MoH might improve their work ethic, and thereby, improve the conditions of health care provision more generally. However, budgetary constraints caught up to the GoT. In July 1997, the government was forced to suspend any new employment within the civil service, as well as disallow existing employees promotions ( The Daily News Dec. 3 1997:1). Capacity building as an ideal existed throughout GoT policy documents and strategic frameworks; yet the financial capital required to actualize those policies remained elusive. Donor Government Relations in the 1990s Until recently, many donors and external consultants characterized Tanzania as being at the f orefront of improving donor government relations. According to a consultant report on the health sector in the country, Tanzania is regarded by many as a model of donor Government partnership (Franz 2004:1).55 There is a history to how relations between donors and the GoT became so cooperative. T he signing of the Paris Declaration in 2005 marked a major commitment on the part of donor governments and organizations to work towards better 55 Similar depictions of the donor assistance environment in Tanzania can be found in other consultant re ports regarding the health sector and a variety of health interventions in the country. Se e also Kelly and Birdsall 2008)

PAGE 157

157 harmonization of their development efforts in recipient countries .56 Yet Tanzania initiated measures to coordinate donor aid and improve aid structures nearly a decade before. The discussion below outlines how Tanzania transitioned from a fragile and eroding relationship with its major foreign financial supporters to dist inguishing itself as an attractive locale for development aid and investment It is largely a story of how the MoH worked to make itself appear attractive for donor investment. As Anna Tsing argues, the possibility of economic performance must be conjured like a spirit to draw an audience of potential investors (2005:57) Therefore, added to capacity building as a term through which donors and the MoH could collaborate, the logic of engagement between Tanzania and its stakeholders would officially empha size coordinationa concept that would strengthen, and later spread to other aid relationships beyond Tanzanias borders through the Paris Declaration. Eroding Donor Government Relations: The Helleiner Report In 1991 the MoH lifted bans on private pract ice and encouraged NGOs to assist the countr y in providing health services, largely precipitating the proliferation of NGOs, FBOs and CBOs offering health related services in the country. Meanwhile, in the mid1990s, donor enthusiasm for providing assistance to the Tanzanian government waned. In the absence of what donors felt was adequate progress towards SAP conditionalitie s, a large proportion of aid was shifted towards NGOs. Donor enthusiasm for supporting NGOs and socalled grassroots development was in line with what Kelsall and Mercer call a global ideological climate hostile to the state (Kelsall & Mercer 2003:293; see also Helleiner 2002). Donors mistrusted the government, while simultaneously, the government perceived donors as too demanding and 56 The key principles underlying the Declaration include improving the capacity of recipient governments in terms of development agenda s, aligning donor and government objectives and utilizing local systems for distribution of aid, coordinating information and procedures between donors to avoid duplicating efforts, and focusing on effecting measurable results. For further information, see OECD 2008.

PAGE 158

158 intrusive into state affairs. Aid to the government was largely frozen, and what assistance donors continued to provide was directed only toward particula r sectors within the government deemed trustworthy (Bagachwa et al 1997 ; see also Gould 2005). In addition, during this period particular health issues (maternal child health, and especially HIV/AIDS) were becoming of heightened concern in global health circles, many of which were brought nearly exclusively under the care of NGOs. Concerned by the cool ing relationship between donors and the Tanzanian government, representatives from Nordic governments longtime donors within Tanzania arranged a meeting with the GoT in an effort to negotiate a better working relationship. As a n outcome of the meeting, Nordic donors commissioned a team two nongovernment employed Tanzanians, and three foreign academics57 to assess the underlying reasons for the impasse. The team was to recommend ways to foster more efficient cooperation and to clarify the issue of what own ership meant in terms of development and participation in the process. The results of the teams research were published in 1995: The Helleiner Report (Helleiner et al. 1995). The team interviewed representatives from among the donors and the GoT. According to the report, the donors concerns were that: [T]he government of Tanzania has lost its momentum and its sense of direction, has little sense of direction, has little sense of ownership of its major programmes, and is unable to exercise fiscal contro l because of declining administrative capacity and increasing corruption. After more that [sic] thirty years of support, donors are disappointed with the Tanzanian performance record and regard their continued support for the GOT as politically unsustainable among their own electorates (Helleiner et al. 1995:5). 57 Of note, although the five consultants hired to compile the report, one of the European consultants later became a World Bank Deputy Country Director, and one of the Tanzanians on the team went on to become the leader of the largest politica l opposition party in Tanzania (Gould 2005).

PAGE 159

159 Further, donors felt that the government did not take their concerns seriously, and regularly responded with rhetoric. At the same time, according to the Report, donors expected that the GoT would become more of an owner of its programs, but simultaneously became agitated when the GoT, in its capacity as an owner, did not favor the projects that the donors themselves preferred. Meanwhil e, the Helleiner R eport summarized the complaints of the Gove rnment thus: The GOT, on its side, considers that the donors are often unrealistic in their demands and their impatience. Its politicians and officials believe that the pace of change in Tanzania is as fast as is technically and politically feasible. They feel that they are being singled out for disproportionate (and negative) attention by the international donor community. They perceive the problem of increased corruption as, in part, a response to reduced real public sector wages and salaries; while seeki ng to lessen it, they see the problem as no more severe in Tanzania than in other developing countries. They regard the donors as driving Tanzanian development programmes and intruding excessively upon matters of domestic policy, and they resent their inability or unwillingness to share information. They do not appreciate the donors evident lack of trust or their consequent efforts to circumvent the government system by creating project islands of their own.They are frustrated by the sheer number of f requent meetings, reports, and contacts that donors require. They argue that inappropriate and impossible donor demands my prejudice some considerable progress that they have so far managed to achieve (Helleiner 1995:5. Parenthesis in original). For the m embers of the consultant team, the perspectives of both sides had merit, and the report advocated for both sides to address the grievances of the other. One of the major critiques on the part of th e consultants who compiled the R eport was that there was a lot of rhetoric on both sides regarding the value of the Tanzanian government owning and coordinating aid priorities, but that this was rarely achieved in practice: There is much more lip service to coordination, it seems to us, that [sic] there is commit ment and action. Time and again we heard from all concerned of the need to reduce the number of projects and to adopt a sectoral focus or concentration. Similarly, many people emphasized the importance of harmonizing country programs with Tanzanias own pr ioritization of projects. Yet effective prioritization on the part of the GOT has still to be developed, and donors willingness to abide by it remains, on the evidence, mixed at best. (Helleiner et al. 1995:15).

PAGE 160

160 Ultimately, the R eport advocated collectiv ely working towards developing the governments management capacity from the top to the bottom, particularly with regards to its projects and programs. Gould (2005) notes an important irony in the Reports recommendations: the Report argued that the government should be trusted that in order to foster ownership and partnership, yet the GoT was plagued by problems of ineffi ciency and its employees frequently accused of unscrupulous acts, making trust difficult to achieve in practice. Planning Reforms As a result of the R eport, after the 1995 election, the GoT worked to improve the macroeconomic climate in the country, and in conjunction with the donors endeavored to implement the Helleiner teams recommendations. In terms of negotiating the actual forms of th e structural adjustments and liberalization of the governing regimes in Tanzania, the 1990s, and after 1995 in particular, marked the beginning of a remarkable degree of cooperation between donors and the Tanzanian government in terms negotiating the parti cularities of the reform agenda (Helleiner et al 1995, and Helleiner 1999). In 1999, after three years of planning, the MoH published and distributed what came to be known as the First Plan of Work (POW 19992002).58 The overall approaches underlying HSR a nd the first POW were divided into eight strategies: 1) providing high quality, accessible, cost effective health services at the district level within a decentralized system; 2) ensuring adequate support of PHC within regional and national (specialist) ho spitals, with a view to improving emergency services; 3) redefining the role of the MoH as policy making body, the main function of which is to effectively communicate the policies, norms and standards underlying health care 58 In my search through the archives of the MoH, while I found the Second Health Sector Strategic Plan (HSSP II), I was unable to locate the first one. It was only in speaking with Dr. Njau that it became clear that the first POW acted as the first HSSP, although it was usually referred to as the Plan of Work 1999 2002.

PAGE 161

161 provision; 4) developing human resources adequately to all levels of the health sector; 5) putting in place a central support system to efficiently provide or oversee the provisioning of staff, drugs and supplies, accounting, auditing, infrastructure, communication and transportation se rvices; 6) implementing a mixed system for financing the health sector, including funds procured from public, private, and donor funds, as well as exploring other options such as national health insurance, community health financing, and cost sharing; 7) e ncouraging partnerships between private and public sectors in delivering health services; and 8) improving the relationship between donors and the MoH in order to improve coordin ation of initiatives. As seen in this POW, the values of coordination, capa city building and even partnership were now to be transmitted throughout the network that linked together individual health facilities, local governments, the MoH, and various extra state actors (donors, NGOs, CBOs, FBOs). This division of policy into s pecific POW strategies was part of the governments economy of appearances, performing development discourse in through a documentation genre recognizable and attractive to donors. Once the POW was in place, the MoH planned to begin the decentralization pr ocess in January 2000. However, the process was delayed because the Ministry was unable to adequately distribute the funds necessary to roll out the reforms, and the structures for disbursing funds were not completely negotiated in time. Actual decentraliz ation began in July of 2000 in thirtyeight districts. An additional forty five districts were decentralized in 2001, and the last thirty one districts began the process in 2003 (Semali 2003). Financing a Mixed Health Sector As HSR was being planned, a ma jor area of concern was how to increase financial inputs to the sector, so that it was not so reliant on the government and its donors. In an effort to transfer some of the burden of financing the sector onto the populace, the MoH decided to

PAGE 162

162 incorporate cost sharing and risk pooling into its funding mix, to include user fees for patients, a Community Health Fund (CHF), and a National Health Insurance Fund (NHIF). All three schemes were meant to increase revenues to the health sector in order to be able to i mprove service quality in government facilities. Particular categories of people were considered exempt from paying fees, including people over the age of sixty, children under six, pregnant women, students, people with disabilities, as well as those patients with communicable chronic diseases (such as tuberculosis and HIV/AIDS). The hospitals collecting fees were permitted to use the money to improve infrastructure or procur e supplies not otherwise available. Between 1999 and 2000, user fees were then intr oduced to all of the other government health facilities in the country (COWI 2007). Paying fees for health services was not entirely new in Tanzania; however, this was the first time that patients would officially be paying for services in public facilities. M issionary clinics and other government approved private facilities had been charging fees for their services for decades. In addition, as discussed in Chapter Two, many health workers charged fees illicitly in government facilities, in particular sinc e the economic declines of the 1980s. When the prospect of cost sharing was official introduced, several concerns were voiced about Tanzanians ability to pay for their services. In a study in 2000, Muela et al. found that while even the poor might be willing to pay for their medical services, many were not able to do so, even if they were in a position to pay for traditional medical services. The authors argued that because traditional healers often took alternative forms of payment such as payment in kind or labor patients were often in a situation where they could pay for services. By contrast, because government health services would only accept monetary payment, the introduction of user fees would put several people at risk. In another study in Ulanga, southern Tanzania, it was found that people

PAGE 163

163 delayed going to hospitals until they were critically ill, due in part to the fees required (Green 2000) I n Greens study, even very poor people were willing to pay for services, but because they had a negative view of the conditions within government health facilities, they were less likely to seek treatment in a hospital than to turn to informal biomedical or indigenous healers. P eoples avoidance of state facilities was in turn influenced by the complex inter action of structural adjustment policies, institutional constraints at all levels of the Tanzanian health service and the strategies of individual health workers to supplement inadequate levels of pay (2000:423). The successful introduction of user fees t o health facilities amid goals for improving healthcare delivery overall would be difficult without local community participation and buyin to the prospect of paying (at least in part) for health services. Yet as we see in the year 2000, the lack of flexi bility in forms of payment and the eroded institutional contexts within which health services were provided served as a major impediment to generating local support for this significant policy change. Developing a System of Aid Coordination The cooperative donor government framework sparked by the Helleiner report and subsequent planning and policy documents was extremely attractive to prospective donors in Tanzania. At least at the macro level, the GoT effectively conjured itself to appear attractive for investment. As various reforms were carried out within the Tanzanian government during the 1990s and the early 2000s, the GoT often in conjunction with external consultants and donors adopted a variety of strategies and initiatives aimed at better coordina ting and organizing development projects and initiatives in the country. The Tanzanian health sector wide approach (SWAp) was implemented in 1999. All donors wishing to contribute to the health sector in the country had to agree, at least in principle, to the MoH coordinating their programs and projects, and to the general principles of the SWAp. Within the SWAp, a large proportion of

PAGE 164

164 donor funding was to be pooled in a basket and distributed through the MoH to the various programs and facilities throughout the country. The SWAp was a means by which to foster the governments capacity to balance its own budget, and increase transparency and accountability. According to a series of case studies of five different African counties that adopted a SWAp, the Un ited Republic of Tanzania has the most advanced sector wide approach (Brown 2000:v) As a technology of subjectification meant to entice donor investment in a particular kind of collaborative engagement based on values of coordination and partnership, the SWAp was a means by which the MoH could alter the emerging aid regime to allow for prevalent development policies to be adapted to a Tanzanian context. In a climate recently hostile to the state, through the SWAp, the MoH claimed itself as a legitimat e partner, with capacity to have an increasing say in how donor funds were allocated. Since agreeing to the SWA p, Tanzania and its donors established policies and agreements to further orient the health sector towards what are apparently common goals.59 In 1999, the MoH published the Health Sector Plan of Action July 1999June 2000 (POA). The POW (mentioned above) outlined key reform priorities, and the POA outlined the specific steps by which these reforms were to be accomplished (Brown 2000). In 2003, the MoH implemented the Second Health Sector Strategic Plan (HSSP II), to govern the years 20032008, emphasizing provision of quality health service (Tanzanian Ministry of Health 2003) again, a carefully articulated value to which donors, the MoH, and loca l facilities could collectively commit, at least in principle. In 2005, the MoH released its Guideline for Reforming Hospitals at Regional and District Levels. T hese reforms aimed to strengthen the management, planning, and financial accountability of hospitals operating at the regional and district level. Within the reformed 59 However, as is discussed below, the amount of agreement with the overall environment of cooperation, and the degree to which individual donor s and government offices subscribe to it, are points of contention.

PAGE 165

165 hospital planning and management structure, hospital administrations would be trained in creating annual and strategic plans for their facilities (capacity building) ; improving the quality of health services (capacity building); developing and implementing monitoring and accountability structures (goodgovernance) ; creating the governing entities required by the National Guidelines in order to manage the facility (goodgovernance ) ; increase management capacity and financial accounting structures (capacity building) ; obtain financial resources in order to support the maintenance and rehabilitation of facilities (through partnership with private organizations and local communit ies) ; and make such funding resources sustainable (good governance) These Guidelines can be understood as technologies of subjectification to bring regional and district health administrations in line with the logic of engagement being fostered at the l evel of the state (Gould 2005). They can also be seen as a serious attempt to implement the legibility effects (Trouillot 2003) of the 1994 Proposals for Health Sector Reform effectively training district and regional health officers to produce good spe cimens of a particular genre (Riles 2000) required in order for the MoH to effectively govern from afar. Subsequently, in 2006, the GoT adopted an initiative for aid coordination and harmonization entitled the Joint Assistan ce Strategy for Tanzania (JAST ). Newly anointed as Development Partners (DPs),60 the bi and multi lateral donors involved in Tanzanian development work to improve the governments planning, budgeting and evaluation processes so that donor specific processes can be streamlined and incorporated into the national planning, budgeting and monitoring system. The main aim of the JAST is to provide a framework for donor government cooperation. It emphasizes shifting aid towards General Budget Support 60 The terminology Development Partners came into being in 2004, coordinated by the Development Partner Group a selection of donor representatives charged with increasing donor communica tion and collaboration in partnership with the MoH.

PAGE 166

166 (GBS) increasing the use of government systems, improving aid predictability, and encouraging open dialogue between the GoT and the DPs. Here we see cooperation, coordination and partnership written into a contract to which all DPs must sign if they elect to conduct any health related activ ities in the country. Helleiners vision of trust and partnership between donors and the GoT became a policy reality at least on paper. Tanzanias Current Aid Modalities At present, there are several types of donors financing development activities in Tan zania. Private aid includes that provided by churches, individuals, charitable organizations or other nongovernmental groups. Official aid is that provided by governments. Among the official aid, there are bi lateral and multilateral donors, and aid may be provided through grants (which are not repaid) or soft loans (having low interest rates and long periods for repayment). Bi lateral aid is provided by one government, whereas multi national is provided by an institution made up of several government s, such as the World Bank or various United Nations entities. Tanzan ia is among one of the largest s ubSaharan recipients of foreign aid, with aid money making up approximately 35 percent of total government spending in 2008/09. (OECD 2010). Aid is provided in three modalities, the philosophies underlying which are subject to debate among donors operating in Tanzania: GBS, Basket Funds, and Project Funds. The GBS pools donor monies into a common fund; the GoT then plans and budgets these funds for its vari ous sectors. Aid is thus integrated directly into the government budgets of recipient countries pending donor approval The underlying philosophy behind GBS is to build the capacity of recipient central governments to manage their own budgets and programs capacity building through trust in the central state. The United Kingdom, Ireland, and the Scandinavian bi lateral donors are major proponents of this modality, with several other donors contributing to GBS while maintaining vertical projects simultaneo usly. However, GBS as a funding strategy is not

PAGE 167

167 supported by all donors: Japan and the United States maintain their own management and oversight of their funds, and do not contribute to GBS (see Moss 2007). Thus, not all donors are so trusting of an aidde pendent state. A second aid modality is the basket fund, adopted in October 1999. Initially, e ight donors agreed to allocate funds to the health basket: DANIDA (Denmark), DFID (UK), GTZ/KfW (Germany), Irish Aid, Netherlands, NORAD (Norway), SDC (Swiss) and the World Bank. While these donors did not supply all of their funds through the basket, they agreed to allocate some of their funds to this modality. The actual operation of the Health Sector Basket Fund (HSBF) is complex. The Basket Financing Committee comprises the MoH, the Prime Ministers Office Regional Administration and Local Government (PMO RALG), the Ministry of Finance (MoF), and the HSBF donors. The Committee evaluates annual budgets and plans, as well as quarterly reports, and also reviews donors contributions to the basket and approves t heir release from the basket funds holding account to the districts. The Basket Financing Committee agrees to release basket funds based on the MoHs satisfactory reports, approved forthcoming cash flow forecasts, and proof of monthly fund releases for previous POAs (Hobbs 2001). Another philosophy underlying financing the health sector is Public Private Partnerships (PPPs). PPPs started to be advocated in the late 1990s within international donor and public health circles. PPPs were seen as a means of distributing the burdens of developing health care in economically disadvantaged countries. However, what PPPs actually entailed was ill understood in the early years when these kinds of partnerships were being advoca ted (Ridley 2001). Through the PPP initiative, g overnment health facilities are widely encouraged to network with a variety of donors and benefactors in order to build their infrastructure and capacity.

PAGE 168

168 One of the most contentious aid modalities in T anzania is that of project support. Although many of the DPs contribute to GBS and/or the health sector basket fund, there are a variety of vertical initiatives to which the DPs variously contribute, which has caused strain on the SWAp structure and the JAST. Despite commitments to donor harmonization, many of the DPs provide significant contributions to projects that are off budget, and therefore elude the ability of the government to properly plan for these infusions of resources in their annual budge ts (see for instance Paul 2005). In 2005, it was estimated that over half of donor funds within the health sector were allocated to vertical projects and programs, outside of official sector budgets Today, many donor organizations continue to subcontract their aid to NGOs or other private organizations, which are responsible for implementing the projects. USAID is one of the bi lateral organizations that subcontracts through NGOs most often (see Moss 2007). This kind of targeted funding runs contrary to th e Paris Declaration (OECD 2005) to which all of the DPs have signed. Thus, despite international agreements and the actual structure of the Tanzanian aid financing modalities which emphasize government ownership, There is a persistent tendency among donor s to earmark funds, not just to the [health] sector but, more seriously, within the sector (Paul 2005:iv). The S afety N et under the B asket Since their inception during the 1990s, the merits of the various health sector funding modalities have been the subject of significant debates among the DPs and various ministries of the GoT. For instance, when a consultant interviewed a variety of donor representatives in 2000, she found that: Interestingly it seems to have been the donors themselves, especially th ose in the Basket, who have been more critical about whether being outside the basket still counts as supporting the SWAp. This was a continuation of a debate during the 1999 appraisal process in which donors questioned each other about how serious

PAGE 169

169 they were about supporting the SWAp if they were not intending to be in the pool (Brown 2000:17).61 As if a display of its commitment to increasing GoT capacity, in 2003, DFID opted to withdraw its support to the HSBF in favor of providing contributions to the GBS overall. DFID donor argued that by investing directly in the health sector, the DPs were perpetuating government reliance on aid instead of investing its own resources in the sector. As a result, the United Kingdom lost much of its influence in the he alth sector. However, those DPs that do not support GBS maintain that the GoT does not have sufficient capacity to manage a system based on general budget support alone (Paul 2005). In 2008, at the Joint Annual Health Sector Review meeting between the DPs and the MoH, this debate continued as a major undercurrent of critiques of particular stakeholders in the sector. USAID, in its refusal to contribute to GBS or the health sector basket fund, was often subject to these critiques and those working for the ag ency often had to defend the United States position. Yet in my interviews and casual conversations with Americans working for USAID or Center for Disease Control (CDC), it was clear that there was an overall sentiment that recipient governments were too inexperienced (or by some accounts, corrupt) to be able to channel funds efficiently from the top levels down to the district level adequately. Further, as one of my informants explained, there was nothing built into the basket funding mechanism that would provide funds during emergencies, unknown contingencies, or shortages. All basket funds were tied to carefully prepared budget line items, with little possibility of flexibility. One officer within USAID Tanzania articulated the dilemma thus: 61 Indeed, in my interactions with various donor representatives at the Joint Annual Health Sector Review in October 2008, the fact that particular donors and the USA in particular refused to support GBS and basket funding continued to be a considerable point of contestation, and there were numerous comments and references both officially within the meeting and unofficially in conversations I overheard about whether or not such donors could truly be commit ted to the JAST and SWAps.

PAGE 170

170 I really fee l that its important to have that safety net under the basket. That there are some donors who are wiling to step up to the plate when necessary, bail out the country when something has failed, when the best plans and the best intentions didnt work, because were talking about human lives here, and not just one or two, were talking about thousands and thousands and thousands. If we run out of malaria medicine, well, people are going to die. If we run out of antibiotics, if we run out of contraceptives, it has a real and intense impact on individual people. And I think we as donors still have a responsibility. Weve sort of developed some of this the dependency on us, and its not a dependency that we like. Id love to see all the countries do what these Eu ropeans think these countries are capable of doing, and Im glad theyre pushing that way. But I think there has to be some safety nets under it, because there are a lot of decisions made by the Minister [of Health], the President, you know, were fi ghting that kind of stuff all the time And if they were able to do whatever they wanted to do on their whims, there would be a lot of consequences. But we want to give the government the opportunity to grow, to mature, to do the proper thing to develop themselves. And they are. And things are getting better. But were not there yet in all places (Interview, USAID Health and Population Officer, 09 2408, English). The kind of stuff to which this representative referred was considered largely irresponsible and unethical, and conversations with aid workers from other bi lateral organizations concurred with this comment. In conversations with two Danish development workers who worked in several development countries for over two decades, they pointed to the T anzanian mainlands MoH as inefficient. One of these individuals stated that working with the MoH in Zanzibar which maintained its own separate Ministry of Health was far easier because decisions could be made at a faster pace. Indeed, he spoke of many occ asions when he wanted to initiate a program on the archipelago, and was able to call the Minister himself and have the process begun within 24 hours. Comparatively, initiating programs on the mainland was said to take months, with numerous stalls in the pr ocess. Inefficiencies in governance were a source of considerable frustration for many of bi lateral donor representatives in Tanzania. Just as not all DP representatives agree with the general budget support aid modality, not all aid workers support the overall notion of SWAp, even if their employers are signatories to the JAST and contribute to GBS. One Scandinavian development worker I interviewed expressed

PAGE 171

171 hesitations about SWAp as an aid strategy. He articulated that, with the advent of SWAps, bi late ral donors replaced their most educated consultants with generalists who could provide very little useful technical support to recipient governments. As a result, he observed an overall retraction of DPs involvement in terms of the number of personnel working within their offices in recipient countries, as well as in donor personnels interest in the development of meaningful programs and sector strategies. He argued that the technical staff of the donor organizations were mere bureaucrats who were unable t o truly collaborate due to their lack of experti se on health sector development : But from the very first day when we discussed [SWAp] in the 90s, I said, youre going to fail. You are not readyto invest what is necessary to be, what I called qualifi ed dialogue partners. In that, the SWAp policy or the SWAp idea is based on a policy dialogue. But that, to my understanding, requires that you have qualified dialogue partners. You and I cant have a meaningful discussion if one of us is very very far from the other one. Either way, under informed or over informed, we only have a meaningful discussion, and also a discussion where we can respect each other, especially if we have to resolve a problem together, we need to respect each other and we can onl y do that if we are also well informed on both sides. What I see, it is that the Development Partners, they dont qualify as dialogue partners because theyre not well informed. And they dont want to be informed. If [the MoH asks] them, Please, come out [to site visits of clinics and hospitals], look, see. They say oh no, we are so few. Ooh, I have my reporting to do. No, no I dont have time to do that. (Interview, 082308, English. Emphasis mine). In particular, this informant noticed that the DPs were using SWAps as a justification for reducing the number of personnel they had within aid recipient countries. In his view, SWAps was a means to save money so that more of foreign aid went directly to the recipient countries and they could report to the ir taxpayers that less was being spent on overhead. The informant saw this strategy as a big mistake: By going in and providing sector wide support, you have to know what is going on in the sector, and how can you know if youre reducing your number of technical staff? According to this informant, who had been involved in development work (including several postings in African countries) for twentyfive years, in order for SWAp to

PAGE 172

172 succeed, recipient countries required a lot of technical assistance in developing the mechanisms that would be appropriate to actualizing the kinds of changes that these countries desired Thus, here was a paradox. The DPs, having extracted the vast majority of their personnel who actually had experience governing health sectors, now expected the Tanzanian government to figure out, largely on its own, exactly how to keep the donors happy. Without technical consultants on the ground, there was effectively no way to really understand how policies and programs affected individual districts and their health facilities. And, as the interviewee above contended on numerous occasions, donor representatives had no interest in actually seeing what was happening. It was no surprise that the USAID representative felt that a safety net was necessary; he mention ed several occasions when the budget system failed, and the country was on the verge of running out of essential drugs or supplies. In these cases, due its refusal to support basket funding, USAID could actually provide those emergenc y funds. R ecently Tanzania again f e ll out of favor with donors. As of May 2010, the DPs involved in General Budget Support decided to withhold nearly US$220 million for the financial year 2010/2011, citing the GoTs in sufficient reform implementation. In particular, the DPs are displeased with the lack of the countrys overall improvements in investment and business sectors as well as the delays in providing equitable access to public services in the country (Lugongo 2010). In December, the DPs met with t he GoT to discuss their concerns over corruption in the country. The chair of the GBS Group of Development Partners in Tanzania, Mr. Svein Baera (Norway) is quoted as saying, It is important to see evidence on progress made on the fight against corruptio n, and that evidence should be provided not only to the DPs but also to the public (The Citizen, 2010) By February 2011, the GoT had to appeal for emergency support from donors in an unspecified amount, due not only to the GBS shortfall due to donors

PAGE 173

173 wit hholdings, but also to food shortages caused by drought and electricity shortages due to poor hydroelectric infrastructure causing the government to have to purchase more fuel to make up for electricity shortfalls (The Citizen, 2011b). The World Bank announced it would provide GBS support of between US$2.5 and $3 billion over the next four fiscal years (Kamndaya 2011). However, in terms of whether or not full donor support for GBS would return in the 2011/2012 fiscal year will not be determined until May 20 11. Notably, it took fifteen years for donors to begin publicly expressing their disapproval of how the Tanzanian government managed its reform procedures and corruption. Given the transnational aid regimes universal vocabularies of good governance and transparency, the exodus of specialized consultants able to collaborate in a meaningful partnership with the MoH, and the disagreement among donors regarding the most appropriate financing modalities for the health sector, one wonders whether the GoT is entirely to blame for failing to live up to donor expectations. Conclusion The 1990s was a period of significant reconfiguration not only in the governance systems of the health sector overall, but also in the aid modalities underlying it. Reliant on dono r investment in the health sector, the MoH produced multiple policy and framework documents that would appeal to emerging donor values at the time. Ultimately, the MoH worked towards a particular logic of engagement, which would configure its relationship with its stakeholders, while also enticing individual health facilities to provide particular kinds of inputs (legibility effects) by which the MoH could demonstrate its accountability and transparency to its donors. This logic of engagement congealed ar ound a remarkably similar set of values to which the government and its stakeholders could collectively commit and through which it could collaborate (Gould 2005): capacity building, partnership, good governance, coordination and

PAGE 174

174 cooperation were key discourses marking the majority of policy documents produced throughout the planning and implementation of HSR. However, when principles had to be put into practice, conflicts arose. As noted by individuals working for donors, collaborative transnational aid c onfigurations also had their drawbacks. Donors who supported GBS and the health basket left little space in budgets for unplanned contingencies; all funds to be spent had to be premeditated and accounted for. The SWAp allowed donors to continue to invest m oney in the health sector, while reducing assistance to the MoH for developing the very governance mechanisms by which the GoTs successes would ultimately be measured. Donors who circumvented aid modalities such as the GBS and health basket also undermine d their credibility to other stakeholders with whom they collaborated; as signatories to both the Paris Declaration and the JAST, these donors vertical programs contradicted the values of coordination and partnership most characteristic of the current tra nsnational aid regime. On paper, the MoH, its DPs and its other stakeholders collaborate in an atmosphere highlighting cooperation and partnership; in reality, there remains a significant degree of conflict among donors, and between donors and the state. I n such a climate, s everal scholars question the degree to which it is possible for an aid dependent state such as Tanzania to truly have ownership over its policies (Green 2003; Gould 2005; Harrison 2008; Semali 2003). However, the reorientation of the he alth sector towards the ideals of partnership has transferred a degree of ownership to locales where before it did not exist: to individual health facilities fortunate enough to be located in places where that ownership would be meaningful. If decentrali zation was meant to increase ownership at the local level, then in the case of Kiunga District Hospital the effects of this ownership are palpable. As reorientations of the health sector

PAGE 175

175 advocated entrepreneurialism on the part of government health facilit ies, ownership could only be meaningful if a facility had access to potential organizations and individuals with whom to network. It is to Kiungas remarkable transformations under HSR that I now turn.

PAGE 176

176 CHAPTER 5 INFRASTRUCTURE AND P ROFESSIONALISM: EXPA ND ING THE HOSPITAL These changes are many, my goo dness, these many changes I see! Nurse, Kiunga District Hospital62 On a warm day in November, nurse Monica and I sat in a storage room of her ward for a formal interview, hoping that our lo cation would allow for privacy as she described the difficulties she experienced as a hospital worker. She traced the many transformations she witnessed at the hospital since she first started working there in 1997. From descriptions in the late 1990s of a working environment marked by stagnation, overcrowding, and paucity, Monicas account turned to the many new buildings and services available at Kiunga District Hospital since Health Sector Reform (HSR). Monica felt that these changes were only possible because the Ministry of Health (MoH) loosened restrictions on health facilities collaborations with potential donors whether non governmental organizations (NGOs), bi lateral donors, private businesses, or concerned individuals. Monica talked about challen ges she faced as a Nurse In Charge (see Chapter Six), but when describing the infrastructure of her current workplace, her tone was positive, animated, and hopeful. When I asked her what she remembered desiring for her workplace during the 1990s, she said working under such scarcity left little room for thinking about the future. It was normal, she said. Yet in 2008, she had many ideas about future directions for the hospital, and for nursing practice within it. Interviews with other staff who had worked at the hospital since before the reforms were similar in this regard; reflections on the past were littered with descriptions of the difficulties of their work. The staff lacked the vast majority of supplies, drugs, technologies and infrastructures that would make their work meaningful. Morale was low. They often described feeling helpless (see 62 Interview, November 2008.

PAGE 177

177 Chapter Three). However, when the conversation turned to the current state of the hospital, most of the staff expressed excitement about what had been achieved, and envisioned possibilities for the future. How can we understand this change? Writing about rural Niger during the late 1980s, Adeline Masquelier (2001) reveals how a new coat of paint on a dilapidated dispensary caused people in the area to assume that th e facility finally received essential supplies and drugs that it desperately needed. Patients flooded the facility in hopes of accessing quality health services. The place went from being seen as a place of death (268) to a place of newly established pr osperity (273). Yet inside the dispensary little had changed; it continued to lack what it needed to provide care, and health workers only doled out what supplies were available to a privileged few. While Masquelier provides compelling details about how c ommunity members envisioned the dispensary after its new coat of paint, her analysis leaves out how this change in working environment impacted the people who worked there. In his analysis of the history of East Africa physicians, John Iliffe argues that one of the many challenges that threaten medical professionalism is the lack of acceptable working conditions (2002). Yet we are left to wonder what kinds of working conditions might be acceptable for medical professionals. What is it about a hospital env ironment that appeals to workers sense of professionalism in Tanzania? What makes a meaningful workplace for health professionals? If a facility is able to move toward that ideal, what effect does this have on the ways that health workers think about, and indeed practice, their profession? This chapter describes the many infrastructural transformations that occurred at Kiunga District Hospital in the wake of HSR. Changes in the hospital landscape had profound effects on how hospital employees felt about th eir workplace and their status as medical professionals. Indeed, while ethically murky practices were common in the years prior to HSR, after the

PAGE 178

178 reforms were implemented at the hospital, there were few reports of employees engaging in such practices.63 For instance, during an exit interview, a patient in her fifties who had been coming to the hospital since the 1980s said, In the 1980s you may have had to wait for a long time or pay someone a bribe to be able to get services [at Kiunga]. But now, even if you offer [a bribe] people refuse! (10 2908, Kiswahili). In this chapter, I would thus like to consider how perceived institutional possibilities affect medical professionalism. This analytical space allows for reflections on how professional desires and institutional possibilities converge. While Masquelier illustrates that a dispensarys nice faade suggests possibilities to patients, I argue that it may also do so for medical professionals. As Sjaak van der Geest argues, Machines and advanced medical techniques conjure up faith, hope and trust, in patients and in physicians (2005:141, emphasis in original) Arguably, it does so for the professional interests of other health workers as well. In my daily encounters with hospital workers, I noted consider able hardships and inadequacies in the institutional landscape. Most of these related to infrastructure. By infrastructure, I refer to physical and material structures required to provide biomedical care. Kiunga lacked paved walkways that would enable easy transfer of patients in wheelchairs or gurneys between departments. Shelving units were often too small to house supplies, and were often cluttered with broken or obsolete equipment. Wards and clinics were incredibly congested. 63 As described in Chapter Three, hospital employees reported that particular individuals would request informal payments from patients prior to providing care. These informal kinds of payments have be en reported in many health facilities throughout Africa, both in the popular press, and in academic studies (see for instance Andersen 2004; Gibson 2004; Martin 2009; Olivier de Sardan 2001). However, in 2008 staff reported that these practices had ceased (although they reported other problematic practices described in the following chapters). Further, when I asked patients during exit interviews about the hospital, all remarked positively on the changes at the hospital, and there no one reported being aske d for informal payments. If these practices occurred, no one reported them to me. The only informal payment I ever witnessed was a family purchasing a beer for a Clinical Officer (CO) in a local market after work. This gift was in appreciation for the CO t reating their relative at the hospital.

PAGE 179

179 Water faucets leaked, and several were nonoperational. These were my own concerns, however, and were rarely echoed by the staff. Susan Leigh Starr argues that infrastructure does not stand alone; instead, it is embedded inside other technologies, social arrangements and structures (1999). Thus, the organizational structures and principles that make up infrastructures are obscured when all of its parts operate together seamlessly. In Starrs view, infrastructure is often taken for granted, and its complexities are only visible when i t breaks down. However, infrastructure in resource poor institutional settings such as health facilities in Tanzania cannot be understood as having this kind of embeddedness. There is little expectation on the part of healthcare workers that infrastructure will not be present, nor that available infrastructure will function as required Instead, such infrastructure is a desire and at Kiunga, until recently, it was more like a dream. The MoHs recent encouragement of health facilities forging their own publi c private partnerships (PPPs) allowed these facilities to look beyond the state for collaborative partners interested in the health sector. Within H SR policies, health facilities that possessed institutional vision and that were advantageously located coul d build infrastructures through (often temporary) networks with donors and other organizations that were, in times past, unimaginable. Kiungas hospital administration sought potential partners quickly. Their successes under PPPs, along with an increasingl y coordinated transnational aid modality had important effects on the facilitys transformation in the span of eight short years. As a result of these successes, staff at the hospital began looking primarily to non Tanzanians to procure funds, supplies, and infrastructure that the government could not provide. By 2008, there was a remarkable confluence in the ways that workers articulated their desires for their institution in the future,

PAGE 180

180 their view of the infrastructural changes that occurred in such a short period of time, and their impressions of donors who helped with those achievements. The infrastructural achievements at Kiunga District Hospital from 2000 to 2008 allowed for the hospital administration to establish a particular economy of appearanc es directed at the staff itself. Prior to HSR, staff was only intermittently committed to providing care, and absenteeism and poor quality of care were common (see also Andersen 2004). However, when the administration found donors to expand the hospital, m edical professionals working there felt that the facilitys capacity was approaching the ideals implied by the designation district hospital. Working in a place that actually looked to them like a real hospital stimulated their interest in investing th eir professional skills into their institution. Thus, not only did the facilitys accomplishments in expanding infrastructure at the hospital increase its capacity, it also incited a variety of professional and institutional desires in the staff working th ere. Hospital workers went from being demoralized during the 1990s to being highly invested in their institution, equipped with hopes for its future, and for their own professional place within it. This link between infrastructure and professionalism was i mportant; believing in their own professional and institutional possibilities made the staff more committed to their work, even though these changes brought them few benefits in terms of compensation (see Chapter Seven; see also Andersen 2004). Drawing on the memories of hospital workers, I outline some of the concerns hospital workers remembered having when HSR was first implemented, and how these concerns were assuaged once they saw the infrastructural capabilities those policies allowed. I then turn to how Kiunga District Hospital solicited relationships with a multitude of non Tanzanian actors and orga nizations, many with wider transnational ties (cf. Gould 2005). I show that under HSR, the

PAGE 181

181 ability to achieve institutional possibilities as a government health facility relies on two major factors: a wider governing system that (paradoxically) advocates entrepreneurialism within state facilities, and just as importantly, on those individual facilities being strategically located in places where there are a plethora of opportunities to network with nonstate actors in their search for business partnerships. From a long history of scarcity and stagnation borne out of a reliance on an under resourced government and the financial limitations of the communitie s it served, Kiunga District Hospital began to increasingly look outward towards expatriate actors both individuals and organizations to help the facility achieve unprecedented growth. From there, I move into a discussion of how hospital workers understood HSR, and how they felt about the MoH and their institution in the wake of significant transformations in their workplace. I tie these sentiments to how these infrastructural changes appealed to their sense of professionalism. Building the Hospital under Health Sector Reform: 1999 2008 Arusha can be characterized as an attractive location in Tanzania, due to a variety of factors. It is close to the Kenyan border, and was the site of several administrative bureaucracies, from being the hub of the East Afr ican Federation, to becoming the main site of the United Nations International Criminal Tribunal for Rwanda (ICTR) in 1995. The region is also home to the Mbuguni Mererani mines, where several semi precious stones are extracted. Mererani boasts the only known source of Tanzanite in the world, and is considered a possible employment opportunity for those whose business ventures fail in Arusha town. Meanwhile, the development of the tourist industry, and the increased presence of expatriate communities with t he establishment of the ICTR have increased formal and informal employment opportunities in the city, from the tailoring industry to petty trade in imported goods (Weiss 2009). Meanwhile, as described in Chapter Three during the 1990s, Arusha and its surr ounding areas were subje ct to a major proliferation of nongovernmental organizations (N GOs ) and other large (largely

PAGE 182

182 expatriate) organizations (Kelsall and Mercer 2003). During HSR, a facilitys location near a variety of expatriate companies and organiza tions opened up possibilities to network in potentially lucrative ways. In addition, changes in the transnational aid domain in Tanzania (see Chapter Four) enticed more bi and multi lateral donors into the sector, each with their own development interests Through PPPs, Kiunga was able to establish relationships with businesses and organizations not tied into the transnational aid domain, and through these relationships, build up the hospitals infrastructure. Yet the proliferation of NGOs in the area also meant that donors attached to large funding streams came into the area with very particular health interventions in mind; these donors also contributed to transforming Kiungas landscape, but in a more targeted way. In this section, I reveal how Kiunga Di strict Hospital managed to build its infrastructure in a matter of a few short years, and how forging relationships with businesses and private donors offered a means of increasing the facilitys capacity I then move to the infrastructural transformations made possible at Kiunga due to its collaborations with donor NGOs receiving funding from bi lateral foreign aid agencies with particular health intervention priorities. Relationships with each of these kinds of donors were never fixed or permanent; the hospital administration never knew if a particular donor would continue to assist them, or when a donor sponsored program would end. However, they nurtured good relations with each of these funding sources when possible, in hopes of continued partnership. I nstitutional Reconfigurations From when reforms were first implemented at Kiunga up to 2008, the facility doubled in size, having constructed a maternity ward, x ray building and minor surgical theatre, laboratory, male ward, and pediatric ward. Existing buildings were also renovated, and were given new functions. Thus the old maternity ward became the reproductive and child health (RCH) clinic.

PAGE 183

183 The old male ward became a district storage facility and the beginnings of an HIV/AIDS clinic, which was extended further in 2007. The outpatient and administrative building was extended to include a new and larger filing area, a bigger laboratory, and the family planning theatre constructed in the late 1990s with USAID funds was converted into a major surgical the atre.64 M any existing buildings were expanded rather than erected from scratch, and thus some departments became convoluted mazes of small rooms, the functions of which may or may not have any connection with those adjacent Several buildings looked rectangular or square from the outside, but were a patchwork of walls, hallways and small rooms within. The uses of buildings or rooms wer e converted as need arose and funds became available. Many of the staff reflecting back on the scope of changes they witness ed over the past decade stated that they would never have thought such growth was even possible prior to HSR; by 2008, building and expanding the hospital appeared to be a never ended project, and members of the staff often spoke about various infrastructu ral developments they hoped to see in the future. However, despite doubl ing in size in recent years, most of the places that were constructed within the hospital were constantly overcrowded with people, furniture, papers, and medical implements, making wo rk within those places difficult. Areas such as the surgical theatre, the patient wards, and the outpatient clinic were relatively new or recently renovated; yet 64 The terminology of major and minor surgical theatre were so termed not only due to size, but also function. The major theatre was actually quite small and, due to the equipment contained within, somewhat cramped. The major theatre also had a room that acted both as the doctors office for writing surgical notes, as well as a changing area and surgical preparation room. The entrance to the theatre was actually too small for a transfer cot to fit through the door so patients often had to get up and walk into the theatre themselves, or be lifted (with considerable difficulty) from the transfer cot at the doorway through to a cot inside the preparation room a task that required several people. The major theatre wa s where most of the more invasive internal surgeries took place: caesarian sections, lapar otomies, appendectomies, removals of various growths, etcetera. Conversely, the minor theatre was where less invasive procedures took place, and most of these were no t surgical. The room was used for setting bones, changing casts, cleaning and stitching wounds, and would act as a triage area in cases of emergency cases such as vehicular accidents or cases of domestic or mob violence when casualties would be brought in to be assessed, and in many cases, sewn up. The minor theatre was arguably one of the busiest areas of the hospital next to the outpatient department, and the three staff members working there would often treat three or even four patients simultaneously th ere.

PAGE 184

184 they were still too small to ease the congestion of people and objects with which hospital sta ff had been struggling for years. The new infrastructure was not taken for granted, nor predictable, and the amount that these transformations eased the burdens of work varied (cf. Starr 1999). However, it was clear from the workers reflections that these changes in the infrastructural aesthetic had profound impacts on how they felt about the facility in present, and its potential for the future (cf. Masquelier 2001). I now turn to how such changes were made possible and through enlisting what kind of assi stancein a few short years. Speakable and unspeakable concerns in the wake of Health Sector Reform When interviewed about their memories of implementing HSR, staff rarely expressed feeling hesitant or apprehensive, despite the major changes that the reforms were set to bring about. This is likely due to the sheer enormity of changes that occurred in the years since HSR in particular, the increased size and capacity of the hospital and the more regular supply chains made possible during the reforms, which changed how many of the employees felt about the possibilities of their work Van der Geest argues that the technology, machines, and expertise associated with biomedical practice conjures not only hope and faith in patients, but also in physicians, and indeed they are critical to the magic and religion underlying biomedicine important to the spectacle of medical care (2005). In a place such as Kiunga in the 1990s, devoid of many of the buildings and technologies necessary to achieving healing, there wa s a contradiction between ones biomedical expertise and ones ability to practice ones craft (see also Andersen 2004; Wendland 2010). With the enormous changes in the hospital landscape, and thus biomedical practice, that occurred since HSR, contradictio ns between professional ideals and the realities of biomedical practice were easier to manage. To the hospital workers, the facility now looked like a real hospital, and apprehensions in retrospect were difficult to

PAGE 185

185 express. They were excited by what had been achieved, the directions in which their institution was headed, and the possibilities of future biomedical practice within that changing landscape One of the few hesitations I encountered was regarding introducing user fees for hospit a l services. Under HSR, one of the major changes felt at the level of the district hospital was the introduction of cost sharing, wherein patients who previously received free services would now have to pay a fee. While this initiative was only implemented at the dist rict hospital level between 1999 and 2000, the idea was not new and was a subject of heated debate for many decades in Tanzania. Even the Titmuss Report, published in 1964, stated that the progress of the [health] services has again and again been limited by the hard realities of finance (Titmuss 1964:67, quoted in Gottlieb 1975:2). At the district hospital level, the user fee to open a file was set at 2,000 Tsh, and 1,000 Tsh for subsequent visits to the hospital if a file already existed Further, patie nts began paying small fees for laboratory tests 500 Tsh for a malaria test, a urine test, a fecal test, and the like. Yet several of the staff members at the hospital were skeptical about introducing cost sharing. According to Dr. Amani, who was a clinical officer ( CO) at Kiunga when HSR began: Dr. Amani: I had hesitations about cost sharing because I was not sure patients would be able to find the money. People could not find the money to buy supplies [in the 1990s] so they could not even help [with healt h care provision]. Patients had no food, no drugs. I saw that it would be hard. But once the reforms started I saw it helped because it was not so strict, so if a person came without money, she was still treated and given a file, and then would look for the money. (Interview, 112408, Kiswahili) Nurse Frida, who left during 1998 to complete additional administration training returned to Kiunga District Hospital late in 1999 when cost sharing began at the hospital: Frida: When I returned, the system of cos t sharing had already begun. So when cost sharing began, we started to finally see changes. That okay at least we can collect money because by that time it was not that services were free. So there had not been a[n official] way by which we could charge patients, like contributions and such, there was nothing. But after starting cost sharing citizens started

PAGE 186

186 contributing. So that really helped a lot, that now we can have that cost sharing money and use it to support the mandate of the hospital. So the hos pital began to plan to use it. So it was easy, that okay what does everyone needs as a hospital. It was not that it [they money] was just thrown in a pool [for the staff], no! It [the money] was contributed here and it stayed here. It is this way up to today. It is contributed here and it stays here. So it has helped a lot, it has brought changes. That at least if you need paper, there are medicines that are not here, now the hospital is able to consider buying medicine, buying stationary buying gas. Of co urse, it does not pay for [staff] allowances nor build [buildings], but if a lock breaks or those little renovation issues, it can do it. So there were some changes, that at least now we could get medicine, we could get work supplies, we could do some smal l renovations to buildings and such, so it started to help. [Interview 071608] Once reforms began, the hospital used income from newlyintroduced patient fees to do minor and muchneeded upgrades around the entire compound, and to buy supplies that the government failed to provide. According to one nurse: Rehema: They started taking contributions [from patients], eeh, because at that time you came to the hospital and you lacked services. Except then the patients and relatives complained because they said we had our nice hospital, that is, for example, the hospital had doctors and nurses but then, they didnt have medicines! And this was the reason behind the governments decision to have [patient] contributions. If you have contributions you are able to buy, medicines are brought! NS: When you knew that people were to start contributing, do you remember what you thought about the contributions? Were you worried or did you think it was good? Rehema: Aah, to contribute, we saw it was good because from the beg inning the contributions helped because you would get a patient having this or that, before you had to send him to the shops ( maduka) to buy you gloves he brought stitching thread instead! Because now we have that money that the patients contribute, its no longer like that [Interview 11 2708, Kiswahili]. In fact, once the staff realized the kinds of improvements possible through cost sharing, they worked hard to ensure that all patients paid their fees. While I never observed any patient in serious or cr itical condition denied services if they did not have the fee available immediately, there were occasions (although the practice was not ubiquitous) when patients admitted to the wards might have an extended stay with minimal care for want of the fees (see Chapters Eight

PAGE 187

187 and Eleven ). This is a telling example of how small improvements in the working environment tapped into sentiments of medical professionalism on the part of both patients and hospital staff. As Rehema articulates, a nice hospital without medicines is not a meaningful place for healingfor patients or health professionals (cf. Masquelier 2001). Once cost sharing brought about the possibility for the hospital to at least maintain the infrastructure, technologies and medicines required for bi omedical practice, staff participated wholeheartedly in perpetuating the cost sharing funding mechanism so that goals for the facility continued to remain possible to achieve. Public private partnerships : Reaching out to build up There is a long history i n Tanzania of communities building their own dispensaries, many of which, once constructed, remained unused due to the lack of staff available to work within them. Yet as described in Chapter Three during the 1980s and 1990s, the central government and communities within a district were the main sources of funds for improving or building infrastructure. When it came to procuring sufficient funds to erect a large hospital building, collecting sufficient resources from among the politicians and populace of t he district could be a long and arduous process. In the early 2000s, with the decentralization of the health sector underway, government facilities were informed that they should attempt to seek out mutually beneficial relationships with private organizations and businesses in order to develop their own capacity rather than rely on the central government. This was part of the overall drive to encourage individual government health facilities to foster PPPs with organizations that could potentially help pro vide resources and funds. However, when I asked about the specifics of how the hospital sought out partnerships with nearby businesses and organizations, I learned that these collaborations largely occurred by happenstance. It was never possible to predict where potential partners funding

PAGE 188

188 interests would lie. It was also never assured that potential partners would come to visit the hospital, and without personal contacts within outside organizations and businesses, few collaborative opportunities with donors were established. This section outlines the ways that the hospital administration brought businesses and private organizations into partnership in order to build up the hospital. The DMO in the early 2000s Dr. Kalete, encouraged her staff to reach out to companies and visitors in the area and bring them to the hospital, in hopes of securing finances to improve the hospitals infrastructure. T his view to networking with private organizations and individuals was a major shift in focus. Due to its locatio n close to a main highway and nearby a plethora of tourist lodges, NGOs, and foreignowned plantations, Kiunga District Hospital was particularly well located to take advantage of the governments PPP initiative by seeking out donors. A nurse administrator described Kiungas advantageous locale thus: Frida: It wasnt that visitors started to come here because of the [policy] changes and whatnot, but you know were on the main road and of course Arusha is a tourist region so many were coming because of tour ism and they also wanted to see a hospital. And the hospital that was close to the road [they were traveling upon] was ours. So many who were coming, maybe they were visiting the wilderness or visiting the various institutions when they arrived to greet th e patients or to see the hospital; they came here because its on the main road and en route. So that was the reason, [the road] brought them, so they were coming because of that reason and when they came to visit the patients, basically they come, they vi sit the hospital, and they see that my goodness, you really need this and this and this. Yeah! We need this and this and this! (Interview, 071808, Kiswahili). Other districts in the country not located close to a variety of potential financiers or partn ers were not so fortunate under the new PPP plan. Frida compared her experiences working in the district of Lindi in the 1990s, which was not close to as many potential donors as was Kiunga: NS: Were the problems you experienced here [at Kiunga] the same as the problems you experienced in Lindi?

PAGE 189

189 Frida: They were the same because the whole country was like that, it was the same Except maybe here [in Kiunga], there had been the ability to get donors friends and such to donate little things. [Interview, 07 1608] Once forming partnerships and soliciting assistance from private companies and organizations became officially sanctioned under HSR, administrators at Kiunga began searching for opportunities to form relationships with potential sponsors as a means to build up the facility: Frida: at that time, something that maybe the DMO and DNO65 could do to help, that is, more than doing supervision and encouraging [staff], at least they could find donors. That is, they could go after various people, to various area s [to get people to] help the hospital. People could donate something that they wanted to donate and such. And remember, a hospital like this [that you see here today] had been planned to be a full hospital but it was without even buildings so you can imag ine to work within an environment that had no buildings or s upplies. Thus, a hospital without infrastructure was not a meaningful workplace that might appeal to workers professionalism and expertise. Under the PPP initiative, staff looked to the hospital administration to make linkages with interested donors, and in doing so, they might create a working environment that approached the idealized district hospital that up to that point was beyond their ability to achieve. Expectations were high, but in the l ate 1990s and early 2000s, little progress was achieved. The administration had to create the economy of appearances that would make Kiunga look like a place of institutional possibility in order to get the staff to invest their expertise and professionali sm into the facility. However, the pressure to find donors was burdensome. A ccording to Dr. Kalete, prior to HSR, locating donors was not so easy. Few donors were interested in contributing funds for constructing buildings before 2000. Dr. Kalete explained we worked with the World Vision as wellbut they were not interested you know [in] any construction or what; they were only interested on HIV and AIDS, that is it! (Interview, 65 District Nursing Officer

PAGE 190

190 61108, English).66 During the 1990s, donors myopic interests proved a major hurdle in building up the hospital. In the mid 1990s, Dr. Kalete left the hospital to complete a degree in public health. When she returned to Kiunga District Hospital, she spear headed the initiative to complete the construction of the xray building and minor theatre (see Chapter Three). S he and her fellow administrators also created a plan for construction and expansion of several other buildings for the facility. According to Dr. Joseph, who was present at Kiunga from the 1980s through to 2008: Dr. Jos eph: First, they wanted to expand the wards to reduce the overcrowding so that there were not three people in the beds. And they wanted to build a[n operating] theatre so that people were given knowledge also. And they brought in four more workers to incr ease capacity. It promoted th e [Kiunga] area to be lifted up. And the big road here, because of all the accidents and whatnot, to refer [to the regional hospital] was a problem. So we contributed so that people did not have to leave this area to get their treatments. So sending people to Mount Meru [Hospital] was seen as an annoyance. (Interview, 111308, Kiswahili). Erecting a surgical theater would entice the MoH to bring medical professionals with surgical training to the hospital, thereby reduci ng the need for Kiunga to transfer its surgical patients to the regional hospital. In bringing people with surgical expertise into the hospital, the facility could also expand the expertise of existing staff and indeed, that is exactly what occurred. Once the surgical theatres were complete, the MoH sent assistant medical officers (AMOs) with surgical training to the facility, who then taught the existing AMOs and several COs how to perform basic surgeries, thus increasing institutional and professional cap acity within the facility. Infrastructure and professionalism were intimately intertwined. 66 There is an irony here, since, as will be seen in the discussions of subsequent chapters, most of the bi laterally funded transnational NGOs running programs at Kiunga continue to have very targeted interests. The differe nce at present is that there are larger, more well funded programs operating within government facilities, as well as more organizations running programs through those government hospitals and clinics. The main difference at present, I would argue, is that the presence of more programs and donors increases the possibilities of a hospital finding ways to negotiate its own institutional needs into the targeted program agendas of the various NGOs.

