Evidence, Ethos and Experiment: The Anthropology and History of Medical Research in Africa 9780857450937

Table of contents :
Contents
List of Figures
Introduction Studying Trial Communities: Anthropological and Historical Inquiries into Ethos, Politics and Economy of Medical Research in Africa
Engagements
Chapter 1 Writing Knowledge and Acknowledgement Possibilities in Medical Research
Chapter 2 Can One Rely on Knowledge?
Chapter 3 Being ‘with the Medical Research Council’: Infant Care and the Social Meanings of Cohort Membership in Gambia’s Plural Therapeutic Landscapes
Chapter 4 Contextualizing Ethics: Or, the Morality of Knowledge Production in Ethnographic Fieldwork on ‘the Unspeakable’
Chapter 5 Testing a New Drug for Leprosy: Clofazimine and Its Precursors in Ireland and Nigeria, 1944–1966
Chapter 6 Elucidating Ethics in Practice: Focus on Accountability
Evidence
Chapter 7 When Physicians Meet: Local Medical Knowledge and Global Public Goods
Chapter 8 The Plausibility Design, Quasi-experiments and Real-world Research: A Case Study of Antimalarial Combination Treatment in Tanzania
Chapter 9 Remember Bambali: Evidence, Ethics and the Co-production of Truth
Chapter 10 Foetuses, Facts and Frictions: Insights from Ultrasound Research in Tanzania
Chapter 11 Healers and Scientists: The Epistemological Politics of Research about Medicinal Plants in Tanzania or ‘Moving Away from Traditional Medicine"
Chapter 12 Parasite Lost: Remembering Modern Times with Kenyan Government Medical Scientists
Chapter 13 Is the Sharia of the Doctors Killing the People? A Local Debate on Ethics and the Control of HIV/AIDS in a Rural Area in Kenya
Politics
Chapter 14 The Historical Interface between the State and Medical Science in Africa: Kenya’s Case
Chapter 15 The Intimate Rules of the French Coopération: Morality, Race and the Postcolonial Division of Scientific Work at the Pasteur Institute of Cameroon
Chapter 16 The Mosquito Taken at the Beerhall: Malaria Research and Control on Zambia’s Copperbelt
Chapter 17 Trial Communities: HIV and Therapeutic Citizenship in West Africa
Chapter 18 Differences in Medicine, Differences in Ethics: Or, When is It Research and When is It Kidnapping or is That Even the Right Question?
Notes on Contributors
Index

Citation preview

Evidence, Ethos and Experiment

Evidence, Ethos and Experiment The Anthropology and History of Medical Research in Africa

Edited by P. Wenzel Geissler and Catherine Molyneux

Berghahn Books New York • Oxford

First published in 2011 by Berghahn Books www.berghahnbooks.com ©2011, 2017 P. Wenzel Geissler and Catherine Molyneux First paperback edition published in 2017 All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher. The printing of this volume has been made possible by the generous support of the Thyssen Foundation and the Heidelberg Center for American Studies. Library of Congress Cataloging-in-Publication Data Evidence, ethos and experiment: the anthropology and history of medical research in Africa / edited by P. Wenzel Geissler and Catherine Molyneux. p. cm. Includes bibliographical references and index. ISBN 978-0-85745-092-0 (hardback : alk. paper) – ISBN 978-1-78533-500-6 (paperback) – ISBN 978-0-85745-093-7 (ebook) 1. Medicine–Research–Moral and ethical aspects–Africa–Congresses. 2. Human experimentation in medicine–Moral and ethical aspects–Africa–Congresses. 3. Medical ethics–Congresses. I. Geissler, Wenzel. II. Molyneux, Catherine. R725.5E93 2011 615.5072'4096--dc23 2011019540 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library

ISBN 978-0-85745-092-0 (hardback) ISBN 978-1-78533-500-6 (paperback) ISBN 978-0-85745-093-7 (eBook)

Contents

List of Figures Introduction Studying Trial Communities: Anthropological and Historical Inquiries into Ethos, Politics and Economy of Medical Research in Africa P. Wenzel Geissler Engagements 1. Writing Knowledge and Acknowledgement: Possibilities in Medical Research Susan Reynolds Whyte

viii

1

29

2.

Can One Rely on Knowledge? Marilyn Strathern

57

3.