PAGE 191

191 In 2000, for the first time, the hospital administration began locating donors whose vision of assistance was not so myopic as those of World Vision or the few mult i national NGO donors working in the area during the 1990s. These donors were private businesses or charitable organizations, but without ties to larger transnational NGOs or bi lateral donors funding targeted programs. Several companies in the area were encouraged to donate funds and resources to government facilities, and BotanaKiunga decided that they would provide some of the funds necessary to construct a new maternity ward, as well as part of the future male ward for the hospital. At the time, the Dis trict Commissioner of Kiunga was what Dr. Kalete referred to as a good sensitizer, and through his efforts the hospital established a relationship with a large flower export company, called BotanaKiunga.67 According to Dr. Kalete, the counselors for each ward68 in the district went to their communities and mobilized people to contribute what little they could to build the maternity ward; the hospital managed to get contributions from all of the nearly forty wards in the district through the efforts of the D istrict C ommissioner the counselors, and the Chairperson of the District Council (DC) When I remarked on the mobilizing abilities of Dr. Kalete and the politicians in the area, she responded, you see, when you tell somebody and they understand, they contribute! W e were joining [the politicians] in mobilizing people to give money. With the money collected from the district as well a s funds from BotanaKiunga, the new maternity ward was constructed beside the existing ward (Figure 51). It opened its doors in 2002 (Figure 5 2) 67 A pseudonym 68 In Tanzania, the administrative organization is organized into a twotier administrative system comprising the Central and the Local Government. The most basic governance unit is the village. In terms of local administration, villages are grouped into units called wards, and wards grouped into units called divisions. Numerous divisions make up a district, and numerous districts make up a region. There are political administrative structures at each level to aid in the governance of those areas.

PAGE 192

192 I n 2002, BotanaKiunga offered to donate bed nets to the hospital, and the matron at the time, Sister Gertrude, refused their offer. She told them that the hospital was able to procure nets from the government, and that what the facility needed was an additional ward BotanaKiunga agreed to fund a large proportion of the construction of the male ward, and also provided the beds needed to fill it. However, the company did not provide sufficient funds to finish the construction, a nd it remained stalled for a few years. In 2000, Sister Gertrude encountered British missionaries through her church and brought them to the hospital. Later, i n 2004, Sister Gertrude was invited to England to visit the missionaries. While there, she talked to them about the possibility of sponsoring the completion of the male ward, the need for a new pediatric ward and a washing machine since all laundry was washed by hand. The missionaries agreed to help. Yet in mid 2004, before the funds were received f or building the new ward, Dr. Kalete was transferred, and the new DMO, Dr. Saidi, was assigned to Kiunga. At the end of 2004, when the male ward opened, he was credited with the accomplishment.69 Under the leadership of Dr. Saidi, infrastructural changes co ntinued at an increased pace. While he maintained partnerships that Dr. Kalete established, Dr. Saidi also convinced the donors already financing small programs at Kiunga, as well as several new donors, to continue to 69 In conversations with me, as well as in encounters with other visitors to the hospital, I often heard Dr. Saidi point to the male ward, the pediatric ward and several other newly constructed buildings at Kiunga and claim that these buildings had only been possible due to his efforts in establishing relatio nships with donors. Meanwhile, the majority of the staff at Kiunga in 2008 had only work ed there for a few years. They also pointed to Dr. Saidi as the progenitor of the major infrastructural transformations at Kiunga. While it is undeniable that the major ity of changes took place at Kiunga during his administration, nearly all of the staff members who were present under the previous administration attributed these transformations to Dr. Kalete. Sister Gertrudes comments are illustrative of many of the sta tements I heard, both in interviews and in casual conversations with long term staff: The changes have been happening very fast over the last ten years. These changes have been planned for a long time, but Dr. Saidi walked into them. These were not his ideas (Interview, 08 17 08, Kiswahili). This note is not to reduce the importance of Dr. Saidis accomplishments, since he was instrumental in establishing many relationships after 2004 that expanded infrastructure and capacity. However, the long term staff felt it important to make a distinction between which transformations he initiated, and those that were planned prior to his administration.

PAGE 193

193 the hospital. Due to the efforts of Dr Mtenga, the Medical Officer In Charge (MOI) at the time, f rom 2006 on, BotanaKiunga provid ed landscaping plants and services to the hospital S he asked BotanaKiunga managers if the company would be interested in donating plants to improve the hospital lan dscaping: When I arrived there I asked them for flowers. I asked for the flowers and I brought many flowers [to the hospital]. Then I planned, I put all of the flowers along the road as you now see them (Interview, 060408, Kiswahili). Indeed, as seen i n Figures 510 and 511, the landscaping was developed throughout the compound so that by 2008, nearly every building and walkway or roadway was lined with shrubs, small trees, or flowers. The aesthetic changes added by the landscaping were a marked contra st to the environm ent I encountered there in 2004. During exit interviews, patients often remarked on the aesthetics of the grounds. A woman discharged from the female ward remarked, since 1992 services have changed so much. Now there are really nice buildings and big improvements. The government is doing its work and doing it very well (Kiswahili, 10 2908). Another woman who had been attending Kiunga District Hospital since the 1980s said there have been big changes and it has been really good. I am happy with the way the hospital has changed. It has a very nice environment now and I will continue to come back (Kiswahili, 10 2908). In 2006, the hospital administration budget ed for the construction of a pediatric ward, and again turned to the English missionary organization and BotanaKiunga to contribute funds to support their efforts. The administration also collected donations from the various wards in the district. C onstruction of the building began in mid2006 (Figures 5 4, 5 5) and it was c ompleted in mid 2007 (Figures 5 7, 58). As seen in Figure 56, a member of the British missionary organization that helped to fund it came to visit and tour the new edifice

PAGE 194

194 Not all of the infrastructural improvements to the hospital consisted of erectin g new buildings. Working under Dr. Saidis leadership, Dr. Mtenga actively wrote letters to the District Council requesting permission for Kiunga Hospital to utilize cost sharing funds in order to renovate some of the existing buildings. The administration asked to renovate the existing dental unit, pharmacy, RCH room, family planning room, and Volunteer Counseling and Testing (VCT) room. The District Council (DC) approved the use of the funds for these purposes, and as a result, several existing buildings were renovated and expanded through 2007 (Figure 5 9) The hospital also managed to procure local government funds to bring critical resources into the hospital. Thus, the DC was also a potential partner; however, while these infrastructural investments cannot be underestimated, they were generally related to infrastructure other than buildings. Of DC investments in the hospital, the one nurses and medical attendants most remarked upon was water. Pr ior to 2005, there was no plumbing infrastructure at the hospital. Challenges relating to the need for water remained t he minds of many of nurses Nurse Officer Monica described the changes thus: NS: When was the water supply improved? Monica: After when was it?two thousand and five! NS: Two thousand and five Monica: Something like that. Otherwise we were using water from the river. We would go to draw water ( kuchota maji ) in the river for cleaning. NS: So you had to walk all the way to the river? Monica: All the way!. .For drinking and whatnot, we woul d have a vehicle to take us to get water up there [to a location approximately five kilometers away] with a Simtank. The trucks went farther up the road there because water was free because there was a lot of it. You went, we put Simtanks in each truck, we collected the Simtanks and carried water, [people] went and brought back water. Now, to take it from the vehicle and bring it to the ward, if that water ran out it just ran out, just like that. But the water for cleaning, we went [on foot] to draw it from the river. Later, yes, it was later that they brought a type of machine to pump water

PAGE 195

195 from the river to bring it here. Later, some kind of machine, and they dug this well, the well you see over there. They said maybe we should make a well and whatnot, so those were the procedures in the past that were here, but the tanks have already been put together, so at least we do n ot have problems again. (I nterview, 111808, Kiswahili. Figure 5 5). The improvement in water supplies was incredibly significant to nurses in particular, as it was they who had to carry the water from the river to the hospital. The addition of plumbing infrastructure allowed them to concentrate more on patient care and other duties within the facility, giving them more time to perform the duties characteristic of their profession.70 Overall, relationships with BotanaKiunga and the UK missionary organization were largely established through staff efforts to find donors. However, this process often happened through chance. When I asked severa l hospital administrators what they were actually doing to attract donors, I learned that while they had prepared a variety of proposals to distribute to potential visitors to the hospital, they were not actively reaching out to many businesses or organiza tions in the area unless one among the staff already had an established contact with that potential partner. Under the PPP plan, businesses could curry favor with the local government by donating funds for service activities or development initiatives. Thi s was the reason that BotanaKiunga first approached the hospital administration about donating bednets. It was only because Sister Gertrude suggested alternative uses of that donor money, and pooled it with other funding sources, that the hospital managed to erect a new ward. Similarly, the relationship with the UK missionary group was established through Sister Gertrudes church. Her ability to 70 The only other DC funded infrastructural development I learned about during my research in 2008 was the upgradi ng of the existing road leading from the main highway up to the hospital gate and into its interior grounds. The road was clearly meant to improve access to the hospital by car from the highway. The road leading from the highway to the hospital gate contin ued beyond the gate up a hill several kilometers into villages that the hospital served. This road remained in a very bad state, often impassible during rainy seasons even by land rover. When the DC funded improvement of the road, the improvements stopped promptly at the gate of the hospital. Those people to the north continued to grapple with difficulties if they were ill and required transport by car, bike, or even foot, to the hospital below.

PAGE 196

196 suggest potential means that these organizations could contribute meaningfully to the hospitals development was the main reason that these partnerships were achieved. They were also temporary. By 2008, the UK missionary organization continued to have ties to Sister Gertrude through her church, but was no longer involved with the hospital itself BotanaKiunga continu ed to provide landscaping plants to the site in 2008, but the business no longer offered to provide funds for additional buildings. PPPs had a major role in assisting the hospital to expand its infrastructure, but these linkages were often temporary. Meanw hile, the DC might procure funds for contributing to certain aspects of hospital infrastructure, but rarely were buildings included in these initiatives. Donors and the Infrastructure of Targeted Interventions Another means by which the hospital was able to expand was through contact with multi national NGOs and bi lateral donors who operated targeted health programs within the facility. On a few occasions, the hospital administration was able to advocate for expanding existing buildings by appealing to the need for additional infrastructure to be able to adequately administer donor sponsored programs. These initiatives established an economy of appearances directed specifically at the donors objective, arguing that their programs were compromised due to t he inadequacies of available space, technologies, or resources required for operating donor programs. The hospital administrations message to these donor partners was clear: Kiunga District Hospital could be an ideal locale for donor investment, able to c ontribute to the success of sponsored programs and projects if only the needed infrastructure was in place. Once the new maternity ward was completed in 2002, the administration drafted plans to convert t he previous maternity ward into a Maternal Child He alth (MCH) clinic. Dr. Kalete

PAGE 197

197 contacted an acquaintance working within a transnational NGO named SIGNAL.71 Dr. Kalete and her acquaintance had gone to medical school together back in the 1980s The American based NGO SIGNAL worked on reproductive health iss ues in aid dependent countries throughout the world, including in Tanzania. Dr. Kaletes acquaintance managed to convince a SIGNAL representative to visit Kiunga District Hospital, where the hospital administration proposed that the NGO donate funds to expand the MCH clinic. After the visit, representatives from SIGNAL agreed to fund the extension, and two additional rooms were added to the MCH (Figure 5 3) By the mid2000s, the acronym MCH fell out of favor because it was seen to exclude men, who were now seen as an important part of reproductive health. Therefore, the acronym changed from MCH to RCH short for Reproductive and Child Health. SIGNAL provided funds to renovate the existing MCH clinic now called the RCH in 2006. With these funds, the hospital created a large waiting bay at the door of the existing RCH clinic, as well as a large waiting bay behind the RCH for the childrens clinic and family planning unit, allowing services for pregnant women, family planning, and children to be separated i nto designated areas. The waiting bay for the maternity clinic also was thereafter used regularly for morning staff meetings. However, expanding the existing building was mutually beneficial; the larger RCH would allow the hospital to better provide the se rvices funded through SIGNALs reproductive initiatives such as family planning and prenatal care. The NGOs participation in infrastructural expansion was programmatically tied in a way not characteristic of BotanaKiunga or the UK missionary organization funding other buildings. 71 A pseudonym.

PAGE 198

198 Similar hospital expansions were made possible due to the hospitals provision of HIV/AIDS related services. As HIV/AIDS services were tied to targeted health interventions sponsored by multi national NGOs, the hospital administra tion was able to initiate infrastructural expansion by appealing to particular goals of the donors sponsoring HIV/AIDS programs at the facility. In late 2002, a small building for volunteer counseling and testing services (VCT) for HIV was erected at Kiung a District Hospital. The structure comprised two small rooms, and had limited capacity to provide services beyond testing. In 2004, the MoH appointed Kiunga as one of its rollout sites for Antiretroviral therapy (ART) provision. The services were dubbed CTC Counseling, Testing, and Care to indicate the services provided there, while concealing their relationship to HIV/AIDS from the general public. At the time, funds for the CTC program were channeled through a web of donor mechanisms. USAID and the Global Fund financed an American transnational NGO dedicated to HIV/AIDS related issues. I call this NGO REFLECT. REFLECT distributed these funds to a variety of (American transnational) NGOs to coordinate the actual CTC activities in the districts where th ey operated In the case of Kiunga district, REFLECT72 commissioned SIGNAL to operate the HIV/AIDS program. SIGNAL was a convenient choice for expanding the HIV/AIDS program at K iunga District Hospital because, in addition to funding reproductive healthre lated activities at the hospital, the NGO sponsored Prevention of Mother to Child Transmission (PMTCT) activities. The short staffed hospital was to supply the necessary workers for the CTC clinic, and SIGNAL was to fund, coordinate, and train the staff wi thin the clinic (see also Whyte et al. 2010) When CTC services began in 2004, the clinic was staffed with one clinician and two nurses, opened once a week, and serviced 10 patients. By June 2006, the hospital serviced over 500 clients. 72 REFLECT is a pseudonym for an actual NGOs

PAGE 199

199 Some of these clien ts receiv ed ARTs, while other HIV positive patients in the early stages of the disease receiv ed pharmaceuticals to treat HIV related illnesses such as infections and tuberculosis ( TB ) In 2006, the United States Presidents Emergency Plan for AIDS Relief ( PEPFAR ) awarded a grant to REFLECT to scale up ART provision. Kiunga District Hospital s administration wrote a proposal to REFLECT to offer the facility as a site of ART scaleup. The proposal suggested that REFLECT finance an extension of the existing C TC building, and provide funds for staff training in data collection and input, filing, and computer familiarization. It also requested additional personnel for the clinic: a minimum of three additional clinicians and four nurses, in order to allow them to open the clinic three times per week. The plan included extending this training to other staff peripherally connected to the CTC clinic, such as those in the wards, and medical personnel within the smaller government health facilities in the district. The proposal stated: Currently, we have 3 rooms with a small corridor: we need to have more rooms to expand our services. Also, we want to improve our dat[a] system by having a compute[r], data clerk, printer, internet and fax machine. Kiunga District Hospi tals proposal was selected, suddenly introducing a barrage of contacts, bureaucratic paperwork, and resources to the site. The facility thus became one of the nearly forty REFLECT funded sites in Tanzania. REFLECT took over CTC activities at Kiunga District Hospital, while SIGNAL continued its work there on reproductive health issues. REFLECT agreed to pay for renovations to the existing CTC clinic, laboratory equipment (including a machine to count CD4 levels), and other supporting equipment. By October 2007, transformations to Kiungas CTC clinic were completed. The changes were astonishing. A large undercover waiting bay was attached to the existing CTC clinic, with two small rooms to the

PAGE 200

200 side: one for housing patient files and the other for a data entr y clerk, to enter information on each patient consultation into a sophisticated PEPFAR database. Another addition functioned as the main clinic. It had a long corridor, flanked by a series of rooms. The first room was reserved for adherence counseling, w here patients who tested positive for HIV would be educated about the disease and the importance of adhering to a strict regimen of drugs. The second room housed a small pharmacy. The third room was a small CTC dedicated laboratory. At the end of the long corridor were two patient consultation rooms. One of the existing clinic rooms was converted into a filing and staff room, and the other reserved for voluntary testing and counseling. Once expanded, the building was filled with objects marking it as a ver y distinctive kind of space for biomedical care. While the rest of the hospital grappled with inadequate or absent equipment, the HIV/AIDS clinic was a place of advanced technologyintensive biomedical intervention, complete with highly systematized model s of patterns of practice to which both patients and personnel were to adhere. As shown in Figure 5 19, t he laboratory was filled with sophisticated technology, all carefully labeled with the name of the donor(s) whose money paid for them. In an interview, Peter, a REFLECT program officer, remarked that Kiungas CTC clinic was one of the ni cest he had seen in the country.73 When I arrived in the beginning of 2008, the expanded HIV/AIDS clinic was the newest building at the hospital. Thus, by appealing to th e goals of NGOs and bi lateral donor organizations, the hospital was able to make itself appear as an attractive place for investing. The economy of appearances established to appeal to donors operating sponsored programs at the site stood in contrast to t he happenstance ways by which the hospital attracted its other partners through PPPs partners through PPPs invested in the hospital directly, but without programmatic attachments. 73 Interview, July 2, 2008, recorded with notes.

PAGE 201

201 Conversely, for donors sponsoring programs within the hospital gates, the only infrastructural investments deemed worthy of funding were those that would improve the immediate success of their targeted interventions. Bilateral donors and large NGOs operating health interventions within government facilities expressed little interest in building up the overall facilities upon which they depended; instead, they concentrated their investments in very specific areas of the hospital. Yet ultimately, for the hospital workers at Kiunga, these two kinds of donors (happenstance partners an d donors sponsoring targeted interventions) had something in common: no one was really certain how long the donors interests would remain at the hospital, nor whether they would be interested in additional infrastructural investments in the future. All th ey could do was continue to propose, in hopes that the donors would see the benefits in investing in their facility. Donors by Happenstance: Building a Laundry Facility While my research of Kiunga District Hospitals recent infrastructural history reveale d the above two kinds of donors (happenstance and donors sponsoring programs), in 2008 I was able to observe and also participate in forging a connection between a potential happenstance donor and the facility. This section points to the complexities and i nstabilities of hoping for, and working with, donors. Just as infrastructure was never taken for granted (cf. Starr 1999), the fact that donors would come, and would provide the entire amount of funds required, was never assumed. Hospital staff and adminis trators hoped that donors would see the benefit of investing in their facility, and as the case below illustrates, that they would provide all of the financing that the facility required. However, hope could turn to considerable anxiety and potential failu re if donors were not forthcoming. And although Kiunga was advantageously located near a main road and in the vicinity of a variety of NGOs and visitors, whether potential partners would enter through the hospital gates was unknown. If they did, however, t he hospital was prepared.

PAGE 202

202 Indeed, early in 2008 the DMOs office put together a proposal for a new mortuary and new surgical theatre to distribute to happenstance guests who might potentially be interested in funding the project(s).74 The desired modern surgical theatre75 would have four operating rooms, two recovery rooms, a toilet, changing rooms, and a doctors room, budgeted at 173,019,000Tsh.76 The proposed modern mortuary would upgrade the 2 cadaver, nonelectrified existing mortuary to construct a new building and include a nine cadaver refrigerator, budgeted at 95,823,800Tsh.77 Of note, these institutional desires were taken up by the majority of the staff, who, during interviews and in casual conversation, marveled at what the hospital would be l ike were a new surgical theatre and mortuary actually built. Indeed, when asked what the staff would like to see built in the future, the most common answer was a new mortuary or a new surgical theatre, and often they answered both, regardless of the rank or expertise of the individual worker asked. For instance, in a conversation with Dr. Ezra, a CO who worked primarily within the CTC clinic but who also assisted with surgeries and outpatient care, I asked what he felt the hospital should build next. He re sponded, a modern surgical theatre would really make this hospital look like a real hospital Can you imagine? Yeah! However, there were some members of the hospital staff who expressed different priorities. When I asked the hospitals Matron, Sister Ami na, what she would like to see built in the future, she immediately responded that the mortuary and a new laundry facility were the highest 74 In fact, the DMO called me to his office to help him make his proposal look more official and therefore potentially more attractive to potential donors. I assisted him in drafting proposals for both projects, but by December 2008, while several of these proposals had been distributed, no visitors expressed interest in beginning to fund either project. 75 The word modern comes from the hospitals own surgical theatre and mortuary proposal, which was handed out to expatriate visitors and organization representatives visiting the hospital. 76 In 2008, this sum was approximately US$147,000. 77 In 2008, this was approximately US$83,336.00, with US$53,000 required for the refrigerator itself.

PAGE 203

203 priorities. Indeed, in pursuit of her institutional desires, Sister Amina was most fond of showing visitors the laun dry building and the machines within it. The hospital went through an astonishing amount of laundry each day, mostly in the form of sheets, and two medical attendants were solely assigned to laundry duty. At the time, the laundry room was attached to a kit chen a throwback to the days when the hospital provided food to patients (Figure 5 12) The laundry room itself was inadequate to the needs of such a large hospital. As shown in Figure 514, i t was about 150 square feet, comprising a series of shelves for clean sheets, and two domestic washing machinesone that did not function consistently and had been repaired several times. Behind the laundry building was a n outdoor area with drying lines for the sheets, blankets, and bednets of the wards (Figure 513). I n rainy season sheets and blankets rarely dried quickly, and often were either damp or in short supply due to the consistent rain. Because one of the machines was broken, the area in front of the laundry building was often littered with plastic washing ba sins, containers of powdered soap, and mounds of sheets stained with bodily fluids of the patients who used them. Staff washed a large proportion of sheets by hand due to the inoperable domestic washing machinea dangerous proposition given the substances on the sheets. Sister Amina was extremely concerned about the safety and comfort of the staff washing sheets; she always spent extra time at the laundry facility when giving tours to visitors, emphasizing the dire need for upgraded laundry machines. In Mar ch, 2008 a group of Canadian university students arrived at Kiunga on tour with their teachersone of whom was a physician. Seeing me working in the RCH clinic with the other nurses, the physician, Dr. Brown ask ed me questions about my activities at the h ospital. When I told him about my research, he invited me to come to speak to his students about health

PAGE 204

204 and healthcare in Tanzania. I readily agreed, and I spoke to the students on a subsequent afternoon. Little did I know how this happenstance encounter w ould affect the hospital. In May 2008, Dr. Saidi announced to the staff that they may have found a private donor to contribute funds to help them replace one of their washing machines; however, the money being offered for the machine was only sufficient t o purchase yet another domestic machine.78 The staff and administrators agreed that instead of purchasing a domestic machine, which would wear out quickly due to constant usage, they should use put the money towards a new industrial sized machine. Such a machine would require an expansion to the existing laundry building, as well as upgrades to water supplies. The hospital needed a donor to help fund the extension. Serendipitously, at the end of May Dr. Brown emailed me regarding whether I knew of any small infrastructural projects in need of assistance, since he recently establish ed a small NGO in Canada interested in funding minor buildings or renovations for worthy causes in Africa. When I received the email, the laundry bui lding immediately came to mind. I responded that the hospital was looking for a donor to contribute to the extension of the laundry building, which would be required in order to house the industrial machine, and provide an undercover drying are a for use during rainy seasons. Dr. Brown a sked me to have the hospital administrators draw up a budget for the building, and I talked to Dr. Saidi and Sister Amina about drafting a proposal. The next morning during the morning staff meeting, Sister Amina announced that they found a donor to assist them in extending the laundry building, which incited a lot of happy comments from the staff. In response, Sister Amina proclaimed, we should all congratulate ourselves and the donors, since it was only because we have worked to tell visitors about our needs that we are getting all of these things donated. We must celebrate our own hard work in bringing these 78 I never l earned who the donors were who were offering to contribute a domestic washing machine; however, as promised, the funds for the machine arrived at the hospital by May 2008.

PAGE 205

205 things to the hospital! Sister Aminas comment acknowledged the amount of effort that the hospital administration and staff were putting into finding donors to support the hospital. Yet I was astonished at the time that she made the announcement prior to Dr. Browns NGO approving their proposal. It was only later when I came to understand the relationships between staff morale and the quality of thei r work that this announcement became comprehensible to me in the absence of the donors confirmation to fund the project. The administration knew that in order to perpetuate the quality of healthcare services at the hospital, they need the staff to collect ively invest in particular institutional desires. When they believed in the possibilities of their facility, they worked with pride despite its hardships. Their sense of professionalism was intimately linked up to the perceived quality of their institutions present, and its future However, despite Sister Aminas proclamation, both the machine and the new building were stalled for quite some time. The administrations confidence that it would find a donor to finish paying for the remaining cost of the industrial machine, as well as the unpredictability of arrival of donor funds caused considerable anxieties among the staff, and the accounting office in particular. Dr. Browns NGO approved the project, but had to fundraise in order to pay for it. Meanwhile, as shown in Figure 515, the industrial laundry machine was purchased on credit, with the first payment installment coming from donors a fact that made the hospital accountant very nervous, since they only paid for a small portion of the amount owed on the machine and the new water tank and pumping system required to make the machine function. The machine arrived in mid July. As seen in Figure 5 16, i t was so large the door to the laundry room had to be expanded and a new gate installed to keep the machine secure when not in use The cost of the machine, its installation, the new water system and pump, and the platform that was constructed for the m achine were extremely expensive. D espite the initial installment from donors, in July,

PAGE 206

206 the hospital still owed the vendor 31 million Tanzanian shillings.79 The subject of how to pay the vendor small installments on the amount owed came up numerous times during monthly Hospital Management Team meetings in the subsequent months, and when I departed in December, the hospital still owed the vendor upwards of 20 million Tanzanian shillings ,80 with no potentially donor on the horizon. By September, the first installment from Dr. Browns NGO arrived and construction began (Figure 5 9) Dr. Saidi wanted to make a lengthy extension to the building, t o create space for undercover drying lines. The DMO asked the contractor to place round hooks in the walls of the building, so that they could attach and remove laundry lines with ease. The ability to remove drying lines would all ow the building to double as a laundry drying area and a training room where the hospital could conduct seminars so that they no longer had to rent a nearby facility for these purposes They could thus save the money for further improvements to the hospita l. Dr. Saidi maximiz ed the possibilities of his newfound partnership with Dr. Browns NGO.81 As seen in Figures 5 17 and 518, when the first installment of funding for the building arrived, it was quickly spent within the span of a couple of weeks, and the second installment did not arrive until January 2011. In the meantime, Dr. Saidi had to manage the staffs impressions of the institutions possibilities, and their own role within it. The stalled funding of the building did little to help Dr. Saidi maintain staff investment in the hospital, and as will be seen in Chapter Seven, problems arose near the end of 2008 that eroded staff morale. Dr. Saidi did not have 79 In 2008, this sum was approximately US$26,315. 80 In 2008, this sum was approximat ely US$17,000 81 According to Frida, while the laundry building was completed in early 2009, the only other major infrastructural development that has occurred between my departure and April 2010 is the construction of a diabetes clinic, which has been cons tructed as an attachment to the outpatient department beside the records department (personal email communication, 4 15 10).

PAGE 207

207 additional donors coming into the hospital at that point, giving the staff little to distract them from their individual professional and personal desires, and the ways that hospital work impinged on those dreams. Reflections on Change: Staff Views of Donors and the Government In light of the magnitude of the transformations in the hospital infrast ructure and capacity at Kiunga under HSR, when staff was a sked to reflect on how they felt about changes in the health sector, the vast majority expressed favorable attitudes about what changes the MoH managed to effect, but felt that more were still needed. To the workers, the impressive transformations in the aesthetics and capacity of their workplace would not have been possible without the approval and encouragement of the MoH. In some sense, through the encouragement of PPPs, the government provided pe rmission for health facilities to look beyond the state for potential resources a dramatic shift from previous policy that asserted the state as the main provider of healthcare. The staff credited the central government for achievements in the health secto r, acknowledged its shortcomings, and praised the MoHs role in coordinating the distribution of donor sponsored health initiatives. F or instance, a general nurse named Anena expressed: Anena: What I know, in truth what I have thought is that they have don e something in the Ministry of Health that is big! That is, we left the place [where we were] and we went, we moved forward very well, but still it is not enough! Mmh, there need to be enough workers, enough facilities They have really tried, truly, on the side of facilities we see that they are trying. (Interview 11 2808, Kiswahili). Similarly, when asked about the importance of the work of the MoH, another nurse stated: Monica: They [the government] are doing big work! Because indeed they are the ones who coordinate right? I know their work is big because the Ministry of Health is indeed our mother and our father. Yes, they coordinate because the donors cannot come here without their permission, right? They are doing the big work also of finding [donors]. That is, they communicate with [the donors] when they [the donors] come here. Its not that they [the donors] come and they just stop

PAGE 208

208 here! It is that they get direction from there [the MoH], isnt it ? [Interview, 111808, Kiswahili]. Overall, staff sentiment towards the donors that funded programs within the hospital at Kiunga was extremely favorable, and there were few critiques. The staff both within and outside the sponsored clinics and programs thought that donors were highly desirable additions to their institution. Donors (whether private, corporate, bi lateral, transnational or NGO) were widely seen as the providers of the infrastructure and development at the hospital over the past decade. While there was some appreciation for the enormous and quick paced steps that the MoH had taken in terms of formulating policies to reform the health sector, it was donors who were seen as the source of what the hospital needed. According to Monica, who had no direct affiliation to the CTC clinic or other donor programs operating at Kiunga: Monica: I see that many [changes] have been pushed by donors. The truth is that we are with the Ministry of Health, and they are giving us services. But if I look at the many changes in development, they have been done by donors! Because if you look at this childrens ward, its donors, right? If I look at th e CTC, its donors, right?. This is why I say NGOs should continue to move forward because truly they are providing big contributions to social development. (I nterview, 111808, Kiswahili) T he views articulated above were the norm among the staff generally, regardless of rank or position. Indeed, of all the informal conversations and formal interviews, I only ever heard one staff member articulate any critique of donors. In an interview, Sister Gertrude called the donors clever [ wajanja]. They dont employ anyone, but use the government workers and steal them from the hospital. Donors set up a project and leave, and they know that if they use government e mployees to operate the project, then the project wont die. (Interview, 9 1708, Kiswahili). The se generally supportive statements regarding donors echo what has been found in other studies in the region on perspectives on donors at the local level. Kels all and Mercer demonstrate

PAGE 209

209 that projects sponsored by World Visionaimed toward empowerment and de velopment of local communities actually r einforced problematic power inequalities in the village. Yet while villagers were critical of those people who were coordinating the donor funds, they were unanimous in their belief that World Vision was a good thing, and a bonus to the community, despite the programs inability to elevate the status of the poor living there (2003:301). Similarly, the resources of the CTC clinic at Kiunga brought a variety of problematic issues to the surface at the hospital, and yet the overarching sentiment was a hope that more such donors would come to the hospital. Conclusion In a country where poor staff morale within health facili ties is often seen as the norm, the infrastructural transformations in Kiunga District Hospital incited many of the staff to invest in the institutional possibilities of their workplace. The increases in technologies (although unevenly distributed. See Cha pter Eight), buildings, and capacity brought about in a short period at Kiunga caused many of the workers to devote themselves more completely to their work of providing patient care than they had prior to HSR (see Chapter Three). By bringing a variety of types of donors to the hospital (private individuals and businesses, transnational NGOs), the administrators effected an economy of appearances that reoriented worker attitudes by appealing to their professional desires for an adequate workplace. Their new ly established working environment made their expertise more meaningful despite continued shortages, the hospital staff had far more of the essential drugs, equipment, and facilities needed in order to use their professional expertise for the good of patie nt care. At Kiunga, infrastructure and professionalism were intimately intertwined. Such rapid transformations in Kiungas infrastructure would not have been possible without the policy changes brought about under HSR, and the emphasis on PPPs in particul ar.

PAGE 210

210 Once PPPs were instituted as policy, health facilities and their administrations were permitted to look beyond the government to procure what they needed in order to provide patient care. Kiunga was remarkably successful at attracting donors in no smal l part due to the willingness of particular administrators to take advantage of existing social ties to individuals within organizations or businesses who might be willing to help, and to Kiungas advantageous location at the confluence of numerous expatri ate businesses, safari lodges, transnational NGOs, and visiting student groups. Located at the gateway to safari parks, near a variety of plantations and newly founded NGOs, and in a region with a long history of missionary investment (see Chapter Three), Kiunga was incredibly well positioned to make connections with private organizations and donors in a way more remote health facilities were not. Reflecting on the immense changes they had experienced in their workplace in the wake of HSR, the staff credit ed the central government and donors themselves for their ability to build as an institution. However, while views of the MoH were largely favorable, as described in the next two chapters, when it came to governance at the district level where politically powerful individuals made decisions that might impact a staff members ability to support his or her family and realize his or her aspirations for the future attitudes were more ambivalent. In order to understand the complexities of balancing responsibilit ies as a health professional with those of realizing ones personal goals, it is important to be attentive to the ways that location impacted the ways that workers understood their lives. It is to the challenges and opportunities presented by living near A rusha that I turn next.

PAGE 211

211 Figure 51. Kiunga District Hospital, circa 20012002

PAGE 212

212 Figure 52. Kiunga District Hospital, circa 20022004

PAGE 213

213 Figure 53. Kiunga District Hospital, circa 2005

PAGE 214

214 Figure 54. Construction of pediatric ward at Kiunga District Hospital. Photo from hospital archives, taken in 2005

PAGE 215

215 Figure 55. Kiunga District Hospital, 2006. Note the addition of water tanks.

PAGE 216

216 Figure 56. District Medical Officer (DMO) of Kiunga showing the new pediatric ward to one of the missi onaries involved with the organization that helped to fund it. Photo courtesy of Kiunga's hospital archives, taken 2007. Figure 57. Kiunga District Hospital's ne w pediatric ward. Photo by N. Sullivan, June 2008. Female ward can be seen in backdrop.

PAGE 217

217 Figure 58. Kiunga District Hospital, 2007

PAGE 218

218 Figure 59. Kiunga District Hospital, circa 2008

PAGE 219

219 Figure 510. Kiunga District Hospital looking from west to east. Photo by N. Sullivan, June 2008 Figure 511. Kiunga District Hospital looking fro m east to west. Photo by N. Sullivan, June 2008.

PAGE 220

220 Figure 512. Kiunga District Hospital laundry and kitchen building, April 2008. The laundry room itself only made up a quarter of the total buildingincluding the door and the window area. The remaining area was the hospital kitchen, which was accessed by a door on the back side of the building. Photo by N. Sullivan Figure 513. Sheets drying during the rainy season at Kiunga District Hospital. August 2008. The rains meant that sheets often took several days to dry, sometimes creating shortages throughout the wards. Photo by N. Sullivan

PAGE 221

221 Figure 514. I nterior of laundry room prior to installation of new machine. April 2008. Photo by N. Sullivan

PAGE 222

222 Figure 515. New i ndustrial washing machine, installed July 2008. Photo by N. Sullivan Figure 516. The door of the laundry room is expanded and a new gate installed to accommodate the new industrial washing machine. July 15th, 2008. Photo by N. Sullivan

PAGE 223

223 Figure 517. Construction begins on the new laundry building extension. Taken October 15, 2008. Photo by N. Sullivan. Figure 518. I nternal structure of new laundry building extension at Kiunga District Hospital, October 23, 2008. The extension was made very la rge in order to provide an interior drying area for sheets and a space for conducting tra ining workshops. Photo by N. Sullivan.

PAGE 224

224 Figure 519. Centrifuge in the CTC Clinic at Kiunga District Hospital. To protect the actual identity of the NGO, its logo has been removed. The REFLECT marks where the original logo was placed. Note the tiling of countertops and walls in this laboratory where the CTC technology is housed.

PAGE 225

225 CHAPTER 6 OF PARALLEL AND HYBRID ORDERS If Global Health is a universal goal, art iculated in current times through the Millennium Development Goals (MDGs), then the application of these universalistic regimes can only occur in particular locations, where they intersect with the contingencies of locality. When new policies and procedure s governing biomedical healthcare provision come into contact with the particularities of place, what happens? What aspects of locality actually remain and even impede state and multinational efforts to manage the implementation of a particular vision of d esired Global Health? What role do individuals personal, professional, and institutional desires play in how these management regimes actually function in practice, and to what effect for biomedical care? In order to grapple with these questions, this c hapter begins to unravel the ways that new regimes for governing hospital work colluded with authority structures that existed at the local level prior to the new configurations establishment. In the pages that follow, the recalcitrance of the local and i ts effects on the nature of everyday biomedical and bureaucratic practice come into view. Anna Tsing (2005) calls for ethnography to emphasize what is productive about particular interactions between global, state, and local forces at their point of connec tion. Following Tsing, below I argue that in Kiunga District, the seemingly universal (and statesanctioned) goal of improving health care provision within government institutions was only achievable through particular gaps or excesses those aspects of loc al practice that state and global policies and plans specifically do not account for, or that are expected to dissipate under the new order. As I will show, these excesses could be productive for some actors and some visions of development of the health sector, but they did not favor everyone or every development initiative. Their effects were uneven. An example from my fieldnotes provides a launching point for this discussion:

PAGE 226

226 It is mid morning. I go to the maternity ward where I have been observing82 for the past several weeks. There are five staff members working that mor ning: the Nurse In Charge, two n urse midwives, and two ward a ttendant s. The Nurse InCharge and the nurse m idwives are busily sweeping the floor in the pre natal part of the ward, where women (several in labor) await de livery. The nurses work quickly. A s the Nurse In Charge sweeps between the cots, she instructs the laboring women on the importance of breathing through pain and not making a lot of noise. The n urse m idwives mop behind the In Charge Nurse on the areas that have been swept. I greet them and move toward the delivery section of the ward, on my way to the post natal area to see what is happening th ere. As I go by, I see the two w ard a ttendants sitting together on a cot, chatting. One of them tells me that her feet hurt her today, and the other asks me questions about living in America. There is a pile of clean sheets beside them on the cot, presumably for changing the patient bedding in the ward. The Nurse InCharge comes into th e delivery section and asks one of the attendants to escort a laboring woman to ultrasound because the f etus heart rate seems too slow. The a ttendant complains that her feet are too sore and suggests that I take the patient instead. The head nurse sighs a nd asks me if I will help, at which point I retrieve a wheelchair, assist the woman into it, and take her to the ultrasound unit. When I return twenty minutes later, the Nurse In Charge is in the delivery area doing a new patient intake. One of the nurse m idwives brings in a box of supplies from the pharmacy and sorts the supplies in to the cupboard. Another examines one of the laboring patients on a delivery cot. I pass the nurses room, where the a ttendants are drinking tea and invite me to join them for s ome chapati83. The sheets remain on the cot where I last saw them The patients beds had still not been changed. [Fieldnotes, August 2008] In the description above, higher rank nurses completed cleaning tasks at the same time as providing m edical advice to patients; lower ranked w ard a ttendants productivity was slower and they negotiate d with their supervisor regarding the speed and types of work that they did. The attendants complained, and their supervisor asked a visiting researcher to assist rather than force rank to entice them to comply with their official duties. Similar scenes to this one occurred regularly at the hospital, in many of the different departments and not merely the maternity ward. Generally, all over the hospital, nurses and m edical a t tendants were all involved in cleaning, and 82 In the maternity ward, I retrieved equipment and medicines for the nurse midwives, helped with wrapping neonates, tying off umb ilical cords, and various other non specialist activities, particularly when more than one birth was occurring at the same time. 83 Chapati is a form of flat bread that originated in South Asia, but that is commonly eaten in Tanzania as both a snack or a s taple accompaniment to a main meal.

PAGE 227

227 this was often a hurried affair. However, while not all ward a ttendants engaged in the practice, it was not uncommon for workers of this rank to avoid a variety of the tasks in which they were expected to partici pate (changing beds, collecting laboratory results, retrieving supplies from the pharmacy or stores departments, escorting patients). It was not uncommon for their work to be exceedingly slow or for workers in this rank to complain about or outright negl ect, particular tasks assigned to them by the InCharge Nurse. Based on the official hospital hierarchy governing the workplace, the ward attendants were required to comply with the orders of their In Charge Nurse. Instead, we find the InCharge Nurse and nurse midwife the two staff in the ward with the most education and medical expertise grappling to balance the necessities of providing care while simultaneously completing unskilled (but critically necessary) labor. How can we explain this paradox? Why we re such working arrangements allowed to persist when they blatantly contradicted the official hierarchical order by which hospital work was supposedly governed? T his chapter turns to professional and personal possibilities and desires of differently positi oned medical workers, and how they intermingled with multiple intertwined logics of daily governance (Blundo and Olivier de Sardan 2006:87; see also Blundo and Le Meur 2008). The concept of daily governance points to the ways that formal policies become necessarily embedded within existing tacit codes and (in)formal norms that already exist within those institutions. 84 Thus, w hen H ealth Sector Reform (H SR ) policies were i mplemented at 84 Blundo and Olivier de Sardans work looks at daily governance in the context of corruption. I, however, do not find this term very useful in analyzing informal and formal practices in the hospital. While there were s ome practices at Kiunga that I may have found ethically murky, the term corruption is highly loaded. Corruption is widely viewed as something to be eradicated, and many global actors (bi and multi lateral donors, multinational organizations and NGOs, etc.) purposely circumvent state apparatuses or pull their funding for particular government programs in aiddependent states based on corruption. However, as Blundo and Olivier de Sardan note, daily practices that make up corruption are often assumed without being attentive to the daily practices and interactions that make them up, and practices that may be deemed corrupt by one actor may be deemed totally

PAGE 228

228 Kiunga, they did not erase existing ways of doing things, but rather w ere variously incorporated, rejected, and adopted, becoming intertwined with the tacit and (in)formal orders that already existed at the facility. In the interstices between policy and implementation, there is a complex intermingling of logics, both offici al and unofficial, which reconfigures the distribution of power and authority in complex, and often contradictory, ways (Chalfin 2010). In an appeal to move beyond common binaries between periphery and centre or global and local in the analyses of governa nce, Blundo and Le Meur call attention to the intermediary perspectives of state agents and how multiple logics intersect within daily practice (2008). Their approach is not only attentive to the ways that state agents navi gate their situations, but also notes the discourses these actors develop in relationship to their daily practices and the strategic behaviors they adopt in the face of other state and nonstate actors. Part and parcel of the approach they advocate is a careful attention to the interactio ns of government techniques and subject making, where Subject must be understood here in its double sense, i.e. both an object of domination and an active agent, in relation to forms of government that do not necessarily emanate from the apparatus of th e state (Blundo and le Meur 2008:10; see also Anders 2010, Blundo and Olivier de Sardan 2006; Chalfin 2010). Seen within this frame, hospital workers within the government health sector are agents of the state and providers of biomedical health care with in the hospital; however, they are also subjects of the state both within and outside of their workplaces. In their daily lives, hospital workers interact with forces of the central state, the local government, the socio economic context of the community i n which they live, and the formal and informal bureaucratic structures of the hospital itself (Martin 2009; Wendland 2010). Within this space of interaction, they acceptable by another or in different circumstances. It is for this reason that I prefer the te rms informal and unofficial and instead seek to trace where and how they operate in practice, and how people think about them,

PAGE 229

229 variously grapple with providing biomedical care to patients. Some are more invested in this care than others. Thus the perspectives of health sector workers provide an intermediary level from which to perceive multiple logics simultaneously how they are lived, experienced, negotiated, perpetuated, and contested (see also Anders 2010; Chalfin 2010). T his chapter explores intermingling logics informing dayto day experiences and bureaucratic and biomedical practices of hospital employees at Kiunga District Hospital. It also addresses the discourses and strategic behaviors they employ in order to m aneuver the intertwined logics that underlie their work. In exploring these multiple and cross cutting logics, I point to what aspects of the daily governance of hospital w ork are negotiable, and to whom Thus, in the first section, I draw upon the interm ediate level of hospital personnel to reveal how parallel orders punctuated dayto day practices of providing health care. This is followed by an analysis of how, at the level of institutional decision making, parallel orders could congeal to form hybrid orders, whereby official and unofficial norms ironically reinforced the very authority structures that HSR policies were meant to stamp out. Ultimately, this chapter points to what is potentially detrimental about the establishment of these orders, but also, and importantly, how they can also be productive. Parallel Orders In his recent work on civil servants in Malawi, Gerhard Anders describes a series of parallel orders (Anders 2005; 2010) the intertwined official and social norms and tacit codes of c onduct (see also Chalfin 2010) that actors in the civil service learn and maneuver on a daily basis. Civil servants thus adjust their behaviors and practices to conform to the numerous entangled conventions that have become a part of their workplace in the wake of governmental reforms. At Kiunga, official hierarchies were challenged by unofficial ones in the realm of everyday hospital routines. Comaroff and Comaroff suggest that in the wake of a move towards

PAGE 230

230 more neoliberal state structures, there have been remarkable transformations between old and new orders, in which, as often as not, the power brokers, bureaucrats, and administrative personnel of the past are either left in situ typically to ensure the confidence of foreign investors or succeed in findi ng less visible ways to keep their hands on the levers of authority (2000:352). Likewise, at Kiunga there are notable continuities between past and current regimes within the hospitalboth in terms of official hierarchical structures, as well as within th e less visible means that particular people had available to them in order to maintain authority. Ones ability to maneuver this system of parallel and hybrid official and unofficial orders and logics was more difficult for some health workers than for others. It was an arena within which the individual and professional desires of differently positioned healthcare workers often clashed. This section outlines the official hierarchical order meant to govern practices within the hospital, and follows with an ethnographic explanation of the unofficial norms and hierarchies with which it operated in parallel. Official Hierarchical Structure of the Kiunga District Hospital The official hierarchical and reporting structure in operation at Kiunga District Hospital is largely echoed in other government owned district hospitals in Tanzania. It is meant to govern the ways that staff within health facilities distributes their labor and expertise. It purports to establish a transparency within biomedical treatment, al locating particular responsibilities and a chain of command within the hospital. Drawing on Anna Tsings ethnographic treatment of global connection (2005), we might see this hierarchical order as a claim to a universal logic of ordering a biomedical workplace. This official structure plays into an economy of appearances that makes Kiunga District Hospital looks similar to other biomedical facilities around the globe administrators are above doctors, doctors are above nurses, nurses are above general worker s, etcetera. Such a system should allow new workers coming into Kiunga to learn very

PAGE 231

231 quickly their official role, and their relationship to others within the facility. This structure provides a sense of how things are supposed to be in the hospital. It i s an ideal, and the histogram depicted in Figure 61 outlining this official work structure hangs at the front door to the District Medical Officers (DMO) office. As decentralization moved the majority of planning and implementation activities from the central government to t he District Councils (DC), the C ouncil ha s become a significant governing body, planning and arranging implementa tion of various state and local initiatives. In the wake of HSR, the overall administration of the health facilities in an y given district of Tanzania fall s to the DC, under the leadership of the District Executive Director (DED). Underneath the DED are the heads for each department of the district. This includes the department of health, which falls under the supervision of the D istrict M edical O fficer (DMO) In terms of the health sector, this administrative system means that all health workers have moved from under the umbrella of the Ministry of Health (MoH) to that of the DC. This change in the administrative apparatus ha s caused a considerable degree of ambiguity in terms of the procedures health workers follow in the pursuit of their rights ( haki ) and desires. Yet within individual health facilities the official hierarchical structure remains largely consistent with the past It is top down, whereby each worker has a supervisor within his or her department coordinating all activities within it. Officially, the DMO is in charge of approving and overseeing all public and private health rel ated activities in the district The DMOs office allocates health workers to the various facilities in the district, distributes medical supplies and pharmaceuticals to those facilities, and supervises infrastructural developments. He is also responsible for representing the problems o f health workers to the DED, particularly relating to

PAGE 232

232 job allocation and payment issues.85 The DMOs administration also oversees all patient complaints and health education programs. His office compiles reports on health activities in the district for the DC as well as for the MoH He also supervises all donor sponsored health related activities in the district. The DMO is the head decision maker, an d any visitors to the hospital N GOs, donor s expatriate student groups, political figures, or other dignitar ies come to his office before entering the compound. Further, the DMO is required to attend monthly and quarterly full district council meetings, where he reports on the healthrelated activities and plans for the entire district. Although the DMO is not t echnically required to be an active administrator of daily affairs in the district hospital itself, in practice the DMO at Kiunga District Hospital, Dr. Saidi, was highly involved in daily workflows and building staff morale. For the hospitals daily oper ations, the main administrator is the Medical Officer In Charge, or MOI. The staff at Kiunga District Hospital often called the Medical Officer In Charge Moi ,86 instead of referring to him87 by name The MOI oversees workflows for the A ssistant M edical O fficers (AMOs) including assigning oncall shifts for afternoons nights, and weekends. He also ensures that the proper paperwork is filed in cases of births, deaths, hospital transfers, surgical procedures, infectious diseases, and other cases that are i mportant to track. The hospitals AMOs bring their concerns to the MOI, and the MOI oversees the administrative duties particular to the hospital. 85 From 2008 up to present (2011), the DMO of Kiunga has been male. However, there are several DMOs elsewhere in the country who are women. 86 pronoun ced moy like the English pronunciation of boy 87 At Kiunga District Hospital, a new MOI was appointed in August 2008, but both of the people in this position were male. However, during all of 2008 there were at least two female AMOs working at the hospital. One female AMO left in July 2008, but was replaced by another female AMO in August. The female AMO who left in July had worked as MOI under the previous DMO administration, but was removed from the position in 2007. I am uncle ar as to whether there was a gender component to this position, as the female AMOs I interviewed (three in total) did not allude to difficulties working as a woman among so many male colleagues of their rank.

PAGE 233

233 An AMO is appointed to each of the four wards of the hospital to conduct ward rounds on weekdays. An AMO or two may also be assigned to the major surgical theatre on weekdays to conduct any of the procedures that are scheduled on a particular day as well as any emergency cases requiring surgical intervention.88 Although an AMO might allow a CO to accompany or assist her during a surgical procedure, ultimately the AMOs conduct the vast majority of surgeries themselves, with the assistance of the nurse officer and medical attendants allocated there. AMOs also rotate on call shifts for nights and weekends in case of emergency surgeries or consultations In addition, during the daily morning staff meetings at Kiunga, the AMOs often educate the entire staff about medical indications, diagnoses and prescribing, and they regularly consult each other on complicated or unusual medical cases. They collectively decide what kinds of interventions or treatments to pursue in complex cases, and whether or not particular patients require transfer to the regional hospital in Arusha. COs are the largest group of doctors within the hospital.89 They handle the majority of outpatient consultations, and a designated CO In Charge coordinates their schedules and activities. A CO is also allocated to the hospital for every afternoon, night, and weekend shift, as the first responder for an y new patients, whether emergency or not. While the COs fall under the direct supervision of the CO InCharge, they also fall under the supervision of the AMOs and the MOI in terms of the hospital hierarchy Several COs at Kiunga work closely with the AMOs in order to expand their knowledge base and many of the COs hope for opportunities and sponsors to be able to go to training to upgrade to the position of AMO. 88 The most common emergency surgical cases were maternity cases requiring caesarian section. 89 For a detailed outline of the use of the term doctor ( daktari ) in Tanzania, please see the note on page 19.

PAGE 234

234 The Matron of the hospital oversees all nurse officers, nurse midwives, nurse attendants and ward attendants She is responsible for arranging the staff roster for all ranks of staff except the AMOs and Clinical Officers ( COs ) She also oversees the stores department and ensures that all hospital supplies are accounted for. She hosts a variety of meetings with hospital nurses in order to inform them of changes to service provision policies and update them on administrative procedures with regards to patients. Herself a nursing officer, the Matron has an active role in mediating disputes and solving problems within the hospital, and she is often the first person with whom individual nurses and attendants consult regarding tensions or problems in the various departments of the hospital. As in the example from my fieldnotes above, w ithin each ward and departme nt of the hospital one nursing officer is designated as the Nursing Officer In Charge responsible for supervising all of the nurses, nurse attendants, and m edical a ttendants in her department.90 Nursing officers have the most expertise and train ing in nursing care, particularly regarding managing complex cases. Nurses are not on call; they rotate shifts based on the Matrons weekly work roster. The only nurses who are oncall during the evening, night and weekend shifts are those assigned to the major operating theatre who, like the AMO oncall, might be called in to work due to an emergency surgical case. While it is d esirable to have more than one n ursing officer working during a shift on any given day, in practice at Kiunga some wards have h igher concentrations of officers than others, and during night and weekend shifts, there may only be one n ursing officer, or sometimes only a nurse m idwife, to supervise the entire hospital. On the day shift, t he pediatric ward ha s the highest concentration of nursing officers. This is desirable for a variety of reasons. Childrens 90 For a detailed outline of the various grades of nurses in Tanzania, please see note on nurses p age 19 to 20.

PAGE 235

235 medical conditions can change very suddenly, and knowing about proper dosages for children of all different ages requires more expertise than dealing with adults. The female, mal e, and maternity wards generally only have one, or less frequently, two, nursing officer (s) working during any given shift, which increases the hardship on the staff in those departments since, as described in the example above, the In Charge Nurse ordinar ily has to complete the same duties as the other nurses, in addition to doing ward rounds with the AMO and managing all administrative duties for her department. The nurse midwives and nurse at tendants make up the largest group of workers at the hospital. D ispensing and administering treatments, in take of new patients, recording in patient files and general nursing duties fall to the nurse midwives. Some nurse a ttendants can also dispense drugs, administer injections, change dressings, and manage some of the less complex treatments of patients. Below them in rank are the ward attendants,91 who receive no medical training, but are allocated to each department to do support duties such as cleaning, delivering laboratory specimens and collecting lab results, r etrieving supply orders from the pharmacy and store department, transferring patients between different departments of the hospital, making beds, and other nonspecialized duties. The Unofficial Hierarchies of the Hospital In practice, this official orde r structuring the relationship between various ranks of workers co exists with a very different, parallel order At Kiunga District Hospital this is one of the less visible ways that bureaucratic hierarchies are negotiated, challenged or maintained (Coma roff and Comaroff 2000:352). This parallel order came to my attention when I observed the cleaning routines of various departments and wards as in the example with which I opened this chapter 91 Ward Attendants were also commonly called Medical Attendants, and I use both terms interchangeably.

PAGE 236

236 While many of the younger ward a ttendants worked at the same p ace as their supervisors and nurses in higher ranks, it was not uncommon for older ward a ttendants to sit or delay participation as their peers and superiors did general tasks. After months of similar observations in many other hospital departments, I was surprised at how much of the nonmedical workload fell to the head nurses, while m edical a ttendants were far less efficient. I was also curious as to why those in charge of the departments said little to the w ard a ttendants about their lack luster work.92 A ccording to Regina, one of the head nurses at Kiunga, the staff under her supervision often maintained an appearance of working while not accomplishing the duties allocated them: Regina: if you look at people they all look busy Everyone here looks busy but you cant know who it is that does it correct and who it is that doesnt do it correct Every person has been working, is busy and if you look by two oclock everyone is exhausted, everyone is tired! A person has not yet sat down, you are busy But if you go to look, yes she has been busy but if you look at what she has done from morning until evening, you dont see it. She will say I started this and it stopped, I started this and stopped, I started this and it stopped! Mmh! (Interview, 11 1108, Kiswahi li) It was not until I started conducting interviews with different ranks of workers at the hospital that I began to understand the underlying role of seniority and community embeddedness in this particular set of routines. When collecting life histories o f some of the older ward a ttendants, I learned that most of them began their work at Kiunga prior to HSR without ever having worked in another health facility. Indeed, many of them worked there as adolescents. Conversely, all but two of the nurse officers and several of the nurse midwives started work at Kiunga at the onset of HSR and a large proportion of them began work there in 2004 or 2005, when the pediatric and male wards were opened. 92 It is important to note that I arrived at the hospital with my own biases regarding the efficiency of work, and so much of my surpris e came from my own subject position and the expectations I had of how labor should be divided and the processes by which it is most efficient to accomplish particular tasks. I often struggled with understanding the logics underlying the working routines of the hospital

PAGE 237

237 I asked several of the head nurses as well as some of the doctor s about communication with the nurses under their supervision. The question elicited a variety of frustrated responses An AMO, Dr. Kweka, was explicit about how the parallel order actually worked in practice: Dr. Kweka: The In Charge Nurses of the wards here, let us say that they are making an effort [ wanajitahidi ]. But even on their own they have a lot of challenges There are several staffs in the wards that have been here a long time, so that one you find the leadership, she is in charge but she has only been here for a short time. She finds in her ward only Nurse Assistants93 and Nurse Attendants, but they were born here, they were here since the time that it was a health center [19681988] so they have a lot of power and they disrupt her a lot. You find those people they have been here a long time and already they have a lot of influence even if they are people at the bottom ranks because they know the leadership at the top that has also been here a long time. They know each other. So the l eaders of the wards [In Charge Nurses] they are trying hard. This challenge to the official hierarchy was wearisome for several of the incharge nurses. Given the influential social ties of the staff under their supervision, many head nurses felt they ha d no recourse This unofficial order made the burdens of being shouldered with both biomedical and bureaucratic responsibilities heavy indeed. Yet they hope d that the distribution of the workload might somehow be changed so that the nursing o fficers would shoulder fewer of the nonspecialized tasks so they could concentrate on patient care. One such nurse was Monica. Monica had started working at Kiunga in 1997.94 When I asked her what it was like to work as an In Charge Nurse, she b egan an agitated account of how cleaning responsibilities took her away from her duties to provide proper patient care. She observed that many of the w ard a ttendants had been allocated to administrative rather than general worktaking patient fees, opening files, counting and sor ting pills, filing x ray films, completing paperwork for ordering supplies and other such tasks. Many of these posts had few or 93 Meaning Ward Attendants 94 Although Monica had been working at the hospital for over a decade, those who worked beneath her had worked there far longer, and she was faced with the same struggles as the other Nurses In Charge.

PAGE 238

238 no patients, and minimal workloads. She said there were even nurse attendants who had some medical training, working in the acc ounting and stores departments: Monica: Is it fair ? They [the ward attendants] just sit there, I tell you the truth! Sometimes a person does not understand what she is doing. I, a professional nurse, I mop [ nadeki ] where they [administrators] could have pl anned for those two nurses95 that are in those [other] stations, that in the morning they come into the wards, they spread out [throughout the hospital] and they clean! But me, I cannot stand there [at the morning meeting] and speak of a thing like this! I see it and I stay silent. (Interview, 11 1810, Kiswahili) Monica continued that several of the medical attendants were assigned to departments such as the tuberculos is clinic and the dental clinic, where there were few patients: Monica: She is there and s he just sits! But also she could help the patients! Nobody can see but this here again, it is for you to see and if you could approach [the DMO] and talk with him, it is not necessary to tell him it was me, say that you have observed this! You go and say that. It is very important If it was possible that those nurse [attendants] be spread out to help with the work so that, goodness [ jamani ] those professional nurses could at least help those patients! Sometimes patients urinate on themselves [ w anajikojolea] and they are supposed to get a catheter, and supposed to get a drip At what time [can you do it] when you are mopping!? When I asked Monica how long this had been happening, she said it had been this way since she had first started working there She desired a workplace in which she would have time to take care of her patientsto put her expertise to use. Her statements are a poignant example of how the standards of biomedical care become eroded due to unofficial orders that impinge on her ab ility to put her professional skills to work. She marks herself here as a professional, as well as pointing to the important care activities that are her responsibility. She feels bad that she is unable to provide the standard of patient care that she wishes to uphold. The envisioned standard of patient care Monica articulates here is not one of high technology or one that is beyond the existing capacity of the hospital to provide; her professional desires point to the simplicity of 95 Here she is referring to nurse and ward attendants, who were often also called nesi in daily conversation.

PAGE 239

239 changing existing workfl ows so that those with more biomedical skill can apply their talents for the good of patients wellbeing. However, her comment also highlights the dangers of challenging this existing order. Monica, like three of the other InCharge nurses of the wards, di d not complain about the situation to her superiors, and spoke of the dangers of criticizing the m edical a ttendants. M edical a ttendants were thus quite powerful according to Monica, so if you talk they will tell you, you newcomer! What do you expect?96 I n three of the four wards, the In Charge nurses resorted to a strategy of trying to lead by example by doing the cleaning work as well as their medical duties in order to demonstrate to those below them that they also participated in nonspecialized work We might see this as a local economy of appearances one that intersects with the unofficial but nonetheless powerful hierarchies that exist within the community in which Kiunga is located. With the silence of the in charge nurses, the workers that were officially under their supervision were able to maintain a less visible hold on their authority. In turn, the head nurses hoped that by leading by example, they might entice those under their supervision to invest their time in the needs of the ward. One of the head nurses, Regina, described the difficulties of being a head nurse: Regina: There are challenges there are differences among the wor kers. There are times when you can tell someone to do som ething and she will not do it. But pe ople can see that I myself do the work it is better that they also do it. You see, eh? So you can see that a person stays at the desk and does nothing but me, I do the work. Even if it is not my job I will do it. (Interview, 11 1108, Kiswahili) Similarly, Elda described the need to show those beneath her that she was willing to do the work in order to motivate them to do it: Elda: To be a leader it is really difficult. Because you come here and you enter and you see that maybe the ward is not very clean. Dustbins are full of things, maybe. So you ask, my goodness! [ jamani ] Please could you do this or do that? because you see that things are scattered . And they always look at you and they think, now you, are you just 96 This phrase was spoken in English.

PAGE 240

240 saying it or are you doing it? Other times it is better you do it yourself, you start by going outside and tending to the garden so that if you want to ask tomorrow for someone to do it they see that you had already done it. You go and you annoy them about the latrine . If you dont lead by doing it yourself first, people will not do it. They will say, that one, shes just an In Charge but she doesnt do it. But if they see me, they see me washing the latrine they see me dusting, so that tomorrow someone might stop and say, wait I must do that too because the In Charge did it the other day. So, sometimes [ hesitates ] to be In Charge is to be overworked more. (Interview 100208, Kiswahili) The majority of nurse officers, AMOs, and COs working at the hospital those with the highe st amount of training and expertise had worked at the facility for the least amount of time. While some had connections to the surrounding community, many did not, and this had an adverse effect on their ability to lead and to motivate those below them to work on less specialized tasks so that they could ensure adequate treatment and care of their patients. Meanwhile, outside of the DMO, most of the administrators at the hospital had also been there for several years or even decades, and many were affiliated w ith the most dominant and influential religious and political leaders in the district. Relationships mattered. The dynamics of social capital emerged regularly, but were often unspeakable except among personnel of the same rank or, if asked, to an anthr opologist who was not immune to the difficulties they faced. New personnel coming to Kiunga learned this parallel order in situ Thus, the parallel order, or what Olivier de Sardan calls the local bureaucratic culture, (2001; see also Jaffr and Olivier de Sardan 2003), was such that those at the very top and the very bottom of the official hierarchy or, more simply, those with more connections to influential people and organizations beyond the hospital were the ones wielding the most sway in terms of dai ly routines of the hospital and the allocation of workers throughout the district (Olivier de Sardan 2001) In effect, there w ere several n ursing o fficers and doctors who expressed concern about being transferred out of Kiunga if they upset hospital employees who were socially tied to influential leaders in the district. Dr. Joseph, a doctor who had worked at the hospital for a very

PAGE 241

241 long time, confirmed that particular individuals at the hospital had more political clout in the district: There are those wh o know people and have more ability to go to them [the head administrators] and ask for their rights or find out more information. So you can see that some of them go there and do that, and they can ensure that someone else can be transferred if they have been fighting with that person. Also, even the politicians get involved in those decisions. Even I know I could be transferred, and that there are others who can decide that you, Noela, you are going to be transferred because those people have more power t o influence things. Eeh. I have seen it. Truly, our condition is better [since HSR ], but quite often, a person who does not have influence, they can be moved and really, it is not going to bring health to do things this way. (Interview, 11 1308, Kiswahili ) Dr. Kweka himself was concerned that he would be transferred, which would jeopardize his personal desires and domestic responsibilities; namely, in his case, the ability to care for his children and the ailing parent he supported: Dr. Kweka: You find tha t one of the people who has been here a long time has been told they will transfer, and then they go to the political leaders. Maybe they dislike someone else. They say they were chosen to transfer but then they tell [the politicians] that there is another person at the hospital who is not useful, who is causing disturbances, who is asking questions and then that other person will be transferred instead. You would be surprised! Me, I started here [recently], it is possible that I will be transferred. I would not be surprised if I got a letter tomorrow or the next day saying that I will move to another place. So it causes people who have been here for a long time, they have influence with the hospital or the political leadership and they go to my boss and the n I am transferred. In the intertwining of orders, the logic of seniority at times trumped the official hierarchy. In terms of day to day bureaucratic and biomedical routines and interactions at the hospital, ones biomedical expertise and training did not carry as much weight as ones social ties to the wider community within which the hospital was situated. AMOs, COs and higher ranked nurses were the most mobile of health personnel As workers who were often transferred they were least equipped to form the kinds of relationships made possible by ones longstanding community ties Although the official hierarchy allocated supervisory powers to these ranks of workers, those with more social capital and clout in the

PAGE 242

242 community were reticent to relinquish their influence. Two distinguishable parallel orders operate in the examples above: there is an official way that things are supposed to be done, and an unofficial but nonetheless powerful way in which work is organized in practice. The distinction betwee n these orders only becomes knowable in practice, and can only be perceived in the interactions between actors that are differently positioned in the official, and the unofficial, orders. Two nurse officers working together would not face the same configur ations of intermingling official and unofficial orders in the same way that a nursing officer or doctor would face them in relation to a ward attendant with seniority and social clout in the community. Parallel orders thus only come about in particular kin ds of interactions, and when they do, they may interfere with each other. The official order marks the Nursing Officer In Charge with a variety of responsibilities that do not fall to ward attendants, and regardless of the unofficial hierarchies, the In Ch arge Nurses felt obligated to complete those official tasks. Indeed, when they could not fulfill their role as professional nurses and provide what they saw as an adequate standard of care, they felt considerable stress. Their professional desires related to a particular standard of patient care, and following the official hierarchy of the workplace, they felt that they should indeed be able to achieve those goals. Yet the unofficial order that dictated In Charge Nurses status in relationship to other hosp ital workers with more seniority meant that the former had to take on a variety of responsibilities that went beyond the responsibilities dictated by their profession. These orders operated in tandem, through particular sorts of interactions. As previously mentioned, the DMO in 2008 had only been in that position for two years. Arguably he would not have the same social capital as the ward attendants and some of the other administrators at the hospital who had longstanding ties in the community. Why would Monica ask me to intervene and speak to the DMO, Dr. Saidi, about these issues if, in terms of the

PAGE 243

243 unofficial hierarchies of the hospital, they had similar statuses? Why would Dr. Saidi comply with changing transfer orders for a particular ward attendant t o someone else working at the hospital? For Dr. Saidi, what could possibly be productive about complying with this tacit order favoring seniority and social capital? In order to answer these questions, we must move from the parallel orders existing at the hospital to the level of institutional and district management of which the DMO was a critical figure. It was in this arena of institutional and district management structures that a particular hybrid order was established at Kiunga. This hybrid order, and the underlying logic of political and social capital on which it depends, was critical to what Dr. Saidi was able to accomplish in terms of institution and capacity building at Kiunga District Hospital. Indeed, rather than attempt to expunge these unoffic ial norms governing relationships within the hospital, Dr. Saidi capitalized on the situation in order to gain the investment he needed to make institutional possibilities into realities From Parallel to Hybrid Orders While Anders (2010) characterizatio n of parallel orders suggests some intermingling, the characterization of them as parallel suggests that they operate in tandem without merging into hybrids In the discussion above, parallel orders operated in just this sort of way: they were perceptibl e only in certain interactions, and not in others. They operated simultaneously in practice. However, they did not fuse together to become a different kind of order altogether. Yet some orders did just this: they amalgamated in particular ways. Thus, I wis h to disarticulate parallel from hybrid orders in the discussion below. I propose that official and unofficial norms may also merge into complex, often unpredictable hybrid orders97 that simultaneously 97 I would like to thank Brenda Chalfin for suggesting this term.

PAGE 244

244 challenge official hierarchies while potentially rein forcing the pre existing authority structures these hierarchies were meant to eradicate (see Chalfin 2010). A t the level of institutional decision making at Kiunga parallel orders merge d into hybrids, with important impacts on who was able to govern, how they were able to do so, and to what effect for the provision of healthcare At Kiunga District Hospital, this hybrid order came about when two official management structures within Kiunga District actually converged with the unofficial order valuing soci al capital, merging into a hybrid order that made several people at the hospital uncomfortable and had a deleterious effect on morale for many of the most qualified medical personnel at the hospital. The first part of this discussion outlines the official management structures governing the health department in Kiunga. The second section contrasts this official structure by demonstrating how it converged with existing unofficial orders favoring social and political connections to the wider community. Within this hybrid order, some workers desires for particular possibilities of delivering biomedical treatments, their imaginings of the institutions future, and for certain (what they felt were achievable) goals for their workplace were often subsumed or sile nced within this hybrid order, leaving little recourse Simultaneously, as the second part of this section contends, it was only through the excesses of hybrid orders that other initiatives to build the hospitals capacity became achievable. Official Dutie s and Membership in District and Hospital Management Teams In planning for the decentralization of the health sector, the MoH conceptualized a series of teams to oversee and carry out policy implementation. At Kiunga, like in other districts in the country, there are two main teams that plan and manage priorities and initiatives at the hospital: the Hospital Management Team (HMT) and the Council Health Management Team (CHMT). 98 98 It is not yet clear to me whether or not the D evelopmen t Partners (D Ps ) had any influence on this managerial order. I am aware that the MoH called for numerous task specific teams to be created in 2008, but whet her or not the

PAGE 245

245 The HMT is particular to the hospital, and the CHMT oversees all health activiti es for the entire district (including the district hospital). These two teams have overlapping responsibilities: developing biomedical healthcare delivery. However, their agendas should be separate. The CHMT sets the tone for the overall planning and budge tary priorities for the health department districtwide; the HMT should take these priorities and determine how to actually implement them within the capacities and limitations of the facility itself. Planning and coordinating the health activities for the entire district, and thus including the district hospital, is the Council Health Management Team (CHMT). Collectively they develop the annual health plans and budgets for the district. With the DMO at its head, t he CHMT is the ultimate deciding body for allocating employees to the various government owned health facilities in the district. While individual health facilities come under the CHMTs oversight, this team is not to be concerned with the dayto day functioning or development of individual facili ties unless particular facilities or wards within the district are to receive funds from the DC or the MoH for particular improvements. The CHMT at Kiunga consists of seven people: the DMO, the District Heath Officer (DHO), the District Dental Officer (DDO ), District Laboratory Technician, the District Health Secretary (DHS), and the District Pharmacist. The HMT mandate is to collaboratively plan for the development of the hospital. They discuss how any additional funds coming into the hospital should be s pent, and plan how patient fees from cost sharing or private donors funds will be allocated. The HMT is charged with collaborating to weigh the needs of the various units of the hospital and negotiating which are prioritized based on the funds available T hey also discuss any infrastructural or staff problems at MoH mandated these alone or the DPs suggested this configuration requires further res earch. Nonetheless, unlike the CHMT, which had a significant amount of decision making power, there was often a dearth of volunteers for specialized teams within hospital management because they required a significant amount of time, provided no compensati on, and had to have all of their decisions approved by the CHMT.

PAGE 246

246 the facility more generally, and come up with strategies to address them. The HMT consists of the persons in charge of each department of the hospital: the head laboratory technician, the head pharm acist, the head physiotherapist, the head dentist, the Matron, the Nurse InCharge of each department, the AMO designated to each department, the Clinical Officer In Charge and the MOI. Similar to the official hierarchy of the hospital as outlined above, this official managerial order separating district and institutional supervisory and administrative duties was part of an economy of appearances. The MoH needed to ensure that management structures were put in place that promised to show its investors (the DPs) that it was taking steps towards achieving good governance through proper management. Indeed, the Acting Assistant Director of Continuing Education within the MoH stated that MoH training programs for health workers emphasized managerial skills, wi th separate training for HMT and CHMT members on management and planning.99 Improving management was seen as necessary to achieve the Millennium Development Goals (MDGs), to which Tanzania was a signatory, to maintaining good relations with existing DPs, and to enticing more donors to invest in the Sector Wide Approach (SWAp) that would increase the MoHs capacity to govern. Merging Management On paper, Kiunga appears to have fully adopted the managerial structures dictated by the MoH. The CHMT and the HMT completed their reports in a timely fashion, following dictated formats. They made plans, created budgets, and kept minutes of their meetings. Thus, these two management bodies played into the economy of appearances requiring the district to comply with ce ntral government dictates in order to garner and maintain MoH financial support. 99 Interview with Acting Assistant Director of Continuing Education, Tanzanian Ministry of Health and Social Welfare, English, 07 08 08.

PAGE 247

247 However, if we follow Tsings (2005) call to understand universals in practice and the ways knowledge move, a very different picture emerges. At Kiunga District Hospital, mana gement in practice was of a different order indeed. T he membership for the HMT and the CHMT overlapped as the leaders of several departments within the district hospital were also official supervisors of district activities overall. This overlap became a particularly contentious issue in August 2008 with a major change in the leadership of the hospital. The MOI of Kiunga up to August 2008 was an AMO He not only attended to his administrative duties but also continued to perform surgeries and consult on diffi cult patient cases. He was well liked and respected by the entire staff, and worked very closely with the hospital Matron to coordinate various initiatives and administrative changes or challenges at the hospital. In August the DMO dismissed him and a ppointed the District Dental Officer (DDO) Dr. Urasa, to become the new MOI at the hospital. The dismissal was due to an allegation that the existing MOI, who was living on the compound, cut the water supply to the hospital in order to irrigate his garden s. This particular change of leadership was highly unpopular among many among the hospital staff, who felt that the DMOs reaction was rash. They felt Dr. Saidi had not adequately investigated the situation leading to the pipe being cut. The doctors at Kiunga also felt that a dentist did not have sufficient biomedical expertise to be able to assist them in consultations on difficult cases. With this change in hospital leadership, there was what was perceived as a dangerous overlap between the CHMT and the HMT. T he head of the laboratory department at the hospital was also the District Laboratory Technician, and now, due to the change in leadership, the MOI of the hospital was also the DDO; thus the people occupying these roles were on both the CHMT and the HMT. This dual membership caused a conflict of interest in terms of prioritizing the

PAGE 248

248 needs of the hospital and those of the district Further, several of the head administrators of the hospital (many of whom were members of the CHMT) had been working at Ki unga District Hospital for decades.100 The overlapping memberships within the HMT and the CHMT created a precarious situation for the department heads in the HMT in their attempts to negotiate meaningful changes at the hospital changes in line with their v isions of institutional and professional possibilities at Kiunga Therefore, if the HMT members openly discuss ed concerns and challenges they risk ed being transferred if their suggestions were not in line with those who were most senior and politically po werful within the hospital administration. Additionally, due to this overlap in membership, the CHMT was largely able to steer the agenda of the HMT such that only particular issues were discussed, whereas others were purposely omitted from the dialogue. I n essence, the CHMT, due to the social capital of several of its members, dictated what was speakable in terms of articulating and addressing problems at the HMT meetings. One member of the HMT described this dilemma at length: HMT Member:101 I can say there are some problems and there is fear because people do not really speak their minds well [in the HMT], they are afraid that their issues will be brought before the CHMT, that they will be brought before the CHMT and people are afraid of each other [ wanaogopana]. For people to be free free. So HMT, the chairman is the Medical Officer In Charge.102 People are not free to speak their issues in front of him. I think there are things that the CHMT plans, but there are other things that the CHMT does not need to do. This is not good. There is no confidence because the CHMT comes into the HMT and the meaning is that the CHMT might talk about it. Maybe there is a meeting and 100 Specifically, the DDO, the DHO, the DNO, the HIV/AIDS Coordinator, the IMCI Coordinator, the Reproductive Health Coordinator, the Malaria Control Coordinator, and the District Mental Health Coordinator had all worked at the hospital since at least the mid 1990s, and several had begun there in previous dec ades. 101 I have removed even the pseudonyms, positions, and interview dates of some of the participants in this section in order to protect them from possible risk from the hospital administration or local government of Kiunga. 102 The MOI at the time of the interview was also the DDO. T his change in leadership was highly unpopular among many of the staff

PAGE 249

249 someone is fearful to attend. CHMT has the confidence to speak, but without the CHMT we would be able to talk about our issues. But with the CHMT [representatives] there we feel we are being watched, watched down at the bottom. So there are people who, before they come to the HMT meetings, are given explanations from the CHMT that they are supposed to have a particular agenda item and only deal with those issues in the meeting. So you see the chair, he is already directed there [at the CHMT] so it is not ours. We cannot speak our minds there. They have entered our space. (Interview, Kiswahili) Here we find an example of how multiple orders converge into hybrids within everyday managerial practice and daily governance. The place of AMOs and Nurse Officers In Charge in the official hierarchy was supposed to be one of considerable import in terms of ens uring that the standards of daily service provision were carried out. The HMT was a space where the leaders of departments should be able to highlight problems staff encountered in their daily routines, and come up with innovative solutions to these challe nges. It was charged with the particularities of the hospital itself. However, in practice, with members of the CHMT some of whom enjoyed considerable political and social capital within the districtimpinging on the activities of the HMT, certain problems became unspeakable. Meanwhile, in his capacity as DDO, Dr. Urasa was a person with considerable social capital in the community and had worked closely with the DNO as an administrator for decades. As the MOI in the second half of 2008, Dr. Urasas presenc e on the HMT controlled the agenda of the team, maintaining a status quo favoring those with more seniority over those with less, at the expense of overall standards of patient care. As a team made up of primarily AMOs and nursing officers those with the f ewest political, religious and social ties in the community HMT members felt particularly vulnerable under the close oversight of the CHMT. They already struggl ed on a daily basis to keep up with their workloads while feeling that those below them with mo re seniority did not contribute their fair share. In theory the HMT should allow the AMOs and Nurses InCharge to collectively address the problems they encountered at the hospital; due the hybrid order allowing membership overlap

PAGE 250

250 between the HMT and the C HMT, their freedom to band together to achieve their desires for the ability to use their expertise for the good of patient care and for upholding workers rights was effectively silenced. Yet all was not egalitarian w ithin the CHMT membership, either. Lik e the HMT, t he dynamics of CHMT interactions were slanted to favor those members with seniority and strong ties to powerful people within the community. T he CHMT included not only senior and very powerful employees, but also junior employees of the hospita l.103 Several CHMT members had only been at the hospital for between four and five years but had considerable biomedical proficiency; their expertise qualified them to official positions representing health departments districtwide While technically they were members of the CHMT, their ability to contribute to decisionmakingparticularly with regards to fund all ocations and employee transferswas stymied due to the influence of particular individuals on the team One member of the CHMT said that they were not a team at all, but rather that all decisions remained in the hands of two or three members of the committee : CHMT Member: Within the CHMT the DMO likes to listen only to one person. DNO only. That is, they are the only two people talking. If the DNO tells the DMO something, he automatically believes it. For example, in the committee, someone brings up an issue, maybe it is that we received money from donors and what should we do with it. If you speak as yourself, they will just listen to you. But lat er when you leave, that is when they will make the decisions. Just those two. So you find, with the CHMT their job is to manage the money of the hospital, development of the district, everything. But you find that really the people who manage are only thos e two. Even if you were to ask me right now, what was done with that money? I am unable to know Because I do not get to participate in the CHMT. It is really just those two, or maybe three with [the DHO]. DMO and DNO are the leaders. They do it themselves So if something is happening I cannot even know about it. I 103 By junior and senior I refer not to rank, bur rather to seniority the amount of time that the employee had been working at the facility.

PAGE 251

251 am in the CHMT, but in name only [ laughs ]. Mmh. So, CHMT, the cooperation there is not good.104 (Interview, Kiswahili) While officially, Dr. Saidi as DMO led the CHMT, ironically he remained under the influence of two of the other members who had the longest standing and most powerful ties to community leaders. However, if we take the context beyond the boundaries of the hospital this kind of structure is not necessarily surprising. DMOs generall y get transferred every four to six years, and must align with powerful people within the hospital hierarchy in order to get things done. The DHO and DNO, as administrators who were affiliated with influential political and religious leaders in the area in the area and having been administrators of the hospital for well over a decade, had the kind of social and political ties Dr. Saidi lacked. P oliticians were key to Dr. Saidis ability to negotiat e funding from district coffers, and all major healthrelate d initiatives required approval from the DC Without political cooperation, Dr. Saidi would be unable to effectively manage the district, and his promotion through the ranks of the Ministry of Health was dependent on the performance of his district in term s of improving health outcomes and successfully abiding by MoH policies. By virtue of their simultaneous connection to the DC, the MoH, and (particularly in the case of the DNO and DDO) political and religiou s figures within the surrounding communities, the CHMT was the embodiment of multiple logics of daily governance. Government policies outlining who count ed as a decision maker in the CHMT aimed to increase accountability and transparency within health sector management. Ironically, by allowing certain heads of units at the hospital to also act as district administrators of particular health departments, those most powerful in the official hierarchy as well as those most influential 104 This membe rs perceptions were consistent with what I observed on the several occasions when I attended CHMT meetings: the majority of speaking was one by the DMO, the DNO, and the DHO, with minimal participation of the other members.

PAGE 252

252 in the community surrounding the hospital could congeal. This merging of the official management structures into a hybrid order caused the authority of those at the hospital with the most training and expertise to be significantly eroded. That said, the establishment of these parallel and hybrid orders served to achieve more t han merely maintaining the levers of authority in the hands of those who had more influence prior to HSR. Indeed, even if the Dr. Saidi was able to oppose the unofficial norms and tacit understandings governing influence and authority in Kiunga, it would have been inadvisable to do so. As described in Chapter Three, the previous DMO at Kiunga, Dr. Kalete, found her hospital development initiatives stalled for nearly a decade due to a lack of political investment by regional and local leaders. It took her ye ars to gain the kind of financial and social investments she needed to make building up the hospital infrastructure into a reality. Conversely, under Dr. Saidis leadership, the hospital doubled in size in the course of four years. The DC had to approve the construction of each building and all hospital plans and budgets Dr. Saidi managed to entice the hospital staff to imagine institutional possibilities that they had never before considered. The staffs morale was remarkably higher than in other biomedic al facilities I visited in Tanzania, largely due to Dr. Saidis ability to bring so many new infrastructural and technological resources to the hospital. Conversely, poor staff morale has plagued many African health facilities, often with deleterious effects on the quality of patient care even utilizing available resources (for examples, see Booth 2004; DelVecchio Good et al. 1999; Gibson 2004; Langwick 2008; Martin 2009; Wendland 2010). Thus, by virtue of his position at the interstices of local, state, and global governance regimes, Dr. Saidi was in a position to play into the economies of appearances required at multiple levels in order to achieve financial, political, and (particularly on the local level) social buy in for his initiatives at the hospita l. Parallel and hybrid

PAGE 253

253 orders could be productive indeed, and Dr. Saidi was fully aware of their potentialities, and their limitations. Conclusion The MoHs official management systems of shared decisionmaking were meant to increase accountability and tra nsparency within the health sector an important initiative to maintain Development Partner investment in Tanzanias health sector However, as others have shown (Anders 2010; Blundo and le Meur 2008; Blundo and Olivier de Sardan 2006; Chalfin 2010), new norms of governance do not necessarily replace those that existed before in a particular locality. Ironically, in the 1990s, bi and multi lateral donors assumed that the local level of government would be more accountable to its people, and more invested in meeting their needs. Regardless of whether individual district councils throughout the country actually attend to their responsibilities to their citizens, one thing is clear: under government decentralization local politicians play a n even more powerful role than prior to governmental reforms This opens up considerable space for the establishment of parallel and hybrid orders, as state institutions at the district level require political and social investment at the local level in order to achieve their agendas. These interconnections at the local level are highly productive, but not evenly so. Being attentive to the meso point of view of health workers at Kiunga we see how various models meant to configure daily routines and decisionmaking regimes in line with HSR and the values underlying the universal goals of Global Health were mapped onto existing logics within the facility itself. There are far more parallels to past structures of authority than a quick glance at management routines and structures at the facility would suggest. Meant to eradicate existing modes of patronage, the policies set by HSR outlining the ways that people should be ordered and who should have decisionmaking power was subverted by parallel and hybrid order s that emphasized seniority and community embeddedness in the daily routines governing

PAGE 254

254 service delivery. The interworking of these multiple logics of daily governance reveal how policies aimed at undermining existing authority structures can, in their implementation, rein force the very power configurations they aim to eradicate (see Chalfin 2010) Within the parallel orders of the hospital, while those at the head of the official hierarchy would appear to exert the most power and influence, they were instead unable to act in the ways that the official order outlined ; they were in some sense domi nated from both above and below. T hose having the most seniority at the hospital were best positioned to become active agents in parallel and hybrid orders governing relationships between workers. While these orders were often helpful in terms of institutionbuilding, the effects were uneven. A t the level of institutional decision making, parallel orders merged into a hybrid orde r in which the official logics of central governance congealed with those of socially embedded and politically salient community governance in an ironic interdependence. The top decisionmaker in terms of health for the districtthe DMO was paradoxically dependent on the DNO and the DDO to provide him with much needed alliances with influenti al political leaders in the dis trict, whose buyin he required in order to make his agenda a priority of the DC. B y permitting the establishment of this hybrid order the DMO situated himself in such a way as to have acces s and some degree of influence within the governance structures of the MoH, the DC, and the local community simultaneously. Through these orders, Kiunga District Hospital could participate in multiple economies of appearances at the same time. At the loca l level, s ince Dr. Saidi had to negotiate his agenda for the health department amid competing departments such as education, engineering, infrastructure, and culture, the ability to align with politicians through a localized economy of appearances could si gnificantly enhance his ability to procure funds and political support Thus, the formation of this hybrid

PAGE 255

255 order merging the agendas of the CHMT and HMT in order to maintain authority structures of the past was critical to achieving particular professional and institutional desires. At the level of state and global interconnections, the hybrid order allowed for the maintenance of existing authority structures, while simultaneously making the district seem compliant with MoH policies and guidelines, which were largely influenced by state desires to work towards the MDGs. As required by the MoH, Kiunga district established a working CHMT and HMT, which collectively (if unevenly) managed daily health care and bureaucratic practices at the hospital. By strateg ically fostering and drawing upon good political and social relationships at the local level, the DMOs office could showcase Kiunga as a success case for Public Private Partnership a major MoH initiative which allowed the district to maintain a relatively steady flow of federal and donor funds into the districts coffers. Thus the hospital was able to sustain an appropriate economy of appearances at the state and global level because of its hybrid order. Attaining local investment in the DMOs vision for i nstitution building allowed his projects to attain support; his achievements through the hybrid order made Kiungas district health department appear compliant and worthy of additional investment from the central government and various donor programs. Yet, while on the measure of institutional development and capacity, patient care benefitted, on the level of daily service provision, the advantages of these orders had their limits.

PAGE 256

256 Figure 61. Official Hierarchical Structure of Kiunga District Hospita l KEY: NO I/C: Nursing Officer In Charge NOs: Nursing Officers (Nurse Grade A) NURSES: Nurse Grade B MOI: Medical Officer In Charge M.ATTs: Medical Attendants

PAGE 257

257 CHAPTER 7 CONCURRENT GOVERNANCE: SALARY, COMPENSAT ION, AND WORKERS RI GHTS I arrive at the hospital for night shift at 6:45 in the evening. The sun is just setting, and the entire night staff is present. The nurse In Charge for the evening is a nurse m id wife, and the three others are n ursing a ttendants. Already I can tell the shift is extremely busy. When I arrive at the maternity ward, three births occur within the span of thirty minutes, with only one m idwife to attend them. I follow one of the a ttend ants and the Clinical Officer (CO) they treat a violence case against an elderly woman, provide prescriptions to their patients, attend new patients, and assess serious cases. Since there is no central communication line between the different units of the hospital, through the course of the evening I am often sent to deliver messages from the staff member in one ward to a nurse or the CO in another ward. Later into the night a woman arrives with her family at the outpatient clinic where new patients are su pposed to be received even during night shifts. I follow the CO and one of the attendants to the scene, where it appears that the woman is having a miscarriage. The woman is quite far along in her pregnancy I estimate she is around thirtyfive weeks pregnant and she is losing a lot of blood. The CO sends me to the maternity ward to get a drug to attempt to stop her contractions. When I return with the vial and a syringe, the CO states that the woman needs to have emergency surgery, since he is concerned tha t she will bleed to death before the placenta passes. I am sent back to the maternity ward, where one of the only working hospital phone lines is located, to ask the midwife to phone the surgical nurse on call and have her come to the theatre to unlock it and assist in the surgery. The nurse officer oncall that evening lives in one of the staff houses on the compound, about 500 feet from the back end of the hospital, and she has the only key to the operating theatre. Apparently, earlier in the afternoon s hift there had been another late term miscarriage requiring the on call surgical nurse to attend the theatre and assist in the extraction of the fetus. When the midwife asks the oncall surgical nurse to return to work again only four hours later, the nurs e refuses. She tells the midwife to call Gloria, another nurse living on the compound, and have her assist instead. The oncall nurse suggests that Gloria come to her house, retrieve the operating theatre key, and then perform the operation with the CO. U nfortunately, Glorias phone is not charged, and repeated calls do not go through. The midwife asks me to go to the administration building and call to the security guard at the front gate, who could then walk to Glorias house and wake her. While the staf f houses on the compound are not far away, the path has absolutely no lighting and it is not safe for any of the staff to walk through the outer boundaries of the compound without an escort. However, when I call to the security shack from the administrativ e building, the guard does not respond. The gate to the hospital is open, and it is obvious that the hospital is not remotely secure given the lack of guards and the facilitys proximity to the local market and a village where

PAGE 258

258 several violent crimes had been reported recently. When I return to the maternity ward to tell the nurses that the guard is not around, I suggest that I could walk all the way to the open hospital gate to see if the guard is sleeping. The nurses become very agitated. They do not want anyone going anywhere near the gate and the guard shack. It is clear no one feels safe, and everyone is nervous. Soon, the CO finds one of the guards, and immediately sends him to wake Gloria. Meanwhile, the Nurse In Charge and one of the midwives are tal king to me about how little sense it makes for a woman to bleed to death for want of a key to the surgical theatre,. They argue that the evacuation kit for conducting fetal extractions should be kept in the minor theatre, for which they have a key on site, instead of the operating theatre so that women do not have to bleed to death waiting for guards to wake up. They ask me to talk to the M edical Officer In Charge [M OI ] and suggest that the evacuation kit be moved to the minor theatre, because, they say, t hey cannot say anything to the administration about it themselves. Gloria arrives within fifteen minutes. The fetal extraction is completed. Yet by that time, the woman had been bleeding for over an hour since arriving at the hospital, not to mention the blood she lost before she was transported here The womans family members need to donate some blood for a transfusion due to the quantity of blood she lost. A few days later, I bring up the issue of moving the extraction kit to the minor theatre up during a morning staff meeting. The MOI and the Matron immediately refuse. They feel that the kit might somehow get misplaced if it is moved to the minor operating theatre. Several of the other staff argue that for particular critical areas, such as the pharmacy or the operating theatre, keys should be available to the night shift in the interest of dealing properly with emergency cases, and a procedure could be adopted to ensure that the keys are not misplaced. The MOI then stands reiterate s that it is a secu rity issue and that the procedure of going to the oncall surgical nurse would remain in place. There is no further discussion and the issue is dropped [ Fieldnotes, August 1218, 2008] I open this chapter with an account that at first glance may appear to uphold many of the existing (negative) characterizations of African biomedical places. It is a description that others might label as an example of staff corruption, abuse, and incompetency. However, I draw on this case not to label such practices as corru pt or morally reprehensible, but to argue for the importance of a consideration of local (personal, professional, and institutional) desires (and disappointments), and their relationship to the quality of patient care within biomedical health facilities in aid dependent states. Despite the major improvements that occurred at Kiunga District Hospital, under Health Sector Reform (HSR), very little attention was paid to the

PAGE 259

259 individual professional and personal needs and aspirations of medical personnel. Yet as the above narrative reveals, this omission can potentially have dire consequences in the context of daily practices at the hospital. The fact that the surgical nurse on call that evening refused to come to the theatre with the key, despite the patients profuse bleeding, struck me that day. What was also noticeable was that while the nurses criticized the guards and the hospital administration regarding their lack of a ccess to the key to the theatre ( and thus to the only equipment at the hospital able to help the patient ) they never said anything, positive or negative, about the nurse who refused to work. It was only in being present during several subsequent night shifts and seeing a number of emergency cases such as this that the nurses actions began t o seem reasonable. Studying nursing practice in Uganda, Helle Max Martin argues that nurses are not merely institutional and professional actors distinct from a wider community (2009). Rather, they are simultaneously social and professional actors, naviga ting professional ideals and realities, and understanding these in terms of the wider social and domestic conditions of living in a particular locality. Like Biehl and Lockes call to be attentive to peoples articulated desires and envisioned possibilitie s as potentially suggestive of alternative futures (2010), Martins work demonstrates that what people want, personally and professionally, and the domestic burdens they shoulder actually matter for the quality of patient care in contexts of scarcity This chapter takes seriously the need for understanding how hospital workers personal and professional desires and interactions impact their daily practices. The nurse oncall that night still had a full shift to complete the next day. Knowing that she would not be compensated for any of the work she did during the night shift, and that another nurse was available who had not yet worked that evening, it was not surprising that she had no

PAGE 260

260 desire to return to the hospital. No matter how many surgeries the onca ll nurse and doctor perform in a given night shift, none of this time receive s any kind of compensation whatsoever. Working in the surgical theatre would not bring her any closer to her personal and professional needs and desires: paying the school fees fo r her children investing in her business ventures paying for the bachelors degree in nursing that she hoped one day to complete As a doctor living on site at the hospital once told me, there was such a shortage of doctors and nurses at Kiunga, it was n ot possible to live on the compound without implicitly agreeing to the additional burdens of extra work whenever asked. The only thing one could do is make oneself unavailable to the hospital not answer the door, have a defunct phone, vacate the compound. There was no incentive to put in additional effort if the government was not going to provide any compensation for it. As the above excerpt illustrates, the lack of compensation of workers not only impacted those individuals who worked without financial ga in, but also adversely affected other employees who were struggling to provide services amid limited resources, and of cours e, the patients in need of care. This is a potent example of the ways that bureaucracy and biomedicine are intertwined within the hospital, and some of the effects of this intermingling. When one does not operate as expected, the other often suffers. The consequences can be potentially rife with anguish for hospital employees, and indeed possibly deadly for patients. Professionalism, Compensation, and Concurrent Governance Within the hospital, there is one overarching issue that acts as an equalizer among all ranks of staff except the top administrators, regardless of ones social capital within the community : that of adequate compens ation. Regardless of the amount of influence any individual staff member wields through parallel and hybrid orders none of the regular staff at the hospital is able to circumvent the dictates of the official system in terms of salary and the largely obsc ure means

PAGE 261

261 that it is distributed in practice.105 Across the different ranks of workers, the most common complaint was the inadequacy of salaries given t he staffs aspirations and responsibilities both professionally and domestically. Workers had to negotiate between personal and professional needs and desires within a context of scarcity (cf. Anders 2010; Blundo and Le Meur 2008; Blundo and Olivier de Sardan 2006; Martin 2009), and criticism regarding pay was often articulated as a matter of their rights (hak i ) as professionals, to adequate and timely compensation (see also Iliffe 2002) These claims to adequate compensation based on ideas of professionalism are not new. Indeed, since independence in Tanzania (as elsewhere in East Africa), medical profession als have struggled with the government for pay increases often to a state insensitive to their material needs. Prior to HSR, Tanzanian physicians argued that if the state would not pay them adequate salaries, they should at least be allowed to practice pri vately. John Iliffe argues that the AngloAmerican model of professionalism emphasizing professions as those occupations possessing specialized knowledge or the power to self regulate within their institutions does not attend adequately to the ambiguities of professionalism in East Africa (2002). He suggests that professionalism requires specialized knowledge, desire for power, self interest, but also, paradoxically, altruism in the provision of care. However, because African physicians (as well as other he alth professionals), have long been dependent on the state for training and employment, their struggles for adequate compensation relative to their professional status put them in direct conflict with the government. Thus, as Iliffe argues, the weakening of East African states since independence eroding public health systems, salaries, and working environments threatens medical professionalism. Martin (2009) makes a similar argument regarding east African nurses, 105 Several were, however, able to gain access to other forms of official compensation, as is discussed in the next chapter.

PAGE 262

262 asserting that ones personal and professional desires are difficult to disentangle, since as whole people, medical professionals often bring their expertise into the domestic sphere, and their familial obligations into the workplace (2009). Both personal and professional aspirations matter in terms of how health personnel provide care, and the ways in which they interact with the institution that employs them. In a complex history of doctors in East Africa, John Iliffe (2002) argues that many of the challenges faced by physicians in the region has related to their struggle to achieve professional status to be recognized for their expertise, perform their duties under adequate conditions, and be compensated accordingly. However, the professional ideals dictated by medical expertise are often in mark ed contrast to the realities they face as health sector workers and community members within an aiddependent country such as Tanzania (Martin 2009; Wendland 2010). Officially, w hile the central government creates the salary grades for civil servants106 and outlines workers rights (including those of compensation), the actual distribution of salaries occurs at the level of the District Council (DC). Following district budgets, the Ministry of F inance releases money to the DC s around the country, where the D istrict E xecutive D irectors (DED) office is expected to distribute these funds based on the pay scales and compensation regulations as defined by the central government. This shared responsibility surrounding compensation constitutes a form of concurrent governance, where public sector salaries are distributed through a series of government offices, to be transferred eventually into the workers individual bank accounts. 106 Including medical professionals, who the central government considers the same as other civil servants. As discussed below, this is a point of contention betwe en Tanzanias central government and MoH officers many of whom feel that the dangers of medical work should receive additional compensation.

PAGE 263

263 In practice, however, the reality was somewhat different. The DC often fails to mee t the ideals set out by the central government. While workers were aware of their official salary scales and the central governments policies regarding their rights to compensation, the offices of the DED and the District Medical Officers (DMO) were their only means of actually making claims on the rights dictated by those policies. As discussed in the previous chapter, the DMO and DEDs interests were not merely tied up with those of the Ministry of Health (MoH) health workers needs were but one of the myriad issues that these offices dealt with, both officially and unofficially. Yet there was an additional layer to the concurrent governance operating in Kiunga District. Resources directed at health workers do not merely flow from a chain of government offices; they also move through networks of donors, particularly relating to donor sponsored programs and clinics operating in the districtor, as articulated in Chapter 5, the enclaves of the hospital. These resources were also distributed through the off ice of the DMO. Therefore, in addition to being positioned at the confluence of MoH, District Council (DC) and powerful figures within the local community, the DMO was at the helm of local articulations of donor policies and programs as well. Concurrent governance collided in the office of the DMO. As discussed below, managing these various government and donor resources was a precarious exercise indeed Scarcity was not evenly distributed throughout the hospital. Certain departments were areas of abundanceabundant technologies, expertise, and, indeed, compensation. This meant that certain hospital workers benefited where others did not. With the DMO and his administration positioned at the convergence of these resources, possibilities for accusations of co rruption and partiality abounded.

PAGE 264

264 The main grievances regarding salary and compensation were expressed along four major themes: the inadequacy of pay, delays in receiving salaries and back pay, the denial of additional compensation such as extra duty (over time) pay, and the allocation of donor and government sponsored workshops through which employees might supplement their incomes. These issues, particularly with relation to extra duty pay, were often spoken of as violations of their rights [ haki zetu ] as medical workers. A n interview or conversation could transition very quickly from praise of the central government for improvements to the health sector overall to utter frustration and outright anger at the ways that their needs as workers and medical pr ofessionals were repeatedly ignored. Reflecting on the ambiguities of ART provision in central Mozambique, Ippolytos Kalofonos pointed to an economy of scarcity, where food aid was made available to the HIV positive clients of AIDS clinics, but in insuff icient supply to distribute to all people on anti retroviral treatment (ART) (2010:371). As Whyte and others (2010) illustrated in Uganda, externally funded HIV/AIDS programs provide opportunities for increased pay through allowances and new knowledge thro ugh workshops and training seminars potentially leading to future promotion (see also Anders 2010). Similarly, the plethora of resources directed at particular departments within Kiunga District Hospital enacted an economy of scarcity surrounding the biome dical and professional possibilities that externally funded programs represent. Ironically, while workers were satisfied with the MoH and changes that reforms had brought about in terms of the possibilities for improved health care provision, donor sponsor ed initiatives and the resources that traveled with them acted as a daily reminder of the failures of the government to provide health workers that to which they felt entitled: adequate pay, upgraded training, and extra duty allowances

PAGE 265

265 The issues of inad equate compensation, delayed pay, and extra duty allowances are discussed in turn below. In the pages that follow, I show how dissatisfactions with the state are often articulated in terms of rights as government employees, in relation to personal and prof essional needs and desires. Frustrations relating to these compensation issues are largely directed at the local government, and the DMOs office. Following the outline of issues surrounding compensation from the government, I move to the issue of donor fu nded worker allowances and workshops one of the only remaining means by which hospital workers can officially supplement their incomes. While the resources and opportunities tied to donor sponsored interventions provide many benefits to both health profess ionals and patients, these opportunities also created an economy of scarcity, with potentially problematic results. In the interest of understanding the whole person as Martin (2009) suggests, the contentious aspects of compensation are then contextualized in terms of socio economic changes in Arusha more generally, and the ways that these transformations have instilled in many individuals a variety of desires that can be understood as neoliberal in character Inadequate Salaries While HSR brought about i mpressive changes to the supply of essential medicines and some basic equipment to health facilities the condition and necessities of government workers themselves, as under colonialism and the Ujamaa system, was largely ignored. When HSR was planned in t he 1990s and later implemented in the early 2000s, there remained little improvement in this regard. Amid increasing scholarly and popular concerns (and criticisms) of medical migration out of Africa (see Astor et al. 2005; Chikanda 2005; Conell et al. 2007; Eastwood et al. 2005; Grant 2006; Hagopian et al. 2005; Iglehart 1996; MacIntosh et al. 2006; Ogilvie et al. 2007; Raufu 2002; Sullivan, Dilger and Garcia 2010), ironically few initiatives within HSR and subsequent policy documents focused on the needs and desires of health

PAGE 266

266 professionals, even though the MoH imbued its workers with particular rights. Tanzanias recent economic history further complicated the issue. In 2008, the government announced salary increases for its employees. However, according t o an officer within the Health Sector Reform Secretariat of the MoH, inflation and increases the cost of living still meant that despite the pay raises, salaries were below where they should ideally be: We are aware that with the inflation that we have had the purchasing power differs, [and] the pay [increase] will not be much visible. So the pay has increased in nominal terms but the purchasing power is less than the previous. During that time for example when we started the reforms [1994] it was 1 US dol lar to 370 T shillings. Today is 1 US dollar to 1700. So if the inflation here has got reach to that level how many times will you increase the salary to meet that inflation and secondly the food prices have gone up, the fares have gone up, etcetera. (Inte rview, 7208, English) S alaries prior to the 1980s were adequate to support families, and few people engage d in informal income generating strategies in order to make ends meet However, since th e 1980s and increasingly since the 1990s, the actual buying power of health sector employees, and their counterparts in other civil service departments, dwindled due to civil service reform, increased cost of living in Kiunga District, and the overall infla tion since the introduction of structural adjustment progra ms (S APs ) The staff at Kiunga District Hospital found it increasingly difficult to make a living. As medical professionals, they achieved a higher degree of education and expertise than people working in nonspecialized employment, yet they felt that thei r salaries were not commensurate with the status and dangers of their profession. The MoH officer mentioned above acknowledged this issue, pointing to the fact that while health sector work was highly specialized and very dangerous, the central government s policy was to pay all civil service workers the same way, without special consideration. While the MoH and its Development Partners (DPs) were attempting to devise a system to circumvent the policies of the government by providing Pay

PAGE 267

267 for Performance i ncentives to health workers, by December 2008 this model of rewards for good performance had yet to be instituted at Kiunga. M any of the staff at Kiunga talked about struggling with their inadequate salaries given the difficulty and dangers of their work. Indeed, the main complaints I heard were about struggling to pay for family obligations: NS: I heard that you will receive increases in your salary. Nurse Rehema: There is nothing! They only gave us 20,000 there! If they give you twenty more, you have hea lth insurance fees taken out. You have [union fees] taken out. T hat is, within that 200,000 [per month] that I get there is nothing left! I had to take out a loan, now I had to take out [another] loan to fulfill needs at home. Now I must pay on the loans, it is cut out of my pay and I am left with only 100,000, even except it isnt 100,000. I have children that go to secondary school and one that is about to start, you are required to pay for the child around 500,000 [per year] so tell me, 500,000, ca n you sign for that? What if you dont have it? You dont even have the mind to understand how to save money!107 Regina described s imilar difficulties: Regina: And then in your salary you have to pay for sugar, for cooking oil, you see eh? So you end up fi nishing all your money before the end of the month and in the end you are forced to take out loans. Eeh. Or you can find that your child is away from school and you are working to try to pay the school fees and then suddenly the child is returned home from school because you have not yet paid all the fees. That salary is just not enough! (Interview 11 1108, Kiswahili) Rehema and Reginas struggles with their salaries were echoed in other interviews, particularly among the lower ranks. While lower ranks had considerable ability due to their political and social ties to negotiate the obligations of their workplace and influence decisions within the administration, due to central government policies governing civil service salaries, they were unable to have a voice when it came to compensation. The struggles with inadequate pay were 107 All values stated are in Tanzanian shillings (Tsh). The exchange rate at the time of the interview was approxim ately 1178 Tsh to US$1. Her pay was the equivalent to about US$170 per month.

PAGE 268

268 further complicated by the delays experienced by several employees in actually receiving the pay that they were owed. Delays in Salary and Promotion As workers within the Health Sec tor, there were two major events that would precipitate significant delays in receiving ones salary: promotion to a higher pay grade, or transfer to a new health facility .108 When new staff came to Kiunga District Hospital to work, while the DEDs office pr ocessed their employment paperwork they were not paid. It could take several months before their salaries arrived. Salary postponements caused considerable hardships for those employees who had to set up new households in the district unless they had another source of income or savings to help them get by. Delays averaged six months.109 During the entire time that their salaries were delayed, the employees worked their fulltime hours with the promise that they would eventually be paid arrears. However, in pr actice, when the DED office finally processed employees paperwork, staff would start receiving their regular salary, but waited on average over a year to receive their back pay. Some employees waited several years without receiving arrears. With the conti nually rising inflation rate, the buying power of salary arrears continued to dwindle and when back pay was finally received it was paid without interest The difficulties of the first months of work at the hospital were echoed across all ranks of worker s. A nurse midwife, Neema, reflected on the differences between the years before and after HSR in terms of salary delays: Neema: In previous years [salaries] were followed up, and that year when we started work the salaries did not give us problems like now. NS: So now there are more problems? 108 This kind of delay occurred for existing government health sector employees who transferred into a facility in a different district, as well as for employees who migrated from the private sector into the government sector. 109 The average articulated by all respondents who I asked

PAGE 269

269 Neema: That is, there are problems that even an employee, until your salary is increased you stay legitimate [ ukae kihalali ] only because of the job that you have. So the salary that you were transferred with, aah, i t takes a long time; even me, my salary, they have harassed me a lot [ wameninyanyasa sana] about my salary! You, as a person that has been an employee for years, for now a nurse assistant is arriving at [a salary of] nearly 200,000! Me, I only approached 100,000 until they came to adjust it. Now I arrive at 152,000. And this is basic [salary]! Because part goes to a loan, so I have 150,000. Now, you tell me, what kind of employee having needs, having a family, if you arrive at 150,000 will you be able to feed your family? NS: But before the changes in the health sector, it was not like this? Neema: In the past in such it was that as a person you were paid your salary and you knew completely [ kabisa ] that this is my salary! You see, unlike now. N ow if I am paid my little salary still I know that it is not my salary, still they are harassing me with salary. But in the past you found through the years when your salary was increased . if it was increased you would find that it [the increase] was t here. It was not bad like now. Since 2003 I have not yet been paid arrears up to today! (Interview 12 108, Kiswahili). Neemas comments reflect frustration not only with the inadequate salary she receives as a nurse midwife, but also the fact tha t lower ranks are paid more due to bureaucratic problems in properly adjusting her pay grade. She was upset that her pay did not reflect her expertise and she still struggled with meeting her familys basic needs. Meanwhile, her rights as a worker were bei ng violated, as she was owed five years worth of arrears. In addition, those employees who had been in the government health sector since prior to the reforms felt that they experienced longer promotion delays than those who entered the sector after HSR. According to Upendo, a general nurse working in the Counseling, Testing and Care (CTC)110 clinic, For people from the old system, promotions are very delayed while newer employees get them quicker. We [from the previous system] have to keep writing letters for it (Interview 11 2008, Kiswahili). S alary delays were especially difficult for those employees who started working at the hospital only recently of which there were many. During the first several months working at 110 This clinic provides services related to HIV/AIDS

PAGE 270

270 the hospital, new employees would work full time without any compensation whatsoever. Based on policy under decentralization, the Hospital Secretary111 is supposed to act as the liaison between the hospital and the DED office. However, at Kiunga, many of the staff were told to go themselves to the DEDs office with their complaints about arrears or lack of promotion. For those staff members who knew people within the DC, going to the DEDs office was not so daunting. However, for others who did not have any social ties to people within the loca l government, attempting to get access to the DEDs office was exceedingly difficult. Regina spoke about the bureaucratic snags she encountered when trying to obtain her arrears. When she told the Hospital Secretary that she had waited six months to even s tart getting paid at the hospital and had not received her arrears, she was instructed to go to the DEDs office herself. However, after repeated failed attemp ts to gain an audience at the DEDs office she gave up. She then returned to the Hospital Secretary, but was simply told that she should be patient. Regina: So you see even this breaks your spirit. You wait six months, six months is a long time and the money will help. You want to send your child to school. If you find that you have worked for six m onths in this difficult situation, you see that you will be paid but you do not know when, and you are told just that you will be paid and even until you are an old person you are not paid! These are problems. NS: So when did they say you would be pa id? Regina: They have not yet said. And we have already filled out the forms, to show when we started and what the salary should have been and that we should be paid with those funds. But still we are not paid. NS: So when did you fill out the forms? Reg ina: Last year. NS: And have you already returned to the DEDs office? When was the last time? 111 This title is capitalized because it is an official office; the Hospital Secretary acts as an of ficial liaison between the DED and the DMOs office. The DMO has his own personal secretary, but this is a general administrative position and not an official office.