Being 'with Medical Research Council': Infant Care and the Social Meanings of Cohort Membership in Gambia's Plural Therapeutic Landscapes Melissa Leach and James Fairhead

77

4.

Contextualizing Ethics: Or, the Morality of Knowledge Production in Ethnographic Fieldwork on ‘the Unspeakable’ Hansjörg Dilger

99

5.

Testing a New Drug for Leprosy: Clofazimine and Its Precursors in Ireland and Nigeria, 1944–1966 John Manton

125

Contents

6.

Elucidating Ethics in Practice: Focus on Accountability George Ulrich

Evidence 7. When Physicians Meet: Local Medical Knowledge and Global Public Goods Steven Feierman

145

171

8.

The Plausibility Design, Quasi-experiments, and Real-world Research: A Case Study of Antimalarial Combination Treatment in Tanzania S. Patrick Kachur

197

9.

Remember Bambali: Evidence, Ethics and the Co-production of Truth Ann Kelly

229

10. Foetuses, Facts and Frictions: Insights from Ultrasound Research in Tanzania Babette Müller-Rockstroh

245

11. Healers and Scientists: The Epistemological Politics of Research about Medicinal Plants in Tanzania or ‘Moving Away from Traditional Medicine’ Stacey A. Langwick

263

12. Parasite Lost: Remembering Modern Times with Kenyan Government Medical Scientists P. Wenzel Geissler

297

13. Is the Sharia of the Doctors Killing the People? A Local Debate on Ethics and the Control of HIV/AIDS in a Rural Area in Kenya Suzette Heald

333

vi

Contents

Politics 14. The Historical Interface between the State and Medical Science in Africa: Kenya’s Case Kenneth S. Ombongi

353

15. The Intimate Rules of the French Coopération: Morality, Race and the Postcolonial Division of Scientific Work at the Pasteur Institute of Cameroon Guillaume Lachenal

373

16. The Mosquito Taken at the Beerhall: Malaria Research and Control on Zambia’s Copperbelt Lyn Schumaker

403

17. Trial Communities: HIV and Therapeutic Citizenship in West Africa Vinh-Kim Nguyen

429

18. Differences in Medicine, Differences in Ethics: Or, When is It Research and When is It Kidnapping or is That Even the Right Question? Luise White

445

Notes on Contributors

463

Index

473

vii

List of Figures

1.1

Questionnaires from Dr Frank's Project (author’s photograph). 31

1.2

The Author Writing This Paper in the Project Office at Makerere University (photograph by Michael White).

38

1.3

Collecting Knowledge, Acknowledging the Other (author’s photograph).

48

8.1

Map of IMPACT Tz Study Districts in Southern Tanzania.

204

8.2

A Strategic Framework for Considering Interventions to Optimise Antimalarial Combination Treatment in Tanzania (adapted from Marsh, Kachur and Mehra, 2002).

205

8.3

Participatory Methods Were Used to Conduct Formative Research to Develop Information, Education, and Communication Materials in Support of ACT Implementation in Rufiji District.

214

8.4

Colour-Coded Dispensing Envelopes Were Developed for Deployment of ACT.

215

12.1

The Staff of DVBD Kisumu, Around 1962 (photographer unknown).

302

12.2

Technologists Bench Training in a DVBD Laboratory, 1960s (photographer unknown).

303

12.3

Officer’s Tent with Kisumu Head of DVBD, 1960s (photographer unknown).

306

viii

List of Figures

12.4

Mobile Laboratory, 1950s (from ‘DDT vs. Malaria’) (photographer unknown).

308

12.5

Portrait with ‘The Mosquitoes of South Sudan’ (author’s photograph).

311

12.6

Portrait of ‘Yellow Fever Inspector’, 1941 (photographer unknown).

312

12.7

Portrait with 1982 Uniform and 1970s Land Rover (author’s photograph).

313

12.8

Plate with Mixed Photographs of DVBD Colleagues, 1960s and 1970s (photographer unknown).

314

12.9

Urban Housing in Shauri Moyo Estate, Kisumu Where DVBD Staff Lived With Their Families in the 1960s (author's photograph).