PAGE 271

271 Regina: I wanted to go but I have not yet gone because I keep being told [by the hospital administration] that I will be paid. NS: So you think to know the DED [personally], that would help more? Regina: It would be helpful because you know if you get there and speak to a person you know, it is different. NS: And the administration here [at the hospital] knows your issues? Regina: They know because they take your letters there to the DED. NS: And you are not alone in this? Regina: I am not alone. There are many of us. We were many who were hired last year and none of us has been paid. NS: So it is the same for everyone that started here last year? Regina: Eh he. [yes] NS: And this year? Regina: The ones who were hired this year, they havent even started to get paid yet. NS: But still you continue to work. Regina: We continue [ sighs ]. (Interview 11 1108, Kiswahili) Such delays were a heavy burden for staff a t all ranks. When they could not meet what they felt were the basic needs for their family (buying food, paying for a childs school fees), the effects were demoralizing. In interviews with those staff not being paid their salary, those who were promoted but did not see increases in their pay, and those awaiting significant months or years worth of arrears, personal and professional desires were discussed rarely. Most employees support ed not only their spouse and children, but also extended family members. There was no possibility to invest in a personal business or think of returning to school for upgraded training in light of unmet domestic burdens. Regina commute d a long distance to the hospital, taking a public mini bus every day, which was very expensi ve. Furthermore, her brother suffer ed from

PAGE 272

272 the late stages of AIDS, and Regina supported one of his children in addition to her own. She also had one ailing parent for whom she was paying medical fees. Other hospital employees stories echoed similar chall enges. Many were expected to support distant relatives in addition to their own immediate families During interviews and in casual conversation, those employees receiving their salaries often spoke of hopes for establishing businesses, sending additional children to private school, or going back to school themselves. Conversely, those not receiving their salary or awaiting arrears described the anguish of taking out loans to cover basic needs, of their children being returned home from school due to unpaid fees, of waiting when the very offices that were the sole gateway to the money they earned continually told them to be patient. The bureaucratic bottleneck affecting timely compensation caused considerable anguish. Extra Duty Allowances One of the most contentious issues surrounding employee compensation at Kiunga District Hospital was the payment of extra duty allowances. Extra duty allowance was a type of overtime pay, to which hospital workers felt they had a right.112 According to some staff members wh o were at Kiunga during the reforms, by 2002 employees began receiving extra duty allowances when they worked oncall or extra shifts due to staff shortages. However, according to some workers, in 2006 extra duty pay stopped completely,113 with no employee having a clear 112 Having heard the repeated use of the term our rights [ haki zetu ] with relationship t o extra duty pay in the interviews, I finally asked during one of my last interviews at Kiunga how workers came to know their rights. The nurse responded that once or twice a year the nurses of the district are brought together for a meeting, where they re read their rights as workers. In addition, all health sector workers were part of a workers union (although the union under which they fell also changed under HSR yet another considerable source of tension). Given the ubiquity of the term rights applie d to the problem of extra duty, it appeared that the staff had multiple sources of official information on which they could base their arguments regarding their right to this kind of compensation. 113 Two informants mentioned receiving extra duty pay in the years prior to 2006. However, the remaining informants did not mention having received extra duty pay in years prior to 2006, and unfortunately the salience of the extra duty pay issue only came to my attention near the end of my study. Thus, I was unabl e to follow up in hospital records or with previous interviewees to confirm whether or not extra duty had indeed been allotted to staff prior to 2006.

PAGE 273

273 understanding as to why they were no longer receiving these funds Frustrations relating to extra duty pay were exacerbated when they learned that colleagues at other government hospitals, including at Mount Meru Regional Hospital in Arusha, reported receiving extra duty pay. According to one doctor114: We are told the money from the government has not come to pay us for the extra duty and when it does come they are really small payments. With the extra duty there are problems. So about the question of doing work, we are dependent on our salaries alone, and they are not enough. When you do extra times, at the end of the month you find the payment is not there. Life here is really high [ maisha ya hapa yako juu], its very expensive! So about lif e, here it is high, and then we find that we do not see the money, so we do our work, extra duty but we are not paid. (Interview, 2008, Kiswahili) During 2008, on repeated occasions at morning and monthly staff meetings, those workers not fear ing retaliat ion from the administration articulated their right to receive extra duty allowances. On many of these occasions the administrators responded with hostility towards the staff or reiterate d what was stated during previous meetings: the money was not avail able to pay extra duty allowances because the District lacked the funds and they should continue to track extra duty hours and send letters to the DEDs office. One doctor described why it was difficult to express concerns about salary and compensation at monthly all staff committee meetings : Doctor: Me, myself, I cannot bring my agenda there. Because if you bring it there it is crushed a lot. That is, if you bring your agenda there, the leaders are not very good because they are truly very abrasive they cannot answer you well, they discourage For example, I remember, that is, after seeing the other day I remember [a doctor] asked about extra duty He spoke about things that are bad, the leaders were really angry and the DMO was not there. Later they [the leaders] went to the DMO and the DMO spoke a lot, that it is not the right of a worker to claim extra duty it is not that you work and then you get extra duty. Here it is not like that! 114 Given the potential for the comments in this section to put the informants at risk, I have omitted pseudonyms in order to avoid indicating the sex of the speaker, and further use more ambiguous professional terms, such as doctor (which could refer to a CO, an AMO, or the DMO) and nurse (which could refer to an attendant, a nurse midwife, a n urse o fficer or the Matron). Finally, I have omitted the dates when these interviews took place although all occurred in the latter half of 2008 In this way I hope to obscure as much as possible the identity of the people speaking while allowing their concerns to be voiced, particularly given the fact that they have been largely unable to have these concerns heard beyond the hospital administration and the DEDs office

PAGE 274

274 You stay there, you sta y there and you will be paid! For exa mple, a worker wants to be given her rights, for example if it is extra duty that she is given it. That she has done her work she be given her rights! Now, to tell that worker that there is no money and then to scold her, it discourages her. (Interview, 20 08, Kiswahili) Further, given that extra duty pay was an item on the district budget every year, the staff did not understand why they did not receive it During an interview, I asked one of the informants why the staff did not take extra duty allowances up directly with the DEDs office The hospital employee replied that all of staff complaints stopped with the DMO and the CHMT, and that they were not allowed to go to the DEDs office to follow up. When I brought up that I heard the DMO tell the staff to go to the DEDs office to find out about it themselves, the informant replied: He was saying that for emphasis. That we think that the money is here but that we are not getting paid. Maybe in terms of coordination it is not there, but we will think it is p art of the policy [to be paid extra duty]. But the administration [ shakes head and clicks tongue ] he said that because [someone] said that at Mount Meru Hospital [they] were paid extra duty but here it is surprising that we are not and he said that w e have not yet been given a good explanation. We are told that the OC those O ther Charges they come but the rates have been reduced. So even if it comes, to pay extra duty, maybe, if it comes, maybe, but the amount is very low so it is not enough to pay the extra duty. But me, my own opinion, I think that even if it comes only a little, at least give them [the staff] the little that comes. Even if the amount is small, at least they will stay, give it to the workers a little at a time, they will stay! (Int erview, 2008, Kiswahili) Particularly for the staff that worked night shifts or were on call, the lack of additional pay for the extra work had an adverse effect on their morale, not to mention the quality of patient care. One doctor spoke about the effect s of low compensation on staff morale: People dont work with harmony and no pleasure T hat they do their work and that at the end of the month they will have their money, their extra duty If you work in that kind of environment you know that you will be paid, that tomorrow you can send your children to school and pay the school fees, because you get your extra duty and you can eat (Interview, Kiswahili). This quote suggests that attention to the personal and professional needs and desires of hospital w orkers has an

PAGE 275

275 important influence on the quality of patient care health workers operate without pleasure in their work, because their needs and aspirations are respected and they can concentrate on patient care. Without this attention to workers needs, pa tient care suffers because there is no satisfaction in the work. The personal contexts of health professionals impedes on their professional life (see Anders 2010; Martin 2009). In the context of daily practice, personal and professional subjectivity were deeply intertwined (Martin 2009). Similarly, another doctor felt particularly desperate given the amount of work and the lack of financial incentives: I am still irritated because the rights of people are still not followed. There is no extra duty pay, no risk pay, nothing. I have thought about running away from my work many times but I have nothing to run to. I do my work with problems, but I still do it. I am tired and demoralized I live in poverty, we keep hearing that the money [for extra duty pay] is not there, but we do not understand! (Interview, Kiswahili). It was common when interviews turned to issues of workers rights and compensation to speak in terms of immediate domestic burdens: food, poverty, the financial capacity to educate ones childre n. While many health professionals wanted to initiate or sustain a small business or return to school themselves for additional training, these dreams appeared too far from the realities they faced. F ew workers at the hospital were bold enough to continua lly i nquire about e xtra duty pay. Many worried that if they agitated the administrators continually, they might be transferred to another facility. This fear of transfer was largely because of family obligations caring for their extended family members or be ing close to their childrens schools A transfer may result in potentially working in a facility with fewer resources than Kiunga, making daily routines more difficult. Significantly, as described above, it also carried the hardship of living with salar y delays for several months at a time when those funds were necessary to support extended family members. There were some hospital staff members who took the risk of fighting for the rights of

PAGE 276

276 the workers before the administration in morning and committee meetings, but most stayed silent. In 2008, I only saw one highly vocal staff member get transferred. When I asked that employee about how she felt about the transfer, she expressed happiness at no longer having to deal with such an abrasive administration when it came to worker rights. But for her too, this transfer presented very real financial, professional, and personal hardship. Overall, when it came to the workers rights relating to compensation, many of the workers were demoralized and this attitude was most reflected when it came to the incentive of many workers to go beyond the minimum required in their job. Many felt caught in a system in which the unofficial order favoring seniority the official system of inadequate pay, and the murkier issue of rights to additional compensation such as extra duty gave them little incentive to do more than the minimum necessary. Of course, the lack of incentive to work created particular challenges for In Charge Nurses, who, as in the previous chapter, struggled to fill in the gaps and continue to care for their patients In addition, another problematic aspect of the relationship between health workers and the state was that the Council Health Management Team ( CHMT ) wa s the main body representing the hospital to the MoH. Therefore, when the MoH sent representatives to the hospital to review the facilitys performance, they bypassed the hospital staff and the H ospital M anagement T eam (HMT), consulting directly with the local governing body that was charged with fulfilling MoH dictates: the C HMT. According to an HMT member: A few days ago there were representatives from the Ministry [of Health] who came here and had a meeting. But who did they meet? Those people in the CHMT alone. CHMT alone! Now, the leadership [i. e. the MoH representatives] comes to CHMT, people cannot know anything about the problems we have here by just talking to them. We have no idea if the CHMT remembered to tell the Ministry about our problems like extra duty. It is possible that they [the CH MT members] were asked, but they could have said we were getting it because they did not want to appear to be causing any problems or that there were any problems here. They do not want

PAGE 277

277 to. They would say yes we are paid. But if the leadership from the Ministry comes for a whole day here or a whole week in the hospital, if they asked the workers they would find out that they are not getting paid their extra duty. So they could see that the policies are not really being followed and that is why the workforce is not fresh (Interview, Kiswahili) While I was not present during the meeting between MoH representatives and the CHMT, it would not have been surprising if the CHMT members did not mention the extra duty issue during their conversations. The DC and the CHMT had a lot of incentive to represent Kiunga to the MoH in particular ways; this economy of appearances was a high stakes game. District health departments found incompliant with MoH policies and guidelines could have their federal funds stalled or rev oked. District offices had to demonstrate that they conformed to all central government policies and guidelines as best they could (see Chapter Eleven); particularly impressive districts might receive additional funds from government or donor coffers to su pport locally relevant capacity building projects. Indeed, as the HMT member above describes, in relationships between the central and district government offices, there was little incentive to articulate problems, particularly as related to individuals n eeds. It was better to appear to be making a good effort, in order to entice the central government to invest further funds in the development of the district. The situation of the MoH consulting directly with the most powerful actors at the hospital in l ieu of speaking to directly to the workers effectively impeded one of the workers few opportunities to do anything to change their situation. The hospital was officially under the authority of the DC, and the DED paid the staff and approved their allocati ons Meanwhile, the hospital was required to follow MoH policies. This gave the hospital staff two masters : the DC and the MoH. At first glance, being accountable to two government bodies presented two possible avenues through which staff might voice griev ances. However, their main liaison to both the MoH and the DED was the same: the CHMT, and the DMO in particular. When it came

PAGE 278

278 to worker compensation, regardless of know who, no one was well positioned to negotiate. Financial flows from the central to the local government were carefully managed. When it came to building the hospital and instituting new programs (particularly in line with the administrations institutional desires) the CHMT and the DMO were very supportive. T he DMO, reliant on the DC, was re ticent to fight for employees rights. W hen it came to fighting for what workers saw as their individual rights [ haki zetu ], conversations were completely restricted or avoided. While ties to the local community could provide a hospital worker with conside rable influence largely allowing them to circumvent the official hierarchies in their everyday practices when it came to salary, all staff members were equally unable to negotiate with the state. Political and social ties had their limitations. There was, however, one potential avenue beyond the state through which individuals might access resources unavailable to them: donor sponsored programs and clinics. It is with the establishment of hospital enclaves (see Chapter Eight) that an economy of scarcity was enacted within Kiunga. It is to donor sponsored resources in the hospital that I now turn. Working Between Abundance and Scarcity : From State to Donor Resources Exploring an AIDS program operating within a government facility in Uganda, Whyte et al. suggest that externally funded health interventions have three important practical impacts on clinical staff (2010). Firstly, they increase the responsibilities and burdens on work without making available additional staff to assist. Secondly, they potentially provide financial incentives. Thirdly, they offer prospects for training workshops and seminars. Yet Whyte et al. point to a forth dimension: the technologies and resources that externally funded health programs instill in health clinics incite new discou rses, hope, and ideals among health workers, as well as particular dilemmas. In this section, I elaborate how hospital staff maneuvered the potentialities of externally funded health programs in light of their inadequate and often delayed salaries.

PAGE 279

279 However the possibilities of donor sponsored health initiatives within government facilities also created particular dilemmas. One such dilemma was determining how to distribute resources among a hospital staff that was financially burdened, that craved updated medical training and that had hope and faith in the kinds of technologies and routines made possible within externally funded departments of the hospital (van der Geest 2005; Whyte et al. 2010). For hospital administrators, the dilemma was how to distribu te resources in an economy of scarcity; for the facilitys employees, it was how to position oneself to access them. The Ambiguous Benefits of HIV/AIDS Work Donors provide resources to institutions and the actors that work within them that are otherwise extremely difficult to access. One of the many benefits offered by donor sponsored programs related to training and monetary compensation. As part of the anti retroviral (ART) scale up, the Kiunga District Hospital administration signed a USAID agreement re garding salary supplementation for those government workers who would be operating the CTC program. The document states: The issue of salary supplements [ sic.] remains sensitive and a matter of both congressional and agency concern. It is USAID policy that salary supplements should be considered an exception to normal USAID practice requiring exceptional justification. Among the justifications for providing salary supplements was the following: the cooperating institution had to show its inability to pay these allocations within the time that the activity was underway, The supplements are judged essential to the achievement of project or program objectives; and The rates and fees paid are in accordance with local standards and are limited to amounts re asonable in relation to an employees pay and, for continuing program, in

PAGE 280

280 amounts which the host country entity could be expected to meet from its own resources within a reasonable time.115 Having met the requirements of the USAID salary supplementation guideline,116 the action plan for the CTC clinic included a variety of budgeted allowances that would be paid to staff for particular activities: overtime allowance, extra duty allowance, travel and per diem expenses, quarterly CTC meeting allowances, and mont hly CTC staff meeting allowances. REFLECT117 also provided funds for hospital HIV/AIDS sensitization seminars, conducting meetings with community leaders, and quarterly meetings with associations of PLWHAs According to Peter, a REFLECT officer, CTC staff we re making from US$80 up to US$200 per month extra from working within the clinic all this in addition to their regular government salary. Peter and his colleagues were well aware of the potential dilemmas associated with the monetary supplementations REFLE CT provided to particular health workers. Peter stated that many of the facilities REFLECT funded were short staffed and already overburdened unable to deal adequately follow REFLECT protocols due to lack of human resources or to handle the ethical dilemm as that ensued when workers in one department were paid for services for which they were not compensated in others. He noted that when HIV/AIDS clinic workers were asked to meet for duties not related to the CTC clinic, they were not paid, causing addition al challenges for the hospitals in motivating the staff to meet on other issues. CTC staff was also paid extra 115 Quoted from USAID Policy guidance on criteria for payment of salary supplements for host gove rnment employees, from hospital files at Kiunga District Hospital. 116 REFLECT argued that salary supplementation was critical to the success of the CTC program, since it was felt that hospital staff would not attend the required weekly, monthly, and quarte rly meetings for the CTC clinic unless they were provided with financial incentive. 117 It will be remembered from Chapter Five that REFLECT was the American NGO who received Presidents Emergency Plan for AIDS Relief (PEPFAR) funds to operate HIV/AIDS clin ics within Kiunga district.

PAGE 281

281 duty allowances for overtime, but as described above, workers in the rest of the hospital were also doing extra duty, without compensation. Seve ral of the hospital staff ( including the CTC nurses, four AMOs, and several administrators ) attended the regular weekly and monthly meetings that REFLECT required. For each attendance at weekly meetings, the staff received an allowance of 5,000 Tsh, and for monthly meetings, they were later reimbursed a 20,000 Tsh allowance.118 The uneven distribution of resources among the wider hospital staff made some of the CTC personnel uncomfor table. According to Marjorie, who worked at the CTC: For example, extra duty we get it because of these donors. But for example, the extra duty for the people who work in other departments, it is really hard. It is really hard. That is, they have not been paid in many years! Eeh. So there is a problem. They say that they have not been paid, not even a little bit (Interview, 111808, Kiswahili). F or the staff at Kiunga the fact that CTC personnel were being paid for overtime hours was not generall y seen as problematic in itself; it was yet another reminder that the overall staff at the hospital was entitled to this pay, which the District Council and hospital administration consistently failed to provide. In this sense, the donor enclave of the CTC exacerbated existing worker dissatisfaction with the inadequacies of monetary compe nsation for their work. The Desire for DonorSponsored Workshops Along with all of the resources it bestowed on the CTC clinic, REFLECT was one of the few providers of regular training workshops for the staff. Training was highly desirable for all levels of the staff for a variety of reasons. Workshops upgrad ed skills and expertise, potentially lead ing to future promotions or additional training. Furthermore, workshops provided very lucrative per diems and travel allowances for participants. Staff members could use payment 118 In 2008, 5,000 Tanzanian shillings (Tsh) was worth approximately US$4.20, and 20,000 Tsh was worth US$16.75, which was more than a days wage for most nurses.

PAGE 282

282 from workshops to build their familys future: school fees for children, additions to a house, investments in small business opportunities. Finally, workshops provided a break from the tedium of daily work with too many patients and not e nough resources. The significance of these additions to the staff salaries were highly significant (see also Anders 2010; Martin 2009; Jaffr and Olivier de Sardan 2003): Rehema: T he other day I completed the PITC seminar.119 There I received 200,000 shillin gs and with those 200,000 you can eve n send your child to school. If another person over there in a seminar got 225,000200,000 you sent pap! [claps hands together]. Twentyfive left to buy food for the home, aah you can see how the month flies by! Y eah!120 REFLECT provided a variety of regular workshops, including training on guidelines for PITC and P revention of Mother to Child Transmission (P MTCT ) In these workshops, health workers were shown through role play how to properly consult with a patient regarding HIV how to be sensitive to his or her needs, listen, and demonstrate empathy. They practiced how to tell patients that they were HIV positive, how to instill them with hope and encourage them to bring forward their spouse and children so that the y could be tested and potentially enrolled into the program. Those who received the training knew protocols for proper counseling and testing for HIV very well. Realizing the potential impact of such workshops on staff morale, capacity, and income, Dr. Sa idi, the DMO, advocated spreading donor and government sponsored workshops widely among the staff: We cared about the employees so that when they saw patients they gave services. We gave them their allowances, cared about their going to training so the e mployees have really changed! (Interview, 120308, Kiswahili). The wide allocation of donor sponsored 119 Provider Initiated Testing and Counseling. This is a program t hat helps medical practitioners to be able to identify patients who potentially are suffering from HIV related afflictions, and provides them with training on how to counsel these patients about the importance of getting testing and the availability of ser vices. 120 Pap was emphasis to indicate speed at which she was able to send and use the funds.

PAGE 283

283 workshops was one of the means by which Dr. Saidi attempted to spread the boundaries of the donor enclaves that had been established at the hospital (se e Chapter 5) It was also part of another economy of appearancesone that would ensure that hospital workers had pride in their workplace and invested in his vision of the hospitals future. However, in an economy of scarcity where inadequate resources are available to allow for equal distribution (Kalofonos 2010), distributing the benefits of donor enclaves could be a precarious exercise indeed. Even by the accounts of staff working in the lower paid jobs at the hospital, such as nurse assistant Rehema, Dr Saidis efforts to distribute training widely did not go unnoticed: Rehema: [Before the current administration], those in charge, they would leave you in the wards, we were left in the wards with the patients. You are humiliated with the pat ients while they go to seminars. Now, when [the DMO] saw this, he saw that this wasnt rights. Why does this person not go and yet you go [to seminars]? (Interview, 11 2708, Kiswahili) Indeed, the DMOs willingness to distribute donor sponsored seminars among even the lower ranks of the staff, as well as the rapid infrastructural transformations he had overseen since his arrival in 2004 (see Chapter Five), created a widespread sentiment among the workers at all ranks that he was truly invested in the development of the hospital. Even some of the most fervent critics of the administration when it came to worker s rights spoke favorably about the DMO with regards to how he built the hospital and elevated staffs capacity to provide meaningful health servicesthe e conomy of appearances that Dr. Saidi established within the hospital was largely successful, even for his most oppositional employees. Unfortunately, this success was short lived. Not all training opportunities were created alike. Some seminars only laste d a few days, whereas others were up to two weeks long. Not everybody felt that the administration was allocating seminars fairly. Staff who, as above, attempted to claim their rights to compensation

PAGE 284

284 through the DMOs office felt that they were being denie d more advantageous training opportunities. One doctor who was often vocal about compensation rights stated: A person who is really bothersome [ anasumbua sana], he is only sent to seminars that last two or three days rather than the ones that last a week o r two. So he will get something new in his head but even income he will not get it. The people who have cars here, people can get compensated salary [through workshops] until they are able to buy a car! Right? Or they can be compensated and ask for another seminar. Those kinds of [long] seminars, you Noela will be given those seminars because you are not a nuisance [ usumbufu]. But me, even as a nuisance I do my work. [Interview, Kiswahili]121 This is a telling example of the limits to the ways th at health personnel can make claims on the state within the hospital. This doctor continually challenged the hospital and district administrators, attempting to claim what he felt were workers rights for the hospital as a whole. Yet because the same admin istrators were responsible for allocating donor sponsored workshops and benefits, they could prevent him from accessing those rights through donor sponsored programs like the CTC. Seeing his coworkers able to access extravagant resources (such as a car), t his doctor felt considerable resentment for some of the staff, because he was trying to fight for their rights. This doctor was not alone in his sentiments. B y mid 2008, the staff fondness for how workshops were being distributed was beginning to wane. Ne gotiating Resources, Rumors, and Accusations: A case study The sheer amount of technical training, resources, and income supplementation that workshops provided caused a variety of tensions to surface among Kiunga staff. While the administration attempted to appear to be equitably distributing seminars by the middle of 2008, accusations of favoritism began to surface. The established economy of appearances meant to garner staff investment in the hospital began to erode. In June, at an all staff committee m eeting, a CO stood to present a previously unannounced agenda item. In front of the entire staff, she said 121 Date and rank of the interviewee have been omitted to maintain anonymity.

PAGE 285

285 she felt that some people were favored and chosen to go to several seminars, and that they used the training as a way to increase their income. It was unfair for only a few select people to have access to extra training and extra pay. It was also her right, she argued, to be chosen to go to a seminar because she, like the rest of the staff, was also poor.122 This proclamation instigated a lot of murmuring among the staff. The vast majority of the staff mistrusted the District Nursing Officer (DNO) Mary, who was in charge of allocating workshops. Mary stood at hearing the accusation and countered that some workshops were only meant for the personnel who w orked in a particular clinic. She said that some people were sent to workshops more than once because particular issues could not be taught adequately in one session. Marys retort precipitated even louder murmuring among the staff, and she started to get defensive. Dr. Saidi saw that problems were escalating, and he quickly walked across the room and whispered something to Mary. Mary sat down. The DMO then attempted to ease the tension by explaining that workshops belonged to everyone, that people would be given workshops gradually, and that some workshops were for particular individuals due to where they worked. As the agitated staff continued to mumble amongst themselves Mary stood again: Lets discuss these problems so that we can finally be rid of them. She rationalized that many seminars were donor funded, and that as the providers of the money, the donors were the ones who had the power to decide what kinds of seminars to provide and who should attend. She suggested that donors allowed her little c ontrol. The DNOs account was only partially true, however, and several staff members were aware of this. Hospital files and interviews with members of the donor sponsored clinics123 told a slightly more complex story: REFLECT and 122 All s tatements in this section were recorded using notes during the meeting. 123 The o ther main donor funded clinic was the Reproductive and Child Health clinic, within which a variety of donor programs operated simultaneously.

PAGE 286

286 other donors providing workshops would often send an email or fax advertising a workshop, and asked administrators to provide the names of a specified number of staff to attend. Some of those workshop announcements mentioned the names of particular employees affiliated with their pr ograms to attend the workshops, while others did not. Further, based on the statements of some of the CTC staff that had been to several workshops already, there appeared to be some flexibility in terms of how workshops were allocated: Dr. Erastus: There is a seminar that is coming next week. They say its integration and issues of TB and HIV that will be done, which I have already gone to. If it comes again I will see that it resembles a lot the one from before, and maybe if I did it not many days ago, I will give it to someone else who has not gone yet. I will wait for that one that I have not yet gone to. (Interview 724 08, Kiswahili) The comments of Dr. Erastus, a CTC doctor, undermines Marys argument that some workers needed to attend the same seminar more than once due to its complexity. His comments, along with those of some of the other staff, imply that workers within donor sponsored clinics were often selected to participate in workshops, and felt able to pass on those seminars they had already c ompleted in order to allow their peers to attend and gain access to that training and the income that came with it. Of course, as is described below, staff that had already attended a seminar would not necessarily feel obliged to give up their place in a seminar for a co worker. However, in articulating the dilemma in a way that put the main responsibility on the donors, Mary was attempting to diffuse some of the tension directed towards her. The effects of the DNOs explanation were short lived. By October, 2008, REFLECT had been providing regular training workshops and seminars for around a year, while other donors or the MoH offered additional training workshops from time to time.124 While several hospital 124 There is an important distinction to be made between donor and MoH sponsored training workshops. In the case of the former, donors paid to operate particular workshops relating to specific programs or clinics they sponsored.

PAGE 287

287 employees benefited from the scaled up HIV/AIDS p rogram in terms of new training and the monetary compensation that came with it, a particular event relating to the allocation of high paying donor sponsored training provisions brought escalated tensions. Catherine, a nurse m idwife who arrived recently at Kiunga, received a letter stating that she had been selected to participate in a two week donor sponsored workshop, which was to take place in the following week at a nearby training facility. She went to the facility on the designated date only to be tol d upon her arrival that one of the nurse administrators at Kiunga took her place in the workshop. Completely shocked and dejected, Catherine returned to the hospital in tears. News of the administrators actions spread quickly among the staff: Marjorie: [T he nurse administrator] created a problem. So now, people do not like her. Because she took a seminar for one of the nurses there in the maternity. [Catherine] was supposed to go, and the [nurse administrator] took her place. She went to a two week seminar It really created a problem, and that woman went there only to be returned to the hospital. She had a letter that she was to go, so she had to cope with the fact that [ the nurse administrator] had taken her place for the entire time. And her letter with her name came from the Ministry [of Health], for her to go and do a training. [The nurse administrator] went there at a time when her name was not on this chance So you see the problem now, and people have begun to hate the [nurse administrator] (Interview, 111808, Kiswahili). Dr. Kweka provided a similar account: Dr. Kweka: Because it can happen that you and I are arguing and we go there [to the seminar], and I have a letter, why if you see my letter should I have to go back and you get to stay there? NS: But why do you think it was this way? Was it because the other person was higher [ in rank] ? Dr. Kweka: Yes, she was higher [ alikuwa juu]. It was [a nurse administrator], and the other one is a low person [ mtu mdogo]. And the issue was that it w as two weeks and the pay amount was good [ interrupted] NS: and what did the [nurse administrator] say? Conversely, MoHsponsored workshops were most often paid for using donor money from the basket fund, and were generally in accordance with the training goals set out by the MoH to be implemented nationally.

PAGE 288

288 Dr Kweka: and the money was good. E h? She has not said anything. NS: She has not said anything? Dr. Kweka: She has not said anything. And the money was good! Others have changed, we are expected to change and others have changed. We here at the hospital, not yet. We have not changed this kind of behavior. (Interview, 111408, Kiswahili) Thus, although both Marjorie and Dr. Kweka were not low rank workers, they, like their peers, were extremely upset that a senior nursing administrator would steal a workshop from a lower level nurse. Their sentiments were echoed widely among the general staff, who felt that the offense was particularly grave given t hat the seminar was a long one (two weeks), thus potentially providing Catherine with muchneeded financial support given that she had started at Kiunga recently and had not yet begun receiving her official salary. The regular full staff committee meeting took place a week later. At the meeting, one of the AMOs stood up and, in front of the entire crowd, spoke out against what the administrator had done. The AMO did not mention Catherines name. The nurse administrator was not present, because she was at th e training that had been allocated to Catherine. When Dr. Saidi heard about what had occurred, he was irate. He had been out of town and had not heard about it until that moment. The DMO stated that the person who had been affected should come to his offic e in private to clarify what exactly had happened. He reminded the staff that often, donors selected specific people for training, and a persons financial need for the seminar rarely weighed into the decision. Dr. Saidi also announced that he would take harsh steps against anyone who showed favoritism in appointing seminars to staff. However, the issue did not end with the meeting. Complaints continued among staff about corruption and favoritism within the administration, with four key administrators bein g the targets of this talk. Dr. Saidi was scheduled for his one month vacation at the time, which took

PAGE 289

289 him out of the hospital. A week after the committee meeting, an article emerged in a national paper accusing Dr. Saidi of corruption. The article, writte n under a false name, stated that he was using donor, government and basket fund m onies for his own private purposes.125 Based on formal and informal conversations with the staff during this period, feelings about the article were mixed. Many, including some of the DMOs sharpest critics in terms of his treatment of workers, refused to believe that he had been involved in taking any money away from the hospital. One of the interviewees who was critical of the DMO on other matters stated: Me, I did not feel good to hear such things about our DMO. It is those little problems of the administration that I can talk about, but in terms of what was written in the letter, I cannot believe he would do that. I cannot believe he would go there. I believe he is not able to do that. Because he has always come before the whole staff to tell us about money that comes in and announces everything officially. And a person who does that, you have to have faith in him. And he has brought many things here. He brings things here and they are used. Those things that were written were about politics only.126 Yet the article had an effect. When Dr. Saidi finally returned from vacation in December, his entrance to the hospital was uncharacteristically quiet. Prior, he was gene rally a boisterous, charismatic leader. Now, amid gossip about corruption and favoritism that he had worked so long to combat, he kept to his office and absented himself from the daily routines of the hospital. On the day when the nurse administrator returned to the hospital from the workshop, she attended the morning staff meeting, but remained silent throughout When the meeting was over, she went to her office, prepared some administrative paperwork, and then left, also on her yearly vacation. With the DMO and the head nurse administrator out of the hospital, the institution was now evacuated of its two most powerful, and previously most well liked, administrators, and 125 Specific details of the article and a citation have been omitted to protect the identity of the hospital and the DMO. 126 In order to protect the identity of this interviewee, all identifie rs and the date of the interview have been omitted. Interview conducted in Kiswahili.

PAGE 290

290 work output of the staff slowed considerably. Many staff members were late or only showed up temporarily to work during this time, and many left early. Several nurses with whom I spoke remarked about the tardiness of others ( wanachelewachelewa ). The MOI, who had recently been appointed and who was highly mistrusted, struggled with staff mor ale. On several morning meetings in the following days the MOI stood and deplored the low attendance at work ( hawapo kazini ) and the overall tardiness of many employees ( wanachelewachelewa ). H is statements had little effect on staff attendance. The MOI was now expected to fill the void in the DMOs absence, but was generally unsuccessful. Patient care suffered tremendously as staff felt that they had no incentive to work for a corrupt administration. Those areas where staff morale was lowest those departme nts falling outside of the donor enclaves were particularly hard hit in terms of the quality of patient care. In sum, the presence of donor enclaves (such as the CTC, and the programs ongoing at the Reproductive and Child Health clinic), and the resources and forms of expertise associated with them, highlighted staffs existing dissatisfaction the shortcomings of the hospital administration in terms of providing them with the kinds of training, salary supplementations, and working environments to which the y felt they were entitled. The staff felt considerable hostility towards the administrators, who at once represented the failures of the state to provide adequate compensation, and the sole means by which hospital workers could access the very resources to which they felt entitled by their profession and their status as state employees. While the MoH acknowledged the rights of workers to updated training and adequate salaries, the government lacked the resources to make these promises into a reality. In an enclaved hospital where resources are abundant in only very specific departments, Dr. Saidi could do little to assuage their disillusionment.

PAGE 291

291 Neoliberal Desires: Negotiating a Commodified World Overall, the joys and dissatisfaction s that hospital employees experienced in relation to their workplace must be understood with reference to their lives outside of the hospital, and the ir domestic burdens and personal aspirations. In Tanzania more broadly, ones personal and familial aspirations must be balanced wi th the considerable pressure of those within ones extended family or social network friends, acquaintances, neighbors to distribute the benefits of ones wealth and status widely. This system of interdependent social networks and the pressures they exert is not unique to Tanzania, and indeed has been discussed in several studies of subSaharan African civil servants (see for instance Anders 2010; Blundo and Olivier de Sardan 2006; Chalfin 2010; Jaffr and Olivier de Sardan 2003; Martin 2009; Wendland 2010) In Arusha region, I was often struck by the commonalities of particular personal and familial aspirations among the wide variety of people I encountered. Within the hospital, and among a wider network of friends living in Kiunga and Arusha, the desires people articulated were very similar: private education for at least one child in the family, building a house to provide property and a secure place to live, and start up funds for establishing a new business or expanding an existing one. Education was se en as a means to secure future salaried employment or to build expertise in business or technology. I never encountered anyone working at the hospital who did not have at least one child in private school, and very often they would ask me to help them find a sponsor to pay for their childrens education. Indeed, this was a common request outside of the hospital from friends, neighbors, or people I encountered by chance. In the years since SAPs, the conditions of government schools in Tanzania have deteriora ted significantly (see Kuder 2005). Thus, despite elementary school being largely free,127 the quality 127 It is important to note that even government education is beyond the means of many Tanzanian families, since families must provide their children with uniforms and text books extremely expensive commodities in Tanzania

PAGE 292

292 of this education is questionable, and many families even those who are poor attempt to find ways to get the necessary funds to pay for their children to go to private school. Recent studies by Martin (2009) and Anders (2010) described similar aspirations among the civil servants with whom they worked. While to Martin, the personal and familial aspirations of nurses in Uganda were middle class (2009), Anders describes the side businesses in which civil sector employees engaged as a desperate attempt to keep alive a dream of modest wealth and material security. The scope of this dream was quite bourgeois : building a house, acquiring furniture, providing th e children with a good education so that they can lead a better life and enjoy the amenities of modern or civilized urban life (2010:71. Emphasis in original). While I agree that many of their attempts were in some sense desperate, since many of the hospital workers took on loans they could scarcely afford to pay back just to keep their children in school, I wish to question the characterization of this dream as bourgeois or middle class. Rather, drawing on Weiss (2009), I wish to argue that these dreams are, rather, neoliberal For the hospital staff, the perception of rights to compensation was intimately tied to larger political economic transformations in Arusha, and Tanzania more broadly. As Tanzania adopted structural reforms and the country b ecame more open to influxes of new kinds of media, ideologies, and commodities that had not had a major presence in times prior, at the same time as employment opportunities were dwindling (see Chapter Two; Weiss 2009) According to Weiss, [t]his sudden c rash on the heels of unprecedented and exhilarating possibilities unrealized by the vast majority of Tanzanians as anything but possibilities made it possible for a broad swath of people to desire the signs and styles of a global order while finding ever narrower means by which to satisfy them (2009:9, emphasis mine). as well as the school supplies the children will need. The requirement to purchase school uniforms and textbooks keeps many Tanzanian children out of school.

PAGE 293

293 Analyzing the proliferation of barbershops in Arusha, Weiss is attentive to the imaginings and practices within the city in the wake of wide reaching and rapid politica l economic transformat ions of the mid 1990s. While he calls attention to these processes as neoliberal he is careful to note the problematic aspects of applying neoliberalism to studies of Africa without tying the term to specific processes and experiences. For his own frami ng of neoliberalism, he points to the common sense among inhabitants of Arusha from across diverse backgrounds and socioeconomic means that the economic, cultural and social transformations since the last years of the twentieth century were extremely profound. As described in Chapter Two, Tanzania underwent important economic reconfigurations prior to the formal governmental restructuring precipitated by SAPs. In the 1980s, particularly in urban and peri urban areas, a wide array of people began turning to informal socioeconomic activities in order to supplement their deteriorating incomes or to gain access to scarce commodities. Thus it was that the informal market preceded the formal liberalization of the state (Tripp 1997; Weiss 2009). Weiss argues that the informal economic activities along with the new commodities available on the market prompted a shift in value towards commodities (and arguably, the money to attain them) rather than other forms of wealth. Similarly, following Trouillot (2003),128 Comaro ff and Comaroff (2000) argue that neoliberal capitalism produces the expectation and desire for consumption of particular things, while simultaneously reducing or threatening the ability of people to achieve those desires. Meanwhile, social wrongs are tr ansposed into an issue of rights (306). Given the massive transformations in Arusha and surrounding areas, and the reduction in hospital employees ability to achieve their desires in light of meager salaries and 128 Original work Comaroff and Comaroff cited for Trouillot was from his 1999 paper presentation, Close Encounters of the Deceptive Kind: The Anthropology of the State in the Age of Globalization, a paper presented on the occasion of the fiftieth anniversary of the founding of anthropology at Stanford University, 9 10 April. This paper was subsequently published in Trouillots 2003 book, Global Transformations: Anthropology and the Modern World. NY: Palgrave Macmillan.

PAGE 294

294 domestic obligations, the issue of compe nsation became firmly tied to the workers views of their rights, which they felt the administration and the local government continually violat ed This shift towards value in commodities and the ability to participate in the markets of Kiunga and Arusha required that one possess sufficient education and financial capacity The desire for money and ability to participate occurred not only among those of higher economic standing, but also proliferated among people with few er economic means. Thus, the staff s desires were not merely tied to notions of becoming bourgeois or middle class, but rather to find meaningful ways to engage with, rather than be excluded from, the market and the forms of value that this transformed market generated (see also Ferguson 20 05). As a result of the inability of their meager salaries to help them achieve their personal, familial, and professional aspirations, the hospital staff, like other Tanzanians, turned to formal and informal economic activities outside of their main job at the hospital. I did not encounter any staff member who did not have at least one other business, whether formal or informal, and most had plans to develop other businesses as well. Similar to what was found in other studies of civil service and health care workers in other African countries (see for instance Anders 2010; Martin 2009; Jaffr and Olivier de Sardan 2003; Blundo and Le Meur 2008; Blundo and Olivier de Sardan 2006), many of these informal activities were carried out during working hours in th eir place of employment. The situation was similar at Kiunga District Hospital. At the time when the nurse administered first immunizations to newborns in the maternity ward, she sold knitted baby hats to the new mothers. A clerk sold cell phone cards out of her admi nistrative office. Those employees living on the hospital compound supplemented income by having relatives prepare and sell m andazi chapati and other snacks to the staff to consume during tea breaks. Several health workers owned private pharma cies. Some pooled money to open a canteen at the gate of

PAGE 295

295 the hospital, selling food and beverages to patients and caretakers since the hospital did not provide them. Many hospital workers had other employment in the private sector on weekends and evenings. Some opened tailoring businesses in town, m anned by relatives or employees; they would often bring kitenge cloth to the hospital to sell. No one not even the administrators discouraged these activities. E veryone knew that making a living on a government s alary was difficult without other businesses on the side. P articipation in additional formal and informal income generating activities was an expected, and indeed universally valued, means of supplementing salaries. Side businesses were one of the strategi c behaviors adopted in the wake of inadequate salaries and compensation in a commodified world. Conclusion While hospital workers were highly complementary of the MoHs efforts and accomplishments in terms of HSR, it was at the level of individual rights and compensation that staff critiques were most fervent. Healthcare workers dissatisfaction with the state was due largely to their personal and professional burdens and their aspirations for the future (Iliffe 2002) These failures of the state were made all the more palpable due to the presence of donor sponsored enclaves in the hospital. These enclaves were one of the few means by which government health workers could access new knowledge, adequate pay, and opportunities for future promotions District health administrators not only had to diffuse tensions relating to salary inadequacies and delays directed at the DC, they also had to determine how to mediate the abundance of the enclaves amid the paucity of the other units of the hospital. At the same t ime, health personnel worked to position themselves so as to make claims on the state through these donor resources, as the only means available to them to access the resources to which they felt entitled (Wendland 2010).

PAGE 296

296 The presence of abundant pay and t raining opportunities relating to donor sponsored programs highlighted the inadequacies of the state. These resources served as an everyday reminder of the persistent inability of the government to live up to its promises. To health workers struggling with inadequate pay and hoping for additional training, donor sponsored programs presented opportunities to achieve personal and professional aspirations unavailable by any other means at the district hospital. However, they also eroded the staffs relationshi ps with the administrators who, within the economy of scarcity, appeared to stand between them and their hopes. Simultaneously, the economy of scarcity precipitated tensions between differently positioned hospital employees, as the economy of scarcity put them in competition with each other. Ironically, the abundant resources of donor sponsored programs are eroding the very institutions and professional relationships on which they depend. The example with which I opened this chapter points to the potential ly harmful consequences of this economy of scarcity, and the importance of understanding health workers as whole people rather than merely medical professionals (Martin 2009). If donor resources are exacerbating dissatisfaction with the state, they also er ode the quality of patient care by eroding staff morale. The nurse in the opening example chose to make her expertise, as well as her key to the only place in the hospital that had equipment necessary for providing care, unavailable because her personal an d professional desires were not respected. When government resources are inadequate for achieving personal and professional (neoliberal) desires, workers may invest their time, and their talents, in other endeavors. While not all workers engaged in practic es of absenteeism when their skills were critically needed, those who did increased the burdens on their fellow workers, and their patients. In an economy of scarcity, few hospital employees criticize such refusals to come to work. Instead, they attempt to position themselves in such a

PAGE 297

297 way as to access donor resources through the state, or circumvent the state completely in their informal income generating strategies. Health professionals, as whole people, do not give up their desires in contexts of scarcit y; rather, they search for alternative ways to make their hopes into possibilities.

PAGE 298

298 CHAPTER 8 BIOMEDICAL PLURALITY BETWEEN THE ENCLAVE AND ITS OTHERS A variety of scholarly work addresses the need to understand biomedicine as a cultural product, rather t han a neutral and objective therapeutic science ( e.g., Burri and Dumit 2007; Good 1994; Lock and Nguyen 2010; van der Geest and Finkler 2004; Wendland 2010; Young 1982). In a themed edition on hospital ethnography, van der Geest and Finkler argue that dive rsity between hospitals throughout the world is due to far more than mere structural differences. Instead, biomedicine, and the hospital as its foremost institution, is a domain where the core values and beliefs of a culture come into view (2004:1996, em phasis in original). This important work incited many medical anthropologists and other scholars to take the hospital seriously as an institution where one can observe wider societal values where culture can be seen through biomedical institutions However my observations of biomedical encounters at Kiunga District Hospital bring this approach into question. Joo Biehl writes that anthropologists must continue to ethnographically chart, understand, and politicize the complex and often contradictory ways i n which neoiberalizing policies, state presence, and health and well being are forged in local worlds where biotechnology and scarcity exist side by side (2005:267). In a hospital such as Kiunga, where infrastructure, expertise, maladies, patients, techno logies, and resources are unevenly distributed where both abundance and scarcity aboundthe notion of a unitary biomedical culture at the local level does not do justice to the complexities of interaction characteristic of biomedical practice at the hosp ital. Tanzania is often cited as one of the many countries in the world marked by medical pluralism. The pluralism suggested by much of this work implies that biomedicine is a singular

PAGE 299

299 medical system that operates in conjunction with other healing moda lities.129 However, I wish to suggest that pluralism in Tanzania is more complex than this interpretation suggests Exploring Chinese medical pluralism, Volker Scheid argues for an ethnography that accepts plurality as an intrinsic and nonreducible aspect of all medical practice, biomedical or otherwise (2002:13, emphasis mine). Indeed, Scheid argues that while biomedicine claims to be systematic and objective, it consistently fails to live up to those standards (2002:26). The notion of a unitary biomedica l culture at the local level does not allow for a nuanced consideration of the plurality of biomedicine within the same health facility, or even, within the same health professional. Below, I take up Scheids approach to Chinese medicine in my own approac h to biomedicine in the hospital, noting how, where, and when global and state forces affect practices, connections, as well as dislocations in biomedical therapeutic practice. I argue that biomedicine in Kiunga, as elsewhere, is an emergent and unstable f orm that varies from department to department, patient to patient, practitioner to practitioner, and encounter to encounter.130 Biomedicine in Tanzania, as elsewhere, is plural, even if as a global healing modality it claims to be otherwise. Yet the influx of government and state programs and policies into the hospital has further intensified inequalities within biomedical practice. Targeted health interventions make the state and a variety of donor entities present within the hospital, largely, as I have ar gued, due to an 129 Tanzanias therapeutic landscape includes a wide variety of non biomedical practices indigenous to Tanzania, as well as others imported from elsewhere (Chinese medicine is a case in point. See Hsu 2002; Jennings 2005; Langwick 2010). 130 This argument is not to suggest that biomedicine is not an institutionalized healing modality. Indeed, many of the protocols and policies underlying biomedical encounters at the hospital are intended to define and move particular patients, ailments, and workers in similar ways. However, while standards and protocols are meant to systematize healthcare provi sion in the hospital, the specificities of biomedical encounters and contexts often undermine these protocols. Stacey Langwick (2007) provides an insightful comparison of the principles underlying hospital care and noninstitutionalized forms of Tanzanian non biomedical healing. Yet I would add to her analysis that while principles may influence hospital practice, they do not necessarily or fully effect a homogenization or total systematization of biomedical practice.

PAGE 300

300 economy of appearances in which the Ministry of Health (MoH) creates particular priorities, programs and policies in order to stimulate donor investment in the health sector (see Chapter Four). Similarly, the daily practices within specific departments of the hospital mark this facility as a relevant place beyond the borders of the district in which it is located. State/donor investments in the hospital modify the material arrangements within the facility; what kinds of (technological, biome dical, human) resources are available during any biomedical encounter are heavily influenced by whether and how they articulate with government or donor interests. Donor/state sponsored programs at the hospital therefore create boundaries within the hospit al, between those spaces and actors linked up with, and those outside of these targeted interests. Biomedical practices and possibilities differ within and beyond the surveillance of donors and the state. Thus, biomedicine as a therapeutic modality is perm eable, emergent, and unpredictable. How might we trace biomedical plurality? John Law and Annemarie Mol argue that attention to intricate and mundane interactions is critical for understanding how globalisation comes about, how it works, what it is. In practice (2008:134). In this chapter, I apply a similar approach as a means for approaching a global form biomedicine and the sometimes banal and sometimes extraordinary processes through which it comes into being within the hospital. I am attentive to the ways that humans, materials, and spaces are interconnected, and the politics underlying which maladies and populations become linked up with donor/state initiatives. In my daily observances of therapeutic encounters between hospital staff and patients at Kiunga District Hospital, the nature of biomedical therapeutic encounters fluctuated based on a patients socioeconomic status and presenting symptoms of underlying maladies; (non) availability and fluctuations of infrastructure, technologies, human resour ces, and expertise relevant to a specific interaction; and (non)applicability of state/donor policies to a particular case.

PAGE 301

301 Tanzania is a particularly useful site for investigating biomedical pluralism. As many scholars in science and technology studies a nd other disciplines have suggested, a wide array of human and nonhuman resources, and an enormous degree of work, are required in order for biomedicine to be made to appear stable and universal. Annemarie Mol argues that reality is multiple, and this mul tiplicity is observable by following the practices or performances through which objects are instantiated (2002). In her work, she follows practices and performances surrounding the disease atherosclerosis, which she shows is performed differently in the c linic versus in the pathology laboratory. The disease is made to appear unitary because these multiplicities are bracketed as if they are the same thing ; a considerable amount of work goes into making a disease that is perceived so differently in various c linical spaces to appear as a unitary entity. As she argues, reality is multiple, so that even the clinic is not a homogenous entity, but is in fact more than one place. A similar multiplicity is true for infrastructure. When infrastructure operates as ex pected, its complexity is taken for granted (Starr 1999), understood as operating as a unitary system, until that infrastructural system breaks down. Science and technology studies scholars are attentive to practices and moments of breakdown precisely because it is often in these moments that the instabilities and complexities of science, and the technologies and infrastructures on which it depends, are perceptible; through tracing practices, the politics of the material ordering of the world become perce ptible (Law and Mol 2008). What makes Tanzania a ripe locale for such an analysis is that no one takes the stability of infrastructure or biomedicine for granted. Technologies are often absent or in disrepair, hospital personnel may or may not be present, beds may or may not be available, biomedical expertise varies from individual to individual and from malady to malady. Doing biomedicine within

PAGE 302

302 Tanzanian health facilities requires significant flexibility and ingenuity, evaluating the possibilities for th erapeutic practice situationally, coping when things break down. What biomedicine is depends on its contexts, and the conditions under which biomedical encounters occur frequently change. This chapter traces biomedical practice, possibilities and constrai nts through three case studies It begins with a consideration of biomedical practices within an area of the hospital that is beyond donor and state interests and interventions. I then employ two additional cases to explore biomedical possibilities that are restricted and are made possible between differently resourced spaces in the hospital: those that are integrated into donor/state programs, and those left out All of the case studies below imply action and movement as patients and medical personnel oper ate within the bureaucratic, technical, and institutional capacities and constraints of the hospital as a whole, and vie for resources that are both scarce and highly targeted. They reveal biomedical pluralism while also pointing to differences in possibil ities for therapeutic interaction within very different hospital spaces. The Hospital and its Enclaves If we are to grasp biomedical plurality within the hospital, we must be attentive to the material arrangements and infrastructures within which biomedic ine is practiced. How resources and personnel are distributed within the hospital opens up and closes down a variety of possibilities for biomedical pluralism. In order to think through the effects of donor funded health interventions on the government hea lth institutions where they are implemented, I wish draw on the concept of development enclaves (Ferguson 2006; Blundo and Le Meur 2008). Exploring the ways that Africa has largely been excluded from the global economy, James Ferguson argues that the not ion of global flows is highly inadequate (2006). Instead, global forms are very selective about where and how they operate; indeed, capital in the select

PAGE 303

303 instances when it penetrates into Africahops between discrete nodes or points across the globe, c reating enclaves of development, and omitting th e majority of spaces in between. Similarly, d escribing the relationship between aid organizations and the states where they operate, Blundo and Le Meur argue that the development aid and humanitarian system have been playing an increasing role in creating overpaid and functional bureaucratic enclaves within the state administration (2008:20). There are characteristics of donor sponsored health programs that fit concept of the development enclave very we ll. Donor funded and NGO sponsored health programs in Africa are generally well resourced, technologyintensive, and systematically governed. The routines and distribution of technological, monetary and other resources related to donor sponsored projects a re highly enforced and bureaucratized, characterized by complex sets of technical rules and overarching values of good governance, transparency and accountability. Further, similar to the ways that financial capital targets particular mineral resources for extraction from African states, donor sponsored health programs generally work to extract particular maladies (such as HIV/AIDS or malaria) or subject populations (such as pregnant women or children under age five) from the general communities and bodies within which they exist in the hopes of making them legible and accessible to biomedical intervention.131 Indeed, within health facilities in Tanzania, and certainly at Kiunga District Hospital, the clinical areas most likely to be subject to donor/state sponsorship and oversight are the reproductive and child health clinic (RCH) and the HIV/AIDS clinic. Donor sponsored programs directed at malaria, tuberculosis, HIV/AIDS, 131 Several ethnographies of biomedicin e are attentive to the ways healing practices and cultures define the objects of therapeutic intervention. See for instance Good 1994; Lock 1993, 2002; Mol 2002; Pigg 2001; Young 1982, 1995

PAGE 304

304 family planning, immunizations, child wellness, and pre natal preventative services ran almost exclusively through these two clinics. Nor is the state absent from enclaves. Indeed, given the remarkable degree of donor state coordination in the health sector, the priorities outlined in the health sector require that the state be attentive to very specific concerns as well. Indeed, the MoHs priorities (themselves often negotiated in the context of transnational governmentality) often embed themselves in enclaves as well. Staff at the hospital do es not associate the enclaves with the donors an d the wider hospital with the state. Instead, the MoH targets many of the same maladies and populations as donors, although the Ministry is not able to provide the same degree of resources as donors to these programs. The enclaves of the hospital are areas where both donors and the MoH make their presence known at Kiunga. The nonenclaved portions of the hospital are, I suggest, beyond the general purview of either the donors or the government. Instead, the expansion of infrastructure and technology in thes e nonenclaved sites depends heavily in the ability of hospital personnel to seek out and take advantage of haphazard encounters with private organizations and individuals interested in building the hospital (see Chapter Four). In the sections that follow I bring together biomedical enclaves and the physical milieus within which they exist. Below, I trace the kinds of movements and possibilities that occur across the boundaries between enclave and the institutional locale where it operates, and what is at stake in these movements. At Kiunga District Hospital, patients, staff, biological samples, standards of practice, administrative documents, technological expertise, biomedical and professional values all move not only within enclaves, but also across their boundaries into the rest of the hospital, and beyond.

PAGE 305

305 While state/donor oversight brought significant bureaucratic routines into therapeutic practices within enclaves, t he logics of biomedical enclaves must be understood as interwoven with the existing logics and conditions of the very institutions and communities where they are established. Being attentive to the intersection of enclaves and the institutions in which they function provides an important vantage point from which to view the politics, contingencies and (in)stabilities of global forms as they interact with the particular contexts where they are implemented. This argument is not necessarily new, and is found particularly in recent scholarship investigating the practices surrounding developme nt aid in poor countries, whether biomedical or otherwise (see among others Carpenter 2008; Ferguson 2006, Kalofonos 2008, 2010; Mosse 2005; Nguyen 2005; Pigg 2001; Rottenburg 2009; Wendland 2009, 2010). What I add to this literature is a consideration of what occurs when biomedical enclaves are established not on the outskirts of health facilities (as in Wendland 2009), but rather within these under resourced institutions, and the kinds of encounters these embedded enclaves enact. Specifically, I suggest that the entrenching of enclaves within the hospital intensifies inequalities within therapeutic encounters creating new hopes and dilemmas for medical personnel (van der Geest 2005; Whyte et al. 2010) and for patients This embedding of NGO enclaves within existing government health facilities is part of a process I will call punctuated development. Such punctuated development approaches targeted at particular ailments, populations, or bodily states break up the body and the biomedical institution that services it, emphasizing particular stages and clinics as ripe for investment and intervention. What of those clinical spaces that are part of health facilities where enclaves exist, but are decidedly outside of donor/state priorities? Biomedical Subjuncti vity Beyond State/Donor Interest In her work tracing healthseeking behaviors in Uganda in the context of medical pluralism, Susan Reynolds Whyte suggests that biomedical health workers provide treatments in

PAGE 306

306 order to relieve symptoms rather than to treat k nown maladies (2002). In Whytes analysis, what is causing the malady and which precise treatment cured it are not necessarily known. Whyte refers to this mode of treatment as a form of subjunctivity of uncertainty and action, of doing something to address a problem without necessarily knowing its cause. She argues that in the context of her study, a malady becomes known through the process of acting upon it, in a similar fashion as the means by which people in Uganda act upon misfortune without necessarily knowing what precise agent or entity is causing that misfortune to occur (see also Langwick 2008; Last 1992; Whyte 1997). She further argues that the laboratory investigations of the biomedical facility in Uganda do not carry the same weight as they might in Denmark (or arguably other countries with access to the resources required for high tech nology biomedical practice). A facility may lack technology or other equipment, and whether or not an adequatelytrained technician will be available to analyze sam ples is never assured. Thus, as Claire Wendland asserts, technologies can be potent actors even when they are materially absent (2010:27). Medical anthropologists working within African hospitals have explored biomedical practices of medical students (W endland 2010) and nurses (Martin 2009; Langwick 2008) These scholars note the uncertainty and improvisation necessary within biomedical institutions in resource deficient health facilities in Africa. Studying nurses in Uganda, Martin notes that many of th e procedures and standards that are well outlined on paper do not play out in practice. This (non)systematic functioning of hospitals was necessary due to the unpredictability of hospital work: staff may or may not be where they are expected to be; drugs a nd supplies may or may not be available; technologies may be in various stages of (dis)repair; hospital staff may or may not have the expertise required to employ particular models of therapeutic intervention or to operate

PAGE 307

307 certain kinds of equipment. The a bility to improvise during clinical encounters was a valued, and extraordinarily valuable, skill for medical professionals (Martin 2009; Wendland 2010). I now turn to one such case of subjunctive, plural biomedicine in practice. Acting upon Happys Malady It is mid April, 2008. It is the middle of the long rainy season, and it seems like the entire hospital has been filled beyond capacity with patients for weeks. I am observing rounds in the male ward with Dr. Mtenga and a European medical student named Ga be. Gabe is one of three such students, who have been at the hospital for the past two months as part of their educational requirements for medical degrees at home. Rounds move quickly. Dr. Mtenga enjoys having Gabe in her ward to assist her; further, she enjoys talking to him about what he would do to diagnose patients back in Europe. She finds the technologies on which he depends back home alluring, but she knows her clinical skills are superior (see also Wendland 2010). Dr. Mtenga relies on few sophistic ated technologies to do her work; she treats patients with her eyes, her hands, her ears, her voice, the essential drugs she has available, and in some instances, orders laboratory or ultrasound investigations that she knows will often not be followed or be uninformative. After rounds, Gabe and I join the other two European medical students interning at the hospital, Kara and Saskia. We are going to the female ward, because Kara heard that there were two pediatric burn cases that had not been seen by a doc tor that day. Kara had a previous bad experience with a pediatric burn patient, who she felt had been neglected by the staff. Ultimately, that child died. She wanted to ensure that these new patients received adequate care. She asked for me to come along a nd act as a translator so she could find out more about the patients. Particularly during the rainy seasons, the pediatric ward is crowded with children, who are most commonly diagnosed with malaria, pneumonia or a combination of the two. Those children

PAGE 308

308 wh o have burns, broken limbs, require longterm care, or whose maladies are difficult to diagnose are often sent to the overflow area of the female ward, in a section behind a small dividing wall as shown in Figure 81 of a child whose broken limbs are in t raction. The pediatric ward doctor and nurses are ultimately responsible for the care of these children, but their patients separation from the pediatric ward means that their needs are often less visible to the hospital staff than those in the pediatric ward. As the four of us climb the long steep ramp into the female ward, and enter into the large, open female ward, we see beds filled with women and their caregivers. When we arrive at the overflow area, we see two girls with significant burns, each on th eir own bed. One of the girls, Dotto, lives with her grandmother as both of her parents had passed away. She burned herself while moving a pot of boiling water, and her burns cover about 70 percent of the front of her abdomen, circling around one hip, the blisters extending to her backside. She arrived two days ago. She is receiving an intravenous drip and some pain medication, but her wounds have not been cleaned in two days, and the students are concerned about infection. The other girl, Happy, is four ye ars old and arrived this morning. Her mother is with her. She explains that Happy burned herself in the courtyard of their home where the mother had been boiling milk on a local stove for tea. Happys burns are very serious, covering from one hand up the e ntire arm and neck, and one entire side of her head including her ear. Her eyes are swollen shut, and the water from the burn blisters oozes down her face. Seeing the four of us at the bedsides of the burn patients, one of the nursing officers, Monica, co mes over from the nursing station. She tells us that she did not believe that the pediatric doctor was informed that the children were here in the overflow area, and asks the students to write her a script for pain medication for them. Monica is very conce rned that these

PAGE 309

309 patients get better care than the previous pediatric burn case that had so upset Kara. Monica recently suggested setting up a protocol to deal with these kinds of cases at the all staff morning meeting, but little had come of it. The pediat ric doctor, Dr. Emmanuel, is very diligent and really cares for children, but because he was not informed about the patients in the overflow, he had yet to examine them. Monica instructs Happys mother and Dottos grandmother that when the doctor and nurse s come, they should say that the children are in a lot of pain. Pain management is difficult among pediatric burn patients, because the AMO and nurses are concerned that the more potent pain medications available at the hospital could be damaging to the ch ildren. By instructing the caregivers to tell the doctor and nurses about pain, Monica attempts to ensure that at minimum the children receive regular doses of paracetamol, to give them a degree of peace. Kara, Gabe and Saskia investigate the childrens b urns and then we all walk down the ramp of the female ward and up the ramp through the back door of the pediatric ward. Kara asks the nurse In Charge to inform Dr. Emmanuel that he has patients to see in the overflow area, and that the wounds on the childr en had not been cleaned. The n urse In Charge agrees to track down Dr. Emmanuel, and to clean the wounds, and we leave the ward. The next day, I arrive at the hospital at 7:30 in the morning, attend the staff meeting, and then go to the male ward to observe morning rounds. As Dr. Mtenga is completing her rounds, Saskia comes into the ward and tells me the other students are finished their rounds and would like me to join them for lunch. I agree and meet them at a local canteen, where they tell me about some of the difficulties they have with practicing medicine in a place where they have so few technologies that will assist them to know or see what it is that they are treating. The laboratory is limited in the tests it can perform, and the only other technologies available to aid in diagnosis are an old ultrasound machine and an xray, both of which are operated by nurses. There are no

PAGE 310

310 specialists available to read x ray pictures or ultrasounds, and many of these investigations come back inconclusive. After lunch, we go back to the overflow section of the female ward to visit the burn patients. Dr. Emmanuel is examining them. Clearly he had a lot of patients today, since he began ward rounds at ten in the morning, it is past noon, and he has yet to break for tea. Both Happy and Dotto received their medicines, and their wounds are clean, raw, and bandaged in several places. Dr. Emmanuel tells Dottos grandmother and Happys mother that they need to buy a small jar of Vaseline and some gauze to dress the childrens wounds. In English, he explains to the students that the ideal wound dressing in burn patients is sterile Vaseline gauze, but it not part of the essential drug package provided by the MoH, and therefore not available at the hospital. Patients relative s must buy it, and at nearly 2,000 Tsh per bandage,132 it is beyond the financial means of most families. The nurses make due using regular gauze and Vaseline, although it is not sterile. Kara suggests that the nurses be asked to prepare Vaseline gauze and s terilize it in the hospital autoclave. However, Dr. Emmanuel reminds her of the hospitals resources. There are two old and very small autoclaves to sterilize instruments in the hospital: the one in the major theatre is constantly in use sterilizing instru ments used in major surgeries, while the autoclave in the minor theatre is all that remains to sterilize instruments for the rest of the hospital and there is an overload on sterilizing in that autoclave as it is. In addition, the short staffed nurses do n ot have time in their schedules to prepare the gauze. The jar of Vaseline that the women could purchase was within their means, and the bandages were free at the hospital. They would make do. 132 2,000 Tsh, the equivalent of about US$1.74. However, the sheer size of Dotto and Happys burns would require that several new bandages be applied each day, and the bandages were only available at private pharmacies in Arusha town, making the distance and price completely cost prohibitive.

PAGE 311

311 Over the next couple of days, I go to visit Happy and Dotto eac h afternoon to see how they are progressing and chat with them about what they like to do. Dottos burns were more severe than Happys, but both children are progressing very well and Dr. Emmanuel says they will likely be able to go home early next week. D uring the weekend, I come down with a cold and have to stay home sick on Monday. On Monday afternoon I receive a text message on my cellular phone from Saskia; Dottos condition is good and she is improving so quickly that she is will be discharged tomorrow. Happy, meanwhile, is struck with a fever, and the students are concerned about the change in her condition. The next morning I feel better, and I arrive at the hospital as usual at 7:30 in the morning. I attend the morning staff meeting and return to t he male ward to observe rounds. Gabe is with Dr. Mtenga today, and Dr. Mtenga stops between patient consultations to tell Gabe about symptoms so he can review what he has learned in tropical medicine and come up with potential diagnoses for the patients. W hen rounds conclude, Gabe and I go to the female ward to see if Kara and Saskia have completed their rounds. Kara has finished rounds in the female ward, and the three of us go to the overflow area where we find Dr. Emmanuel and Saskia at Happys bedside. We learn that Happy is in very serious condition. Dr. Emmanuel and Saskia cannot seem to determine what is ailing Happy. As is common at Kiunga, the laboratorys power has cut out, so the samples they sent to the lab could not be processed. As Dr. Emmanuel and Saskia examin e Happys weak and unconscious body, she begins to convulse in front of us. The doctor quickly asks the mother how many times Happy convulsed that morning. The mother figures over ten times. Happys pupils are nonreactive, her breathing i s extremely labored, and just when Dr. Emmanuel and Saskia are discussing what could be ailing Happy, she stops breathing. Dr. Emmanuel rushes to Happys side with a stethoscope. Her heart has stopped. Saskia, Gabe, and

PAGE 312

312 Kara jump in to aid Dr. Emmanuel to resuscitate Happy. I spent a lot of time playing with Happy last week and am unable to fathom that this child who was well on her way to recovery is now dying. As someone with no medical training, I feel in the way. N one of my anthropological training has prepared me to deal with observing the death of a child. I leave the bedside. I pace outside of the building. I return fifteen minutes later to the nurses station in the female ward, thinking that from there I should be able to see what is happening with Happy without being in the way. When I enter, I learn that they have successfully restored Happys breathing and heart rhythm with CPR I am incredibly relieved. As Dr. Emmanuel is assessing Happy, the medical students are looking through the resuscitatio n kit that the nurse brought to Happys bedside. They are not speaking in English, but, I can tell from their tone and body language that they are very displeased at the drugs and equipment located in side it. Dr. Emmanuel motions to leave. He says that he believes Happy is suffering from cerebral malaria. Happy is already receiving intravenous quinine, and there is nothing more to be done but allow the quinine to work. However, Kara and Saskia suggest that Happy may be suffering from meningitis, due to the way she is positioning her body. Given that they do not have the equipment to run multiple IV drugs through the existing intravenous line in Happys hand, they suggest placing another intravenous line in her other hand and administering the powerful antib iotics for meningitis. Since they have no way to know for sure what is ailing Happy, this way they might be able to target multiple maladies at the same time. Dr. Emmanuel agrees to let them try, and leaves to attend to his other patients. Medical student s would never be permitted to perform such interventions on children at their hospital back in Europe, they tell me. Pediatrics is a specialization. It requires years of additional training.

PAGE 313

313 Students are not allowed to work with children. However, here at Kiunga, allowing Happy to pass away because of regulations that govern their biomedical practices at home is unconscionable. The students take turns attempting to place another line in Happys limp limbs. However, Happys body is too dehydrated and they al l fail to locate a usable vein. Suddenly, Happy stops breathing again, although at this point Dr. Emmanuel is not around to do the resuscitation. Gabe, who has never performed a resuscitation before, let alone on a pediatric patient, begins compressions on Happys chest and succeeds quickly to re establish a heartbeat. Happys shallow breath returns. I leave to try to locate Dr. Emmanuel, but he is nowhere to be found. I am determined to find any assistant medical officer (AMO) to help and begin running f rom ward to ward. None can be found. The Matron, Sister Kupaza, sees me and asks what is the matter. Even during emergencies, staff rarely ran on the compound. Continuing to run, this time toward the administration block, I yell back to her that a pediatric patient is in need of resuscitation and I am looking for an AMO to assist the European medical students. She calls Dr. Emmanuels cell phone, but there is no answer. I realize that at this time of day, ward rounds are complete and the AMOs may be out for lunch or off the compound. There is no intercom system at the hospital by which a doctor can be paged in emergencies. I find no one. Deflated, I return to the overflow of the female ward. Sister Kupaza comes in behind me with a Clinical Officer (CO) she m anaged to locate. Over the course of the next three hours, the students and the CO work on Happy, trying to get an additional line into a vein or, later, into a bone by which they might be able to administer the antibiotics for meningitis. None of us have eaten since early that morning, and finally, when Happys condition stabilizes and it appears that they will not be able to achieve a secondary intravenous line, the students determine there is

PAGE 314

314 nothing more to be done but wait until the malaria treatment i s complete to administer the meningitis medication. They instruct me to tell the mother to find a nurse or yell for help if Happys condition changes again. The mother, Dotto, and her grandmother were present to watch helplessly during the entire time that the doctors worked on Happy. Their expressions are remote, and they are clearly exhausted. The students and I leave to get something to eat, hoping Happys condition will improve. After dinner, we return to the female ward to check on Happy and see if th ere are any improvements. The nurse tells us as we enter the ward that Happy died an hour before. All of the deaths here seem to happen at night, Kara observes with a tone resembling both fury and resignation. I am in disbelief. Kara asks me to talk to H appys mother, to find out what the mother was told about Happys condition and what happened in their absence. They want to know if their instructions were followed. They want to know that everything possible was done for Happy, that Happy was not, as the y saw it, neglected because it was the night shift.133 As the mother of a child not much younger than Happy, I feel ill equipped to talk to a grieving mother, but given the students concerns, I feel compelled to try to compose myself enough to at least cons ole her. We go outside, where Happys mother is sitting on the side of the ramp leading up to the ward, rocking back and forth, comforted by Dottos grandmother. I translate between the students and Happys mother. We learn that Happy stopped breathing aga in, but that no one was able to find a CO to help resuscitate her again. I pass on our condolences, as Dottos grandmother tells the mother not to cry. The students and I return to our respective homes The next morning I arrive to the morning staff meeti ng, where a report is being made about the events that had occurred during the afternoon and evening shifts. This report is conducted in 133 For a description of condi tions during the night shift at the hospital, see Chapter Seven.

PAGE 315

315 English, and Kara, Gabe, and Saskia understand it. Ordinarily, the head CO of the night s hift reads one report, and the nurse In Charge reads another. Both are to give the staff an update on serious patients. However, this morning the Nurse InCharge must be dealing with some kind of emergency, because she is not there. The CO reads his report aloud, and at the end, says deaths: nil. The MOI asks the staff if there are any comments. Saskia stands and challenges the report. She says that deaths: nil is false, because Happy died last night. Thus, how can the report say that there were no deaths? The CO responds he is not aware of any deaths and that perhaps it was in the nurses report, which lay beside him. The MOI asks the CO to read the nurses report. Same conclusion: deaths: nil. There is no reaction among the staff, and the MOI moves on to general announcements. Th e Subjunctivity of Emergent Biomedicine Happys case points to some of the gaps, uncertainties, and scarcities that characterize the therapeutic practices of biomedical health facilities in resourcepoor settings. The ways that doctors attend to patients s uch as Happy also highlight that what biomedicine is can only be understood through interaction and practice. In well resourced and staffed biomedical facilities in richer countries, biomedicine as a science and a healing modality may be understood as universal.134 The routine maintenance and repair of technologies, the seemingly endless supply of simple and sophisticated medical equipment, the standardized and continual training of medical staff, the hierarchical ordering that determines which medical profes sional is responsible for what aspect of care, the ceaseless cleaning of human and non human surfacesthe very ordering of hospital spaces and the people and objects that inhabit them can, under the right 134 However, there are health facilities in purportedly rich countries that grapple with lack of resources, staffing, and training. Despite this, even under resourced hospitals in these count ries tend to have basic equipment that is lacking in facilities in aid dependent countries.

PAGE 316

316 circumstances, give the impression of a stable, con trollable, universal healing science in countries that are not aid dependent It is only when something goes wrongwhen a physician forgets to remove gauze from an open abdomen during surgery, when a pharmacist gives out the wrong prescription, when a rout ine operation goes awry, when a patients ailment eludes biomedical diagnosis that the instabilities and pluralisms of biomedicine come into view in such resourceand personnel rich hospital spaces. However, in resource poor biomedical settings, such sta bility cannot be and is not, assumed. Hospital staff, patients, and caregivers that inhabit such locales are all too familiar with uncertainty. Happys malady was acted uponthe AMO and European students tried out and innovated with what they had availabl e in that given moment. What characterizes biomedicine in a given moment depends on the equipment available, how the patient responds to treatment, and the expertise (or lack thereof) of the particular medical professional(s) present during the therapeutic encounter. Further, the socioeconomic status of the patient and his or her relatives influences the kinds of treatments that will be pursued. Knowing that Happy and Dottos relatives could not afford pre packaged sterile Vaseline gauze, Dr. Emmanuel found another alternative that was not ideal, but that would provide the best care possible under the circumstances. However, the malady eluded their interventions. In places such as Kiunga District Hospital, biomedicine is ongoing, relational, unstable, and ri fe with unpredictability. It is plural. S carcity does not necessarily stop biomedical healers from acting on a patients ailment. Like other forms of healing, biomedicine is in this way r elational and plural through the processes of subjunctively trying out. As Gabe shows, this subjunctivity of biomedical practice can also be learned. However, in an environment of uncertainty that permeates hospital practice, not all bodies or maladies can be made legible to biomedicine.

PAGE 317

317 No one knew what finally killed t his little girl who, only days prior, was recovering well. Because her malady was not made legible to biomedical intervention, it was also effaced her life and death were not recorded in the administrative apparatus of the hospital, making her treatment at the hospital practically disappear. Beyond her patient file, which would remain in the hospital, Happys case would forever remain beyond the purview of the administration, the donors, or the state. Thus, there are therapeutic encounters that are never made biomedically or bureaucratically legible. Punctuated Development: The Confluence of Enclaved and NonEnclaved Biomedical Practice In 1994, Byron Good suggested that biomedicine constructs its objects Analyzing the biomedical practices within a mater nal child clinic (RCH)135 in a district hospital in southwestern Tanzania, Stacey Langwick traces the ways that the clinic cards of pregnant women and patients attending the clinic translated these people into a discrete, predictable, intelligible patient. That is, they materialize the biomedically understandable body as an object of therapeutic practice (2001:141). Langwick notes the long history of state involvement and interest in maternal child health (s ee also Richey 2008), and that R CH clinics are sep arate from other parts of the hospital. However, she makes no distinction between biomedic al practice characteristic of the R CH, where, due to resources available, bodies are more easily made legible to biomedicine, and the characteristics of therapeutic p ractice in other areas of the facility. In contrast to Langwicks approach, the R CH clinic might be understood as part of an enclave subject to infusions of state/donor interest, and expectations that such biomedically 135 In the mid 2000s maternal child health services (MCH) were re designated as reproductivechild health services (RCH) in order to be more inclusive of mens repro ductive needs and try to encourage more men to take responsibility for their reproductive health. In the interest of simplicity, I have elected to use the same acronym consistently throughout this chapter, since the services provided at the MCH clinic in L angwicks study are highly similar to the RCH services provided at Kiunga in 2008.

PAGE 318

318 understandable bodies will be transl ated into reports that will be exported to the offices of the MoH and its interested donors (see Chapter Ten). Compared with other areas of the hospital, donor/government policies and oversight impinge far more intensely on biomedical practices in enclaves such as an RCH. However, even within enclaves, where attempts are made to systematize biomedical routines, biomedical plurality exists. This plurality is most perceptible when things break down, or biomedical encounters do not go as expected. While biomedical encounters in enclaves may be (precariously, unstably) systematized, the contexts in which enclaved biomedicine is enacted do not permit an easy predictability Further, as I show below, there is considerable movement between enclaved and nonenclaved departments of a health facility. This is due in part to the punctuated development characteristic of how enclaves function. Health professionals designated to work in the RCH and HIV/AIDS clinic do not work exclusively there; often they have responsibil ities or engagements in other parts of the health facility as well. Meanwhile, as described in Chapter Seven, medical personnel working in other parts of the hospital often attempt to gain access to enclaves because enclaved resources rarely spill out into other parts of the health facility. The sophisticated resources, training opportunities, and compensation available in enclaves appeal to workers personal and professional aspirations. Some equipment tied to donor sponsored programs may seep into other parts of the health facility, where staff must then determine how best to integrate those supplies into their provision of care given the overall resource paucity of nonenclaved portions of the hospital. This section provides two case studies to illustra te the processes by which patients bodily states and maladies allow them to become meaningful for enclaved biomedical care, and what states are beyond their purview. Importantly, biomedical practice in the enclave punctuates

PAGE 319

319 patients bodies, so that only particular moments in a patients interaction at a health facility are deemed suitable for intervention. These cases illustrate the politics surrounding material arrangements in the hospital. Only certain patients or bodily states are able to gain entry i nto enclaved biomedicine. While numerous guidelines exist within enclaves to direct the course of biomedical interventions, and health workers at Kiunga attempted to comply as best they could with those protocols, the unpredictability of infrastructure, te chnology, expertise, and patients agency during therapeutic encounters threatened the systematization of biomedical routines. From Pregnancy to Birth: The Case of Miriamu It is May, 2008. Miriamu is among the crowd of about twenty women who arrive today for their prenatal services. All are sitting in the waiting bay, awaiting the nurse to call them in for services. Miriamu is approximately eighteen years old. This is her first pregnancy. Her pregnant belly is barely perceptible, as she is dressed in multi ple of layers of colorful kikoi cloth. Like all new clients of the maternity clinic, Miriamu is registered. The nurses here regularly ask me to help with registrations. When it is Miriamu s turn, the nurse and I ask her a variety of questions Her answers are entered carefully into three different registers as well as onto a maternal clinic card. Miriamu pr ovides a variety of information about her previous pregnancies, her level of education, occupation, where she lives, the father of the fetus, and histori es of prior pregnancies. The nurse also enters on the card an estimated date of delivery, based on her report of last menstruation. It is determined that Miriamu is due to give birth in late July. Once the registration process is completed, the nurse hands Miriamu her filled out maternity clinic card, as well as a voucher. The nurse says that the voucher will give her a discount on a mosquito net, which she should go to use in the nearby market so that she does not risk her pregnancy by getting malaria (see Chapter Eleven) I then escort her into the clinic, where another nurse awaits her.

PAGE 320

320 Miriamu is welcomed and told to sit in the empty chair beside the table. The table is covered in a clean white sheet, atop which are two registration books as well as seve ral small instruments. The nurse tells Miriamu that it is important for her to know her health status. She should be tested for HIV. Not knowing exactly what kinds of services were offered at the maternity clinic, Miriamu is surprised by the screening, but agrees because she heard good things in her village about the quality of services here. The nurse copies some of her information into the two registration books that are on the table. The nurse then takes a small blood sample, and tells Miriamu she will k now the results within the hour. She gives Miriamu three pills, two of which she says are to prevent malaria. The third is a folate supplement The nurse says she needs to check the baby, and escorts Miriamu into a screened off area. There is a cot inside, and Miriamu lies upon it. The nurse uses her hands to feel the position of the fetus, assesses the height of Miriamu s belly with a measuring tape, takes her blood pressure, tests her blood sugar, and writes information on the clinic card. She asks Miria mu if the baby plays [ mtoto anacheza?] regularly in her belly, and checks Miriamu for visible evidence of any sexually transmitted infections that might pose a risk. Satisfied, the nurse helps Miriamu to a sitting position, and escorts back to the waiting bay until the results of her HIV test are ready. The nurse moves on to the next client.136 About thirty minutes later, the nurse asks me to call Miriamu back into the clinic room. Miriamu comes in, a stoic look on her face. The nurse informs her that she doe s not have HIV. Miriamu thanks the nurse. The nurse tells her to return again in four weeks for a follow up visit, when she will receive another malaria preventative and be screened again for any health issues that may jeopardize her health or that of her unborn child. 136 In the enclaved portions of the hospital such as the HIV/AIDS clinic and the RCH, a clinic attendee is often referred to as a client [ mshitiri ] or customer [ mteja ] ra ther than a patient. Indeed, a doctor at Kiunga remarked to me, Before [these donor programs], we had patients [ wagonjwa ], now we have customers! [wateja ]

PAGE 321

321 Unbeknownst to Miriamu the services she receives are funded by a variety of transnational NGOs, most of which receive funding from the American government (see Chapter Ten) It is early August, and I am conducting research in the maternity ward. It is night shift. The ward is a long, rectangular building. It has three internal sections: antenatal, birthing area, and post natal. As can be seen in Figure 82, t he ante natal section houses ten beds, but often caters to twice as many women. When women need cervical exams during active labor or feel birth is imminent, they walk themselves to the birthing area: a section in the middle of the building comprising three cots just a few feet from each other. The post natal ward also houses ten beds, but when it is too overcrowded women who feel well enough to leave often depart without formal discharge, to make room for other new mothers and their infants. The birthing section regularly goes from empty and quiet to an area bustling with activity especially during night shifts. During day shifts there are usually at minimum a senior nurse and two nurse midwives available. At night, as often as not, only one midwife is stationed in the maternity ward. While midwives can keep apprised of the labor progress of the women in the ante natal ward, other pregnant patients can arrive at any moment from villages nearby, in very active labor or experiencing complications. Night shifts are never predictable, and never well staffed. There is no operable communication system between the wards for emergencies. Midwives who need assistance often rely on laboring or postpartum women or their relatives to walk to one of the other wards in search of staff members to help. In the maternity ward, one cannot abandon a patient in active labor or showing complications. While there are doctors on call for the night shift, whether one will answer his or her phone for an emergency caesarian section is never assured. Most doctors do not live on the compound, requiring that a security guard take a car to pick up the doctor and bring him or her to the hospital. Emergencies never

PAGE 322

322 move quickly, despite best efforts. As a result of the precariousness of the birthing process and the regular short staffing of the maternity ward, the nurse in c harge of the night shift comes to the ward periodically to ensure that the midwife is able to handle the cases. When I arrive in the birthing area, the women in the labor ward are collectively singing a hymn in hopes that God will give them strength through their pain and delivery. Their song can be heard in Object 8 1.137 T he nurse midwife assigned to the ward, Gertrude is attending to one woman in the middle cot. The babys head is crowning. Another woman is laying on the next cot, clearly in very active labor, curled in the fetal position. Gertrude tells the woman she is attending to stop making noise. Push into your bottom! [ sukuma hapo chini! ] she says. When the birthing woman calls to Jesus to help her, Gertrude responds, Jesus cant help you right now. Push! The woman pushes, and out comes the shoulders followed by the rest of the baby. Gertrude uses the kanga cloth the woman brought to wipe off the baby, wraps it in another kanga, and passes it to me to put on the nearby neonatal table. The baby is pink and looks healthy. A boy. I turn on the bare bulb lamp above the babys head to keep him warm. There are no incubators or heating lamps here; the heat of the highwattage bulb is the only available means of keeping the baby warm while the midwife a ttends to the mothers placenta and any subsequent tearing. Suddenly the woman in the next cot shrieks loudly, Nesi! Nesi! As she rolls over I recognize the womans face. A quick look at her clinic card at her feet, I recognize my 137 Regularly on night shifts I heard laboring women singing together in the maternity ward, but thes e were always Christian hymns, even when a Muslim woman was present. Relations between Christians and Muslims at the hospital whether between hospital workers or between patients were always highly cordial. On Muslim holidays, Christian hospital workers wo uld call their Muslim friends to acknowledge the holiday, and vice versa. The Tanzanian government abolished questions about religious affiliation from the census. Today, the most commonly cited estimate is that the country is 33% of each religion (Muslim, Christian, Indigenous), although distribution is uneven. For additional information about religious freedom in Tanzania, see US Bureau of Democracy, Human Rights and Labor 2007.

PAGE 323

323 handwriting. I see it is Miriamu who I helped register into the maternity clinic several weeks ago. Gertrude snaps at her that she is not ready to push yet, but Miriamu protests. Flustered, Gertrude tells Miriamu to wait. She catches the placenta of the woman who has just given birth and tosses it into a nearby bucket. Gertrude pushes on the womans belly to ensure that the bleeding has stopped, then instructs the woman to take out the cotton wool she brought with her to the hospital, place it between her legs, dress herself an d go to the antenatal ward to find a bed. Miriamu is calling the midwife again, with a long hissing Nesi! Gertrude drops her bloody gloves in a trash bin, dons new gloves, and examines Miriamu s cervix. The woman who just gave birth places the cotton wool between her legs and ties a kanga around her waist. I tell her I will bring her the baby shortly. She slowly shuffles off toward the postnatal ward. Gertrude tells Miriamu it is time for her to push. Miriamu begins to push and lets out a high shriek as she does it. Gertrude slaps her knee, you are not pushing properly. Push! In your bottom! Quit making noise and push! Miriamu breathes and tries again, but Gertrude is still unsatisfied with how she is pushing. The contraction subsides and Gertrude tries to coach Miriamu Miriamu begins to cry. Meanwhile, the woman who just gave birth hobbles back in. All of the beds are full, where should she lie down? Gertrude asks me to find her a space, and I escort the woman back into the postnatal ward. I ask one of the women to make space for another person, and she moves over to accommodate. They lie with the feet of the other facing their heads. I bring the woman her infant, who she immediately tries to breast feed. I return to the birthing area. The nurse I n C har ge of the night shift, Safira, is now standing at Miriamu s side. She came from her rounds in the male ward to see if any assistance was needed While Gertrude stands near her legs, Safira instructs the crying Miriamu that she must control herself and push because

PAGE 324

324 her baby is in distress. This baby is not too big for you! she says. Miriamu hisses as another contraction starts, and both Gertrude and Safira tell her to push again and again. Finally, the baby crowns and with a last push, the shoulders emerge The baby is large and blue, and Miriamu has a major tear. Once the cord is cut, the baby, another boy, is not breathing. The fluid from the amniotic sac contains a lot of meconium, indicating that the baby defecated in utero and is at risk of infection. When Safira sees the condition of the baby, she quickly wipes the baby as clean and dry as she can, wraps him up, and takes him to the neonatal table, where an old, foot pump controlled suction device lays waiting. Thankfully, the suction machine is ready for such a situation. On other occasions, the midwives do not have time to prepare it in advance of a birth. Safir a suctions the baby, then puts on the stethoscope hanging nearby and checks his heart rate. The heartbeat is strong, but the baby is still blue and has not yet breathed. She grabs an ambubag to blow air into the babys lungs, but the hose to connect it to the oxygen machine is missing, and the mask is pediatric size too big for the neonates small face. The mask reaches to the babys forehead. S he asks me to hold my hands around the edge of the mask to try to create a seal, but to no avail. I offer to run to the pediatric ward to get the only ambubag on site small enough to potentially work, but Safira tells me to wait. She unwraps the infant, pi cks him up by his feet, and smacks him firmly on the back several times. The baby starts breathing, but he does not cry. His nostrils are flaring; he is clearly in distress. With no neonatal nasal prongs available and the hose for the oxygen machine missing, there is little more to be done but wait. An hour later, cold under the bare highwatt bulb of the lamp, the baby finally cries. Gertrude tells me to take him to his mother, who waits in the postnatal ward, sharing a bed. The midwife tries to fill out the partograms and forms in her registers and on the womens clinic cards. Yet after Miriamu another woman delivered. So late at night and having stood for hours

PAGE 325

325 tending to laboring and birthing women, the details of backto back births were starting to b lur together. I take the cold baby to Miriamu who places him between her and the other mother and covers him with the blanket. She tries to nurse him. Two days later I return to the maternity ward during the day shift. Miriamu was discharged yesterday, b ut she sits in a chair in the birthing area, showing the nurse her baby. He stopped nursing. He shakes with convulsions. The nurse calls the ward doctor, who immediately identifies that the baby has an infection. He prescribes valium to stop the convulsion s, to be administered rectally so that the effects do not wear off as quickly as they might with intravenous or intramuscular administration. However, the catheter needed for the rectal administration is out of stock. The nurse improvises with an IV cathet er and a syringe. The rectal administration does not work with the IV catheter, however, and the baby starts to seize. The doctor prescribes phenobarbital by IV. The nurses do not have neonatal IV needles, so use a pediatric sized one. They manage to get a line into the babys foot, and soon the seizing stops. The nurse puts a pediatric sized feeding tube down the babys nose, which makes him wail in pain. The government does not regularly supply neonatal sized supplies, and overall they are difficult to or der privately in Tanzania. The nurse asks me to get a cup and help Miriamu to express some milk that they can place in the tube for a feeding. I hold the cup in place and Miriamu distressed at the sound of her baby crying, tries to express some milk. She is exhausted. Discussion. Thankfully, Miriamus baby lived. Miriamus body was given a certain legibility though RCH clinic registration and screening. Yet this legibility did not extend to childbirth. Her complications were not foreseen. The legibility a chieved at the RCH made Miriamu appear to be a pregnant woman lacking the kinds of indicators that would mark her as being at risk, or mark her body as a site of uncertainty. However, even in this apparently

PAGE 326

326 systematized biomedical space, there is pluralis m. There is no certainty that a patient will comply, that the HIV test will provide an accurate result, that the nurse checking fetal position will have sufficient expertise to be able to identify potential risks clinically, that the supplements provided w ill be used as directed, or that the patient will tell the truth when undergoing registration. Meanwhile, when it came to biomedical care within the maternity ward, Miriamu was subject to a similar biomedical pluralism as Happy in the previous section. An RCH screens for potential issues, and registers indicators. Women found to be at risk based on those indicators are told to give birth in the hospital rather than at home (see Richey 2008). Punctuated development in the enclave only deems prenatal screenin g and registration as suitable sites of investment. Considerable uncertainties remain as a womans bodily state transitions from pregnancy to childbirth. In this liminal state, biomedicine is subjunctive and plural. How a mothers or fetus body will react to interventions (or noninterventions) in a health facility is never certain. Tanzania has one of the worst maternal mortality ratios in the world. According to the most recent report jointly compiled by United Nations organizations and the World Bank, 1 in 23 women in Tanzania will die during childbirth during their lifetimes (WHO/UNICEF/UNFPA/ World Bank 2010). The country stands as one of the lowest ranking in the world, above only Somalia, Sierra Leone, Niger, Guinea Bissau, Afghanistan, and Chad in terms of country maternal mortality rankings. As of 2003, Tanzania ranked as one of the top twentyfive countries with the highest infant mortality rates (UNICEF 2002). Ironically, Tanzania is also one of the top recipients of foreign aid in subSaharan Af rica, much of which is distributed through bi laterally funded transnational NGOs, targeting primarily maternity services, HIV/AIDS and malaria.

PAGE 327

327 By narrowly targeting pregnant women prior to childbirth, the punctuated development modality of NGO enclaves disarticulate maternal health from maternal experience, prenatal services from overall maternity care, and the maternity clinic from the wider institution upon which it depends. Pregnant women move through the hospitals enclaves, but also beyond their boundaries into the under resourced and over burdened biomedical spaces beyond them, where, when it comes to biomedical rather than bureaucratic encounters, plurality is the norm. Biomedical Pluralism within and Beyond the Enclave: The Case of Tumaini We ar e in the male ward. It is a Tuesday in May, and I am in the ward to observe ward rounds now that the European medical students (like Gabe) have left Kiunga. It is the end of the long rainy season and there are a lot of malaria and tuberculosis cases coming through the hospital. The ward is completely full; beds each have between one and two patients. If there are two men in a bed, they lay with the others feet at their head. Some have blankets, while others do not. In the rainy season, air drying laundry outdoors makes dry blankets a rare commodity. There are hooks on the ceiling above each bed, some of which suspend a mosquito net while others are empty. Some of the mosquito nets are in good condition. Others have many holes, to match the sheets that cover the worn bed mattresses. The dividing walls in the male ward make movement between beds difficult, because each section of the ward is filled to the brim with beds, patients, and IV lines. One of the patients there today is Tumaini. He is in his mid 30s, and is a casual laborer living in Kiunga district. His wife sits with him on the bed, which he shares with another elderly patient. His wife is expected to bring him food, help to feed and wash him, and buy prescriptions. Tumaini was admitted a few days ago because he has been extremely fatigued lately, and on the day he was admitted he was unable to walk. He has some rashes and diarrhea. He also suffers from persistent sores in his mouth.

PAGE 328

328 Tumaini has no patient file, and until today, no doctor has review ed his case beyond the clinical officer who first admitted him. He is not receiving any prescriptions. Regina comes by and reminds the w ife that she must pay the fee, 2 ,000 Tsh,138 to open a file for her husband. He cannot receive proper treatment without it she says. The wife explains that she cannot afford the fee for the file as well as the cost of prescriptions and of transportation by mini bus to cover the twelve kilometers to the hospital from their home. Regina encourages her to find a way. Visiting hours end. The guard from the front gate enters, ringing a bell, to indicate to all visitors it is time to leave. Regina passes by the bed of each patient, and when she sees the wife at Tumainis bed, and tells her she must leave. This is a male ward, only male relatives may stay to look after the patients (cf. Brown, forthcoming) Does her husband not have a brother or son or father who can come to look after him? It is inappropriate for the wife to stay. The wife begins to move towards the door. As she wa lks by, Regina reminds her yet again about the importance of opening a file for her husband. Dr. Mtenga is about to begin her rounds. When the doctor arrives at the foot of Tumainis bed, she asks Regina when he was admitted. Regina does not remember and asks to see the clinical notes. Frustrated, Dr. Mtenga picks up the patients piece of paper, glances at it, and says she will wait until the family opens a file before examining him. Once ward rounds are complete and Regina has filled in the patient reco rd books by hand, she returns to Tum ainis bed. I follow. She looks carefully at the clinical notes on Tumainis admitting form. She sees that Tumaini has not been examined since the Thursday when he arrived. She motions me to follow her to the nurses sta tion and tells me that she has decided that his symptoms were erratic enough that his case might call for PITC. PITC stands for Provider 3Tanzanian shillings, the equi valent in 2008 of about US$1.74.

PAGE 329

329 Initiated Testing and Counseling. Despite many of the nurses having received training about it in previous months, the order in which a patient should get an HIV test is sometimes confusingdoes she need his consent to do the test first? Regina decides she will do the test. If it is positive, she will pretend she has not done one yet, encourage the patient to have one done, and reveal his results later. Regina goes to Tumaini, draws a small amount of blood and tells him it is for a lab test. She does not say for what, and Tumaini does not ask. She scribbles some acronyms on a piece of paper in the file. She uses a dropper to put some drops of blood into a rapid HIV test a device resembling a pregnancy test. A few minutes later, the result appears : positive. She goes back to Tumainis bed after some time, and encourages him to submit to an HIV test so that he can know hi s status and then, if he is HIV positive, obtain the treatments he needs. Tumaini nods consent. Wanting to allow some time to pass so that Tumaini does not know he has already been tested, Regina and I go to the nurses office for tea. We share some fried bananas and drink the thick, sweet, milky tea. When the snack is over, we walk out of the room and back into the general ward. Now that visiting hours are over and many of the patients have gone outdoors to get some fresh air, Regina and I visit Tumainis bedside again. She tells him very quietly that he is positive, and that she will be sending someone from the CTC (Counseling, Testing and Care)139 clinic to speak to him. Tumaini closes his eyes, nods, and turns his head to the side. Regina and I leave th e bedside and return to the nurses station. She asks me to help her transfer the patient to the CTC Clinic to see a doctor there. We retrieve one of the wheelchairs that are parked outside 139 As mentione d in previous chapters, due to the stigma surrounding HIV/AIDS and how this stigma made people reluctant to seek services, CTC was adopted as the acronym for HIV/AIDS services in health facilities in Tanzania in order to obscure the kinds of services provi ded there from the general public and thereby increase patient attendance at these clinics.

PAGE 330

330 at the back door of the male ward. As shown in Figure 83, t hey ar e in various states of (dis)repair, several having broken wheels. The one we choose is one of the few that has intact wheels, but the chair, like many others there, has no foot rests on which patients can place their feet. The CTC clinic always has staff available to consult on HIV cases during the day. In most PITC cases, when the patient is found to be HIV positive and consents to speaking with clinic staff, the patient is sent to the clinic for a consultation. There, they learn about the disease, and re ceive adherence counseling in order to ensure that the patients understand the importance of when and how to take their medications. I help Regina transfer Tumaini carefully into the wheelchair. He is very weak, but he still manages to sit and keep his feet balanced precariously on the frame of the chair, because there is nowhere else to rest his feet. We wheel him towards the front door. When we get there, Regina moves around to the front of the wheelchair, and while I support the handles at the back, w e negotiate the wheels and Tumainis weight down the short, steep ramp of the ward, to the outdoor walkway below. Regina returns to her ward, and I wheel Tumaini across the unpaved walkway, which rattles both his weakened body and my hands as I push. Around a corner we go, to the building adjacent to the male ward: the CTC clinic. Arriving at the CTC Clinic, it is clear that we were entering a very different kind of place than the ward. The clinics large, undercover waiting bay has built in benches along the walls for patients to sit and watch the television, which is labeled with a REFLECT logo the American NGO operating the CTC Clinic within Kiunga District Hospital At that moment, a football game is showing, but the waiting bay, as usual, is completely empty. As we go through the entrance into the main part of the clinic, there is beautiful furniture in the spacious foyer, with carved

PAGE 331

331 wood embellishments and padded seats. There is a coffee table in the middle, stained mahogany, with brochures about nutr ition and health for the patients waiting there. The emblems of several different donors and the Tanzanian Ministry of Health grace the back of the pamphlets. A CTC nurse directs me to wheel the patient into the room with the words adherence counseling written in English on the door. Adherence counseling is the gate keeper to the clinic; without assurances that patients will comply with the drug regimen, they will not be eligible to receive anti retroviral therapy (ART). Those patients caught not adhering properly are sent through adherence counseling again prior to getting access to any additional anti retroviral drugs. Repeat offenders are denied further treatment. Inside the room, a clinic nurse sits across from Tumaini and, in Kiswahili, begins to ask him what he knows about HIV and whether any of his family members have been chronically ill. He answers all of the questions, as asked, but in short sentences in a very low voice. He offers no additional information, and asks no questions. She says that t he disease is completely incurable, that all the medicines that he will take are critical to his survival, but that they will never make the disease go away. The medicines will give him a longer life and allow him to feel better, but it is crucial that he take them as directed, every day, without missing a dose. He also learns that many of the drugs have unpleasant side effects, but that he will have to take the drugs despite these. He is informed that every so often, they will measure his blood [ pima dam u], and make sure his blood was not too low, [ damu si chini sana], which he learns is the means by which they determine how sick he is. The nurse notifies Tumaini that he will have blood drawn today, and that blood will be tested over the next few days t o tell them what stage he is whether or not he is sick enough to need the ARTs or not.140 Meanwhile, they will treat his other, HIV 140 In Tanzania, anti retroviral treatments start being administered to HIV positive patients when their CD4 cell counts are lower than 200 per cubic millimeter.

PAGE 332

332 related problems, but until his CD4s fall below 200 he will not receive the anti retroviral drugs. Tumaini nods. I wheel hi m out. All of the staff members at the clinic greet Tumaini and I warmly. I am directed to take him from the adherence counseling room to the consultation room. A clinical officer (C O ) assigned specifically to the clinic, Dr. Ezra, is awaiting him behind a desk. The room has two nice desks put together in the shape of an L, as well as an examination cot and two padded chairs, one facing each desk. Behind the chairs are additional large, padded and handcarved chairs like those lining the foyer. Behind the desks there is a sink with liquid hand soap. The desks are covered with bright white sheets, and each has neat piles of reports and files on it. Dr. Ezra introduces himself as one of the main doctors that he will see regularly, and proceeds to take a patient history: Is Tumaini married? How many children does he have? Are he and his wife still together? Where does he live? What tribe is he from? What does he do for a living? Does he have friends [ rafiki ]141 also? How is he feeling? When did he first noti ce he was getting sick? Are any other family members sick? There are many, many questions and as Tumaini answers, Dr. Ezra writes notes in a new file bearing Tumainis name. Inside the file, there is a blue, thick cardstock form with the Ministry of Health s emblem on top. There are several other different colored papers within the file as well, but the doctor seems most concerned with the blue form and with learning details about Tumainis life Dr. Ezra appears genuinely concerned, as if he has the time t o listen, and makes eye contact with his patient. He waits patiently for Tumaini to respond to the questions, and listens closely despite Tumainis quiet responses. Dr. Ezra encourages his patient to divulge his HIV status to his wife, because 141 I n Kiswahili, the term friend [ rafiki ] is not gendered, and can be used to speak about ones acquaintances as well as one who is not a spouse but with whom a person has an intimate relationship. The question in this case is meant to inquire about whether or not Tumaini has sexual partners other than his wife.

PAGE 333

333 she might al so be sick. Tumaini hesitates, but tenderly the doctor assures him that his wife will be relieved to know her status and that it is better to know and get treatment than to leave his family in a bad situation. After Dr. Ezra completes his notes in the file, he hands Tumaini a small blue card that matches the one on which he wrote in Tumainis new, free file. He then makes a tick mark on a piece of paper at the side of his desk, which bears the REFLECT logo. Dr. Ezra tells Tumaini that this blue card is his government identification card, to identify him as HIV positive and a s a client of this particular clinic It gives him exclusive access to services that are provided at the clinic, including drugs, and if he ever moves he needs to make arrangements with them for the transfer paperwork to enroll him into a different clinic. The card bears Tumainis name and certain numbers, with spaces on the back to record dates of visits. He is instructed to bring this card with him when he comes for clinic visits. Dr. Ezra informs Tumaini that from now on, his drugs and treatments at any government health facility will be free of charge, but that once he is discharged from the hospital he will need to come to the clinic every month to consult with the clinic staff and r eceive his drugs. Then, Dr. Ezra writes some words on a yellow pad of paper and tells Tumaini that they will draw some blood in the laboratory in the next room. The doctor wheels him out to the foyer with file and the other yellow paper on Tumainis lap. We go into the laboratory next door. It is for the exclusive use of the CTC Clinic. Two lab technicians are there. In contrast to the general laboratory for the rest of the hospital, this lab is filled with sophisticated machines, each one carefully label ed with donor acronyms and logos. There is a computer in the corner where one of the technicians writes an email with two fingers. The lab technician looks expertly at the form on Tumainis lap, tells him they are going to take big blood [ damu kubwa], an d wraps a long rubber tape around the top of his arm. He is

PAGE 334

334 instructed to make a fist, and she inserts a thick needle into his vein, extracting blood into two different vials. The technician explains to Tumaini that they will have the results soon. Suddenl y, the electricity goes out and the technician using the computer groans. This will put their lab activities behind for a while. Since they can no longer work because all of their machines require electricity, the staff goes for tea. After the visit to the laboratory, the laboratory technician asks me to take Tumainis file to the filing room before returning him to the male ward. His information is now contained in a smooth and shiny folder with the letters CTC on the front marking him to those familiar with the acronym as HIV positive, and as entitled to free health services at Kiunga. As we exit the laboratory, we pass another door that looks like a second consultation room. On clinic days, two doctors usually occupy the same consultation room, each sit ting at one of the desks and talking to his or her respective patient. They see multiple patients at the same time. The other consultation room is unofficially considered the doctors office, where they hang their coats and might go to read a religious book or plug in their cell phones to charge them. Administrators from the other departments of the hospital often bring their laptops to that office as well, since the electricity cuts out of the administration building more often than it does the CTC. I e scort Tumaini to a small door across the undercover waiting bay that we passed on the way in. I knock, and a door opens, but the room is so cramped with a photocopier, a fax machine, a desk and shelf that the person inside has to open the door all the way, and then come around it to be able to see who is there. This data clerk is not a member of hospital staff. She works directly for REFLECT. She has a newer computer in this tiny room and a pile of CTC patient folders at its side. Before we knocked she was recording all of the details that had been written in a patient folder into a sophisticated database on the computer the screen is littered with various

PAGE 335

335 cells that she must fill. She asks me if I need to purchase any phone cards for my personal cell phone. Like so many other staff members at Kiunga, she supplements her official salary with informal income generating opportunities. I decline the phone card. She passes me a pile of folders and directs us to drop them at the CTC recording room on our way out. As we turn around and move towards the exit of the waiting bay, we stop by a doorway that is not attached to the main CTC clinic, but adjacent to it. Inside it is a tiny foyer with a couch and two other doors. This area was what remained of the old CTC cl inic T oday, the two rooms have a slightly different, but related function. One is the counseling and testing room where clients who voluntarily c ome for testing can see a CTC clinic nurse. The other is the r egistration room, where I enter and handed the p ile of folders to the CTC nurse inside. There are shelves along the wall, lined with binders labeled with REFLECTs logo, and various other labels such as lost to follow up and monthly reports. At the moment that we enter the room, the nurse is handwri ting information from patient folders into a large register bearing the emblem of the Ministry of Health. It is the same color as the cardstock inside Tumainis new patients file. Tumaini, like all the others here, will be recorded his information added to the MoHs annual reports on HIV/AIDS (see Chapter Ten) Returning to the male ward, I wheel Tumaini along the bumpy walkways and up the steep ramp, past the ward nurses station and back to his bed. Regina administers an IV line to Tumaini prescribed, at last, by Dr. Ezra in the CTC clinic. From now on, each day one of the CTC clinic staff will come to his bedside to see how he is progressing through their treatments. He will receive counseling if he is depressed, and will be encouraged about his future If his condition improves, as it is expected to, Tumaini will eventually be discharged and then be expected to come back to the clinic every two weeks at first, then every month, to be weighed

PAGE 336

336 and to consult with the doctor. He will have blood dra wn ever y six months to check his CD4 cell counts. Once the counts dip below 200 cubic millimeters, he will have access to free anti retroviral drugs for the remainder of his life, or the remainder of time that the PEPFAR program continues operating there, whichever ends first.142 Meanwhile, the drugs needed for his treatments in the ward will also be free provided the doctor remembers to write the letters CTC on the top of his prescription form and the hospital pharmacy has drug requested. Here, with the exceptio n of antiretroviral drugs themselves, free does not always coexist with available. 143 Discussion. The biomedicine suggested in the beginning of Tumainis case is one marked by uncertainty and by making due and innovatively coping with technologies that are absent, inadequate, or otherwise unavailable. There are limited resources available to investigate a patients ailment the clinical encounter itself, a laboratory test, an ultrasound exam or an xray were the main means by which a malady is diagnosed. At the same time, the places where particular kinds of investigations could be undertakensuch as the laboratory or the ultrasound and xray department were unpredictable. The electricity often went out, technicians might be absent, or (particularly durin g rainy seasons when there were far higher proportions of patients) there may be a significant delay in receiving the results of investigations. Further, bureaucracy and economics can impinge on a patients access to care. Tumainis ailment was not deemed necessary for intervention or care because he had not opened an official patient file (see also Chapter Ten). When a caring nurse decided to test Tumaini, his malady opened up possibilities 142 At the end of 2009, Mdicins sans Frontires published a report suggesting that donor enthusiasm for funding HIV/AIDS treatment, prevention and research is beginning to wane. See Mdicins sans Fronires 2009; McNeil 2010. 143 I am indebted to a volunteer expatriate medical student who was working at the hospital in 2008 for this phrasing. Working in the hospital for 3 months, seeing how often the free drugs for senior citizens, children under 5 and ma ternity cases were out of stock, free but unavailable was her astute phrasing to characterize the system.

PAGE 337

337 for a standard of care denied him in the male ward. HIV is the gat ekeeper to the CTC enclave. Had a different nurse been present that day, Tumaini may not have been made meaningful to enclaved biomedicine at all. Biomedical practice in the CTC enclave at first glance looks systematic. Bodies are screened as in the RCH, and there are protocols for how to treat each patient that falls within a particular range of CD4 measures. Enclaved biomedicine is in some sense more bureaucratic than it is therapeutic, although people do receive meaningful treatment in these areas. Howe ver, enclaves are not exempt from uncertainty. Patients understandings of their own maladies may be incommensurate with the questions health practitioners are told to ask. Patients may not comply with treatment regimens. Patients may lie. Patients may com e to the clinic complaining of maladies that do not fit within known diagnostics surrounding enclaved biomedical care. The electricity and water supplies on which the CTC enclave depends may cut out. The expertise imparted on hospital staff within and outs ide of the enclave may not be understood in the ways the NGOs intend, as in the nurse doubting protocols relating to the PITC procedure. Staff may use enclaves to conduct their own private business activities, or use the spaces for activities unrelated to the priorities of the enclave Hospital workers remain uncertain as to how long these enclaves will continue to exist at the hospital Due to the uncertainties prevalent in the institution on which the enclave depends, and the unpredictability of the actor s within these spaces, although care may appear systematized, it is not necessarily systematic. Conclusion The biomedicine of the enclave may be multiple in the way that Mol (2002) suggests. For instance, the HIV as perceived through the quick HIV test in the male ward, the HIV as measured in terms of CD4 counts in the CTC lab, the HIV perceived in the clinical encounter with the doctor or nurse and the HIV as experienced and articulated by the patient in the clinical

PAGE 338

338 encounter i n all of these cases, HIV is made legible to some extent and different HIV objects are bracketed together, to appear as if they are one entity: HIV. By an analysis similar to Mols, we see biomedical multiplicity. Yet I suggest that within Kiunga District Hospital, other pluralities also exist. As the therapeutic encounters surrounding Happys case suggest, intervention can occur without the creation of bodies or objects (diseases) amenable to these interventions. Bodies or diseases may call for interventions or expertise that are not available. B odies and their maladies may remain completely invisible to biomedical intervention, whether purposely or haphazardly. Patients may not be made meaningful for intervention, or the interventions required may be beyond the staffs expertise, or the infrastructure and technology available for therapeutic encounters. Trying out may fail. Beyond the bureaucratic biomedicine of the enclave that systematically registers and screens patients, i t is common at Kiunga for doctors and nurses to try o ut a variety of different therapies and treatments without being certain about what malady is afflicting the patient (Whyte 2002). Within the outpatient clinics and inpatient wards, diagnoses are often not recorded until a patient is discharged, once the doctor can evaluate which particular medication appeared to have the best result. The notion that a patient will become intelligible to scientific medicine is only possible in those places of the hospital that have available the technologies and routines that make such intelligibility feasible. Even then, intelligibility does not necessarily equate with certainty. In her ethnography of medical students in Malawi, Claire Wendland notes that biomedical technologies were presenced through medical textbooks, at medical school in the form of lectures from visitors from more wealthy countries, or within enclaves surrounding the hospital to which medical students could send samples or patients, but to which they had no access. The

PAGE 339

339 medical students of Wendlands study were aware of the possibilities of a hightech biomedicine, but had no way to access it. Yet when such enclaves enter into the very under resourced and under funded biomedical facilities such as those where the students of Wendlands study works, med ical personnel do indeed have access to multiple possibilities of biomedicine, simultaneously. The challenge was not whether or not they could access this kind of biomedicine, but rather procuring the kinds of expertise or afflictions that allowed one to c ross into the enclaved biomedical regime. Dr. Mtenga might practice a form of biomedicine marked by subjunctive tense and institutional scarcity, but if she entered into the CTC clinic to do consultations, she was adept at code switching to conform to the routines and procedures that REFLECT had so adeptly paid her to learn in a previous workshop. This kind of code switching was also common for Dr. Ezra, who would often work in the outpatient department or to do night shifts. In the context of the nonencla ved portions of the hospital, the kind of biomedicine that Dr. Ezra practiced on a daily basis in the CTC was aspired to due in large part to its technological capacity and the status ascribed to such enclaved expertise but it was largely unattainable in t he non enclave D epending on ones sex, age, socio economic status or ailment, certain people can be made meaningful as patients in the hospital, while others remain nonpatients. Yet while the biomedicine of the CTC or RCH clinic was different than that practiced within the male ward and maternity ward, as shown by the vagaries of electrical current and the informal income generating activities of the staff, the enclave was unable to completely circumvent the logics, logistics, and realities of the cont ext within which it operated (see also Malpas 2003). As Tumainis case illustrates, the politics of the spaces of biomedical practice, where the context is marked by scarcity and inadequacy of expertise, technology, and infrastructure, where the

PAGE 340

340 professionals who are to enact the practices by which maladies come to be known have poor morale and are not only under resourced but also under compensated, there are no assurances that the bureaucratized biomedical practices of one place at the facility with autom atically be observable for the institution as a whole. Through the cases above it is possible to discern the parallel standards of patient care that exist at Kiunga due to the presence of enclaves. I n resource and personnel deficient areas of the hospit al, the standard is to ensure payment from patients for services in order to be able to ensure that the health facility can become financially self reliant; in the enclave, the standard is to facilitate the bodys translation into a manageable and measurab le entity upon which the enclave can intervene, while simultaneously encouraging the person about a future possible only due to the generous procurement of life saving (and sometimes, life long) biomedical therapies. It is to the politics of standardizatio n and bureaucracy within the health sector that I now turn.