315

12.10 Younger DVBD Staff Members, Mid 1970s (photographer unknown).

320

12.11 Organigram of the Entire DVBD Displayed at the Director’s Office, DVBD HQ, Nairobi (author’s photograph).

320

12.12 Group Picture of DVBD Staff Families on the Occasion of the 1963 Christmas Party (photographer unknown).

321

12.13 Portrait of Mr O. 2005 and 1945 (author’s photograph).

324

15.1

Cameroonisation as Technical Emancipation. A Laboratory at the Centre de Recherches Médicales du Cameroun (Future Pasteur Institute) in 1957.

376

15.2

Professor Etoundi, Head of the Centre Pasteur/IMPM, Interviewed for the National Daily Cameroon Tribune (28 December 1978).

380

ix

List of Figures

15.3

‘We Were Wearing the Coat of a Mechanic’. A Laboratory Technician in a Dark-Green Coloured Coat, c1986.

391

15.4

Unskilled Staff Wearing Dark Blue Uniforms at the Centre Pasteur du Cameroun, c1986.

392

16.1

Garden City – Roan Mine. (no date given) (Plate 31, in Malcolm Watson, African Highway; with permission from John R. Murray).

413

16.2

European House, 1935. (Plate 32, in Malcolm Watson, African Highway; with permission from John R. Murray).

415

17.1

Slogans at a ‘Community Meal’ of Positive Women in Ouagadougou (author’s photograph).

435

x

Introduction

Studying Trial Communities: Anthropological and Historical Inquiries into Ethos, Politics and Economy of Medical Research in Africa P. Wenzel Geissler

Overseas Medical Research This book is about medical research carried out in Africa, by African institutions and their collaborators from Europe and the USA. It is thus about what used to be called ‘overseas’ medical research, a term which – unlike more recent terms such as ‘transnational’ or ‘collaborative’ – recalls its imperial origins as well as the assymetrical topography of power and resources it still involves. Overseas research is shaped by its geographical and political-economic frames, as well as by colonial history and by the process of nation building, and decay, that marked the postcolonial era (or, as Ombongi, below, distinguishes, the ‘postcolonial’ and the ‘post-postcolonial’). This is why the authors of this volume, participants of the conference ‘Studying Trial Communities’, held in 2005 at the Kenyan Medical Research Institute (KEMRI) Centre for Geographical Medicine in Kilifi, Kenya, include historians among the majority of anthropologists, and why many of the anthropologists here draw upon historiography or historical sources for the purpose of their ethnography. Medical research in Africa is an area intensely shaped by history, and the fact that it often is oblivious to its own origins and genesis makes it particularly important that we combine ethnographic and historical-archaeological investigations. 1

Evidence, Ethos and Experiment

The chapters below focus on contemporary medical research endeavours and, to a lesser extent, their postcolonial prehistory. They cover a range of African countries, and diverse types of medical research: clinical studies, drug development and randomised controlled trials, entomological surveys and vector control, ethnobotany and phytopharmacology, even medical anthropology. What they have in common is a commitment to understanding how medical research is shaped in the interactions — set within stark political and economic disparities — between global scientists and their institutions, national and transnational forms of government, and people who contribute time and effort, and often also bodily substance, to research projects, either as temporary employed staff, or as study subjects without direct, formal remuneration. The contributors ask why, and under which circumstances, scientific medical work takes on particular forms, and how the people and institutions involved determine what is right and wrong, and which direction to take. Underlying these inquiries is, for many of the authors, the question of how scientific investigations – as well as the public good that medical science makes possible – could be realised in a more democratic and equitable manner. Thus, their scholarly occupation with science in Africa is intertwined with reflections on the politics and ethics of medicine, in the ‘overseas’ situation of unequal scientific and technical capacity and great disparities in power and wealth.

The Ethos of Medical Science This book could be said to be about the ethics of research, but it does not limit itself to the domain of discourse that recently has been cast as ‘research ethics’ or ‘bioethics’ in the regulatory sense of guidelines and principles of ‘good clinical practice’ (see e.g. CIOMS 2002), and which has given rise to some fruitful debate in public health and attendant social sciences. This literature blossomed in particular during the past decade, stimulated, among other sources, by the human rights discourse that evolved around HIV/AIDS, and, more specifically, by the debates provoked by the 1997 discussion among scientists and ethicists on appropriate ‘standards of care’ in African HIV research (see Angell 1997; Lurie and Wolf 1997). Simplifying a rich debate for the purpose of positioning this volume, we can discern a continuum across this literature, ranging from contributions that aim to produce better regulatory frameworks or to implement rules more systematically (e.g. Leach et al. 1999; Emanuel et al. 2004), to social research that enriches the debate through the use of sociological and anthropological method (e.g. Molyneux et al. 2004, 2005a,b; Fairhead et al. 2006; Parker et al. 2008). 2