PAGE 341

341 Figure 81. Pediatric overflow area in the female ward, Kiunga District Hospital. On the other side of the dividing wall lies the area dedicated exclusively to female patients. Photo by N. Sullivan. Object 8 1. Women in the maternity ward of Kiunga District Hospital sing a hymn during the night shift. August 2008 (.wav file 784 KB).

PAGE 342

342 Figure 82. Neonatal section of the maternity ward at Kiunga District Hospital. Note several beds have two women per bed. Photo by N. Sullivan.

PAGE 343

343 Figure 8.3. A worn wheelchair without footrests sits beside a newer chair with a broken wheel. Kiunga District Hospital. Photo by N. Sullivan.

PAGE 344

344 CHAPTER 9 APPLES WITH APPLES : THE AESTHETICS OF GOVERNING WITH NU MB ERS It is October 2008. Over 200 representatives from among the D evelopment Partners (D Ps ) the Ministry of Health ( MoH ), and other stakeholders (nongovernmental organizations [N GOs ] faith based organizations [ FBOs ] community based organizations [ CBOs ] representatives from the private sector) sit together in a grandiose conference room within the luxurious Kunduchi Hotel a five star resort overlooking the beach, complete with east Africas largest water park. Tables are strategically placed in long, con centric rectangles so that the individuals representing each organization can easily see the head table where sits the Mo H s Permanent Secretary, the Minister, and the Chair of the Development Partner Group ( DPG) as well as the faces of the other representatives in the room. They are assembled for the Joint Annual Health Sector Review (JAHSR) meeting to discuss the draft of the Health Sector Strategic Plan III (HSSP3). This looming air conditioned space, with its windows draped with heavy embroidered curta ins, conjoined tables donning bright white tablecloths, and servers in flowing AfroArabic inspired uniforms offering bottled water and samusa snacks, is the place where the framework for health policies governing the next five years will be negotiated. The decadence of this meeting place stands in stark contrast to the under resourced and often neglected places where those policies will be implemented. The HSSP3 was drafted three months ago, by a team of officers from the Health Sector Reform Secretariat, along with a few appointed individuals from among the DPs consulting. In the conference room, the overall tone is tense as, page by page, line item by line item, the MoH defends its draft, and the DPs and various other stakeholders question its validity a nd practical applicability. Several representatives make suggestions that they feel will make the HSSP3 into what they call a better document.

PAGE 345

345 It is the second day of the meetings, and the one that for some DP s members is the most important. Today, the group is delving into the various indicators that are included in the HSSP3, and ensuring that whatever vertical program or sector wide objective they support is adequately represented in the drafts stated goals and the indicators by which those objective s will be measured. The HSSP3 will become a policy document that guides the entire sector, but will be especially applicable to district level health institutions, where district health plans and budgets are crafted, policies carried out, and raw data coll ected. As the DPG Chair recognizes each representative to speak to the assembly with a microphone brought to his or her table, another suggestion for improving the phrasing of an indicator is put forward. After nearly a dozen bi lateral donor delegates ha ve offered ideas for changes, Dr. Faustin Njau, the Head of the Health Sector Reform Secretariat, stands and addresses the crowd. He reminds the participants that the HSSP3 must be written in such a way that the district administrations will be able to understand it. The indicators must be applicable on the ground. Subsequently, an officer representing CBOs says that he wants to know what assumptions are behind the table of indicators for monitoring and evaluati ng the sector. He questions whether the listed allocated funds adequately correspond to the overall health goals of the HSSP3. A few minutes later, the conversation reveals an interesting paradox. The Swiss Development Cooperation representative stands and states that the framework for the monitoring and evaluation table is not sophisticated enough to reflect the actual system. A Department for International Development144 representative adds that some of the indicators have been disaggregated by gender, and she feels that more i ndicators should be se parated out by gender. An Irish Aid officer follows that the participants need to ensure that the indicators they decide 144 DFID is Britains official foreign aid agency.

PAGE 346

3 46 upon for inclusion in the HSSP3 can be measured every year and included into the national Health Management Information System (HMIS) The officer feels strongly that only those indicators that can be measured statistically annually should even be included in the HSSP3. At this point Dr. Njau stands and takes the microphone. He reveals that several of the indicators on the monitoring a nd evaluation table are related directly to the Millennium Development Goals (MDGs). Since Tanzania is a signatory to the MDGs, they are required to maintain these indicators in the HSSP3, even if they are not measurable. Therefore, some indicators simply cannot be rejected because they reflect multi national agreements to which Tanzania and the DPs are signatories Dr. Njau illustrates by pointing to several empty cells on the table being discussed. It is difficult to turn goals like the MDGs into measurab le indicators. He says that ultimately they were filling too many forms, many of them for HMIS and he feels that ultimately they should move towards single indicators that are simpler to track in such a way as to reflect what progress has been made. Clearly, power over health sector policy is not solely in the hands of any one of the organizations and entities present in the room. It is contingent, negotiated in a room where a variety of different interests converge. T his chapter takes documents statistical representations, and policies seriously as meaningful objects of ethnographic study. They are that through which a variety of different power structures are embedded, recontextualized, and even circumvented. Below, I outline an existing lexicon that i s helpful for thinking through the relationship between documents, facts, representation, and governance. I then turn to the ideals underlying the use of medical statistics, highlighting the perceived role of statistical representations in governance, and the aesthetic criteria that guide them From there, I describe how donors shifting visions of the ideal health information data collection system based largely on universalistic notions of the proper

PAGE 347

347 aesthetics and roles of statistics for governance com plicated the ability of the MoH and its regional governments to generate data useful for health sector planning. The chapter then moves to consider the discordances between the ideals of health statistics and the paradoxes of the ways that health statistic s are used in practice to create policies, and to represent Tanzanias health sector on a global scale. Overall, this chapter demonstrates the considerable discordances between idealized view of medical statistics and the complexities of how they are enact ed and engaged. I show that even facts that are hotly contested by those who use them still do immense work in the world, whether through governance systems, or through representations of reality to the world. Further, the controversies surrounding data and the policy documents they inform demonstrate that within the universal languages employed by transnational aid regime there is considerable instability. A Vocabulary for Thinking about Documents S ince at least the 1980s, s cience and technology studie s have explored how very powerful forcesscience, technology, globalization actually work in practice.145 One of the most well known of these scholars is Bruno Latour, who created a vocabulary to describe the minutiae of practices and actions that collective ly produc e scientific facts. What emerges from his work in laboratories (see Latour and Woolgar 1986; Latour 1987) is that scientific institutions do not create facts in and of themselves, but rather produce a variety of documents that become the basis o f scientific depictions of reality. In these studies, Latour describes in detail the ways that certain depictions of the world become black boxed that is, they are no longer questioned or scrutinized but stand alone as facts. I draw upon Latourian voc abularies regarding 145 In addition to work in science and technology studies, several anthropologists and other schola rs have looked at the social relations and practices that surround objects (documents, medicines, technologies) in order to draw attention to ways that they bring the local and the gl obal simultaneously into focus, and highlight practices that actualize th e global forces within local sites, and local forces within global sites (see for instance Anders 2010; Mitchell 2002; Riles 2000 ; Whyte, van der Geest and Hardon 2002).

PAGE 348

348 measurement and documentation in order to describe the practices surrounding documents at the macro level (Latour 1987), and how these are variously engaged by different actors drawing upon them. For some actors, the documents and stati stical measurements relating to the health sector in Tanzania are black boxed. For others, they continue to be plagued by controversy. Latour developed a lexicon that assists us in understanding how certain measurement s are made into representations of pa rticular and vastly separated sites, in order to permit domination at a distance (1987:223) .146 Thinking about the graphs, tables, and charts characteristic of scientific publications, Latour trace s the processes by which these summarized exhibitions of reality come to be constructed. Based on his terminology, an inscription is the final visual display of data that is presented to the scientific community the graph, chart, table, histogram, or other diagram that comes to depict fact or reality. The basi s for something being defined as an inscription is that it creates a final visual display that is later taken up in the final presentation of a scientific reality. Inscriptions can therefore be graphs, charts, figures, or other such representations published in scientific texts as the basis of making claims about the world. Latour calls for a deliberate look behind the inscription to the actions involved in its creation. Taken together, the variety of actions, objects, technologies and people that are mobi lized to create a particular inscription are what Latour calls the instrument. Thus, what makes up an instrument can be a variety of different resources, technologies, objects, and actors, all working to contribute to creating that particular display tha t is later published in a scientific report or text.147 Within the instrument are a variety of measurements that become summarized, synthesized and re contextualized in the process of creating the final inscription. These 146 Miller and Rose (1990) have a similar argument, that complex technological and proce dural mechanisms had to be put into place to instill self discipline and self control in populations and individuals in order to be able to govern at a distance. 147 How the instrument works at Kiunga District Hospital is covered in Chapter Ten.

PAGE 349

349 measurements that are not visibly r epresented in the final inscription but nonetheless informed its display are called intermediate readings. Thus, he states, The instrument, whatever its nature, is what leads you from the paper to what supports the paper, from the many resources mobilis ed in the text to the many more resources mobilised to create the visual displays of texts (Latour 1987:69). Importantly, for Latour, the variety of int ermediate readings that underlie inscriptions remain invisible provided that the validity of those read ings does not come into question. At the point when an inscription is considered a fact, the plethora of intermediate readings that contributed to its creation are black boxed, because the fact is not questioned. Based on this lexicon, the first measurement of a particular object or process of interest, which later informs the intermediate readings, is called a primary trace. Thus, to illustrate for a hospital context, a primary trace would be the tick mark put on a tally sheet tracking the number of p atients utilizing a particular clinic. Those tally sheets might be collected for several facilities in a given area and then summarized on a graph to represent the total number of patients using those services in a given region. This graph would be an int ermediate reading. If the graph is sent to the central government and collated with similar graphs from all of the other areas of a given country, and then summarized in a histogram or pie chart that the government then publishes as representing those ser vices overall for the entire country, at that level it would be an inscription. That inscription might be taken up by scholars or others, compared with similar inscriptions from other governments elsewhere in the world, recombined and made into still more inscriptions that are said to represent something about the world Behind the Indicators : Ideals of Governance and Aesthetics I became interested in these global health statistics about Tanzania when I was at Kiunga for a simple reason: beyond providing biomedical care, health workers generate a lot of data. They write in files, produce reports, fill in registers, and tick off tally sheets. At Kiunga, what I

PAGE 350

350 noticed was that different kinds of documents required different kinds of inputs, and some of the processes for of filling in forms and creating documents were more technologyintensive and regulated than others. Many of the official forms and registers for tracking particular kinds of data at the hospital had the emblem of the MoH on their headers or covers. Some of the forms and registers had emblems of other organizations, like REFLECT,148 for instance. Still others were created by hand and did not bear the emblem of any given organization. As I observed the countless and mundane practices surrounding documents, I began to notice that not only did the engagements with and importance placed upon these documents differ, but also, some data inputs were compiled into larger documents and reports, which were sent out of the hospital. In October, when I atte nded the JAHSR meetings and observed the considerable anxieties that donors and MoH officers express ed regarding data collection processes, suddenly the Kiunga staff s engagements with documents took on a new significance. At JAHSR, one of the overarching complaints articulated during the meeting as well as informally among the many donor representatives I encountered, was the poor quality of the available data to depict what was actually going on, on the ground. I discovered that donors hotly debated th e very kinds of indicators that were uncontroversially displayed on the W orld Health Organizations (W HO ) Country Health Profile for Tanzania those relating to progress on the MDGs for instance. As in the preceding example, members among the DPs saw that it was not possible to measure MDG indicators with the existing HMIS system. Suddenly, the streamlined tables of indicators produced by the WHO and the U nited N ations (UN) seemed particularly intriguing, if not suspect. I f so many of the donors financ ing l arge proportions of the health sector complain that the data available are inadequate or inaccurate, where is this inadequate data coming from? Is 148 As w ill be remembered from previous chapters, REFLECT was the American NGO operating the HIV/AIDS clinic at Kiunga with funding from the Presidents Emergency Plan for AIDS Relief (PEPFAR).

PAGE 351

351 this inadequate data the same data that later appeared on WHO and UN websites and reports, or do they are t hose reports based on other data ? If the WHO and UNs data is of good quality, why is it not being shared with the DPs and governments of these aiddependent countries? These were some of the questions that became of interest to me when observing the ways that the health sector was governed from above. I soon discovered, these questions were not so easily answered. In order to delve into these questions, first I had to understand what accurate data meant to those actors who perceived paucity in the existi ng data. Within the transnational aid domain (Gould 2005:63; Chapter Four), there is a remarkable universality of language employed by the entities funding development initiatives throughout the world, in health as well as other sectors (see Anders 2005; G ould 2005). This section delves into the assumptions underlying the universal language relating to data and documents characteristic of the transnational aid domain. The ideals underlying the DPs and UN organizations faith in numbers (Porter 2005) were based on assumptions regarding governance, and aesthetic criteria that, within the culture of transnational aid, were evidence of compliance. Health Statistics, Quantitative Reality and Governance S tatistics were first used in biomedicine beginning in the 17th century (Jorland and Weisz 2005). At that time, some European countries began collecting statistics about causes of mortality. By the 18th century, medical employed numbers to record the diseases and environments of various areas, especially in c ities where epidemic outbreaks were common In the 19th century, European governments looked to statistics to evaluate epidemics and other diseases, and these statistics were seen as a way to assess social dangers and the degree of progress achieved in a p articular state: This was not merely a vehicle for the guidance of public interventions and the expansion of state power. Statistics became also a tool of liberalism, a way of informing the people about the condition of society and about the effectiveness of their

PAGE 352

352 governments (Weisz 2005:395). Health statistics held the potential to inform, but also to assert accountability to subjects being ruled. Today, given the sheer number of health facilities, maladies, environments, and health programs operating i n a country like Tanzania, one of the main means by which governments and donors shape their understandings of conditions at the local level is through numbers and the statistical representations used to summarize them. Indeed, because health sectors are t ied to myriad other governing entities in most countries in the world (governments, insurance companies, law firms, banks, donors), the interactions between patients and doctors are punctuated by a variety of technologies and practices meant to quantify an d thus objectify the extent of particular maladies and the effectiveness of interventions to address them : The relationship of patient to doctor, once relatively unconstrained, is now bound up with myriad systems of regulation and oversight and intricate problems of trust and authority. If the expanding role of quantification in medicine is part of the history of science, it is also one of the forms of knowledge that mesh most effectively with the prevailing contemporary forms of administration and government. (Porter 2005:400) At the top levels of transnational governmentality from the level of the state to the level of the DPs and supranational organizations that inform its policies there is considerable commitment to and anxiety about the power of numbers to not only reflect reality but more importantly, as Weisz suggests above, to demonstrate the effectiveness of governance initiatives. Here again, an economy of appearances, allowing governments to appear accountable to their citizens so that th e latter continue to invest in the state Thus, biomedicine is no longer merely about a therapeutic encounter, but an avenue to allow the MoH and the DPs to govern from afar, and to present an image of their efficacy to stakeholders (tax payers, citizens) at home. Underlying the frameworks of medical statistics and reports is a theoretical commitment to a universal la nguage that will allow multiple, disparate particularities to appear compatible in

PAGE 353

353 order to be able to compare them. The universal language used to describe the realities of the world is negotiated, and once decided upon, is used to make comparable generalizations about reality. Anna Tsing suggests that collaborations [between differently situated actors] create convincingly agreed upon obse rvations and facts that then appear to support generalization directly that is, without the prior mediation of the collaboration. The contingency of the collaboration, and its exclusions, no longer seem relevant because the facts come to speak for themselves (2005:89. Emphasis in original). Thus, through collaboratively agreeing upon the indicators of the HMIS or the HSSP3, differently placed actors within the MoH, the DPs and the other stakeholder organizations create a space where generalizations acro ss space are made possible, in order to be able to recognize particular figures as reflecting certain realities These quantitative representations then in form what courses of action are planned and later published in policy documents As Stirrat articulates related to the philosophical assumptions underlying development documents, to count it is to know and to control (2000:36) To donors, data presented in preset ways produces the appearance of accountability, transparency, and good governance. Data ena bles rule across disparate spaces, from afar. If one can organize a complex world into a set of seemingly objective indicators, one can determine how best to change that world in predetermined ways based on set formulas (Stirrat 2000:37). The power of obj ective measures to reflect reality is articulated across medical statistics reports at multiple levels from the offices of the government of Tanzania (GoT) to the publications of the UN umbrella organizations and beyond. For instance, despite MoH and donor s at the JAHSR expressing concerns about poor quality of data, the Annual Health Statistical Abstract Tanzania Mainland 2008 a document made up of the very statistics that were so controversial at the meeting attributes a specific quality to the health d ata contained

PAGE 354

354 within In the Foreword of the document, the MoH Permanent Secretary writes that the MoH created the abstract with the aim to provide a clear picture on the performance of routine health delivery (MoH 2008:8, emphasis added). There is no indication here that there may be anything amiss in statistical representations (inscript ions) produced in the abstract, although the terminology aim to provide subtly suggests that it may be a goal not fully attained. A similar assertion about the power of medical statistics can be found in the recently published World Health Statistics 2010. If one downloads the current report, one can find an interesting set of contradictions. In the third paragraph of the documents introduction it states, Taken together, these indicators provide a comprehensive summary of the current status of national health and health systems (WHO 2010:8, emphasis added). The data are portrayed as complete wide ranging, comparable, and able to provide a clear picture of the health status of populations and the state of biomedical systems in place to address them.149 The Foreword of the Annual Health Statistical Abstract 2008 also acknowledges the efforts of those individuals and institutions that contributed to the data inputs on whic h the abstract was based. In Latourian terms, these individuals and institutions are part of the instrument producing the intermediate readings that make up the final inscriptions found in the Abstract Parties mentioned as having contributed to the produc tion of these inscriptions include: [T]he health facility personnel, Council and Regional Health Management Teams (CHMTs and RHMTs) at different times they contributed to data collection and compilation at the national level inputs were made by directors o f hospital services, preventive services, policy and planning, administration and personnel, human resources, Chief Nursing Officer, programme managers and heads of Sections. (MoH 2008:8. Parenthesis in original ). 149 Notably absent from the World Health Statistics compiled by WHO are the other therapeutic systems that provide care throughout various countries.

PAGE 355

355 Clearly, obtaining the intermediate readings required to produce inscriptions of health statistics for Tanzania required that a number of individuals and resources provide necessary inputs individual personnel and district CHMTs collected data, and these readings then moved up the scale where they were compiled with still other readings produced and aggregated by other individuals with other technologies. The Role of Aesthetics in the Economy of Appearances Since HSR was initially conceptualized, the MoH has collaborated continually with the DPs t o be able to create, and effectively implement, a data collection and policy distribution system that appeals to donor aesthetic interests. The structure of documents like the HSSP3 outline a path for how biomedicine and bureaucracy should be practiced wit hin health facilities countrywide ; however, this is not their sole function. In addition, policies generate a plethora of guidelines for the procedures that should underlie certain kinds of data collection, and the aesthetic criteria by which it should be presented (see Gould 2005; Riles 2000; Stirrat 2000). As a part of the MoHs economy of appearances to make itself attractive to potential investors, these standards are not only meant to measure particular aspects of the health sector, but also to ensur e those measurements are configured in a form that is recognizable, and appealing, to the DPs and other stakeholders (Dunn 2005; Stirrat 2000 ). Exploring the ways that environment advocates learned and later employed rhetorics of conservation, Anna Tsing states that if they do not portray resources and communities in terms that are recognizable by the state, national elites, and international experts, their advocacy will be entirely ineffective (2005:199). In a similar vein, not only the MoH, but also individual health facilities, had to learn the aesthetics and rhetorics of health policy and development initiatives to make themselves recognizable to their primary stakeholders. The MoH adopted, and at times subtly adapted, indicators and report aesthetics that would be appealing and recognizable to

PAGE 356

356 donors in order to secure their investment. It was a deliberate establishment of an economy of appearances directed at securing investment from the international donor community. In Tanzania, at the level of the district health offices throughout the country, districts that learned how to produce good specimens of a particular genre (Riles 2000) and submitted these reports and documents on time found that they were subject to more donor and government investment than those districts that did not produce and send these documents. Returning to Tsings argument, it was necessary for health offices to learn and adopt the rhetorics of global health and development that the Tanzanian MoH made national policy based on global health policies and the priorities of the transnational aid domain. Had they not, their facilities would have remained beyond the purview of state or donor interest and investment. Learning not only to provide data, but to amass it based on specified aesthetic criteria, was paramount to accessing the resources health facilities required. The MoH had similar motivations for drawing so heavily upon global health rhetorics and donor interests doing so made the Tanzanian government appear like a good investment. At the levels of the state and the transnational aid domain, the production of documents is an aesthetic exercise. Importantly, the aesthetic style of particular documents is not merely about appreciating a certain beauty, but is also crucial t o the ability to secure donor investment. The formats of various development policy documents is prescribed, and the internalisation of a rigorously formalised aesthetic for the production of such documents is considered a prime indicator of improved capa city (Gould 2005:79). In essence, the capacity of a country or project to be able to develop is critically tied to its ability to present itself convincingly in a pre determined way. During the JAHSR, the various DP representatives were not questioning the validity of the genre itself, but rather the ways to best fit the complexities of Tanzanias health

PAGE 357

357 sector to the aesthetic of a prescribed policy document. I now turn to how donors and the MoH attempted to create an idealized health statistics collect ion system within Tanzania, and the difficulties in achieving this idealized standard. Setting up the Ideal Health Management System This section turns to the establishment of the HMIS itself in order to provide a point of comparison for the data systems c haracteristic of donor programs During the 1980s, Tanzania had several datacollection systems in place. A team was assembled to assess existing data systems. The systems were found to be highly fragmented and often overlapping, thereby increasing the bur dens to the health facility personnel who were expected to generate primary traces and intermediate readings. After Tanzania agreed to the SAPs, various stakeholders in the health sector began exerting pressure on the Tanzanian MoH to design a new datacol lection system that would pave the way for more reliable, integrated decentralized health sector governance. In 1989 1990, an external consultant joined a team of Tanzanians in designing an HMIS, which was piloted in 1991 in Iringa and Mbeya regions. The pilot experiences were used to hone the HMIS, which was then implemented in phases beginning in 1993 (Rubona 2001). In Tanzania, this HMIS was called the MTUHA ( Mfumo wa Taarifa za Uendeshaji wa Huduma za Afya ) Initially meant to be a paper based system, the district w ould be the main operating center compiling data from its individual health units It would then send this data to the regional level, where it would be collated with data from other districts and sent to the MoH (Kimaro and Nhampossa 2005). Computerization of the MTUHA system began in 1993, but the software, based on a dBASE system, had numerous glitches. Ironically, after creating the software, the original developer was no longer available to fix the glitches so a second was hired. Yet by this time, DANIDA had done an evaluation of the dBASE system and decided that several fundamental

PAGE 358

358 ch anges needed to be made to the types and frequency of data collected (Kimaro and Nhampossa 2005). In essence, reorientation of donor interests undermined t he original data collection system that the Danish government helped to initially establish. Ironically, despite the decision having been made to create a new database system, i n 1997, the dBASE system being implemented within a ll of the regions, and healt h workers were already trained on how to use it (Rubona 2001). However, w ith financial backing from DANIDA, a new system based on Microsoft Access database software was introduced in 1998, but it also suffered numerous problems, and, with funding from the German bi lateral donor ( GTZ ) the system was overhauled again in 2003. S everal of the Tanzanian health workers at the regional level complained that while they received training on the dBASE system of MTUHA, they received the conf igured MS Access system without supporting training be able to understand the new softwares use or capabilities. Further, when designing the database, software developers did not consult with e ither the MoH or the regional governments about their needs for the system, resulting in complaints that the data was of little relevance to the local levels it was supposed to serve. A regional health information officer interviewed in 2004 stated: Our work is based on the higher level needs, for example, the districts and regions write their reports based on the national level guidelines. We are not able to write based on our own capability. We use the Ministrys guidelines as our own needs. We tell the district that we need this kind of report and then they make that report for us. This is the kind of behavior we have created in our society. (Quoted in Kimaro and Nhampossa 2005:286, emphasis added ). Under HSR, while responsibilities for planning health programs shifted to the district level, according to this officer, the design of the system did not account of the kinds of information that might be most useful at lower level s (regional or district levels) where health plans and budgets would be implemented. Instead, the regional office was required to produce a report following the aest hetics outlined at the national/donor level (the higher level needs) but this genre was of

PAGE 359

359 little relevance to those creating it. Ironically, from the very beginning, donors consultations and decisions regarding establishing statistical oversight in Ta nzania undermined the very efficacy of the systems they wished to establish in the country. They successfully taught regional health workers to produce the correct types of reports, but the data itself was not meaningful for planning health activities at the local level. It was aesthetically correct, but lacked relevance to the very offices meant to utilize them for planning and budgeting at local levels (Stirrat 2000). Paradoxes of Health Statistics in Action If we apply Latours ideas about inscriptions to the myriad reports and statistics that are produced about aid dependent countries like Tanzania, important questions come to light. Many bi lateral and multilateral donors and other global organizations draw on statistics to talk about the conditions i n socalled developing countries. Indeed, one might visit the website of any bi lateral donor involved in the health sector in Tanzania and find that their reports draw on statistics that multinational institutions such as the WHO or the UN generated and published. These statistics are supposed to tell us something about what is happening in Tanzaniahow healthy the population is, how developed it is. These inscriptions travel widely, but leave essentially black boxed the efforts undertaken in creating them, and the uncertainties underlying them. For instance, on the WHO country profile of Tanzania,150 under the heading mortality and burden of disease one finds the category country profile on regional site. If one clicks on this link, the WHO website opens another page where it is possible to download the WHOs Country Health Profile of Tanzania 2006. Once downloaded, we see inscriptions in this case, a series of tables that depict particular indicators that are supposed to illustrate to us 150 http://www.who.int/countries/tza/tza/en/ accessed August 12, 2010.

PAGE 360

360 impor tant information about the state of the health system in Tanzania This document provides inscriptions on the total fertility rate, life expectancy at birth, probability of dying within a particular age range (between fifteen and fifty years, before age fi ve), infant mortality rate, neonatal mortality rate, maternal mortality ratio, causes of death for children under five, HIV and tuberculosis ( TB ) prevalence, and further down in this publication, a series of indicators related to the MDGs for which data is available for Tanzania. For each row dedicated to a particular indicator, there is reference to the year in which those statistics were collected. The years span from 1990 to 2005. Upon closer review, we see that there are a variety of indicators for wh ich data are not filled in These are marked by three asteri sks and a reference at the bottom of the form that data not available or not applicable. The citation for these data is the World Health Statistics 2006 Strikingly absent on the inscription (in this case, a series of tables published by the WHO of the health profile of Tanzania) or on the World Health Statistics website is a detailed explanation of how these data were amassed, by whom, under what conditions. Wanting confirmation about where t he WHO and the various UN agencies retrieved their data and failing to find more than an opaque reference to data sources on the agencies websites, I consulted a United States Agency for International Development ( USAID) Tanzania development worker, Robe rt. He had worked in various countries on healthrelated programs for nearly three decades. A t USAID, Robert and his co workers rely on quantitative data to make programmatic decisions regarding what projects the agency will support and to evaluate what e ffects their funded interventions have within sites hosting these programs. In many cases, the quantitative data consulted is that generated by the MoH. In an email, I asked Robert about

PAGE 361

361 where the WHO got its data that subsequently went into their country profile reports and other publications. He responded: Robert: God only knows where WHO and UNICEF get their data. They try to use similar sources all over the world, so they can compare apples with apples, and since UN Agencies (including WHO), are governe d by the countries as a whole, they do use a lot of HMIS data In most of the countries where I worked, HMIS data is garbage, with data which is fabricated, or just not collecte d in over half the facilities. Garbage in, garbage out. (Email correspondence, 09 2710) Roberts characterization of the data as garbage indicates that he felt that the data did not represent what it should represent based on the aesthetics and ideals of governance underlying the transnational aid regimes universal language of da ta and representation. Along the same lines, a study interviewing various stakeholders involved in the health information system in Tanzania found that there was a general mistrust across all levels regarding the data being put together in the HMIS (Smith et al., 2008). In this study, MoH officers reported site visits where staff forgot to use the appropriate forms during ward rounds: They [the MoH officers] believe that at the end of each month the health workers hazard a guess at the appropriate data an d fill in the HMIS books accordingly (Smith et al., 2008: 9). In my own research discussions with MoH officers also depicted an unflattering picture of the data collected on the health sector. Getting districts to comply with reporting regulations has be en an elusive process, particularly in the most under resourced rural districts. As one expatriate development consultant within the Ministry of Health described to me151: We dont have a very precise health information system in Tanzania because we have not been able to create very effective mechanisms to secure timely and high quality data. If you go to the districts in many developing countries at the subdistrict level facilities, you will find that sometimes they are not filling in the forms. And then you ask yourself: Okay, if they dont fill in the forms, what then? 151 This consultant was employed at the MoH and technically worked for the MoH, but the bilateral donor paid his salary. That said, he reminded me on numerous occasions that he did not represent the bi lateral donor, and that he disagreed quite often with donor aid policies more generally. He considered himself an employee of the MoH.

PAGE 362

362 What happens? And you will find out in some countries, that some of the health facilities just copy from last time when they actually filled in figures. It happens, its not half of the fac ilities, that are not what I am saying, but it happens that you see it. Because they dont have a supervision system that takes the information seriously (Interview, English, 0923 08). The consultants depiction of facilities not generating the correct da ta suggests that there are implications for not doing sowhat then? If the data are required in order to govern from afar, but are not reflective of the predetermined and standardized portrayals of reality that the MoH and donors negotiated into policy, what then for its applicability to policymaking and governance? What does the use of such inaccurate and inadequate data suggest in terms of how governance is negotiated and implemented? Controversial Facts It turns out not all donors are willing to implement programs and monitor and evaluate their progress by drawing on data deemed garbage. It is very common for bi lateral donors (particularly those funding vertical health interventions) and transnational NGOs152 to produce their own data parallel t o that produced by the central governments of aid dependent countries. For instance, in 1984, USAID began funding an alternative data collection system to inform American foreign assistance policies and programs This data allows USAID to monitor and evalu ate their programs in the specific sites within the countries where they operate. These data are collected in a variety of ways. Every four to five years, a Demographic and Health Survey (DHS) is conducted in a host country,153 and unlike the HMIS, which rel ies on under resourced health facilities to diligently report data, USAID has a degree of control o ver the quality of the DHS data; USAID contracts individuals to collect data at the local level through facility or communitybased surveys. These data are often used to inform USAID affiliated programs, but 152 Often running programs w ith funding from bi lateral or multi lateral donors. 153 See http://www.measuredhs.com/. Accessed Sept. 27, 2010.

PAGE 363

363 are not generally representative of averages at the national level. According to Robert, We do constantly look at data, and compare national reported information with the local information [we collect on the ground], and one survey with another and all with HIMS. But it is not easy. He provided an example from the Presidents Malaria Initiative (PMI) in Tanzania to illustrate his point: The GOT and PMI wanted to know what is the national prevalence of m alaria this year [in Tanzania]. We did six or seven surveys in 2008. These were all good surveys. But not all were national, and they were taken at different times of the year. As you know, malaria is regional, and seasonal, and seasonal regionally. The r esults were all over the place. The MOH and PMI were so upset that no prevalence studies were done in 2010. This is stupid because we need data; we just have to acknowledge that it doesnt make sense to state a number for yearly average of national data. (Email correspondence, 09 2710). Roberts comments point to an interesting and somewhat concerning paradox in the ways that global health governance in aid dependent states is actualized. The actors involved in creating policy (the DPs and the MoH) are aware that the data upon which they rely is faulty, yet they build policies based on those data measurements, and create indicators tied to policies and global health agreements (such as the MDGs) that they are aware cannot be measured effectively. Fur thermore, Roberts statement that it doesnt make sense to state a number for yearly average of national data acknowledges that the problem lay not only with the infrastructure underlying the creation of data (since this data is garbage in, garbage out), but also, significantly, with the fact that some kinds of data could not be made to reflect certain realities, no matter the quality of the software and system used to produce it. Nonetheless, w e see here little to suggest that this data is garbage. The data in MoH and UN reports is asserted to be a clear picturedevoid of any controversy. The data stands as fact, even though numerous members of the transnational aid domain find that very same data suspect. Critiques of data inadequacy and inaccurac y are due to the reports providing little of the

PAGE 364

364 idealized, universalistic precision necessary in order to be able to govern effectively at a distance (Miller and Rose 1990). Despite these controversies, within global health publications and HMIS reports data represent. It is simply that the stakeholders involved with governing global health as well as health policy and budgeting in aiddependent states are aware that the data does not represent what their ideals dictate it should. The ideals do not fit the reality. Yet this data works. It travels. It is compared and combined with data generated elsewhere. It is employed to rank Tanzania among other countries in the world in terms of health status, health systems, development. This data provides the foundation for policy and budgetary collaborations between the MoH, the DPs and other stakeholders. The very basis of an inscription being perceived as fact in Latours study is that these inscriptions are not surrounded by controversy. However, unlike the neatly packaged inscriptions characteristic of Latours study, the inscriptions created in Tanzania are rife with controversy for those very governing bodies that aim to administer to the country at a distance. During the JAHSR, t he MoH and DP officers expr essed considerable anxieties over the kinds of indicators and quality of data. But no one challenged the applicability of HMIS statistical measurement s for governing the health sector even though development consultants such as Robert were well aware of t he discordances between faith in data and data in practice. As Stirrat argues, the frameworks of statistical documents contain a shared underlying assumption that the world can be known objectively by applying rational measurements to predetermined categor ies. These universal categories are laid down within which the world can be neatly arranged (2000:36). The main assumptions underlying this approach to governing are never questioned publicly, although in private they are the subject of grave concern. Of ficially, failure to achieve

PAGE 365

365 universalized standards of data collection is seen as the problem; there is no question that the epistemological assumptions underlying the data are sound. Purging Uncertainty for the Public Meanwhile, the erasure of debates i n the publication of the WHO Country Profile brings to light an other interesting paradox: contentious inscriptions at the level of health sector governance produce anxiety, yet stand uncontested at the transnational l evel in WHO statistical reports. How mi ght this be explained? Tracing controversies about the use of neviraprine in HIV positive pregnant women in South Africa, Didier Fassin notes that when it came to providing information to the public about the drug, all aspects of uncertainty were purged: The consolidation of new facts into truths and their translation from scientific circles to the public arena are mutually reinforcing. Hesitation gives way to certainty and margins of error disappear. Truth solidifies (Fassin 2007:85). The process appears similar for HMIS data. At the level of donorstate governance in Tanzania, the data is contentious and cause for serious concern. When it moves to the transnational sphere and becomes publicly accessible, as in the World Health Statistics reports mentione d above, the controversies underlying the data are pushed to the margins or disappeared completely. In a study of institutional activities and knowledge production among networks of international aid agencies, NGOs and government offices in Fiji, Annelise Riles isolates two kinds of documents that were produced at the interstices of these organizations (2000) The first type was meant for consumption within the network, whereas the second was meant to circulate an image of the network to the public, or to organizations outside of the network. If, following the Latourian line of reasoning, Riles first type is in our case the kinds of documents that are necessary for collective governing from afar (through the HSSP3, for instance), then the documents circulating among the MoH and the DPs must be of the kind that in some way reflect

PAGE 366

366 what is happening on the ground, in order to assure the potential for their policies to be effective. Because the inscriptions were known by all parties to be inadequate based on t heir own standards of what constitutes good data, the entire process of collaboratively creating policies risks perpetual collapse. At any point, the complexities of biomedical and bureaucratic practices in chronically under resourced biomedical faciliti es might reveal the fragility of the very inscriptions upon which the MoH the DPs and stakeholders rely. That there is considerable anxiety about the quality of data for the purposes of governing should therefore not surprise us. Yet as evidenced by the seamlessness of WHO statistical representations of Tanzania, considerable efforts go into obscuring the controversies surrounding data (Rottenburg 2009), and into projecting an image of success (Mosse 2005a; Stirrat 2000 ). Thus, the inscriptions intended to circulate an image of the realities of health in Tanzania to the public necessarily obscure the contestatio ns behind those inscriptions The WHOs World Health Statistics Report 2010 provides an example. In the introduction to the publication, well below the claim that the report provides a a comprehensive summary of the current status of national health and health systems (2010:8), is the only place where the uncertainties of these data come into view: Because of the weakness of the underlying empirica l data in many countries, a number of the indicators are associated with significant uncertainty. It is WHO policy on statistical transparency to make available to users the methods of estimation and the margins of uncertainty for relevant indicators. Howe ver, because of space restrictions, printed versions of the World Health Statistics series include uncertainty ranges for only a few indicators. Further information on the margins of uncertainty for additional indicators will be made available at the Globa l Health Observatory web site. (WHO 2010:8) I only found this reference to uncertainty in the data because I was looking for it. I was unable to find similar depictions of WHO data on the website, nor in donor publications or website inscriptions depicting information about Tanzanias health sector. Anyone interested in further

PAGE 367

367 scrutinizing the data beyond the Report was directed to another reference to yet another website, where maneuvering the data in order to get at its uncertainties is extraordinarily difficult. Yet statistics such as those produced by the World Health Statistics reports often form the basis for a variety of other publications and papers, scholarly and otherwise, about Tanzania and other countries in the world, and the state of health an d development in these places. These facts travel widely, but the processes behind creating them remain black boxedat least, to all but the most savvy who know the inner workings of their production or are willing to scour numerous reports in order to find a small reference to original sources. Ironically, although at the JAHSR all of the parties involved complained about quality of data and the lack of effective monitoring and evaluation possible under the HMIS, in the subsequent presentation after the i ndicators were agreed upon, these very (problematic) data were those upon which all parties relied in order to inform the directions in which they planned to take the health sector in the next five years. The inscriptions on the PowerPoint slides shown at the JAHSR to depict the state of affairs in the health sector in Tanzania conformed to the frameworks necessary to allow them to be compared to data in other countries: comparing apples with apples. Nevertheless, the data did not really depict in a desir able way what was happening on the ground in Tanzania generating considerable clamoring among those parties involved in health sector governance at the level of the state These inscriptions permit little quantitative familiarity with the places they cl ai m to reflect. The MoH and DPs govern partially, even somewhat blindly. Yet they still govern. Importantly, the fact that these numbers are incomplete, that from the time they were created until the time that they are mobilized in order to govern they were tainted by doubt and contestation never appears on the clean inscriptions made available by the WHO. Here we see an important divergence from Latours conclusions for facts.

PAGE 368

368 Even facts surrounded by controversy are powerful. Even controversial facts can be come black boxed. Conclusion By taking medical statistics, and the documents and policies through which they travel, seriously as ethnographic objects, a variety of paradoxes come into view with regards to what numbers claim to do as idealized representat ions of an objective reality, and what they actually do in practice. The universal language of the transnational aid domain requires particular aesthetic criteria be met in order to be deemed credible as representations of what occurs on the ground. Adherence to these aesthetics was necessary in order to allow Tanzanias health sector to be recognizable to donors, and thereby bring about their investment in the state. Aesthetics were key to the economy of appearances enacted by the state in order to sti mulate donor assistance. Further, those same aesthetic representations were central to the donors ability to demonstrate accountability and transparency to the government whose taxpayers funded donor foreign assistance programs. Yet the reality of data co llection in Tanzania also threatened to undermine this very appearance of accountability and transparency. It is thus not surprising that in policy and statistical documents produced both by the MoH and by multinational organizations such as the WHO, data are depicted as comprehensive, providing clean pictures of particular realities on the ground. The contestations underlying data only exist within the network of donors, stakeholders, and government entities in Tanzania working within the health sector. When a quantitative representation of Tanzanias health sector is provided to the public, these controversies are removed. While Latour argues that facts are able to travel because they are no longer subject to debate, I show that even facts that are highly and hotly contested have considerable power. While the facts produced and circulated about Tanzania may not represent what the MoH and DPs wish to be reflected, they nonetheless do

PAGE 369

369 considerable work. The data are used to inform policies and programs and, their inconsistencies obscured or removed, they are compiled at the global level to circulate and be compared with similar data from other distant locales. Facts, I show above, do not need to be above question in order to circulate. However, the controversies surrounding quantitative representations also demonstrate the fragility of global health and state health governance. Desperate to appear accountable, adherent, and transparent, both the MoH and the bi and multi lateral donors involved in healt h related activities in Tanzania are highly concerned about the inability for data about the health sector to meet the ideals set by the transnational aid domain. Governing with data that does not represent what the donors and the government want it to represent, the MoH and DPs govern only partially, with little idea whether the data they employ reflects conditions within the sector. It is to the conditions of data collection, and the meanings of documents at Kiunga District Hospital that I now turn.

PAGE 370

370 CHAPTER 10 THE SOCIAL LIVES OF DOCUMENTS AT KIUNGA DISTRICT HOSPITAL The aim of this chapter is to trace what documents can tell us about global and local power configurations in practice. Since Health Sector Reform (HSR) began and externally funded health pr ograms were instituted at Kiunga, the number and types of documents with which health workers engage have increased exponentially. Whereas prior to 2001, records were scarce, by 2008, there were many documents circulating around the hospital: patient files meeting minutes, laboratory investigation orders, prescription pads, employee files, policy documents, reference manuals, informational pamphlets, district health budgets, national disease registers, clinic cards, data sheets, donor assessments, schedules, correspondence, supply and drug ordering sheets, stocking records, a wide variety different logs, and more. Attention to the biographies of objects first emerged out of Science and Technology Studies (STS) (pace Latour 1987; Latour and Woolgar 1979), but anthropologists have also taken up this approach. Arjun Appadurais edited volume, The Social Life of Things suggests the analytical benefits of tracing how material things travel across multiple settings, how they are valued or exchanged (1986). Drawi ng on this perspective, Whyte, van der Geest and Hardon apply such an approach to medicines (2002). The analytic power of this approach is that it allows for the consideration of the lives that medicines have with people and between people. These lives ar e imbued with the practical artfulness and purpose that characterize technology. They are lived in relation to problems and contexts (Whyte et al. 2002:14). In a similar vein, this chapter explores how people within the hospital engage and are engaged by documents: in short, how documentation practices mediate various interactions in the hospital, and what this tells us about how local, state, and global forces become unevenly embedded, distributed, and mobilized within biomedical settings.

PAGE 371

371 People at Kiunga District Hospital engage with documents every day. Drawing again on the Latourian lexicon outlined in Chapter Nine, s ome of these documents such as tally sheets, registers, clinic cards and supply forms are primary traces. They provide initial informatio n, which later the DMO office will compile, process, and recontextualize as part of the district health plan, budget, and quarterly or annual reports. However, there are other documents such as patient files, prescription forms, and laboratory investigatio n sheets that circulate largely between health workers and patients and some few administrators, but which are not generally networked into the instruments required to produce inscriptions Those documents that contribute primary traces for creating interm ediate readings I will call aligned documents154 in that they are associated with state and donor networks through which inscriptions are created and circulated Of these, we can make a distinction between those inscriptions that are aligned with governanc e as negotiated at the level of the state, and those aligned to donor sponsored vertical health initiatives. While both are networked, by observing how these documents are aligned and what kinds of meanings and practices surround them, we can observe very different kinds of power configurations that get established through documents in decentralized compared to vertical forms of governance. The documents that are not pivotal to creating intermediate readings I will call nonaligned documents. At Kiunga Dis trict Hospital, the presence or absence of aligned documents, such as disease registries or donor produced pamphlets related to health, mark the areas of the hospital that are subject to higher or lower degrees of government and donor investment. Whether o r not the hospital staffs practices conform to the aesthetic frameworks intended for a particular document, 154 I take the terms aligned and non aligned from Stacey Langwick 2010.

PAGE 372

372 the presence or absence of aligned documents mark which spaces of the hospital are included as part of development enclaves within the hospital and which are excluded. Whether a document was aligned or not had important impacts on the ways that it moved, how it shaped and was shaped by social relations in the hospital, and what it meant for workers and patients at the hospital. Nonaligned documents presented opportunities for particular staff at the hospital to demonstrate their status and expertise to others workers or to patients, meanwhile select staff could use these documents as a means of circumventing established hierarchies at the hospital. M eanwhile, aligned documents adhere d to certain aesthetic conventions, although there was a degree of flexibility with some of these reports. Similar to what Peterson finds in his study of Gikuyu negotiating British rule through engaging in practices of government (2004), the workers and especially the administrators of the hospital could use their proper reproduction of good specimens and strategic display of the facilitys virtues as a means to negotiate with the DC, the MoH, and especially the transnati onal NGOs of the enclaved clinics. As Peterson reminds us, documents are not merely sources or texts in and of themselves, but [do] work within the human realm of ambition, action, and politics (2004:242). Engagements with Aligned Documents: Registers and Donor Reports In the enclaves of the hospital where both government and externally funded programs operat e simultaneously, there is an interesting configuration of aligned documents that mark the presence of the variety of governing bodies networked w ith these clinical spaces. Below, I describe engagements with aligned documents within two enclaves of the hospital: the Reproductive and Child Health Clinic ( RCH) and the Counseling, Testing and Care clinic

PAGE 373

373 ( CTC) .155 In both units, government registries and donor reporting practices collude, and clinic workers scramble to collate and translate the data from primary traces recorded during clinical encounters and registrations according to the individual aesthetic preferences of the governing entity requiring it. Engaging aligned documents could generate anxieties for staff, particularly when patients were not forthcoming with the information required to create the primary traces needed for registers and tally sheets. B y exploring engagements and meanings surrounding documents affiliated with the decentralized government system and those aligned with vertical health interventions, we can see two very different modes of governance at work. Those documents aligned with the MoH were subject to a more diluted system of regulation. Health workers had little flexibility when it came to whether or not they would provide the data demanded by MoH affiliated registers; however, they were able, when necessary, to produce a good enough specimen (Riles 2000; see also Stirr at 2000) of the document rather than adhering strictl y to the guidelines outlining the data to be collected. Conversely, the structures governing data collection and reporting in the CTC enclave was of a highly vertical and concentrated form, where surveil lance came straight down from the Dar es Salaam ensuring consistency with regulations (Chalfin 2010). However, while little negotiation was possible with enclaved reporting guidelines, as shown below, Kiunga administrators used their compliance with autho rity and demonstrated interest in the priorities of their enclave benefactors as a means negotiate the facilitys own needs into the priorities of the organizations funding the hospitals enclaves (cf. Peterson 2004). Their (variable) compliance with gover nment and donor data collection requirements was an important aspect of the economy of 155 As noted in previous chapters, the CTC clinic conducts HIV/AIDS ca re at the hospital. This acronym is meant to obscure the function of the clinic to the general public, so as to avoid stigmatizing those individuals who attend the clinic for services.

PAGE 374

374 appearances that the hospital administration actively established to secure and maintain government/donor investment in the facility. Thus, while the quality of the da ta compiled for the MoH and various donor entities varied, the aesthetic criteria underlying that data were meticulously followed. Registers in the R eproductive and C hild H ealth C linic It is February 6th, 2008. Christine, the head of the clinic, introduce s me to the maternal child health assistants (MCHAs) who are mopping the deck of the long waiting bay. When I tell Christine that I would like to assist with simple, non specialized work at the clinic as part of my research, she tells me that once the MCHA s are finished mopping they will show me what to do. Once the cleaning is complete and the cement floor is dry, two MCHAs, Anena and Gertrude, bring in a long wooden table and a dozen wooden benches and set them in their designated places in the bay. Like some of the classrooms I have seen in Tanzanian public schools, the benches all face towards the head table, where I assume the MCHAs will sit facing their audience. Women begin to drift into the waiting bay, placing themselves in what appears to be no particular order on the benches. Some are chatting. They are beautiful with their kangas or kikoi wrapped across their upper bodies, concealing their pregnant bellies. The scene set, Gertrude and Anena leave, and return with a stack of hardcovered notebooks, pamphlets on nutrition and breastfeeding, and booklets that appear to have some kind of tickets inside them. There appears to be around twenty women waiting for the clinic activities to start. The intermingling voices are a subdued cacophony as each cli ent156 awaits her turn. 156 As noted in Chapter Eight, in the CTC and RCH clinics areas of high donor/government investmentpeople attending clinic services are called clients ( washitiri ) or customers (wateja ), rather than patients (wagonjwa ).

PAGE 375

375 Gertrude gives me quick instruction. Apparently I am to fill out the registers and clinic cards for any pregnant women who are coming to the clinic for the first time. There are three registers in all, plus the clinic cards for each woman. The first register is a hard covered notebook with penand ruler drawn rows and columns. This has a space for the womans name, and the name of the village, hamlet, and ward within which the woman lives, and the names of the political leaders at each of these levels. The second register, which is nearly identical on the outside, records the womans name, age, number of prior pregnancies, medical indicators of issues that may present complications with the pregnancy, number of live births, date of las t birth, and a column for recording what treatments or supplies the woman will receive that day. The third register looks similar to a coupon book, bearing the emblem of the United States Agency for International Development ( USAID) and the Presidents Ma laria Initiative (PMI) on it.157 This is a registration book for mosquito net vouchers. Women get the vouchers and can use them at local participating maduka158 to get a discount on a mosquito net as a malaria preventative, which again records the name of th e womans village and hamlet, and their political leaders. A voucher is detached from this register for each new client leaving a small piece of paper inside the register with her particulars and the number of the voucher she receives. 157 President George W. Bush introduced PMI in 2005 as a major initiative to combat malaria deaths in A frica. PMI funded indoor pesticide spraying and mosquito net distribution in Zanzibar in 2008, largely eradicating malaria from the archipelago. However, beyond the indoor spraying on Zanzibar, the majority of PMI funds are directed towards preventing mala ria in pregnant women and children under five, who are considered most vulnerable to death from the illness. 158 A duka (plural maduka) is the Kiswahili term for a shop or store. In this case, the shops referred to are in the local market, an many of them a re not permanent structures. Particular shop owners agreed to register for the PMI voucher program, but all vendors did not accept the vouchers. Often, women receiving their vouchers would have difficulty finding a participating shop amid the mass of other shops and the informal sellers peddling their wares in neat piles on tarps or sheets of plastic on the ground between the shops. Many women tucked their vouchers carefully into their clinic cards but did not use them. This often became a point of contenti on when the women came to the RCH for subsequent visits. Nurses would often chastise the women for not using their vouchers and threatened to give them to another woman, at which point it often came out that the woman could still not afford the mosquito ne t, even with the discount from the voucher.

PAGE 376

376 Finally, there is the womans clinic card, which she will take with her. Like the second register, this one records the name of the woman, her age, type of work and educational level attained. The same information is recorded for the father of the fetus on the card. There i s another box on the card recording the number of live births, number of still births, number of abortions,159 estimated date of delivery, and a variety of other information. This card also includes information on the village, hamlet, division and ward, each with corresponding political leader. There is a space to record the number of the mosquito net voucher the women receives. Another box labeled PMTCT (for Prevention of Mother to Child Transmission) is used to record her HIV status, which is tested when it is her turn to enter the clinic after registration. The number 1 indicates she is HIV positive, number 2 is negative. This testing is part of the REFLECT program, funded by the United States Presidents Emergency Plan for AIDS Relief ( PEPFAR ) Wome n who test positive will be escorted to the CTC, where they will begin another barrage of tests and be counseled about living with HIV/AIDS and the procedures by which they access ARTs. In addition, there is a box for SP (sulfadoxine pyrimethamine), a ma laria preventative, with two rows for recording dates.160 Inside the card is a partogram a chart used to monitor the womans labor and a box for recording birth outcome. The use of partograms has been part of a major initiative by WHO to reduce maternal and neonatal mortality rates in under resourced settings. 159 Abortions in this context refers to miscarriages or spontaneous abortions. Deliberate abortions are illegal in Tanzania, although there were many rumors of private practitioners and healers in the villages p erforming these procedures, and often when a woman came to the hospital during a miscarriage, she would be given a pelvic exam afterwards to determine if the abortion was self inflicted. If there was any trauma to the cervix, the police would be consulted; however, in practice, I observed several nurses and COs refrain from doing the exam so as to avoid being obliged to contact the police. Feelings about induced abortions varied among the staff considerably. 160 SP is a drug used to prevent malaria in pregna nt women as part of the PMI program, and they are supposed to receive two doses before giving birth, during the second and third trimesters.

PAGE 377

377 As I sit asking the women balozi ? (councilor), kata ? (ward), kitongoji ? (hamlet), I struggle to hear their response. The women answer in very low voices and because I am unfamiliar with the names of all of the villages in the district, I have the impression I am making mistakes. To my relief, Gertrude hands me the voucher registration book and says that she will take care of the other registers. Anena is inside the clinic consultation room, doing f undal height161 measurements, HIV testing, and providing the women with SP. Between the three registers and the clinic card, it takes us nearly ten minutes to fully register a woman to the clinic. Gertrude puts a lot of pressure on the women for the proper i nputs for the registers. Many of these young women hesitate to answer, particularly for the names of political divisions and leaders. This would set Gertrude to asking a barrage of terse questions. What is the name of the really well known man in your war d? How can you not know? Think, Sista, think! You must know his name!162 Some of the women would mumble a name or something under their breath. Gertrude would just sigh, and on occasions when she was unable to coax an answer out of the woman, she would flip backwards in the register to find the name of another woman from the same hamlet or village, and copy the names of the leaders and wards from the previous entry. Christine tells me that I am making a lot of mistakes in my recording of wards, hamlets, and political heads and that it might be better for me to write in pencil so that I am not scribbling things out so often in pen. She says that by scratching things out on the registries, it looks like forgery. I wonder what this means. They are very a nxious about filling every box and line on 161 A fundal height measurement is done on pregnant women to measure fetal development. A measuring tape is placed at the top of the pubic bone and followed along the apex of the womans abdomen to the top of the uterus. With each visit a fundal height measurement can be taken to monitor fetal growth and, when a woman is due, provide a rough calculation of fetal size pri or to birth 162 While it is not possible to distinguish gender in Kiswahili pronouns, in this case the nurses were referring to males. Though there were several women leaders in political roles within the district, nurses invariably asked the women what the name of the well know man was in her village or ward.