Studying Trial Communities

The authors of the latter kind of texts often critique and qualify the terms set out in standard bioethics guidelines, or resituate research ethics in their political-economic context – thereby critiquing their inherent ‘anti-political’ effects (e.g. Benatar and Singer 2000). By contrast, the former – social science that remains within the delineations of the existing bioethics discourse – tends to reify these limitations of regulatory bioethics, excluding, at times explicitly, questions of political and economic inequality and interest from the purview of ethical ‘worries’ (see Emanuel et al. 2005). The contribution of this literature to critical debate about medical research and scientific knowledge, and thus to the creation of egalitarian and democratic science, is therefore limited by an implicit acceptance of the status quo. Setting themselves in particular apart from this normative version of the scholarly debate on medical research ethics, the anthropologist or historians below situate themselves outside the frames of existing normative ‘bioethics’, and many critique these frames or their epistemological and political premises (see e.g. Dilger, Heald, Strathern, White). In as far as they do directly refer to research ethics, they prise open the taken for granted closures of existing ethical discourses, in order to create the possibility of thinking up alternatives. In other words, the chapters collected in this volume wish to extend the critical and analytic end of the spectrum delineated above.1 While most contributors below acknowledge the importance of legal instruments to regulate medical research, the ethical impulse behind their chapters is different from that which drives bioethics guidlines in the sense embodied by ‘Good Clinical Practice’.2 For the sake of distinguishing it from ‘ethics’ in the restricted sense that is now often taken for granted in the context of medical research – emphasising the protection of individual rights at the expense of a wider societal project including the pursuit of justice – one could refer to this different concern as the ‘ethos’ of medical research, that is, the visions and projects that orientate and direct the discourses and practices of different actors and groups, in different places, situations and periods.3

Ethos in Time Scientific ethos, or articulations of ethos in relation to medical science, is shaped by historical and political-economic circumstances, and revealed by close attention to how different actors, in different localities and times, produce scientific evidence in particular ways, how they express motivations and aims, take decisions, identify and solve problems, chose ways forward and evaluate past actions. Ethos can be articulated on diverse 3

Evidence, Ethos and Experiment

levels of scale – including, for example, commitment to thoroughness and pride in technical inventiveness, faith in progress and enlightenment, commitment to freedom, welfare, equality or democracy – and linked to different wider societal projects – such as the nation or a global commonwealth, the market or the welfare-state. In contrast to narrow definitions of ‘ethics’ in terms of guidelines and rules, the broader study of ethos emphasises the problem of direction over that of particular choice, the openness of intention and anticipation over the closure of a right or wrong action. Asking for the ethos of science is searching for the projects implied in it, the imagined futures and pasts, and thus it is as much about temporality as it is about morality (for the centrality of time for diverse articulations of ethos, see also Badiou 2002). Articulations of ethos shift over time, and not infrequently the broad frames of one historically situated ethos serve as shared ground, upon which struggles about diverging interest can be carried out. A key example of this, which several of the chapters below attend to (e.g. Ombongi, Schumaker, Geissler), is the specifically modern, mid-twentieth century ethos which took modernisation, progress and development for granted, and which shaped science and drew upon it to expound visions of society and government. The outlines of this particular ethos may be particularly visible to us, today, for two reasons. It has been analysed, critiqued and deconstructed, most fruitfully by the Frankfurt School and later by the Foucauldian tradition, giving us a sense of distance from it; and yet, as we inhabit an increasingly uncomfortable and destructive sort of after-modernity, we cannot help but rediscover some of the attractiveness of the outdated hopes of our modern forebears (whom we now know have never been modern, leaving us with a task at hand). Importantly, this modern ethos served also in the African medical context as a shared frame of reference for politically and economically radically opposed interests: colonial medical administrators could draw upon it to justify, for example, racial segregation, while anti-colonial freedom fighters could demand medical equality and redistribution of the fruits of progress. Claims and counter claims could thus be made with reference to this one overarching ethos. Choices could be contested, even fought over, within this frame, which did not provide simple moral answers. This potential inclusiveness, and ambivalence of ethos, which allows for diverging interests, conflict and dialectics, is an important difference to ‘ethics’ as conceived of in regulatory research ethics.

4

Studying Trial Communities

Ethos and Politics In a similar way, a contemporary ethos based on individual freedom and rights and the value of life can be drawn upon to defend very different interests, for example those of the pharmaceutical industry claiming intellectual property rights, and those of treatment action campaigns demanding equal rights in HIV treatment; those of industry-sponsored ‘patient advocacy’ groups demanding high-cost drugs for unproven improvements of patients’ lives, and those of ‘right to die’ activists, who oppose the dictate of life-extending medicinal technology (see Ong and Collier 2005). Thus, within the space of an overarching ethos different interests can be both disputed and brought together. Indeed, as most of the authors below will agree, contestation, consensus-making and contradictions are inevitable dimensions of the historical process, and must be named and articulated in the political engagement of science. At the same time, different articulations of ethos can coexist, compete and interrelate within one historical situation. For example, in the field of medical science and healthcare, individual rights provide for a very different ethos from that of older government public health; yet, both coexist around many of today’s key health issues. If we, for example, want to engage with the ethical challenges that HIV research and intervention pose, we need to understand how these different kinds of ethos overlap, mix and conflict across this field; we must consider the different political and societal projects entailed by, respectively, emphasising individual rights to confidentiality and voluntariness, or the responsibility of government public health to minimise suffering and maximise citizen’s welfare. Commitments in the sense of ethos thus go beyond rule-set ‘ethics principles’. The latter distinguish right and wrong, an inside and an outside, divided by a moral line. Such an apparently clear moral separation obscures the more ambiguous political and economic struggles that shape medical science, and the society that it exists in and brings about. Discussions about the ‘ethos’ of science include interest and conflict into the purview of ethical reflection rather than excluding them by the sanitising morality of principles. Ethical action can thus be anchored again from where it has been unmoored: if science is about truth and value – with all the term’s moral-cum-economic ambiguity (see e.g. Kelly and Geissler 2011) – these are inseparable from power and resources. Whether science harbours the possibility of change – the improvement of knowledge, lives and societal order – depends crucially upon whether and how the linkages between 5

Evidence, Ethos and Experiment

ethics, epistemology and political economy are rendered visible, negotiated and contested. Struggles about ethos are thus linked to questions of truth as well as to matters of interest.

Trial Communities Anthropologists are usually called to work on medical research with a clear remit: to study ‘the community’, that is what medical research protocols define as ‘study populations’ and ‘participants’ or ‘volunteers’. Their task is, then, to describe certain characteristics of this group, such as ‘preparedness’ and ‘response’ to a clinical trial, or ‘cultural’ concepts and behaviours of significance to the trial or intervention. The aim of the anthropologists assembled in this volume, and their historian colleagues, is different. Their unit of analysis is, in diverse ways, not a particular, delimited group, but sections of a wider network that is constituted not through attachment to a place or ‘culture’, but by the work of collaborative medical research.4 When we met at the conference in Kilifi, Kenya, whence this book derives, we referred to this network of actors engaged in medical research – for some of us including non-human actants – as the ‘trial community’. This definition of the subject has two implications: it opens the frame to include everyone concerned, and it shifts focus from the concerns arising from a particular social grouping to those arising from the relations between points, nodes and groups in the network. This allows discerning new collectives and solidarities – such as trial volunteers, data collectors and entry clerks and other workers in scientific production sites – and new connections, distinctions and separations, for example between Zambian and US American doctors, or between medical anthropologists and medical ethics. Finally, a focus on the production of associations destabilizes taken for granted boundaries – such as the line between ‘community’ and ‘researcher’, the difference between ‘cultures’, between scientific and traditional medicine and so on.5 Since we met in Kilifi in 2005 to study ‘trial communities’, which we then felt were an exciting, largely unexplored field, several important works have been published, which in diverse ways apply anthropological knowledge to overseas medical research, if not all of them in Africa, and which inspire many of the authors below in their ongoing work (e.g. Biehl 2005; Nguyen 2005; Petersen and Folayan 2005; Petryna 2006, 2009; Rajan 2006; Cooper 2008). The fact that these important books speak a somewhat different language from that of many of the authors below is not only due to the editor’s 6