PAGE 378

378 the registers properly, but at the same time they do not have all of the information and often have to reference another patient entry. Even then, they say it is a best guess. I am registering a new client, and she appears very young, possibly not yet twenty. When I ask her the name of her ward and she replies, a few of the women nearby start chiming in that she is incorrect and that her ward is actually called something else. By this time I have already written th e name of the ward that the client provided to me. Hearing the women correcting the client on the name of the ward, Anena turns and yells at this young woman, Why did you not give the correct name to begin with! Now youve made this mzungu write a wrong a nswer This is a government register! Now the government will be angry for your mistake! Discussion : The clinic card is a visual representation of the multitude of donorsponsored health programs into which the woman has been enrolled by virtue of her at tendance at the RCH. Boxes on the clinic card represent the confluence of programs paid for by PEPFAR, PMI, and WHO. The clinic card for children under five, which a woman receives after a live birth, marks a similar convergence of externally funded progra ms, but more donor funded programs are represented in the inscriptions: boxes to record dates of immunization (provided by the Canadian International Development Agency [ CIDA] refrigerated by units shown in Figure 101, donated by the Japanese Internation al Cooperation Agency [JICA]), doses of vitamin A (CIDA, to be administered nine times between age nine months and age five), PMTCT status of both mother and child (PEPFAR through REFLECT), number of mosquito net voucher (PMI), and a growth chart to track the development of the child through time based on a universal standard of growth (Figure 10 2) Significantly, although these numerous funding and governing entities programs are allocated their own boxes on the clinic card, their emblems are conspicuously absent Similar to the way donor agencies erase reference to their influence on government policies and

PAGE 379

379 development agendas (Anders 2010), they similarly erase reference to their inputs within this program targeted at prevention and surveillance of legi ble pregnant bodies on the RCH clinic card. While the clinic card summarized the kinds of assemblages within which the RCH was entwined, each registry represented the parallel bureaucracies ( Blundo and de Sardan 2006:87) to which the clinic was encumbere d due to being networked with those entities. Each registry was a place where an elaborate series of primary traces could be collected. Filled registries that were aligned with the HMIS would be sent to the DMO office. There, they would be compiled into in termediate readings, plugged into the HMIS software, and translated into inputs within the various reports that the district level was required to produce. Here, if primary traces were not available because patients did not know the answers or refrained fr om providing them, a suitable and convincing substitute for that trace might be found by some quick research of previous primary traces in the registry. While the MCHAs were unable to negotiate which categories in the registers needed to be filled, they were often able to wra ngle some kind of input for each column from the client herself, and if necessary, refer to traces left in the registers from encounters with other patients in the clinic. The voucher registry was a different story, however. They, like the MoH registers, could be negotiated because they were not subject to direct oversight. Often the MCHAs looked to data from the MoH registries to fill in the few categories needed on the voucher booklets, and there was little hesitation about doing this when necessary. From time to time a representative from SIGNAL would arrive at Kiunga to collect the voucher registers that had been completed, and provide the RCH with more books of vouchers to distribute. The main difference between voucher registers and those of the MoH was that the former would not move through the HMIS

PAGE 380

380 and be included in district data to go to the MoH. Instead, the primary traces on voucher booklets would be taken to SIGNAL offices in Tanzania, input into a database, and transferred to USAIDs PMI office. Eventually, these intermediate readings from all of the PMI funded clinics in the country would be compiled in large reports and statistical inscriptions, which would be sent to Washington for the consideration of the head USAID offi ce. There, they might be reinscribed and recontextualized, summarized and co mpared to inscriptions depicting other places in the world, and compiled into yet another inscription often headed to federal government offices, so that politicians and their aide s could scrutinize the inscriptions to measure the impacts of their funding on malaria rates The voucher registries were aligned to a donor organization and were transferred through donor networks, but a degree of flexibility was possible. Thus, while bot h types of documents were aligned, the alignment was not all encompassing, in part because the donor programs operating within the RCH were somewhat diffuse. We now turn to a place where the hands engaging and producing aligned documents were far less unen cumbered and authority far more concentrated: the CTC. The P olitics of D onor R eporting at the CTC The CTC was a department where the multiplicity of donor and government interests was most obvious. This was a place where there were no handmade registers of the kind we saw above for the RCH. Everything was official. T he burdens associated with the parallel bureaucracies of the MoH and REFLECT were heavy indeed It was also an area under regular surveillance because a REFLECT Program Officer, Peter regular ly visited Kiunga In addition, twice during 2008, dignitaries from REFLECTs head office in Washington D.C. also visited. Unlike at the policy level, where according to an informant donors dont want to be informed (see Chapter Four), in the NGO administ ered and bi lateral funded vertical interventions at the

PAGE 381

381 hospital, there was a considerable interest in enforcing the parameters by which information would be compiled and presented. Lucina, the nursing officer In Charge of the CTC and the most educated of the nurses at Kiunga, spent the majority of her days filling out the registers and reports required by both the MoH and REFLECT. She spent extraordinarily little time caring for the clinics clients or putting her nursing expertise to work. According to Lucina, due to the sophistication of REFLECTs data collection and oversight, deviating from reporting regulations was extraordinarily difficult. Lucina: With the reports, for [REFLECT] I told you they are systematic so when you send a report [to them] w hen you write a report we have had them for a long time. So even if you have a report you must send it there to the end. So if you have a report that is a lie, [REFLECT] always knows. They will know that this is not correct, because the data they have it. So if you send a report that is not correct they call you on the phone. They tell you that this report is not correct and they can even tell you in the report this here and that there in this section is not correct. So directly you correct everything and you send it by mail so that the issue is resolved immediately. So for reporting for [REFLECT] there is no problem. Their reports are always correct. Because if the report arrives [to their main office] in Dar es Salaam or even to America, it must be correct. The Program Officer of the site must look at it carefully and if there are problems he calls us and we communicate and he tells us here, there is a problem and you correct it and send it back. NS: And [REFLECT employee on site] is doing a lot of work L ucina: [REFLECT employee] has the job of data clerk. She has the job of entering all of the files. Every file, every clinic. After clinic, she takes all of the files and enters them in the computer. All of them. All of the sick go into the electronic system. NS: Is it you who creates all of the reports for [REFLECT]? Lucina: All reports, I create them. But it is not just one person. We all do it. It is too much work for one person to do. So we do all the reports everyone together, and we give them to [the R EFLECT employee on site], she looks them over and then she is the one who sends them all out by email (Interview, 101708, Kiswahili) Clearly, a highly managed system was put in place to ensure that all of the inputs going into these traces and readings followed the regulations set by REFLECT, and by PEPFAR more widely. Of interest here, the data clerk in the CTC was employed directly by REFLECT, and not

PAGE 382

382 by the MoH. Her inputs from patients files went directly into a PEPFAR database, to which REFLECT had access. However, the reports generated from the PEPFAR database are not public. PEPFAR does not share the considerable data it has with the MoH, other bi lateral donors, or multinational global health organizations. PEPFAR reports are compiled within the USAID Tanzania office, then sent to Washington to be scrutinized by Members of Congress, who decide whether or not to continue funding the program based on evidence from the report that American taxpayer money is transparently being allocated as directed.163 What is more, REFLECT required that CTC staff compile a set of regular reports that related to its own HIV/AIDS priority areas, above and beyond the PEPFAR database. Therefore, the REFLECTemployed data clerk handled PEPFAR database burdens, but REFLECT r equired that hospital employees compile reports and inscriptions directly related to its own interests as an American NGO working on HIV related issues. REFLECT officers could scrutinize those reports by referring to inputs from the PEPFAR database in orde r to determine whether the staff was producing the correct data in the reports. The more technology data technologies available, the more the hospital could be subject to direct and very authoritarian oversight. I asked several other CTC employees about t heir role in compiling reports and the characteristics of their interactions with REFLECT officers. Communication between those workers who created the data and those who received reports was continually described in terms of supervis ion and oversight. T ha t supervision came in the form of scrutinizing reports (see also Anders 2005; Gould 2005; Stirrat 2000) Peter rarely observed clinical interactions between COs 163 Indeed, in October 2008, the PEPFAR office within USAID Tanzania was busy compiling the yearly report to send to Washington. According to conversations with three USAID officers (two who worked unde r PEPFAR, one who worked on health issues outside of HIV/AIDS), the report was upwards of 1600 pages in length. This report was for PEPFAR related activities in Tanzania alone.

PAGE 383

383 or AMOs and patients. Instead, he looked at reports and files to ensure that regulations were being followed. I became curious whether the communication between REFLECT and the CTC workers was as unidirectional as it appeared based on Lucinas description. I began asking about how CTC personnel felt about the kind of communication they had with Pet er and the official REFLECT visitors who came through the hospital from time to time. Each time I tried to ask about whether CTC workers felt able to suggest particular ideas to REFLECT regarding improving or tailoring the CTC program to the specific needs of the communities they served, the conversation returned to the fact that they were supervised, and told what they needed to improve. I tried to rephrase my question in several different ways with four different workers, but the responses were remarkably similar. An extended quotation from an interview with Dr. Erastus illustrates: NS: If you have ideas about improving the services here [at the CTC], like improving the working environment here, do you feel that that those people of [REFLECT] are listening and trying? Dr. Erastus: Yes they try as supervisors. They look at our performance [ utendaji ] here, they do evaluation they give us feedback about where we have done well and where we are supposed to reinforce [the policies] and whatnot. But the feedback we are given at the office of the DMO later. We sit all of us as a team and they [from REFLECT] tell us those shortcomings that they see. T hey tell us if there is a new pr oject, something like that. So in time we are wanted to [ tunatakiwa] imp rove, each time and to take out [ kutoa] mistakes and whatnot. At the end of the day you will find that everyone has already done total improvement of that quality of care . 164 NS: So do you give REFLECT your feedback as well, though? Dr. Erastus: Aah, they come to supervise us. Now if they have already supervised they look many times they interrogate us, they go in the files and look more at 164 Total Improvement was the name of the program that REFLECT was introducing to the Tanzanian sites in order to streamline the processes of providing care across all of their clinics. All CTC employees were on the list to be given training seminars (most often conducted in English) with the intention of ensuring a particular standard of quality of care was upheld at each clinic hosting a REFLECT coordinated HIV/AIDS program.

PAGE 384

384 the reports and whatnot, they go into our data and whatnot, they way we do enrollment and other things the y look at. (Interview, 7 2408, Kiswahili). Discussion: An important context for understanding the presence of the multiple registries on the ground at the CTC is that donors targeted programs require measurable and clear objectives by which they can just ify their programs and their spending. With the proliferation of externally funded programs at the hospital, there was a need to manage the web of complex donor requirements and networks which further strains Tanzanias already overstretched resources (S mith et al. 2008:8). Yet even though REFLECT presenced itself regularly at Kiunga not only through registries and reports, but also through regular visits to the site at the ground level, the concern was that the reports and registries were being properly compiled, holding to the data gathering standards set out by REFLECT guidelines. Communication on issues beyond policies and reports was not sought out by Peter, or by the CTC staff. The relationship was clear: Peter was there to supervise, not to solicit advice on how to make the REFLECT sponsored program be more conducive to the needs of the hospital or the communities it served. The burdens of paperwork were tied to the resources that the program made available. Without proper paperwork, financial flow s could be delayed or halted. Compliance was important to everyone involved. The documents they produced had to conform to a particular aesthetic (Riles 2000; Stirrat 2000 ). Those aesthetics even emerged in the way Dr. Erastus talked about his work at the clinic. Total improvement and quality of care (terms that he articulated in English during the course of his interview in Kiswahili) are terms CTC personnel are taught in REFLECTfunded workshops, and that show up ubiquitously as indicators to be measured within CTC reports to the NGO. In the CTC, the donor reports were meant to achieve a remarkable degree of transparency and accountability in the program a prerequisite to good governance. However, as has been

PAGE 385

385 suggested elsewhere, by periodic visits f rom Peter and from REFLECT representatives from the very top of the NGOs hierarchy, the CTC reports show how good governance and other anticorruption initiatives contrary to their explicit goals both utilize and augment authoritarian leanings of the bure aucratic apparatus (Chalfin 2010:103). Within development enclaves, governance from afar had a decidedly different character than the governance characteristic of the HMIS system. There was considerably more flexibility and less oversight when it came to reporting burdens related to HMIS and RCH donor reports. For CTC reports, supervision was far more direct. What occurred, then, when staff had to work with aligned documents such as those of the CTC at the same time as producing reports for the HMIS? Worki ng between Government and Donor Reporting Burdens Within the CTC and the RCH arguably the two most enclaved units of the hospital at the level where primary traces were made into reports (intermediate readings), workers at both clinics spoke about the intr icacies of properly preparing data according to two different sets of standards within the parallel bureaucracies of the hospital. The MoH and donors had vastly different standards and expectations when it came to reports and data, which meant that adminis trators working within enclaved clinics had to learn the aesthetics criteria for how each type of document was expected to look. Administrators in the RCH were required to prepare a daunting number of reports about clinic activities. One administrator expr essed the extraordinary pressures she felt as the coordinator of the reproductive and child health services in the district: so many reports, so many activities. So many facilities. Too much movement (Interview, 112408, Kiswahili). The systematic na ture of donor reports stood in stark contrast to the guidelines and data indicators required in the HMIS reports. Still, in the CTC at least, despite the fact that the reports for the MoH and the individual organizations running vertical programs had diffe rent

PAGE 386

386 requirements, it was possible to be strategic in their preparation Lucina developed a system by which she was able to create the reports for REFLECT, which were very detailed, and extract from these the data necessary to compile reports for the HMIS When I asked her about how difficult it was to compile separate reports for the MoH, the DC and REFLECT, she responded: Lucina: The monthly reports are small but the quarterly reports, they require a lot of information and you have at the same time the re ports for [REFLECT], the reports for the government and the reports for the District Council. That is, you have a lot of reports. So with that many reports it means you keep busy because you are doing all of those reports well into the evening because those reports must be sent. (Interview, 101708, Kiswahili) Lucina was deliberate in the order in which she compiled her reports for the various parties. When I asked her about whether there was duplication of work across the various reports, she concurred, b ut it was clear she had devised ways to reduce her difficulties in amassing the correct inputs: Lucina: the reports for [REFLECT] for quarterly you do those first because the other reports, all the information you will need for the other reports is within the report for [REFLECT]. That means, everything you need for the district report you can already see in the [REFLECT] report and you can fill it [the district report] out. So it means that once you have finished the report for [REFLECT], the data is that same data for the other reports. The differences are that these groups want this information, maybe it is male/female ratios or something, these other ones want some other different thing, but what is wanted is that exact thing and so because [REFLECT] ha s all of that stuff it is easy to take it from there. And even if there is a problem, like why is the government one saying this thing but the district one says something different? You can go to the [REFLECT] report and determine where the problem is. So then you can correct it. So it helps to verify that you have done the reports well. NS: So that [REFLECT] report has everything in it. Lucina: Everything! that is [ emphasizes word ] everything! NS: So why doesnt [REFLECT] use the same reports that the g overnment uses? Lucina: Aah! [REFLECT] is different because it has a lot of different things it wants to know. Many many many many things, every little thing, different from the government. For instance they have many many categoriesand you find with the

PAGE 387

387 government it is not like this. But [REFLECT] there are many things that they need, that is, pregnant mothers, that is many things. Different than the government. As Lucina described, while there was a duplication of information in the MoH and REFLECT re ports, there were opportunities to ease the burdens of preparing data for each intended audience. And the most sophisticated data was in the REFLECT reports. In terms of generating data for the government and the donor programs operating at Kiunga, the pl aces where these were particularly salient objects in the daily routines of workers were the administration block and the enclaved units o f the hospital. In the enclaves, nursing officers and in charge staff painstakingly fill ed out rows of registers by ha nd, and later us ed computers to generate reports. The administrative block was also a place where primary traces from the various hospital divisions would be compiled into electronic forms from the paper registers and tally sheets brought in from the diffe rent hospital departments as well as other health facilities in the district. On a day to day basis, as data was accumulated and translated into digital forms on computers, it seemed like the burdens of data collection occupied a disproportionate amount of time in the overcrowded hospital. However, when I spoke to COs and nursing officers about the burdens of reporting, they always couched these responsibilities in terms of their importance for securing funding. A ttending to these documents allowed for pat ients to obtain better care with more resources. Receiving assistance from donors and the MoH was tied to proper reporting. Without proper reporting, these entities could stall or completely halt financial streams coming to the hospital, which were critica l to being able to maintain an adequate supply of essential drugs and equipment Frida: You could be without more work and patients would not get services. You could relax But is it not better you suffer, you do more of your work but you help society? I think if it is possible that all programs are coordinated together, that is, those verticals werent here, but still Noela, even if they were not vertical but still

PAGE 388

388 you would have to give them reports if they were at the ministry [level], and still you are giving them reports and it is possible to format them so that maybe they were not very different and maybe those of the donor I do not know. But to me if those programs bring improvements, for me to get more work is not a problem. Me, as Lucina. (Interview 07 1608) Dr. Ezra expressed a similar sentiment when I asked about the challenges of providing care when so many reports were required. He described that the burdens were shared among all of the staff in the enclaved units, and were a necessity in order to assure that Tanzanians continued to be able to access meaningful health services. The duplication of work for Dr. Ezra seemed inevitable if they were to continue to work with the government and the donors at the same time, and working wi th both was a way to ensure that services were available to the community. Dr. Ezra: Ya h, it brings other work, it is a lot [of work] [ laughs ] NS: So, what importance do you see in doing this kind of work? Dr. Ezra: Now, you know the government itself coming from above, they know this. Eeh. Me, I could not even do anything because those donors indeed help so that my ndugu165 get help. So as my ndugu, they get help coming from the donors right eeh? It is necessary I do the things that will help the donors, you see! So they can help me, you see, eeh? And me, when they help me they help my ndugu, you see? (Interview, 071708, Kiswahili) Given that understaffing was a problem even in the enclaved units of the hospital (albeit, shortages were not as severe as i n other units), and the major burdens that reporting added to the already difficult working conditions of the hospital, the staffs reflections on reporting struck me. The process of having to complete many reports, often duplicating work, did not appear t o alter staffs interest in complying with these burdens. The staff was committed to meeting the required aesthetic criteria of donors and the government because it would assist the hospital to bring in the resources it needed not only within the enclaved parts of the hospital, but significantly, in the low priority areas of the 165 Ndugu in Kiswahili implies sibling or comrade. This was a term often used under Ujamaa to refer to fellow citizens of one family under the nation.

PAGE 389

389 hospital as well, where the majority of the staff and patients interacted. The constant stream of essential drugs and supplies that came into the hospital and that were most used i n low priority areas only persisted because the hospital complied with reporting requirements. Having those supplies available allowed for more possibilities for biomedical carefor the health workers to be able to try out more interventions within non e nclaved spaces within the hospital. In this sense, compliance with documentation procedures was critically tied to the overall ability of the facility to provide care, even though the care available was unevenly resourced throughout the hospital. Having come from a past where the hospital lacked even gloves and medicines, hospital personnel had little interest in returning to a context of overall paucity. They would navigate the abundance and scarcity of the current hospital in hopes that they could attract more donors, more government interest, more investors, more expertise, more resources. If we take the aesthetic criteria of donor and state documents discussed in Chapter Nine and consider how these play out within the spaces of Kiunga District Hospital, important aspects of flows across these multiple scales come to light. As illustrated in Figure 10 3, the hospital is a space where multiple scales intertwine in complex flows of resources, guidelines, and data. Through the HMIS, the health facility sends quantitative data to the MoH inputs required in order for the MoH to govern from afar. In return for data compliance, the MoH ensures that essential supplies and pharmaceuticals continually flow into the district, resourcing both highand low priority areas of the hospital. These resource flows are particularly important for the nonenclaved spaces of the hospital, since these spaces are beyond the direct purview of the MoH and donors, and therefore data compliance is one of the sole means available (beyond securing unpredictable Public Private Partnerships166) to procure the objects necessary for 166 Fo r information on Kiungas establishment of Public Private Partnerships and their effects on low priority areas of the hospital, see Chapter Five.

PAGE 390

390 providing meaningful care in those spaces. At the level of the MoH, HMIS data from various districts are compiled into reports used to create policies collaborativ ely with Development Partners (DPs), which then are also sent down to the various districts in the country. The MoH sends its data up to United Nations (UN) umbrella organizations, careful to include all of the data needed based on a select group of indica tors that will allow Tanzanias measurements to be compared to data inputs from other countries all over the globe in World Health Statistics reports. In return the UN entities make available to the MoH global health policies and programs (such as the Mill ennium Development Goals) and data from other places in the world that might help to inform MoH policies and guidelines. As seen in the figure, the other circulation relates to bi lateral donors such as USAID, and to the NGOs that administer bilaterally funded programs such as those related to HIV or RCH services. Part of the parallel bureaucracies wrought by the infusion of NGOs and targeted donor sponsored health interventions, these circulations move in carefully managed ways. The donors and NGOs provi ded guidelines to direct practices related to their programs, and supply the necessary, and often highly expensive and sophisticated, technologies and resources required to make those programs operate as intended. In return, health facilities create numero us reports catered to the very specific interests of those donors funding the enclaves. What we do not see in this figure, and the circulation that specifically does not occur, is any distribution of donors data to the MoH, or to the UN. In fact, donors are highly protective of the data they collect relating to their specific targeted interventions. That data is meant to show donor accountability to the citizens of the countries funding the program. It is not provided for the scrutiny of other donors, nor for the government in which those programs operate. Government health facilities thus navigate parallel bureaucracies, determining exactly how

PAGE 391

391 (in)flexible donor reporting requirements are, and strategically complying to those guidelines while working to minimize their overall reporting burdens. Ironically, many of the donors that criticize the quality of HMIS data (see Chapter Nine) are the very same that compile their own data reports within various health facilities scattered throughout Tanzania. Yet a s will be seen in the next case, there were other incentives for taking great care in the preparation of donor reports. By engaging donor related data collection, administrators and workers learned quickly about donor priorities. This knowledge allowed the m to propose ways to extend the boundaries of the enclaves and thus its resources to other units within the hospital. It was incorporated into the hospitals economy of appearances towards donor groups. Further, Kiunga administrators combined their good re putation for recordkeeping compliance with the pleasing aesthetics of the clinic. These performances helped to convince donor officers of their legitimacy and commitment (see Peterson 2004). Largely, these attempts to precisely abide by the regulations an d reporting guidelines of REFLECT and the government were not so much indicative of a desire to follow the rules per se as they were meant to create the impression of a trustworthy partner for foreign funders (Annist 2005:150). Staging Partnership: An H onored Guest in the CTC It is July 2008 and we heard two weeks ago that the Vice President (VP) of REFLECT would be coming to Kiunga District Hospital to tour the facility. Dr. Saidi and the Medical Officer In Charge ( MOI ) have been encouraging the staff every day to ensure that every corner of their departments is clean, because one never knows when a visitor will look inside one of the clinics that are not part of the official visit. The administration wants the hospital in top shape for the visit, and s ince REFLECT provides more money to the hospital than any other donor, maintaining this relationship is very important.

PAGE 392

392 The official visitor is supposed to arrive late in the morning today. It is a Tuesdaynot a clinic day, so there will not be many patie nts coming in. T he large SUV with REFLECTs logo on the side arrives and parks near the administration block. A man comes out of the vehicle and walks straight to Dr. Saidi. It is clear that they have met before. A second man emerges from the SUV and the f irst introduces him to Dr. Saidi as the VP. They shake hands. Dr. Saidi introduces me to the guests as a researcher from America. The first visitor, who I will call Mr. Hansen, tells me that he is the Tanzania Associate Technical Director of REFLECT, and i s remarkably good at speaking Kiswahili. It turns out that he implemented REFLECT programs in other African countries as well, and has been in Tanzania for five months. Mr. Hansen tells me that the VP has only been with REFLECT for six weeks, and is visiti ng various REFLECT projects all over the world to familiarize himself with what the organization is doing. The VP is particularly excited about doing medical research with REFLECT, and he asks Dr. Saidi several questions about what kind of research might be possible here at Kiunga. Dr. Saidi has several ideas about ways to integrate some of their other services into REFLECTs program, and Mr. Hansen offers that Kiunga was the first district in Tanzania to integrate PMTCT services within the RCH clinic by co ordinating with the NGO SIGNAL. They hope to use Kiunga District as an example of how their programs might be scaled up to the rest of the country. The VP has many questions about medical research, which is one of the reasons REFLECT hired him. Apparently the NGO is looking to scale up research projects within its sponsored clinics. The VP asks Dr. Saidi if they have done research on how many HIV positive mothers are delivering at home versus at the hospital. Do they know, of those who deliver at home how many bring their infants to the hospital for testing? How does Kiunga Hospital ensure that mothers who deliver at home take their Nevirapine before delivering? Dr. Saidi responds by

PAGE 393

393 talking about all of the research ideas that they have thought about doing through the hospital: Theres a lot of things to be learned with the [REFLECT] program. We learn as we go. Catering directly to the stated research interests of the VP, Dr. Saidi says that the hospital is very motivated to develop some of these research programs. As the VP and Mr. Hansen walk around, they continued to comment this clinic is so nice!. They are very pleased that their clients do not have to go far to get access to the services REFLECT provided all of their services are concentrated wi thin the CTC except PMTCT, which is administered a short walk away at the RCH and maternity ward. The visitors seem impressed with the organization and layout of the CTC and of the hospital in general. We move through different parts of the clinic until finally we reach the doctor consultation rooms. The AMOs and COs consistently use one of the rooms for patient consultations. The other is most often used as an office for the AMOs, but as we reach the second room I see that all of the furniture has been r emoved from the office except one desk, which is covered in a tablecloth. On the desk is an impressive spread of food: meat stew, cabbage salad, cucumbers, fried starchy bananas, and tea with milk and sugar. Two of the nurse attendants are behind the table in aprons, waiting to serve their guests (Figure 10 4) Given that most of the staff rarely have time to eat lunch while on shift, this is a feast indeed! The CTC staff is quick to join the line to enjoy the free food. They pile their plates high with food. One of them turns to me as we are eating and say s we should have guests more often! We both laugh. The VP finishes his meal and passes his empty plate to the nursing assistant. More? she asks him in Kiswahili. The VP puts his hand up to decline. H e walks into the hallway and begins looking around. At this point Dr. Saidi leans towards Mr. Hansen. Have you heard about the car? Dr. Saidi asks. Apparently he wants REFLECT to donate a car to the hospital so that they

PAGE 394

394 can more easily go to patients v illages and follow up with those clients who stop coming to the clinic for their ARVs. Mr. Hansen responds, the donor is my boss. I think that [Kiunga] needs the car, but I have to follow what the donor says. He promises to get back to Dr. Saidi about it. Discussion While the reporting requirements of both the government and the donors increased the burdens of the hospital staff without providing additional workers to help reduce the workload (see also Whyte et al. 2010), there are advantages to strict adherence to donors data collection and reporting guidelines. Having experienced a history of chronic shortages in all units of the hospital prior to HSR, engagements with vertical health programs and with donors more generally opened up the possibility f or health workers to practice the therapies they were trained to perform Working compliantly with documents was a means to be able to ensure that people got access to care. It made their biomedicine more meaningful because they finally had (some of) the r esources they needed to practice it. Such spaces appealed to the staffs sense of professionalism. REFLECT kept the clinic space full of sophisticated technologies and medicines. They continually updated the staffs skills through workshops. They provided additional compensation. Working in an environment where these kinds of opportunities and resources were not available through the MoH, the CTC was a workplace that recognized their expertise and professional status. These resources and prospect s carried a burden in terms of reporting and adherence, but it was one about which few ever complained. An institutions compliance with reporting aesthetics and inputs allowed for opportunities to impress donors, and opened the possibility of bringing additional re sources into the hospital. The VPs visit was carefully staged: the manicured hospital landscape (due in large part to efforts begun two weeks prior), the orderly filing, the initiatives to integrate PMTCT with RCH services, the clean and open workspaces w ere all laid out in such a fashion that this honored

PAGE 395

395 guest could not help but be impressed. In advance of the visit, Dr. Saidi and the CTC staff were highly familiar with REFLECTS priorities in part because of careful attention to the kinds of information the NGO collected. When the VP mentioned research, Dr. Saidi was quickly able to suggest a plethora of ideas for ways to bring research initiatives (and therefore resources) that would cater to REFLECTs interests into departments within the hospital outs ide of the CTC. Having exceeded expectations for an official visit, Dr. Saidi was able to open negotiations with Mr. Hansen in a way that was not possible for his employees who were carefully monitored with regards to their reports and data collection pro cedures. The CTC not only followed REFLECTs rules; it surpassed the visitors expectations. It was at this level of showcasing the hospital to the visitorand importantly, expressing interest in the donors priorities that the hospital administration was able to negotiate with its well funded NGO partner (cf. Peterson 2004). The established economy of appearances directed towards the donor was highly effective. They were able to maintain their mutually beneficial relationship with their donor because they went to enormous efforts to nurture and preserve good relations. Non Aligned Documents There are a variety of other common engagements with documents at Kiunga District Hospital, beyond the purview or state and donors interests, which open up space for c onsiderable improvisation within a difficult working environment. Documents that are not aligned with donor and state priorities have an important presence at the hospital. They are often an important means of communicating with chains of personnel across various shifts. Doctors or nurses writing orders in patient registers or files provide instructions to their counterparts who take over when their shifts are complete. Inscribed on paper, these documents record actions taken (or not taken) through time and across space in the hospital, as patients move in and out of the outpatient department (OPD), to the ward, to the laboratory, to the ultrasound clinic, to the

PAGE 396

396 minor surgical theatre, to hospital administration offices, etcetera. Documents such as laborato ry orders, x ray reports, prescription forms, supply orders from the stores department, injection registries, patient registers and the like are intended to make patients and health personnel move through the hospital and engage with one another in delinea ted ways. However, in practice, nonaligned documents could mediate or constrain biomedical encounters in ways not originally intended. Below, I draw on select examples of engagements (and nonengagements) with the adult patient file, as well as considera tions of a select few additional nonaligned documents that tend to circulate within the hospital. Nonaligned documents could generate a considerable amount of anxiety for the individuals charged with filling them out However, these documents could also be variously mobilized to procure funds for the hospital, assert professional status in relationship to other hospital personnel, or rank patients when workers were particularly overburdened. The Social L ife of the Adult Patient File167 Prior to 1995, there were no patient files at Kiunga. According to Sister Gertrude, before then, patients were coming with their notebooks and left with them (Interview, 8708, Kiswahili). Clinical notes on patients were taken in the notebooks that patients brought on ran dom sheets of paper if they were available, or not at all. Once user fees were instituted, patient files, and a system by which to organize them, began having daily significance at the hospital. Once files and a storage facility were introduced to Kiunga D istrict Hospital, there were two staff members whose responsibilities were entirely dedicated to creating, organizing, delivering, collecting, and re shelving patient files at the hospital. 167 This section describes interactions surrounding patient files for adults only, and specifically, those adult s who are not pregnant or HIV positive. Given that services are free for HIV positive patients, maternity services, and for children under five, the engagements characteristic of therapeutic encounters with these patients varied considerably from those adu lt patients who did not fit these categories.

PAGE 397

397 In 2008, when a client came to the hospital for the first time (o r at least, the first time since health sector reforms were instituted), he or she had to go to the outpatient department (OPD) and pay a user fee of 2,000 Tsh168 in order to open a file. This fee covered the cost of the consultation with the doctor, and if the patient was admitted, the hospital stay would be free of charge. Upon subsequent return visits, the patient was required to pay 1,000 Tsh to have the ir file retrieved, and therefore access services at the hospital. The patient registry was nonaligned, in that the patient information placed in its hand traced rows did not get incorporated into other reports or documents that circulated up to the MoH, NGOs, or other donors. The columns drawn in the registry followed a format that was outlined in one of t he practice guidelines by the MoH. Details about the new patient were recorded: his or her name, name of father, tribe, religion, date of birth, village, hamlet, ward, and name of the political leader of each of these geographical divisions.169 However, unli ke official government registers, there was less urgency to fill in every category, since these registries and files remained at the hospital and were not sent beyond to the DC, the MoH, or any NGO or donor affiliate. Patient files were but one of the doc uments through which hospital attendance was recorded. Throughout their treatment, patients bodies, symptoms, and treatments found their way into a variety of documents throughout the hospital. Within the male and female wards, those patients too ill to leave but apparently not in serious enough condition to require immediate 168 In 2008, 1000 Tanzanian shillings was the equivalent to about $0.87 USD. 169 Under Ujamaa in order to promote unity and combat favoritism based on religion or tribe, the central government did not record the tribe or religion of its citizens. However, not only at Kiunga but also at every other health facility (public and private) that I visited in Tanzania, details about a patients religion and tribe were always part of the information recorded when opening a file. Wondering why this information was collected given the government policy on unity, I emailed Frida who had been in the health sector for over twenty years regarding why she felt this information was collected. She was unaware of an official reason, but had ideas about it: For a hospital, we ask the tribe and religion so that we can give our clients [ wateja wetu ] freedom to pray and and maybe if there is are unique cultures that the client has so that you can advise her. Also, in giving services, if a client has spiritual needs it is good to give him services and know to call a Priest, Pastor or Sheik. Again these are my ideas, as Frida. (email communication, 04 15 10, Kiswahili).

PAGE 398

398 attention could wait in the ward for days without being evaluated further for lack of a patient file (see also Chapter Eight ). On many occasions I observed both doctors and nurses put ting enormous pressure on patients and family members regarding opening files. Patients cost sharing fees were an important source of income for the hospital, supplementing the funding from the MoH and donors to pay for small infrastructural improvements or supplies that the government did not provide. Much of the cost sharing money went to low priority areas of the hospital, given that the government provided fewer necessities for biomedical care to these areas of the hospital. An example from observations made in the male ward in April 2008 is illustrative of these exchanges. If We Do Not Help Each O ther What Will We D o? It is a Monday in May, 2008. A patient is in a bed, having presumably arrived on the weekend, when no (Assistant Medical Officer) AM O was available to see him. AMOs in charge of the wards are not on duty on nights or weekends unless the case is an emergency, in which case one AMO is designated on call. This patient has a clinic form with notes attached, but has yet to open a file. Duri ng ward rounds, due to having no file, Dr. Mtenga refuses to examine him and says he can wait until the next day and should have a file by then. The patient says nothing during the entire exchange. The next morning, when Dr. Mtenga arrives at the same pat ients bed for ward rounds, the patient still does not have a file. She yells at him, If we do not help each other what will we do? ( Kama hatusaidiani tutafanyaje?) The doctor threatens to discharge him for not complying with her wishes, and then continues on with a lecture to the rest of the patients in the ward Files are important! she exclaims ( Mafaili ni ya muhimu eti! ). In this sense, files could mediate access to (timely) medical services. Furthermore, in a ward where the doctors and nurses are overburdened, and there are two patients per bed, files provided a means by which staff delay

PAGE 399

399 services to particular patients in order to lessen their overall workload on a very busy day, so that services can be delayed until a less busy day in the hospital .170 However, not all patients without files were refused treatment. On the contrary, several patients who never open files continue to receive care from nurses and doctors, without much mention of the fact that they need to open a file. The criteria on whic h decisions about refusing or providing care through the patient files had me perplexed for a long time. Initially it appeared arbitrary that one patient would receive care having only a clinic sheet while another was refused services. However, after month s of observations in various units within the hospital, I noticed a correlation.171 It appeared that refusal to provide treatment to those patients without a file correlated to how ( in ) conclusive their apparent symptoms were. If notes written by the admittin g Clinical Officer (CO) were thorough and a plausible diagnosis was recorded, AMOs appeared more willing to continue on with treatments. For example at the same time as the patient above was refused treatment, another in the same ward, under the supervis ion of the same doctor, had had two surgeries in the previous three weeks, with no patient file ever opened. He was not a relative or friend of any hospital employee nor did he appear to be a man of means.172 His was a complicated case, but also a medically 170 At Kiunga District Hospital, Mondays and Wednesday were particular ly busy days. Between Fridays morning tea and Monday morning, no AMOs conduct ward rounds, so that by Monday the wards are usually full, and they have to examine larger numbers of patients Wednesday was a local market day, which meant people with various maladies were more likely to make the journey (often by foot) from their village to the hospital because their family members were already going to a nearby location to sell or buy food. Patients too sick to walk to the hospital were often carried on the backs of their relatives to the hospital, or, if they could afford it, on the back of a bicycle or in a wheelbarrow forhire. Wealthier families usually arrived by taxi, but the overwhelming majority of patients came on foot. 171 It must be noted that the explanation that follows is speculative. The ways that the staff talked about their work and how they practiced were often at odds. 172 It was common for the family members of hospital workers to be provided with private rooms or files for free. Treatment w as the same for politically or religiously influential people in the district and their relatives, much to chagrin of nurses. There were several occasions when the relative of a politician was admitted to the hospital, and the politician complained to the DMO about the quality of nursing care. Nurses resented these accusations, given the additional services these powerfully aligned patients were granted.

PAGE 400

400 interesting one. In a workplace that provided little of the technology and upgraded training to which health workers felt entitled due to their expertise and profession, cases that were academically interesting and that Kiunga was equipped to handle might be treated without a patient file. I n the end, this patients medical history made his case interesting, and on numerous occasions AMOs and COs congregated at his bedside to assess his condition, collectively strategizing on next steps. F inally after a few weeks, multiple surgeries and little improvement, he received a transfer to the regional hospital. In the weeks that the patient stayed in the ward before finally being transferred, he never opened a file, and in daily observations of ward rounds during this period, I never heard anyone mention to him or his family the necessity of doing so. In cases when patients are readmitted soon after they were discharged, the file can become a barrier to accessing services. The hospital is not adequately staffed t o be able to spare a nurse from a ward to go to the filing area to retrieve a patient file Thus, in many cases, if a patient is able to walk (even if doing so is painful or difficult), he or she is sent to the filing department to retrieve his or her own file and bring it back to the ward. If returned after a recent admission, the file may not have been re shelved, causing a further delay in services as filing attendants may delay looking for a file due to other demands on their time. Patients who go to the filing area to retrieve their files may be sent from the filing area of the OPD to other areas of the hospital in search for their file, and many were asked to return to the filing area later to retrieve their files, thus delaying their services even fur ther. It was common for filing clerks to treat patients tersely, but they were also very abrupt with hospital staff. On more than one occasion, I was asked to get a file from the clerks. There was a perception among the staff that the clerks would not be so curt towards me. One In Charge nurse told me, If I go to the filing, they will refuse. But if you go,

PAGE 401

401 as mzungu,173 they will give you the file. Please get the file so we can care for our patient! Thus, both staff and patients had difficulties negotiati ng with filing clerks for files. The filing clerks had post secondary education and considered their work to require particular expertise also. By refusing patients and hospital staff access to files, they were able to demonstrate their own importance to t he daily routines of biomedical work, and assert their own status as professionals in their domain. Burdens of Work: Non Compliance with Non Aligned Documents Patient files are the objects through which select workers are able to challenge official hierarchies and procedures, and a means of maintaining professional status in the face of gaps in knowledge. More than any other worker s nurses record the highest proportion of information in the patient file. Upon arrival at the hospital, those patients who ar e being admitted are supposed to arrive in the wards with a sheet of clinical notes provided by the CO in the outpatient department who first evaluated and admitted the patient. The CO may also order preliminary laboratory investigations. Particularly sinc e the patients being admitted are not likely to see the ward doctor until the next day (or if admitted on the weekend, not until the following Monday), patient care is highly dependent on thorough clinical notes of the admitting CO. In practice, however, m any of these notes are inconclusive or incomplete, and AMOs and nurses in the wards may argue with COs during morning staff meetings about the inadequacy of their clinic notes. This often launches into a general staff debate about whose job it is to provid e such 173 Mzungu is the Kiswahili word for white person or European, but holds no negative connotation in Tanzania. Indeed, walking on the street, it is very common for people to a visitor on the street by saying mzungu to get their attention. As mentioned in the Introduction, despite my protestations to the contrary, it was ubiquitously assumed that I ha d access to powerful donors or benefactors. This is likely the reason that staff felt that I was able to negotiate in ways they could not. I was also regularly asked to anonymously make suggestions to the DMO and the Medical Officer In Charge about changin g workflows or small routine improvements that would ease the burdens of work. I did my best to honor these requests, and overall everyone was incredibly gracious toward me. However, I cannot assume that this was merely because I was well liked. The reputation of mzungu carried weight, whether I agreed with that weight or not. See Masquelier 2001; Richey 2008 for similar observations.

PAGE 402

402 details. The outpatient department, where the majority of COs work,174 is generally inundated with dozens of other patients, and there are too few doctors to serve the patients needs (Figure 105) However, there are many demands also on AMOs and nur ses in the wards, and the lack of clinical notes often delays timely services to patients. Nurses are also expected to chart information in the patient chart at the end of each shift. There are three shifts during a 24 hour period: the day shift goes from 7:30am until approximately 3:30pm, the afternoon shift goes from 3:00pm until approximately 6:30pm, and the night shift goes from 6:00pm until approximately 8:00am the next morning. In the course of her175 shift, a nurse in each ward is expected to track ur ine output of any patient who had a urine catheter chart the time each medication was given, take vital measurements of each patient (such as blood pressure and pulse) at regular intervals, and write the results and measurements on specifically designated forms contained in the file. In practice, however, this process can get quite complicated, particularly in an understaffed ward. The forms in the patients chart are duplicated in a general patient binder, which nurses are supposed to use as a quickrefe rence of all inpatients care and vital readings. In theory, this allows all nurses access to the most important information about patients from one source. 174 Unlike AMOs, who do not work weekends or nights unless on call, for every afternoon, evening, and weekend shift, there is at least one CO working. On weekends there were often two COs working in the outpatient clinic. On afternoon and evening shifts, the CO is technically supposed to remain in the outpatient department awaiting emergency cases. In practice, however, the CO i s often difficult to locate if issues arise during these shifts. This is due in part to the lack of intercommunication system between departments, and intermittent cellular phone signals. On some occasions when a CO was being sought, he or she would be in a ward caring for a patient. In others, he or she may have found a private room in which to sleep, without informing the nurses of his or her whereabouts. 175 At Kiunga District Hospital during 2008, all of the nurses and the nurses assistants were female save one. The sole male nurse was the main anesthesiologist at Kiunga, and worked almost exclusively in the major surgical theatre, where he interacted mostly with a surgical nurse and the AMO doing the surgical procedures I never saw this nurse nurse wo rking in other areas of the hospital. One of the male COs at Kiunga during 2008 was a nurse in the 1980s, but quickly upgraded his training to become a CO. There were two male ward attendants working there, who were present in the wards but did not do any medical procedures. Their responsibilities were cleaning the wards and, in some cases, assisting to transfer patients to oth er units of the hospital, such as from the ward to the surgical theater. For a discussion on gender in nursing, and the challenges of being a male nurse in East Africa, see Martin 2009.

PAGE 403

403 However, in application, this model of expectations of the nurses is cumbersome and takes nurses aw ay from their already limited time for patient care: Regina: So you have to fill out all the vital signs in the patient file, and then a separate sheet in the binder for the patient as well. And in other places you only have the form for the patients, no others. But having to duplicate everything means that we have shortages of workers to do the work. It takes time. You fail because you are alone, to give the medicines, to check vitals, to measure temperatures, so you find it is really difficult. (Intervie w, Kiswahili, 111108) Non aligned documents also could get considerably stalled, particularly those documents ordering laboratory or ultrasound investigations. M ore than one nurse spoke to me about the pain of losing a patient because the laboratory or u ltrasound and xray units did not respond to paper orders marked urgent. In the parallel order, in which the nursing officer in charge generally had less power in the unofficial hierarchy over the people below her or even adjacent to her in official rank ,176 it was difficult to compel someone to follow written orders. Nurses often felt powerless. For instance, in her role as the In Charge nurse of her ward, Monica found it particularly infuriating that patients died of preventable deaths because the laborat ory did not pay attention to written orders: Monica: Other departments do not go along as they are supposed to. For instance, if a patient is admitted, it is wanted that all activities like what things in the laboratory should be done, down there in the O PD everything done so that the patient comes to the ward, okay! But you find the patient comes, it is written urgent Hb,177 she [the patient] comes all the way with her paper [lab request form] to the ward! Now, it is the job of the patient herself, she cannot walk, others have been brought in wheelchairs it is better that I, nurse of that ward I do that work that should have been done in the out patient I take the blood and bring it again to the laboratory. You understand me eh? [ umeelewa, eh?] You find the patient is supposed to get an X ray, but instead on admission she right over there [at the reception] directly she is sent to me, already I am just waiting for the report [from the lab] about how things should proceed, [the patient] comes to the ward 176 Parallel orders and hierarchy in the hospital are discussed in Chapter Six. 177 An abbreviation for hemoglobin, calling for an urgent laboratory test to check the patients hemoglobin levels.

PAGE 404

404 and again is supposed to have an X ray, I come, I take her in a wheelchair that is, it is work! (Interview, Kiswahili, 11 1808) Monicas frustration stems from the fact that the laboratory is right next to the OPD, and that officially it is the OPD nurse s job to ensure that the patient, who requires an urgent laboratory test, be given proper care prior to being brought to the ward. Yet in practice, paper orders from one department were shuffled (often in the hands of the ill patients, who regularly had difficulties moving due to their conditions ) to different departments rather than following official procedures regarding who should move the patient through the departments to ensure prompt care. Monica, like other nursing officers InCharge had no way to force her co workers in the OPD, let alone the nurses of whom were far below her in rank, to complete their duties Official complaints to the Matron had little effect. As a throw back to the ideals of worker unity under Ujamaa and with chronic staff sho rtages at the hospital, a bad worker was better than no worker at all. I D o N ot A sk the N urse! I A sk the P atient There are, however, some ways that nurses were able to mobilize non aligned documents to challenge official hierarchies of the hospital. Rar ely were hostilities between co workers voiced out loud, and discretion was highly valued.178 Yet through a patients file, a nurse could communicate a lot to an AMO. AMOs were often curt with nurses in the wards, and the highest ranking nurse in particular. I observed this in the male, female, and pediatric ward: an AMO would ask a nurse what was going on with a patient or yell at her for not anticipating the needs of the doctor at any given moment. If an AMO was particularly harsh with the nurses working in the ward, nurses may respond through the patient files neglecting any and all documentation responsibilities within the files. As the ward doctor and the head nurse goes through rounds 178 This was not only for the hospital, but overall in Tanzania, respect and good manners are highly valued. Public altercations are frowned upon, although the extent to which this is followed may vary from region to region.

PAGE 405

405 during the day, a frazzled AMO may yell or complain to the nurse that vital measurements have not been recorded for a particular patient. A common response on the part of the nurse is silence, or stating, how could I know about that? I started my shift this morning and I was not present during the night shift to be able to know what happened or why those papers were not filled out. Dr. Ezra, who was very well liked by nurses and AMOs alike, spoke of the ways that documents mediated the encounters between doctors, nurses, and patients, and how nurses could hinder patient ca re in order to communicate her disapproval of a doctors treatment: Dr. Ezra: For instance a patient is in the ward, I have written medicine [a prescription], okay? I expect that tomorrow morning if I come I will find that the patient has swallowed that medicine, but I come in the morning and I ask the patient, I do not ask the nurse! I ask the patient, have you swallowed your medicine? he says, I have not yet swallowed any medicine, how do you think I will feel? Right? Now, if you go and reprimand the nurse directly, she will build for you [a reputation] that such and such a doctor [ daktari fulani ] is bad in that environment [of the workplace]. But it is not that the doctor is bad, the doctor is good, but the nurse has not carried out his orders! (Inter view, Kiswahili, 07 1708). As D r. Ezras statement eludes nurses can negotiate with the official hierarchy by failing to complete doctors written order s Thus, for as much as the nursing officers complained about the OPD and laboratory staff not follow ing procedures, or about the parallel order that undermined the hospitals official hierarchy,179 it was not uncommon for nurses themselves to resort to this tactic. In the cases of nurses, it was not so often a lack of incentive to complete their work as it was a lack of respect for the particular doctors attitude towards them at any given time. Given that the nurses felt that they were under compensated and that they worked in an environment that did not honor their status as professionals, if doctors exhi bited distain for the status of nurses, they could find nurses particularly uncooperative and experience considerable impediments to their ability to provide care. Doctors therefore often rely on their patients or 179 For further information, see Chapter Six

PAGE 406

406 patients family members to tell them whether or not they received their medications or had particular treatments, due to deficiencies of charting and to unresponsive nurses.180 In other cases, nurses may neglect to follow treatment orders within a file in order to maintain her professionalism in r elation to a gap in her expertise. For instance, after each surgery, the operating AMO provides a list of post operative instructions that nurses are supposed to follow and report on in the morning report the next day. In one case in April 2008, a man who had his groin operated upon was returned to the ward with a file listing a set of post operative orders at the bottom of the clinical form. Among the orders was for the nurse to test the patients random blood sugar with a new machine that was introduced at the hospital only recently When the operating doctor came back later in the day to check on the status of her patient, she looked in the patient file and saw no notation of the random blood sugar reading. She approached the ward nurse, who told her that she did not realize that it was supposed to be done. When the doctor showed the order to the nurse, the nurse replied the order was written at the bottom of the form, rather than the top where she argued it should have been. The writing was in large re d lettering and marked urgent, the doctor said, unsatisfied by the nurses response. It was one of the many post operative orders that were listed for the patient, and other orders had been carried out. The doctor was perplexed. She ordered the nurse to do the test there and there. At that moment the nurse revealed that she did not know how to use the random glucose machine .181 180 These observations stand in stark contrast to those reported by Mulemi for hospital work in Kenya (2010). Mulemi observed that the hierarchical structures between doctors and nurses were starkly maintained. 181 Much to the vexation of the hospital staff and the nurses in particular, updates in clinical training have not been part of the MoHs training initiatives. Workshops and training seminars that are sponsored by the government and/or the DPs tend to surround issues of budgeting, monitoring and eval uation procedures, or to deal with set priorities of the MoH or donors such as maternal and child health, HIV/AIDS or malaria. For further information, see Chapters Seven and Eight.

PAGE 407

407 Non aligned documents were often present at the hospital due to MoH practice guidelines that were supposed to govern clinical inte ractions. However, unlike aligned documents which were tied into governing structures beyond the hospital nonaligned documents w ere not generally subject to oversight In light of the parallel orders undermining official hierarchies, and the burdens of pa rallel bureaucracies brought on at the hospital due to the presence of externally funded health programs, this lack of supervision was not so surprising. Conclusions Tacking between practices within and outside the purview of state and donor priorities and programs, this chapter uses documents as a lens on interconnections between multiple scales as they are embedded within the hospital. Indeed, donors, government ministries, and NGOs largely entrench their presence in health facilities through documents a nd the insertion of technologies, policies, and guidelines by which their interests are ideally to be carried out in practice. Some documents are within, and some specifically beyond the surveillance of donors and the MoH. An approach attentive to the circ ulations of and engagements with various documents within a health facility, and from a health facility to a governing body (MoH, NGO, donor) sheds considerable light on the flexibilities and constraints within the bureaucratic and biomedical configurations that make up hospital work. Taking documents seriously as ethnographic objects allows for significant insights into how resources are distributed, what possibilities and constraints they offer, and how they might be engaged in order to achieve institutional and professional desires of individuals working in the health facility. Documents aligned with development enclaves mark the presence of governing bodies and reveal considerable information about the contours of their priorities within the health sect or. Donors and governments equally desire particular quantitative representations of hospital work, in order to govern from afar and to be able to monitor and evaluate the progress of their

PAGE 408

408 interventions on the ground. Yet the state and the donors do not have equal capacity to be able to actualize the kind of surveillance of report making necessary to make data represent what it is that they want it to represent. Above I show that hospital workers endeavor to adhere to the aesthetic criteria of various dono r and government reporting guidelines, and as in the case of the RCH clinic cards, express considerable anxiety about securing the necessary inputs to make their reports appear authentic and compliant with reporting protocols. That said, there was consider able variation in what compliance meant within the hospital. Relating to reports on low priority areas of the hospital or on programs within which multiple donor interests merged, there were considerable reporting burdens, but not necessarily considerabl e apparatuses of surveillance from above. Conversely, donor reports within heavilyfinanced bi lateral donor funded programs such as that operating at the CTC, reports came with a high degree of authoritative surveillance. The data generated at the hospital for donor and government audiences did not necessarily circulate evenly among all actors involved in health sector governance. While HMIS data circulated widely among numerous donors and multinational entities, reports relating to targeted donor sponsore d programs were fiercely protected, traveling mostly among bi lateral donors and the governments who supported them. Hospital workers complied with the burdens of donor and government reporting in order to ensure that they continued to have access to the resources made available from that compliance. Their compliance was deliberate; they learned quickly that fulfilling those documentation requirements made them appear as an attractive site for donor and state investment. Dependent on resources from the government and the state, submitting to reporting burdens became a kind of currency through which the hospital could at least maintain its current resource flows, if not expand them.

PAGE 409

409 With regards to documents not aligned with state and donor priorities, peop le mobilized these objects in order to navigate official and unofficial hierarchies; maintain professional status in relation to patients, other staff, or gaps in their expertise; and cope with their difficult working environments. At the same time, as nonaligned documents called for inputs from various health professionals (and nurses in particular, who shouldered the majority of documentation burdens in nonaligned spaces), people experienced considerable anxiety trying to meet the demands of numerous (a nd often redundant) registers and records while simultaneously meeting their obligations to patient care. The entwining of bureaucracy and biomedicine, meant to systematize therapeutic interactions and communicate between workers across time and space, oft en failed to assist health workers to achieve the kinds of care they hoped to achieve even with available resources.

PAGE 410

410 Figure 101. A refrigeration unit at Kiunga for vaccines in the RCH. The white label has an image of the Japanese flag, and below is written "From the People of Japan". Photo by N. Sullivan

PAGE 411

411 A B Figure 102. Under five child development clinic card. A) Front cover of clinic card. Lower box provides space for recording immunizations and doses of vitamin A. B) Interior of clinic c ard. Bottom left has PMTCT (Prevention of Mother to Child Transmission) to indicate HIV status, as well as hati punguzo the Kiswahili term for the mosquito net vouchers. There is also a space for ngao, which is an insecticide provided at intervals t o treat nets. Vouchers and insecticide are part of the Presidents Malaria Initiative (PMI) program. PMTCT sponsored by Presidents Emergency Plan for AIDS Relief (PEPFAR) funding. Growth chart is part of a World Health Organization ( WHO ) initiative.

PAGE 412

412 Fi gure 103. Resource, policy, and data circulations between Tanzanian health facilities, the Ministry of Health, donors, and multi lateral organizations such as the United Nations.

PAGE 413

413 Figure 104. The Vice President of REFLECT (far left) is served a fresh Tanzanian lunch in one of the doctors offices of the Counseling, Testing and Care clinic. Photo by N. Sullivan

PAGE 414

414 Figure 105. Clinical Officers share a small room, consulting with multiple patients simultaneously. Photo by N. Sullivan.