Studying Trial Communities

tardiness (most chapters were written before these works became available to the authors) but also due to slightly different starting points. Most of these recent publications on clinical research aim to shift the location of anthropology to levels of scale that explode the notion of locality, as discussed previously, for example by Gupta and Ferguson (1997); they take a principal interest in what Ong and Collier described as ‘global assemblages’ (2005), working through the global construction, and transformation, of scientific production. The contributors to this volume come at similar issues and concerns from a slightly different angle, one perhaps grounded more in the empirical, ethnographic British-European tradition of social anthropology, starting out from concrete social realities – which not only, but also and importantly, consist of concrete, localised social engagements and experiences. Moreover, these recently published works share an interest in the political economy of overseas research – which is what makes them so inspiring against the backdrop of less clearly political forms of medical anthropology and ethics. Partly therefore, they focus mostly on commercial, industrydriven transnational medical research or on the effects of private, capital interest in bioscience (e.g. Petryna 2005, 2009; Rajan 2006 Cooper 2008), and most of them take a particular interest in clinical trials – as the most advanced and industry-endorsed form of medical research – and less in the wider field of public health research. The papers in this volume do not pursue such a coherent agenda. In the open spirit of the Kilifi meeting, they are more diverse and eclectic, exploring different possibilities of making medical research and medical knowledge in Africa the centre of anthropological attention. Moreover, many of the authors below focus their interest mostly on medical research in the public domain, funded by government or bilateral arrangements and conducted by government or parastatal organisations. Although publicly funded and governed medical and public health research is of course increasingly affected by ‘partnerships’ with private enterprise and charity, and by reformulations of academic intellectual property rights in the wider neoliberal context, public scientific institutions and the research they conduct pose slightly different challenges from the (important) critique of commercial, openly profit-oriented research, and might also help us to discern diverse directions and intentionalities in the conduct of science today, as well as discover alternatives to the neoliberalisation of science that Cooper (2008) and others have successfully explored.

7

Evidence, Ethos and Experiment

Critical Ethnographies Underneath the diversity of geographical areas, scientific domains and historical periods, and different political viewpoints and disciplines, the papers assembled here share a commitment to ethnography in the sense of tracing relations and separations that shape social space. For most of us, ethnography implies both studying social relations across domains and levels of scale, and beyond locality, and understanding people’s, including scientists, lives; both fascination with ‘global assemblages’ and attention to intimate social encounters and engagements. Both are needed to question formations and habits that have become commonsensical, to make visible structures and processes that are obscured or ignored, and to contribute to the critical analysis of science and society. Most of the texts in this book could be said to be ‘critical’ studies of medical research and its ethos, in the sense not of ‘anti-science’ – all contributors are committed to scientific medical research and progress, and none has an issue with science as such – but of allowing us to see scientific research in a different way, and positioning science within relationships that at times are overlooked, at other times downplayed and ignored. First, attention is fixed within the material realities, both of the research work and of the lives of people involved in research, including political and economic inequality and deprivation. Second, authors attend to the institutional relationships beyond the purview of research itself, within given localities such as between government healthcare system, pharmaceutical industry and medical research – and across levels of scale, between national government and transnational institutions and corporations. Analysing such networks means tracing visible as well as unseen connections, and exploring the aggregations and separations produced by social, spatial and epistemological categories. As such the critical study of medical research aims to prise apart the pre-existing and newly formed social relations – such as the ‘community’ that performs much work in contemporary understandings of research ethics – and to problematise separations such as that between researchers and research participants, which projects a problematic epistemological distinction onto social space and overlays, sometimes obscures, other social distinctions. Last but not least, contributors aim to reveal alternative solidarities, material similarities, overlapping interests and tensions attendant to these groupings – such as, relating to the previous example, the material similarities and overlapping interests among overseas research participants and some categories of technical re8

Studying Trial Communities

search staff, and between different groups of scientists within transnational scientific collaborations. In other words, the social studies collected in this volume are not positioning themselves opposite medical science, aiming to cast doubt over or relativise its value and validity. Most contributors would indeed defend the superiority of a scientific approach to medicine, and the achievements of its tools, like randomised controlled trials, and its products, like antibiotics and childhood vaccines. And neither do most of the contributors take the side of a particular group in overseas research – such as the research participants, or the underprivileged community – vis-à-vis the researchers and their institutions, as certain older anthropological solidarities would call for by default. Our aim, then, is not so much to ‘take sides’, but to make sides: to discern contradictions and separations, bonds and cohesions, that need to be debated in order to produce better science and better public medicine. The overarching deontological question is: where do we want science to take us? What futures are we aiming for through science? As such, ‘critique’ means here a search for openings, not definite positioning and closure (see Geissler 2011).