PAGE 415

415 CHAPTER 11 CONCLUSION This study focused on the convergence of transnational governmentality, global health discourses, and development aid, and how these impacted both bureaucratic and biomedical practices within a government owned Tanzanian hospital. It argued for th e importance of considering the hospital as a totality, and movements beyond its boundaries. Such an approach allows for explorations of the politics of global and state governance apparatuses and priorities, and how these encroach upon biomedical encounte rs within aid dependent countries. Existing hospital ethnographies tend to focus their attentions on specific departments, maladies, or populations within hospitals, providing little understanding of how they potentially link up to other forces or actors, and the effects of these connections and movements on the possibilities and limitations within therapeutic spaces. Combining insights from medical anthropology and history, anthropology of the state, anthropology of policy, and science and technology studies, I demonstrated that multiple scales (global, state, local) are linked up within the confines of the daily practices within health facilities. The embedding of state and donor priorities within under resourced and fragile health facilities matters. The past decade marked a significant divergence from the stagnation that health workers and patients experienced over previous decades. Although now a decentralized government, Tanzanias health sector is subject to more donor and state surveillance and direc t intervention than ever before. As a result, technologies, resources, professional possibilities, bureaucratic burdens, and forms of expertise are unevenly distributed in the hospital, offering significant opportunities, but also highlighting important in equalities, within the daily practices of hospital work. In the era of transnational aid regimes, donor sponsored and well resourced health interventions, and important reorientations of government health sector policies, the

PAGE 416

416 notion of a unitary hospital culture within a health facility is no longer tenable. It is high time for an ethnography that takes account of the politics of uneven resource distributions within health facilities, and how hospital actors engage and manipulate governance and biomedical regimes in order to achieve their own desires whether personal or professional, whether for healing or being healed. This dissertation is an endeavor to account for unevenness, and explore its implications for health workers, patients, and the regimes meant to govern these unwieldy places. In this study, I focused attention on the interface of desires and performances along multiple axes, in order to explore the motivations of differently situated actors (whether individuals or collectivities) for engagin g and collaborating across space. The desires of individuals, institutions, donors, government ministries all of these matter for the politics of how collaborations operates in practice, and the effects of these alliances. Tacking between global, state, and local scales permits an analysis that goes beyond merely describing the conditions within which people interact in medical facilities. It offers explanations of why these contexts are as they are, and what possibilities are opened up or closed down as a result of punctuated interconnections. Biomedical Pluralism Sites such as Tanzania are analytically useful places to investigate what biomedicine is, in practice, and how it is transforming under changing modes of state and global governance regimes. Cont rary to the universalistic notions of a medical science that travels neutrally across space and time, investigating biomedicines instabilities sheds light on the flexibility and adaptiveness of biomedicine when applied to different contexts. Standardizing those practices across time and space potentially erodes its therapeutic value, particularly in locales where infrastructure, technology, expertise, and maladies are never taken for granted. What was seen in

PAGE 417

417 this study was not a unitary, culturallyspecif ic biomedicine, but rather a multiplicity of biomedicines, punctuated by the uneven materiality of practice that enacted both abundance and scarcity within the same health facility. The machinations of donor sponsored health interventions and state ident ified priority areas caused considerable inequalities within the hospital. Attention to the pluralism of biomedicine unearths its plasticity. While biomedicines endeavors to eradicate other therapeutic systems and claim a kind of hegemony throughout the globe has concerning consequences, its other power, and one critical to its role in aid dependent countries in particular, is its adaptability to multiple contexts not because it has achieved standardization and universal applicability, but because its plu rality allows for multiple avenues to try out a variety of therapeutic interventions to be creative within a context that is always transforming and that is transformed by biomedical practice itself Increasing the inequalities within the hospital, development enclaves guard their resources well, making them unavailable for therapeutic encounters beyond enclave boundaries. These resources are carefully kept within the enclaves, for use in carefully delineated therapeutic encounters for the few whose expertise, maladies or bodily states allow them access. These enclaves do provide meaningful care to those patients with HIV, or who are pregnant or under five years of age. But even then, the punctuated nature of state and donor priorities can disarticulate bodily states from bodily experiences. A pregnant woman is only able to access enclaved biomedical care while she is pregnant. Once she transitions into labor, her body is no longer under donor/state purview. The majority of patients who find their way in to hospitals suffer from non prioritized ailments, many of which could benefit from similar resource investments as those available in

PAGE 418

418 the enclave. More resources would allow for increased plurality, expanding the capacity for health workers to improvise c are appropriate to the situation. Instead, these resources are restricted to enclaved spaces. Meanwhile, a considerable amount of health personnels time is spent compiling data for the government and donors, reducing the amount of time they can spend cari ng for their patients in an environment already chronically under staffed and over burdened. The data compiled may conform to the proper aesthetic criteria dictated by guidelines and policy mandates, but its applicability to meaningful governance from afar remains in question. As Volker Scheid argues regarding attempts to standardize Chinese medicine, but equally applicable here, it is only through creating normalization processes (such as measurement indicators and their required statistical inputs) that governments (and donors) can implement statesponsored and state controlled schemes for the improvement of society (2002:269). However, due to the attempt to standardize what is, by its very nature, a pluralistic and flexible therapeutic modality, failur e to achieve that very standardization becomes the norm instead of the exception. Never have the MoH or its donors managed to achieve the degree of standardization desired. It is an unachievable ideal, in Tanzania, and everywhere else. The futility of this kind of exercise is most visible in places such as Tanzania, where practitioners and patients coming together to address maladies can never take for granted the contexts on which that encounter depends. Implications for Hospital Ethnography Global health policies and donor sponsored programs are increasingly implicated in daily healthcare provision in aid dependent countries. In such a context, existing ideas about hospital culture as reflecting the core values of a local society are no longer tenable, simply because health facilities are punctuated by so many entities and interests that are far outside the local communities where these institutions are located Diverging from existing hospital ethnography

PAGE 419

419 approaches that disarticulate particular departm ents, populations, or maladies from the institutions in which they exist, I argue that studies should account for the multiplicities and interconnections that make up hospitals and hospital practices. In doing so, one can reveal the politics underlying the uneven materiality of hospital spaces, what movements are allowed or restricted, what possibilities open up or close down for health professionals and patients, and the intricacies of how these facilities become embedded in connections with state and glob al entities outside of their specific borders. Drawing on approaches from anthropology of the state, anthropology of policy, and science and technology studies can provide significant insights about how ethnography might achieve this. As Lock and Nichter a rgue, medical anthropology needs to start documenting increasing inequalities, uncertainties, rigidities, and violence directly associated with globalism, modern economies, and scientific rationalism that informants everywhere are subject to today (2002: 26). As global health policies, donor priorities, and national health agendas collude and become embedded in health institutions, it is vitally important to investigate how biomedical pluralism interacts with these forces, and to what effect for healthcare provision. As demonstrated in this study, the decentralization of the state has, ironically, increased donor and state influence in the health sector, which substantially increases inequalities and uncertainties for both patients and health professionals within biomedical spaces. Looking for Resources Beyond the State One of the major transformations highlighted in this study was a reorientation in where health professionals sought out resources. In the period before HSR and public private partnerships (P PPs), government owned health facilities largely relied on the central state and the local communities in which they were situated in order to procure the resources they needed. As a result of HSR, PPPs, and increasing infusions of NGO and donor sponsored vertical health programs, the hospital administration, as well as individual health professionals working there,

PAGE 420

420 began looking beyond the state for that which was aspired to. On an institutional level, the hospital administration managed to formulate rela tionships with a variety of what I called haphazard donors, and as a result succeeded in constructing several new buildings in the span of a few short years. The administration also took advantage of the interests of donors sponsoring vertical initiative s in the area, convincing them to expand existing hospital buildings in the interest of improving the success of donor programs. This reorientation of focus away from the state and towards donors was also significant among health workers themselves. While the consensus was that the MoH had an important role in creating policy and coordinating donor efforts, it was the donors who provided workers with the resources and training that the state consistently failed to offer. Staff had their own ideas about wha t might be most useful in terms of training when I asked them, but largely they would take whatever training opportunities were provided, so long as they were able to increase their skills and certification, and access the hefty compensation that such trai ning allowed. Donor programs appealed to professional aspirations for upgraded training, working conditions nearing or even exceeding their professional expectations, and extra duty allowances. Despite my own concerns about the distribution of donor resour ces, hospital workers expressed few reservations about donors or their priorities. They simply hoped that more donor interest would come into the hospital, and they worked hard to cater to donor interests. Individuals invested a lot of time and energy into following donor guidelines and procedures. These relationships were carefully nurtured, institutionally, and professionally, in hopes to maintain access to additional resources for the hospital, and to increase personal access to training and additional c ompensation. Donors were perceived as the possessors of what workers needed. Never prior to HSR were workers so oriented to the interests of donors.

PAGE 421

421 Looking beyond the state in such ways also had its drawbacks. Donor programs increased resources, but they were so targeted at particular areas and interests in the hospital that certain individuals benefitted more than others. Given that there were not sufficient workshops and donor related compensation opportunities to go around, it created an economy of sca rcity in which staff competed against each other to gain access, eroding the fragile morale that had been built up in previous years due to the many infrastructural improvements at the hospital. Donor sponsored workshops created fissures between workers, a nd fostered suspicions that administrators were implicated in corruption and favoritism. By the end of 2008, several staff members resumed practices that had not reportedly been a problem since before HSR: absenteeism, tardiness, lackluster work. Reminded of the failures of the hospital to provide them with professional opportunities that would help them achieve their personal and professional aspirations, the staffs commitment to offering the best quality services they could in both highand low priority areas waned. The development enclaves that donors established within the hospital did little to reinforce the fragile foundation of the institution on which it depended to enact their programs. The Importance of the Local As revealed by tacking between t he hospital and the spaces where health policies and programs were planned, there is considerable interdependency between the state, donors, and the local facilities where policies are implemented. I argued that the local matters considerably in terms of t he ways that policies and programs are enacted. Ideally, all districts within the country would invest their time in not only providing biomedical care, but also providing the data inputs. These were necessary for the MoH and its Development Partners (DPs) to create policies aimed at improving biomedical healthcare within Tanzania. However, as was discussed, commitment to data varied among health facilities in the country. Kiunga District Hospital, having learned about

PAGE 422

422 the linkage between data inputs and re liable resource circulations, invested significant attention to preparing government and donor reports, but still, the quality of these reports differed based on the audience for whom the reports were being compiled. The machinations surrounding documents within the hospital mattered for making policies; however they also mattered in terms of what kind of representation of Tanzanias health system filtered up to the multi national organizations that compile comparative global health statistics. The fact that the districts were not fully participating in proper data collection generated considerable anxiety among both the MoH and its various DPs as they realized that they were creating policies and drawing upon numerical representations that were, as one donor employee called it, garbage. Given the hardship characteristic of low priority areas within health facilities, and the uneven distribution of resources overall, the expectation that health facilities will be motivated to be data compliant is perhaps unrealistic because the infrastructures and resources required to make that possible were lacking in most facilities. Often the unofficial norms informing daily practices at the local level get subsumed under the label of corruption, yet ironically even these supposedly corrupt practices were critical to hospital administrations ability to improve the infrastructures at the facility, which in turn increased staff investment and compliance in the standards meant to govern their workplace. Indeed, consistent with other anthropological studies of the state, what was found at the local level in Kiunga was that policies aimed at eradicating existing systems of patronage at the local level often served to reinforce them (pace Chalfin 2010). The parallel and hybrid orders outlined in this study favored continued attention to politically and religiously powerful individuals within the local community. Indeed, it was only by nurturing relationships within the existing patronage system that the District Medical Officer (DMO) at Kiunga was able to achieve the sort

PAGE 423

423 of local political buy in required for his infrastructural aspirations to become realities. Without local political support, the DMO would have failed to have his projects approved by the local gover nment, even if he managed to acquire the financial support from donors. Indeed, the DMO at Kiunga prior to Dr. Saidi found her initiatives considerably stalled, even when she had finances available, due to the lack of local political support. In order to a chieve this political investment in building up the hospital, the DMO had to tolerate an unofficial order that allowed for those at the top and very bottom of the hospitals official hierarchy to have the most power. These unofficial orders based on exist ing authority structures in the local community had ambiguous effects. The DMO was able to garner political support for his infrastructural endeavors, which then caused the hospital staff to believe in a collective institutional future that increased their own professional investments in their workplace. Based on the perspectives of staff who worked at the hospital from before the implementation of HSR, this commitment to their work and the improved standards for healthcare provision were largely a result o f the facilitys successes at building itself up. Infrastructure and professionalism were intimately intertwined. However, when it came to day to day practices, the people at the hospital with the most professional expertise were also the least powerful w ithin the parallel and hybrid orders. Staff in the lower ranks exhibited little interest in doing their share of the work, requiring that highranking and better trained nurses spend much of their day doing nonspecialized tasks. This had a negative impact on patient care. Patients, hoping to receive adequate treatment at the hospital, may have to wait upon admission to a ward for several days before receiving services because nurses and Assistant Medical Officers (AMO) were entirely overloaded. Higher rank ed nurses experienced considerable hardship attempting to meet the minimum requirements for their

PAGE 424

424 wards, often without the assistance of the other personnel below them in rank. While two of the head nurses adopted a strategy that allowed their departments to better distribute the workload, the In Charge nurses in the other wards feared retribution if they attempted to authoritatively assert their role as the leader of the department. Even AMOs feared they might be transferred if they were too harsh with low er ranked nurses who had significant social capital. These unofficial norms governing hospital work were learned in situ when new staff came into the hospital. They were also learned very quickly. Social capital remains highly important and relevant to pe oples needs not only within the hospital, but importantly, outside of it. When ones workplace does not provide opportunities for individuals to meet their personal and professional aspirations, or even their basic needs, it is often through social capita l beyond the official workplace that people find economic opportunities that can help them achieve what is important to them. Finally, there is a way that the local level matters that remains to this day largely obscured from view. It draws on Biehl and L ockes argument (2010) that peoples hopes and desires offer alternative paths. In Tanzania, hospital administrations, the donors, and the MoH missed out on important opportunities to take advantage of the perspectives and experiences of local health worke rs. There was a considerable bottleneck in the person of the DMO. The DMO was situated at the confluence of the local politicians, the MoH, and donor programs. This meant that any ideas or concerns that workers felt should be made known had no recourse if the DMO would not listen. Because the existing, official management structure of Community Health Management Teams (CHMT) and Hospital Management Teams (HMT) catered to the interests of unofficial authority structures in the district, the needs and innovat ive ideas of the workers were effectively silenced. When MoH representatives visited the hospital, they spoke only to the CHMT and HMT members; there was little workers to do to speak their concerns or raise their

PAGE 425

425 (often very novel and achievable) ideas. D onors and NGOs operating through the hospital also missed out on this opportunity. They were supervisors, not collaborators. However, given how resourceful the staff was, this was a missed opportunity. Staff often talked about small and largely inexpensive (or even free) initiatives through which they could reorganize the hospitals workflows or resource distributions in ways that did not violate state or donor guidelines, but would go far to improve their work, and especially, the overall quality of care they were able to provide to their patients. Workers were proud about the hospitals achievements, and had a collective vision of a way forward (a surgical theatre, a mortuary). Moreover, they felt considerable pride when they were able to provide the kind of patient care that was consistent with their professional ideals. They were also incredibly disappointed when the capacity of their workplace failed to allow them to assist a patient in need. The many institutional accomplishments they witnessed in a sh ort period of time generated in them numerous ideas about what more they could do to help their fellow Tanzanians or their ndugu, as one doctors articulated it. Because their ideas too often went ignored, they were often spoken only in private. On a few oc casions, workers approached me about them, in hopes that I could bring them before the administration. My efforts often failed as well. The administration was too busy working to garner local political support through the unofficial orders, and to meet the donor and MoH reporting requirements that ensured that resourced continued to come into the hospital. The case of Kiunga District Hospital demonstrates that health professionals invest more of their efforts in an institution where they believe that possibilities exist for their own professional and personal growth. The fact that their desires and ideas received no effective audience was a missed opportunity. Improving the working conditions improved the quality of work.

PAGE 426

426 An Exit Story Despite my protestati ons, on my last day at the hospital in December 2008 the entire hospital staff, save the DMO and the Matron who were on vacation, pooled money from their own pockets to throw a farewell party. At the end of the afternoon shift on that sunny Friday, the ent ire staff descended on the undercover area of the pediatric health unit within the reproductive and child health clinic. Two of the ward attendants had spent the day cooking a feast, and boxes of colorful sodas sat to the side of a head table, draped in be autiful kitenge cloth and covered with a brand new hospital sheet. They raised enough money to rent a sound system, crackly though the acoustics were. I was directed to sit at the head table, between the District Nursing Officer (DNO), Mary, the Medical Of ficer In Charge (MOI) Dr. Urasa, and the acting Matron, Laura. They made speeches. They sat me in a chair as they paraded gifts to music. I bought every department a new tea set. Tea breaks were incredibly important to the staff a chance to sit, socialize, and share small snacks before going back to their difficult work. We ate the splendid food, and even patients who meandered by were invited to join in the feast. Once the meals were finished and all the food was gone, the music began to play and we danced. The last song was called Nibebe Carry Me. As it played, the staff took turns carrying each other on their backs (Figure 11 1). The message was profound for a hospital. They all needed and relied on each other. The hospital did not work unless everyone contributed. The same might be said for the health sector as a whole.

PAGE 427

427 Figure 111. N. Sullivan and acting Matron Laura participate in Nibebe dance.

PAGE 428

428 REFERENCES CITED Abel Smith, Brian and Pankaj Rawal 1992a Health Sector Financing Study for the Gove rnment of Tanzania: Report on the Potentiality for Cost Sharing. First Report: User Charges. Overseas Development Administration 1992b Can the Poor Afford Free Health Services? A Case Study of Tanzania. Health Policy and Planning 7 (4) :32941. Adams, Vincanne 1998 Doctors for Democracy: Health Professionals in the Nepal Revolution. Cambridge: Cambridge University Press. Alger, Edith H.S. 1981 A Survey of the Health Services in the Arusha Region. S.n. Anders, Gerhard 2005 Good Governance as Technology: Towards an Ethnography of the Bretton Woods Institutions. In The Aid Effect: Giving and Governing in International Development. David Mosse and David Lewis, eds. Pp. 3760. NY: Pluto Press 2010 In the Shadow of Good Governance: An Ethnography of Civil Service Reform in Africa. Leiden & Boston: Brill. Andersen, Helle Max 2004 'Villagers': Differential Treatment in a Ghanaian Hospital. Social Science and Medicine 59:20032012. Annist, Aet 2005 The Worshippers of Rules? Defining Right a nd Wrong in Local Participatory P roject Applications in SouthEastern Estonia. In The Aid Effect: Giving and Governing in International Development. David Mosse and David Lewis, eds. Pp. 150170. NY: Pluto Press Appadurai, Arjun, ed. 1986 The Social L ife of Things: Commodities in Cultural Perspective. Cambridge: Cambridge University Press. Astor, Avraham, Tasleem Akhtar, Maria A. Matallana, Vasantha Muthuswam, Folarin A. Olowu, Veronica Tallo and Raidar K. Lie 2005 Physician Migration: Views from Professionals in Colombia, Nigeria, India, Pakistan and the Philippines Social Science and medicine 61(12):2492500. Bagachwa, Mboya and Colleagues [sic] 1997 Aid Effectiveness in Tanzania, with Special Reference to Danish Aid. In Foreign Aid in Afric a: Learning from Country Experiences. Jerker Carlsson, Gloria Somolekae & Nicolas van de Walle, eds. Pp. 168193. Uppsala, Sweden: Nordiska Afrikainstitutet.

PAGE 429

429 Beck, Ann 1977 Medicine and Society in Tanganyika 18901930: A Historical Inquiry. Philadelphia: The American Philosophical Society. 1981 Medicine, Tradition, and Development in Kenya and Tanzania, 1920 1970. Waltham, MA: Crossroads Press. Benson, John S. 1996 Primary Health Care Options in Tanzania: The Case of Arusha and Arumeru Districts. P hD Diss., Department of Geography. University of Minnesota. Biehl, Joo 2005 Technologies of Invisibility: Politics of Life and Social Inequality. In Anthropologies of Modernity: Foucault, Governmentality, and Life Politics. Jonathan Xavier Inda, ed. Pp. 248271. Malden MA: Blackwell. Biehl, Jo o and Peter Locke 2010 Deleuze and the Anthropology of Becoming. Current Anthropology 51(3):317 351. Biermann, Werner and Jumanne Wagao 1986 The Quest for Adjustment: Tanzania and the IMF, 19801986. Afri can Studies Review 29 (4) :89103. Blundo, Giorgio and JeanPierre Olivier de Sardan 2006 Everyday Corruption and the State: Citizens and Public Officials in Africa. With N. Bako Arifari and M. Tidjani Alou. Susan Cox, transl. London & NY: Zed Books. B lundo, Giorgio and Pierre Yves Le Meur 2008 Introduction: An Anthropology of Everyday Governance: Collective Service Delivery and Subject Making. In Governance of Daily Life in Africa: Ethnographic Explorations of Public and Collective Services. Giorgio Blundo and Pierre Yves Le Meur, eds. Pp. 138. Leiden and Boston: Brill. Booth, Karen M. 2004 Local Women, Global Science: Fighting AIDS in Kenya. Bloomington and Indianapolis: Indiana University Press. Brown, Adrienne 2000 Current Issues in Se ctor Wide Approaches for Health Development: Tanzania Case Study. Geneva: WHO. Brown, Hannah Forthcoming Hospital Domestics: Care Work in a Kenyan Hospital. Special issue, Hospital Heterotopias: Comparative Ethnographies of Biomedical Places. Alice Street and Simon Coleman, eds. Space and Culture.

PAGE 430

430 Burri, Regula Valrie and Joseph Dumit, eds. 2007 Biomedicine as Culture: Instrumental Practices, Technoscientific Knowledge, and New Modes of Life. NY: Routledge. Chikanda, Abel 2005 Medical Leave: The Exodus of Health Professionals from Zimbabwe. Cape Town: Southern African Migration Project. Chalfin, Brenda 2010 Neoliberal Frontiers: An Ethnography of Sovereignty in West Africa. Chicago and London: University of Chicago Press. Charters, A .D. 1985 Reminiscences of East Africa and Western Australia in the Milestones of a Doctors Life, 19031984. Kalamunda, Western Australia: A.D. Charters. Citizen, The (Dar es Salaam) 2010 Donors: Govt too Slow in Fighting Corruption. Dec. 10. 2011 Govt Seeks Emergency Support from Donors. Feb. 10. Clyde, David F. 1962 History of the Medical Services of Tanganyika. Dar es Salaam: Government Press. Comaroff, Jean 1993 The Diseased Heart of Africa: Medicine, Colonialism, and the Black Body. In Knowledge Power, and Practice: The Anthropology of Medicine and Everyday Life. Shirley Lindenbaum and Margaret Lock, eds. Pp. 305329. Berkeley: University of California Press. Comaroff, Jean and John Comaroff 2000 Millennial Capitalism: First Thought s on a Second Coming. Public Culture 12(2):291343. Commonwealth Local Government Forum 2007? Country Profile: Tanzania: The Local Government System in Tanzania. Electronic document, http://www.clgf.org.uk/index.cfm/PageID/126/ViewPage/ Tanzania, accessed May 3, 2010. Conell, John, Pascal Zurn, Barbara Stilwell, Magda Awases and Jean Marc Braichet 2007 SubSaharan Africa: Beyond the Health Worker Migration Crisis? Social Science and Medicine 64(9):1876 91. Cooksey, Brian 1986 Evaluation of Swis s Funded Health Projects: Health Policy and Donor Policy in Tanzania. Report for Swiss Development Cooperation. Dar es Salaam: Afro Aid Consultants Ltd.

PAGE 431

431 Cooper, Frederick 2005 Colonialism in Question: Theory, Knowledge, History. Berkeley: University of California Press. Coser, Rose L. 1962 Life in the Ward. East Lansing: Michigan State University Press. COWI, EPOS Health Consultants, and Goss Gilroy, Inc. 2007 Joint External Evaluation of the Healt h Sector in Tanzania 19992006. Copenhagen: Ministry of Foreign Affairs of Denmark. Daily News, The [Tanzania] 1972 Some Public Services are Deteriorating. May 6, no. 10:9. 1972 Kondoa Hospital. May 8, no. 11:9. 1973 I Sympathise with Nurses. Jan. 20, no. 543:5. 1973 Medical Salaries Gap Jus tified. May 31, no 31:9. 1997 Chiduo Pledges Doctors Allowance Arrears. Aug. 27:3. 1997 Medics Assess Health Delivery. Sept. 25:3. 1997 Govt Suspends Promotions. Dec. 3:1. DelVecchio Good, MaryJo, Esther Mwaikambo, Erastus Amayo, and James MImunya Machoki 1999 Clinical Realities and Moral Dilemmas: Contrasting Perspectives from Academic Medicine in Kenya, Tanzania, and America. Daedalus. Vol. 128(4):167 196. Dibble, J. Birney 1965 In this Land of Eve. New York: Abingdon Press. Dilger, Hansjrg 2001 Living PositHIVely in Tanzania: The Global Dynamics of AIDS and the Meaning of Religion for International and Local AIDS Word. Afrika Spectrum 36(1) :73 90. 2009 Doing Better? Religion, the Virtue Ethics of Development, and the Fragmentation of Health Politics in Tanzania. Africa Today 56 (1) :89 110. 2011 Claiming Territory: Medical M ission and H istories of I nequality in a R eligiously D iverse C ity. Paper presented at workshop, Medical Diversity and its Spaces, Max Planck Institute for the St udy of Religious and Ethnic Diversi ty, Gttingen, March 28 29. Dunn, Elizabeth C. 2005 Standards and PersonMaking in East Central Europe. In Global Assemblages: Technology, Politics, and Ethics as Anthropological Problems. Aihwa Ong and Stephen J. Co llier, eds. Pp. 173193. Malden, MA: Blackwell. Eastwood, John B., Rachel E. Conroy, Sarala Naicker, Peter A. West, Roger C. Tutt, and Jacob Plange Rhule 2005 Loss of Health Professionals from SubSaharan Africa: The Pivotal Role of the UK. Lancet 365: 1893900.

PAGE 432

432 Elyachar, Julia 2003 Mappings of Power: The State, NGOs, and International Organizations in the Informal Economy of Cairo. Comparative Studies in Society and History 46:571605. Evans Pritchard, E. E. 1976 Witchcraft, Oracles, and Magic a mong the Azande. Abridged. Oxford: Clarendon Press. Fassin, Didier 2007 When Bodies Remember: Experiences and Politics of AIDS in South Africa. Berkeley and Los Angeles: University of California Press. Feierman, Steven and John M. Janzen 1992 Int roduction. In The Social Basis of Health and Healing in Africa. Steven Feierman and John M. Janzen, eds. Pp. 123. Berkeley: University of California Press. Ferguson, James 2006 Global Shadows: Africa in a Neoliberal World Order. Durham: Duke Universi ty Press. F inkl er, Kaja 2004 Biomedicine Globalized and Localized: Western Medical Practices in an Outpatient Clinic of a Mexican Hospital. Social Science and Medicine 59:20372051. Foucault, Michel 1991 Governmentality. In The Foucault Effect: Studies in Governmentality. Graham Burchell, Colin Gordon and Peter Miller, eds. Pp. 87104. Chicago: University of Chicago Press. Gibson, Diana 2004 The Gaps in the Gaze in South African Hospitals. Social Science and Medicine 59:20132024. Goffman, Ervi ng 1961 Asylums. Garden City: Anchor Books. Good, Byron J. 1994 Medicine, Rationality, and Experience: An Anthropological Perspective. Cambridge: Cambridge University Press. Gottlieb, Manuel 1975 Health Care Financing in Mainland Tanzania. Syrac use, NY: Maxwell School of Citizenship and Public Affairs, Syracuse University.

PAGE 433

433 Gould, Jeremy 2005 Timing, Scale and Style: Capacity as Governmentality in Tanzania. In The Aid Effect: Giving and Governing in International Development. David Mosse and David Lewis, eds. Pp. 6184. NY: Pluto Press. Government of the United Republic of Tanzania 2005 National Strategy for Growth and Reduction of Poverty. Dar es Salaam: Vice Presidents Office 2008 Annual Health Statistical Abstract Tanzania Mainland Dar es Salaam: Health Information and Research Section, Department of Policy and Planning. Grant, Hugh M. 2006 From the Transvaal to the Prairies: The Migration of South African Physicians to Canada. Journal of Ethnic and Migration Studies 32(4):681 695. Green, Maia 2000 Public Reform and the Privatisation of Poverty: Some Institutional Determinants of Health Seeking Behaviour in Southern Tanzania. Culture, Medicine and Psychiatry 24:403430. 2003 Globalizing Development in Tanzania: Policy Fr anchising through Participatory Project Management. Critique of Anthropology 23:123143 Guardian, The (Dar es Salaam) 1995 Minister Warns on Hospital Corruption. Dec. 13:1. 1997 Govt Should let Doctors Strike, says Ndimara. July 22:5. Gupta, Akhil a nd Aradhana Sharma 2006 Introduction: Rethinking Theories of the State in an Age of Globalization. In The Anthropology of the State: A Reader. Akhil Gupta and Aradhana Sharma, eds. Pp. 141. Malden: Blackwell Publishing. Hagopian, Amy, Anthony Ofosu, Adesegun Fatusi, Richard Biritwum, Ama Essel, L. Gary Hart and Carolyn Watts 2005 The Flight of Physicians from West Africa: Views of African Physicians and Implications for Policy. Social Science and Medicine 61(8):175060. Hall, Budd L. 1978 Mtu ni Afya (Man is Health): Tanzanias Health Campaign. Washington, DC: Clearinghouse on Development Communication. Harrison, Graham 2008 From the Global to the Local? Governance and Development at the Local Level: Reflections from Tanzania. Journal of M odern African Studies 46(2) :169 189. Havnevik, Kjell J., Finrby, Ruth Meena, Rune Skarstein, and Ulla Vuorela 1988 Tanzania: Country Study and Norwegian Aid Review. Bergen, Norway: Centre for Development Studies, University of Bergen.

PAGE 434

434 Helleiner, Gerry 1999 Changing Aid Relationships in Tanzania. Ms. Dar es Salaam. Cited in Helleiner, Gerry, Tony Killick, Nguyuru Lipumba, Benno J. Ndulu and Knud E. Svendsen 1995 Report of Independent Advisors on Development Cooperation Issues between Tanzania an d its Donors. ms Dar es Salaam. Electronic document http://www.tzdpg.or.tz/uploads/media/Helleiner_Report_1995.doc accessed April 10, 2010. Heggenhougen, Harald Kristian 1986 Health Services: Official and Unofficial. In Tanzania: Crisis and Struggle for Survival. J. Boesen, K. J. Havnevik, J. Koponen and R. Odgaard, eds. Uppsala, Sweden: Scandinavian Institute of African Studies. [pages?] Hobbs, Graham 2001 The Health Sector Wide Approach and Health Sector Basket Fund. Final Report. Dar es Salaam: Economic and Social Research Foundation. Electronic Document www.tzdpg.or.tz/uploads/media/Basket_Fund_Study.doc, accessed June 23, 2010. Holden, Pat 1984 Nursing Sisters in Nigeria, Uganda, Tanganyika/Tanzania, 19291978. Oxford, England: Oxford Development Records Project. Horsley, Philomena 2008 Death dwells in spaces: Bodies in the hospital mortuary. Anthropology and Medicine 15(2):133 146. Hsu, Elizabeth 2002 The Medicine from China has Rapid Effects: Chinese medicine patients in Tanzania Anthropology and Medicine 9(3):291 313. Hunt, Nancy Rose 1999 A Colonial Lexicon: Of Birth Ritual, Medicalization, and Mobility in the Congo. Durham: Duke University Pr ess. Iglehart, John K. 1996 The Quandary over Graduates of Foreign Medical Schools in the United States. New England Journal of Medicine 334:167983. Iliffe, John 1979 A Modern History of Tanganyika. New York: Cambridge University Press. 2002 E ast African Doctors: A History of the Modern Profession. Kampala, Uganda: Fountain Publishers. Originally published 1998 by Cambridge: Cambridge University Press.

PAGE 435

435 Imperato, Pascal J. 1964 Doctor in the Land of the Lion. New York: Vantage Press. 1966 Witchcraft and Traditional Medicine among the Luo of Tanzania. Tanzania Notes and Records 66:193201. Inhorn, Marcia C. 2002 The Local Confronts the Global: Infertile Bodies and New Reproductive Technologies in Egypt. In Infertility around th e Globe: New thinking on Childlessness, Gender, and Reproductive Technologies. Marcia C. Inhorn and Frank van Balen, eds. Pp. 263282. Berkeley: University of California Press. 2004 Privacy, Privatization, and the P olitics of Patronage: E thnographic C hallenges to P enet rating the Secret W orld of Middle Eastern, Hospital based In V itro F ertilization. Social Science and Medicine 59(10):20952108. Jaffr, Yannick and Jean Pierre Olivier de Sardan 2003 Une medicine inhospitalire: Les difficiles relations entre soignants et soigns dans cinq capitales dAfrique de lOuest. Paris: APAD, Karthala. Janzen, John M. 1978 The Quest for Therapy: Medical Pluralism in Lower Zaire. Berkeley and Los Angeles: University of California Press. Jennings, Michael 2005 Chinese medicine and Medical pluralism in Dar es Salaam: Globalisation or glocalisation? International Relations 19(4):457 473. Jilek Aall, Louise 1979 Call Mama Doctor: African Notes of a Young Woman Doctor. Seattle, WA: Hancock House Publisher s, Inc. Jorland, Grard and George Weisz 2005 Introduction: Who Counts? In Body Counts: Medical Quantification in Historical and Sociological Perspective/La quantification medicale, perspectives historiques et sociologiques. Jorland, Grard, Annick Opinel, and George Weisz, eds. Pp. 315. Montreal and Kingston: McGill Queens University Press. Kaiser, Paul J. 1996 Culture, Transnationalism, and Civil Society: Aga Khan Social Service Initiatives in Tanzania. Westport, CT: Praeger. Kalofonos, Ippolytos A. 2008 All I Eat is ARVs: Living with HIV/AIDS at the Dawn of the Treatment Era in Central Mozambique. PhD Diss., Department of Anthropology. University of California, San Francisco and University of California, Berkeley. 2010 All I eat i s ARVs: The Paradox of AIDS Treatment Interventions in Central Mozambique. Medical Anthropology Quarterly 24(3):363 380.

PAGE 436

436 Kamndaya, Samuel 2011 Tanzania to Get Sh4tr World Bank Grants and Loans. The Citizen (Dar es Salaam), Feb. 9. Kelsall, Tim and C laire Mercer 2003 Empowering People? World Vision and Transformatory Development in Tanzania. Review of African Political Economy 30( 96) : 293 304. Kimaro, Honest C. and Jos Nhampossa 2005 Analyzing the Problem of Unsustainable Health Information S ystems in LessDeveloped Economies: Case Studies from Tanzania and Mozambique. Information Technology for Development 11(3) :273298. Kiwara, A. D. 2003 Gender Implications of Structural Adjustment in Health. In Against Neoliberalism: Gender, Democra cy and Development. Chachage S. L. Chachage and Marjorie Mbilinyi, eds. Pp. 183195. Dar es Salaam, Tanzania: Tanzania Gender Networking Programme. Langwick, Stacey A. 2001 Devils and Development. PhD Diss., Department of Anthropology, University of North Carolina at Chapel Hill. 2006 Geographies of Medicine: Interrogating the Boundary between Traditional and Modern Medicine in Colonial Tanganyika. In Borders and Healers: Brokering Therapeutic Resources in Southeast Africa. Tracy. J. Luedke and Harry G. West, eds. Pp. 143165. Bloomington and Indianapolis: Indiana U. Press. 2007 Devils, Parasites, and Fierce Needles: Healing and the Politics of Translation in Southern Tanzania. Science, Technology and Human Values. Vol. 32(1):88117. 2008 Articulate(d) Bodies: Traditional Medicine in a Tanzanian Hospital. American Ethnologist 35(3):428 439. 2010 From NonAligned Medicine to Market Based Herbals: Chinas Relationship to the Shifting Politics of Traditional Medicine in Tanzania. Medical Anthropology 29(1):15 43. Last, Murray 1981 The Importance of Knowing about not Knowing. Social Science and Medicine 15B:387392. Latham, Gwynneth, and Michael Latham 1995 Kilimanjaro Tales: The Saga of a Medical Family in Africa. London and NY: T he Radcliffe Press. Latour, Bruno 1987 Science in Action: How to Follow Scientists and Engineers through Society. Cambridge, MA: Harvard University Press. 2005 Reassembling the Social: An Introduction to Actor Network Theory. Oxford and New York: Oxford University Press.

PAGE 437

437 Latour, Bruno and Steve Woolgar 1986[1979] Laboratory Life: The Construction of Scientific Facts. Princeton, NJ: Princeton University press. Law, John and Annemarie Mol 2008 Globalisation in Practice: On the Politics of Boi ling Pigswill. Geoforum 39:133143. Livingstone, David 1874 The Last Journals of David Livingstone in Central Africa: From 1865 to His Death. Horace Waller (ed.). 2 vols. London: John Murray. Lock, Margaret 1998 Menopause: Lessons from Anthropology. Psychosomatic Medicine 60:410419. 2002 Twice Dead: Organ Transplants and the Reinvention of Death. Berkeley: University of California Press. Lock, Margaret and Vinh Kim Nguyen 2010 An Anthropology of Biomedicine. Malden, MA: WileyBlackwell. Luga lla, Joe L.P. 1995 The Impact of Structural Adjustment Policies on Womens and Childrens Health in Tanzania. Review of African Political Economy 22( 63) :43 53. 1997 Development, Change, and Poverty in the Informal Sector during the Era of Structural Adjustments in Tanzania. Canadian Journal of African Studies 31(3) :424 451. Lugongo, Bernard 2010 Tanzania: Treasure in Crisis Meeting as Donors Cut Sh286 Billion Aid. 14 May. Dar es Salaam: The Citizen. Electronic Document, http://allafrica.com/stories/2010051402 98.html, accessed May 14, 2010. MacIntosh, Maureen, Kwadwo Mensah, Lerio Henry and Michale Rowson 2006 Aid, Restitution and International Riscal Redistribution in Health Care: Implications of Health Professionals Migration. Journal of International Development 18:75770. McNeil, Donald G. Jr. 2010 At Front Lines, AIDS War is Falling Apart. NY Times, May 9. Electronic document http://www.nytimes.com/2010/05/10/world/africa/10aids.html?partner =rss&emc=rss, accessed May 9, 2010. Mallo y, Patrick Thomas 2003 Holding [Tanganyika] by the Sindano: Networks of Medicine in Colonial Tanganyika. PhD. Diss, Department of History, UCLA. Malpas, Jeff 2003 Biomedical Topoi the Dominance of Space, the Recalcitrance of Place, and the Making of Persons. Social Science and Medicine 56:23432351.

PAGE 438

438 Martin, Helle Max 2009 Nursing Contradictions: Ideals and Improvisation in Uganda. Diemen, Netherlands: AMB Publishers. Masquelier, Adeline 2001 Behind the Dispensarys Prosperous Faade: Imagini ng the State in Rural Niger. Public Culture 13(2):267 291. Mdicins sans Frontires 2009 Punishing Success? Early Signs of a Retreat from Commitment to HIV/AIDS Care and Treatment. November. Geneva: Mdicins sans Frontires. Electronic document http:/ /www.msf.org/msfinternational/invoke.cfm?objectid=C34D1CE215C5 F00A 25AFA94BDAF99F59&component=toolkit.pressrelease&method= full_html, accessed May 9, 2010. Miller, Peter and Nikolas Rose 1990 Governing economic life. Economy and Society 19(1):131. Mitchell, Timothy 2002 Rule of Experts: Egypt, TechnoPolitics, Modernity. Berkeley and Los Angeles: University of California Pres. Mol, Annemarie 1998 Ontological Politics: A Word and Some Questions. Sociological Review 46:74 89 2002 The Body Multi ple: Ontology in Medical Practice. Durham, NC: Duke University Press. Moss, Todd J. 2007 African Development: Making Sense of the Issues and Actors. Boulder, Colorado: Lynne Rienner Mosse, David 2005a Global Governance and the Ethnography of Inter national Aid. In The Aid Effect: Giving and Governing in International Development. David Mosse and David Lewis, eds. Pp. 136. NY: Pluto Press 2005b Cultivating Development: An Ethnography of Aid Policy and Practice. NY: Palgrave Macmillan Muela, Anna H., Adiel K. Mushi and Joan M. Ribera 2000 The Paradox of the Cost and Affordability of Traditional and Government Health Services in Tanzania. Health Policy and Planning 15(3) :296 302. Mulemi, Benson 2010 Coping with Cancer and Adversity: Hospital Ethnography in Kenya. Leiden, Netherlands: African Studies Centre.

PAGE 439

439 Nationalist, The (Tanzania) 1970 Use Local Herbs in Hospitals MP. July 10, no. 1932:1. Nguyen, VinhKim 2005 Antiretroviral Globalism, Biopolitics, and Therapeutic Citizenship. In Gl obal Assemblages: Technology, Politics, and Ethics as Anthropological Problems. Aihwa Ong and Stephen J. Collier, eds. Pp. 107123. Malden, MA: Blackwell. Nkya, Ananilea and Agoro Anduru 1993 Always Serving: A Portrait of Thecla Grace Mchauru. Dar es Salaam: Publishers Association of Tanzania. Nsekela, Amon J. and Aloysius M. Nhonoli 1976 The Development of Health Services and Society in Mainland Tanzania. Dar es Salaam: East African Literature Bureau. Nyerere, Julius 1977 Introduction. I n Stirling, Leader. Tanzanian Doctor. Montreal, Canada: McGill Queens University Press. Obrist, Brigit 2006 Struggling for Health in the City: An Anthropological Inquiry of Health, Vulnerability and Resilience in Dar es Salaam, Tanzania. NY: Peter Lang. OECD (Organisation for Economic CoOperation and Development) 2008 The Paris Declaration on Aid Effectiveness and the Accra Agenda for Action. Paris: Organisation for Economic CoOperation and Development. Electronic Document, http://www.oecd.org/d ataoecd/30/63/43911948.pdf, accessed June 24, 2010. 2010 Tanzania Aid Statistics Fact Sheet. Electronic Document, http://www.oecd.org/ dataoecd/24/21/1882894.gif, accessed Sept. 2, 2010. Ogilvie, Linda, Judy E. mill, Barbara Astle, Anne Fanning and M ary Opare 2007 The Exodus of Health Professionals from SubSaharan Africa: Balancing Human Rights and Societal Needs in the Twenty First Century. Nursing Inquiry 14(2):114 24. Olivier de Sardan, JeanPierre 2001 La sage femme et le douanier. Cultures professionnelles locales et culture bureaucratique privatise en Afrique de lOuest. Autrepart 20:6173. Paul, Catherine 2005 Mission Report: Tanzanias Health SWAp: Achievements, Challenges, and Lessons Learnt. Research Group on Cooperation Instrume nts in Support of Sector Policies: GRAP SWAP. Electronic document http://www.districthealth service.com/cms/ upload/policycat_95_3073.doc, accessed August 1, 2009.

PAGE 440

440 Peligal, Rona E. 1999 Spatial Planning and Social Fluidity: The Shifting Boundaries of Ethnicity, Gender, and Class in Arusha, Tanzania, 19201967. Ph.D. thesis, Columbia University, New York. Peterson, Derek R. 2004 Creative Writing: Translation, Bookkeeping, and the Work of Imagination in Colonial Kenya. Portsmouth, NH: Heinemann. Pi gg, Stacey L. 2001 Languages of Sex and AIDS in Nepal: Notes on the Social Production of Commensurability. Cultural Anthropology 16(4) : 481541. Porter, Theodore M. 2005 Medical Quantification: Science, Regulation, and the State. In Body Counts: Medi cal Quantification in Historical and Sociological Perspective/La quantification medicale, perspectives historiques et sociologiques. Jorland, Grard, Annick Opinel, and George Weisz, eds. Pp. 394402. Montreal and Kingston: McGill Queens University Press Ranger, Terence O. 1992 Godly Medicine: The Ambiguities of Medical Mission in Southeastern Tanzania, 19001945. In The Social Basis of Health and Healing in Africa. Steven Feierman and John M. Janzen, eds. Pp. 256282. Berkeley and Los Angeles: Univer sity of California Press. Raufu, Abiodun 2002 Nigerian health Authorities Worry over Exodus of Doctors and Nurses. British Medical Journal 325, 13 July:65. Richey, Lisa Ann 2008 Population Politics and Development: From the Policies to the Clinics. NY: Palgrave MacMillan. Ridley, Robert G. 2001 Putting the Partnership into Public Private Partnerships. Bulletin of the World Health Organization 79(8) :694. Riles, Annelise 2000 The Network Inside Out. Ann Arbor: University of Michigan Press. Ro bertson, Bridget M. 1993 Angels in Africa: A Memoir of Nursing with the Colonial Service. London and NY: Radcliffe Press. Rottenburg, Richard 2009 Far Fetched Facts: A Parable of Development Aid. Transl. Allison Brown and Tom Lampert. Cambridge, MA : MIT Press. Originally published in German under title Weit hergeholte Fakten. Eine Parabel der Entwicklungshilfe. Lucius and Lucius 2002.

PAGE 441

441 Rubona, Josibert J. 2001 Routine Health Information Systems that Operate in Tanzania. In The RHINO Workshop on Is sues and Innovation in Routine Health Information in Developing Countries. Pp. 183193. Potomac, MD: The Bolger Center. Available electronically, http://www.cpc.unc.edu/ measure/publications/html/rhino2001/theme3/Tanzania _paper.pdf. Accessed Sept. 20, 2010. Rusimbi, Mary 2003 SAP for Whom? Grassroots Perspectives. In Against Neoliberalism: Gender, Democracy and Development. Chachage S. L. Chachage and Marjorie Mbilinyi, eds. Pp. 99109. Dar es Salaam, Tanzania: Tanzania Gender Networking Programme. Schaffner, Kenneth F. 2002 Assessments of Efficacy in Biomedicine: The Turn Toward Methodological Pluralism. In The Role of Complementary and Alternative Medicine: Accommodating Pluralism. Daniel Callahan, ed. Pp. 114. Washington: Georgetown Universit y Press. Scott, James C. 1998 Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press. Scheid, Volker 2002 Chinese Medicine in Contemporary China: Plurality and Synthesis. Durham: Duke University Press. Semali, Innocent A. 2003 Understanding Stakeholders Roles in Health Sector Reform Process in Tanzania: The Case of Decentralizing the Immunization Program. PhD Diss., Faculty of Natural Sciences, University of Basel. Smith, M., S. Madon, A. Anifalaje, M. Lazarro Malecela and E. 2008 Integrated Health Information Systems in Tanzania: Experience and Challenges. Electronic Journal on Information Systems in Developing Countries. Electronic document http://www.ejisdc.org/ojs2/index.php/ejisdc/article/view/395 /240, accessed January 14, 2010. Standard, The (Tanzania) 1971 Some Private Doctors are Dangerous. April 5, no. 12.581:2. 1972 The Patient Comes First. April 21, no. 12.900. Starr, Susan Leigh 1999 The Ethnography of Infrastructure. American Behavioral Scientist 43 (3) :377 91.

PAGE 442

442 Stirling, Leader 1947 Bush Doctor: Being Letters from Dr. Stirling, Tanganyika Territory. Westminster, England: Parret and Neves, Ltd. for the Universities Mission to Central Africa. 1977 Tanzanian Doctor. Montreal, Canada: McGill Queens University Press. 1987 Africa: My Surgery. Worthing, West Sussex: Churchman Publishing Ltd. 1995 Come Over and Help Us: A Doctor in Africa. Dar es Salaam, Tanzania: AMREF Tanzania Publishers. Sti rrat, R oderick L. 2000 Cultures of C onsultancy. Critique of Anthropology 20(1):3146. Street, Alice Forthcoming. Affirmative Infrastructure: Hospital Landscapes of Hope and Failure. Special issue, Hospital Heterotopias: Comparative Ethnographies of Biomedical Places. Alice Street and Simon Coleman, eds. Space and Culture. Street, Alice and Simon Coleman Forthcoming. Introduction: Real and Imagined Spaces. Special issue, Hospital Heterotopias: Comparative Ethnographies of Biomedical Places. Al ice Street and Simon Coleman, eds. Space and Culture. Sunday News (Tanzania) 1972 Hospital Matrons Encounter Numerous and Tough Complaints. April 2, no. 1006:11. 1972 ED Three Days without a Glance at my Body. June 18, no. 1017:6. 1974 That Confuse d, Careless Nurse. March 3, no. 1101:4. Sunday Times (UK) 2003 Leader Stirling: Missionary Surgeon who Became Health Minister of Tanzania and Married a Princess. April 8. Electronic Document http://www.timesonline.co.uk/tol/ comment/obituaries/article1127962.ece accessed June 14, 2010. Sturmer, Martin 1998 The Media History of Tanzania. Tanzania: Ndanda Mission Press. Electronic document http://tanzania.at/mht/The_Media_History_of_Tanzania.pdf accessed March 29, 2010. Sullivan, Noelle, Hansjrg Dilger and David Garcia 2010 Negotiating Professionalism, Economics and Moral Obligation: An A ppeal for Ethnographic Approaches to African Medical Migration. African Diaspora 3(2) :237 254. Tanganyika Development Commission 1946 A TenYear Development and Welfare Plan for Tanganyika Territory. Dar es Salaam, Tanzania: Government Printer.

PAGE 443

443 Tan zanian Ministry of Health 1992 Primary Health Care Strategy. Dar es Salaam, Tanzania: Ministry of Health. 1994a National Health Policy. Dar es Salaam, Tanzania: Ministry of Health 1994b Proposals for Health Sector Reform. Dar es Salaam, Tanzania: Ministry of Health. 1995a Strategic Health Plan 19951998. Dar es Salaam, Tanzania: Ministry of Health. 1995b National District Health Planning Guidelines, version 1.0. Dar es Salaam, Tanzania: Ministry of Health. 2003 Second Health Sector Strategic Plan (HSS P) (July 2003 June 2008). Dar es Salaam, Tanzania: Ministry of Health 2005 Guideline for Reforming Hospitals at Regional and District Levels. Dar es Salaam, Tanzania: Ministry of Health. Titmuss, Richard, Brian Abel Smith, George MacDonald, Arthur Wil liam and Christopher H. Wood 1964 The Health Service of Tanganyika: A Report to the Government. London: Pitman Medical Publishing Co. Tripp, Aili Mari 1997 Changing the Rules: The Politics of Liberalization and the Urban Informal Economy in Tanzania Berkeley, LA and London: University of California Press. Trouillot, Michel Rolph 2001 The Anthropology of the State in the Age of Globalization: Close Encounters of the Deceptive Kind. Current Anthropology 42(1):125 138. 2003 Global Transformations: Anthropology and the Modern World. NY: Palgrave Macmillan. Turner, Victor 1969 The Ritual Process: Structure and Anti Structure. Chicago: Aldine. Turshen, Meredeth 1984 The Political Ecology of Disease in Tanzania. New Brunswick, NJ: Rutgers Univ ersity Press. UNAIDS/WHO 2008 Tanzania Country Profile. http://www.who.int/globalatlas/predefinedReports/ EFS2008/short/EFSCountryProfiles2008_TZ.pdf (accessed December 29, 2009). UNICEF 2002 State of the Worlds Children 2003. NY: UNICEF. United Nations n.d. United Nations Millennium Development Goals website. http://www.un.org/ millenniumgoals/bkgd.shtml. Accessed Sept. 12, 2010.

PAGE 444

444 United States Bureau of Democracy, Human Rights and Labor 2007 International Religious Freedom Report: Tanzania. Electronic document http://www.state.gov/g/drl/rls/irf/2007/90124.htm Accessed February 22, 2011. van Amelsvoort, Vincent (compiler) 1976 Medical Anthropology in African Newspapers: An Annotated Facsimile Edition from the Third World. Oosterhout, The Netherlands: Anthropological Publications. van der Geest, Sjaak 2005 Sacraments in the Hospital: Exploring the Magic and Religion of Recovery. Anthropology and Medicine 12(2):135150. van der Geest, Sjaak and Finkler, Kaja 2004 Hospital Ethnography: Introduction. Social Science and Medicine 59:19952001. van Etten, Geert M. 1976 Rural Health Development in Tanzania: A Case Study of Medical Sociology in a Developing Country. Assen/Amsterdam, Netherlands: Van Gorcum. Vaughan, Megan 1991 Curi ng their Ills: Colonial Power and African Illness. Stanford, CA: Stanford University Press. 1994 Healing and Curing: Issues in the Social History and Anthropology of Medicine in Africa. Social History of Medicine 7(2):28395. Weck, Dr. (ed. and transl by Alison Redmayne) 1969 Hehe Medicine. Tanzania Notes and Records: The Journal of the Tanzania Society 70:2940. Originally published as Weck, Dr. 1908. Der Wahehe Arzt und seine Wissenschaft. Deutsches Kolonialblatt 1908. Pp. 10481051. Wendland, C laire L. 2004 Making Doctors in Malawi: Local Exigencies Meet Global Identities in an African Medical School. PhD Diss., Department of Anthropology. University of Massachusetts Amherst. 2009 Research, Therapy, and Bioethical Hegemony: The Controversy over Perinatal AZT Trials in Africa. African Studies Review 51(3):1 23. 2010 A Heart for the Work: Journeys through an African Medical School. Chicago and London: University of Chicago Press. Weiss, Brad 2009 Street Dreams and Hip Hop Barbershops : Global Fantasy in Urban Tanzania. Bloomington and Indianapolis: Indiana U. Press. Weisz, George 2005 From Clinical Counting to Evidence Based Medicine. In Body Counts: Medical Quantification in Historical and Sociological Perspective/La quantificati on medicale, perspectives historiques et sociologiques. Grard Jorland, Annick Opinel, and George Weisz, eds. Pp. 377393. Montreal and Kingston: McGill Queens University Press.

PAGE 445

445 White, Luise 1995 They Could Make their Victims Dull: Genders and Genres Fantasies and Cures in Colonial Southern Uganda. American Historical Review100(5) :13791402. 2000 Speaking With Vampires: Rumor and History in Colonial Africa. Berkeley: University of California Press. White, Paul 1952 Doctor of Tanganyika. London: Paternoster Press. WHO/UNICEF/UNFPA/World Bank 2010 Trends in Maternal Mortality: 19902008. Geneva: WHO Press. Whyte, Susan Reynolds 1997 Questioning Misfortune: the Pragmatics of Uncertainty in Eastern Uganda. NY: Cambridge University Press. 2002 Subjectivity and Subjunctivity: Hoping for Health in Eastern Uganda. In Postcolonial Subjectivities in Africa. R. Werbner, ed. Pp.171189. NY: Zed Books. Whyte, Susan Reynolds, Sjaak van der Geest and Anita Hardon 2002 Social Lives of Medicin es. NY: Cambridge University Press. Whyte, Susan Reynolds, Michael A. Whyte, and David Kyaddondo 2010 Health Workers Entangled: Confidentiality and Certification. In Morality, Hope and Grief: Anthropologies of AIDS in Africa. Hansjrg Dilger and Ute L uig, eds. Pp. 80101. NY: Berghahn Books. Whyte, Susan Reynolds, Michael A. Whyte, Lotte Meinert, and Betty Kyaddondo 2006 Treating AIDS: Dilemmas of Unequal Access in Uganda. In Global Pharmaceuticals: Ethics, markets, Practices. Adriana Petryna, And rew Lakoff, and Arthur Kleinman, eds. Pp. 240262. Durham: Duke University Press. Wind, Gitte 2008 Negotiated Interactive Observation: Doing Fieldwork in Hospital Settings. Anthropology and Medicine 15(2) :79 89. WHO 2006 Country Health System Fact Sheet 2006: United Republic of Tanzania. http://www.afro.who.int/index.php?option=com_docman&task=doc_download&gid=58 &Itemid=2111, accessed Aug. 12, 2010. World Bank 2008 World Development Indicators 19602008. Available at http://data.worldbank.org/ indicator. Accessed Sept. 13, 2010. Yoder, P. Stanley 1980 Health Sector Assessment, Arusha Region: Report. Manuscript prepared for Development Alternatives, Inc., Washington DC.

PAGE 446

446 Young, Allan 1982 The Anthropologies of Illness and Sickness. Annual Review of Anthropology 11:257285. 1995 The Harmony of Illusions: Inventing Post Traumatic Stress Disorder. Princeton: Princeton University Press. Young, T. Kue 1986 Socialist Development and Primary Health Care: The Case of Tanzania. Human Organi zation 45(2):128 134 Zacher, Mark W. 2007 The Transformation in Global Health Collaboration since the 1990s. In Governing Global Health: Challenge, Response, Innovation. Andrew F. Cooper, John J. Kirton and Ted Schrecker, eds. Pp. 1527. Zaman, Shaha duz 2004 Poverty and Violence, Frustration and Inventiveness: Hospital Ward Life in Bangladesh. Social Science and Medicine 59 (10) :20252046. 2005 Broken Limbs, Broken Lives: Ethnography of a Hospital Ward in Bangladesh. Amsterdam: Het Spinhuis.

PAGE 447

447 BIOGRAPHICAL SKETCH Noelle Sullivan, ne Colquhoun, was born and raised in British Columbia, Canada. She went to the University of Victoria (UVic) and in 2000 earned her bachelors degree with a double major in anthropology and history. Upon graduation, she m oved to the United States. She completed a Master of Arts in African and African American Studies from the University at Albany, SUNY in 2002. Upon completing the degree, she began the PhD program in Anthropology at the University of Florida (UF). Motivate d by the anthropologists with whom she studied at UVic, Noelle specialized in medical anthropology and African s tudies at UF. She first traveled to Tanzania, east Africa in 2004 for intensive Swahili training. While she originally intended to research HIV/ AIDS in east Africa, her visits to Tanzanian hospitals and clinics in 2004 inspired her to broaden her scope, and consider the wider health care system in the country, how it transformed under government reforms, and how this impacted the kinds of care ava ilable. She completed eleven months of fieldwork in northern Tanzania in 2008. Upon returning to Florida in 2009, Noelle began teaching in the anthropology program at UF while she worked on her dissertation. She received her PhD from the University of Florida in the spring of 2011.