Overview of the Book The chapters below, which were discussed at the 2005 Kilifi conference, were grouped into three sections: engagements, evidence and politics. Since all of them deal with relations, epistemology and power, these headings serve not so much as compartments, but as three intersecting lines of inquiry.

Engagements The chapters of the first section took the conference theme most literally, focusing on the relations that make up medical research, and some of them have left a particular imprint on other, subsequent, papers in this collection. Whyte’s exploration of a Ugandan epidemiologist’s research, and of the role of paper in it, sets up a distinction between epistemological and ontological concerns; between knowledge and acknowledgement, ‘knowing about’ and ‘knowing’ people. Whyte pursues the relational possibilities of scientific work by showing that paper, apart from its capacity to contain information, can also embody substantial relations. While contrasting objectified knowledge and ‘being together’, she shows that both inhere to the same practices and may be pursued by the same people within the same 9

Evidence, Ethos and Experiment

work; all ‘situated social actors’ are motivated by both ways of knowing and by both capacities of the practice of writing knowledge. Thus, Whyte implies, we should not attribute epistemological motives to scientists and relational ones to study subjects. All members of trial communities learn new abstractions and gain understandings, and all must get to know others in the process of research work; and knowing and relating changes those involved. In conclusion, Whyte calls for ethnographic attention to the interfaces where making knowledge creates relations, and knowledge relies upon relational practices. While this is clear advice to ethnographers of research, the intertwining of knowing people and making knowledge that Whyte describes should also have implications for scientists conducting research: instead of the widespread tendency to exclude or silence the relational dimension of knowledge-making, for example through bureaucratic apparatuses and regulatory standards, it should be taken seriously, not only as a complementary foundation of ethics, but also as a contribution to science itself. Approaching concerns with epistemology and ontology from a different angle, Strathern takes issue with the notion that the engagement between researchers and researched is predicated upon divergent, respectively scientific or local, ‘perspectives’ (onto one reality) that have to be aligned through ‘information’ and ‘communication’, in order to achieve effects – namely scientific results. Such ‘perspectivalism’ might label the ‘non-scientific perspective’ as ignorance (to be corrected through information), or – more fashionably – as local ‘cultural’ knowledge, which scientists should understand and appreciate for the sake of better communication. Despite the different degree of political correctness in these two renderings of the ‘local’ or ‘lay perspective’, the assumption is the same: the other has a different way of seeing; shifting perspective – in the first case ‘theirs’ in the second case ‘ours’ – is proposed as the solution. Both approaches focus on epistemology. Instead, Strathern suggests to consider the cultural, ‘Euro-American’, nature of perspectival knowledge itself: ‘perspectives’, and the concomitant notion of ‘context’ (supposedly shaping perspectives) is for her not a solution to the problem of difference, but ‘part of the problem’. To make this point, Strathern examines ethnographic cases in which engagements among humans and the world are not conceived of as epistemic exercises directed at one substantive reality, but as substantial relations producing diverse forms of being – the dispute about ownership and belonging of museal human remains is a case in point. Instead of knowledge and information, Strathern suggests to ‘rely on relations’, notably of bodies. The 10

Studying Trial Communities

social relations around field research, which many of the other papers in the book explore, are thus not peripheral, secondary – not ‘the social context of data’ – but substantial, central to knowledge generation. As witnessed by my summary, this argument is easily reduced to the very pattern of opposite ‘world-views’ that it critiques: epistemological scientists versus ontological natives. The ethnographic labour consists in discerning the mutual articulation of these modes in the practices of research; here, I think Whyte’s material above, and many of the chapters below can be of use. Leach and Fairhead’s exploration of ‘being with’ research in The Gambia deals more directly with the problems of knowing and being, or rather with the intertwining, sometimes collision, between the epistemological concerns of scientific knowledge production and the material struggles and conflicts of the people engaged in research. They show that the uncertain world in which Gambians live is partly specific to local social forms and cultural understandings, as well as, in an important sense, an outcome of global economic processes. Therefore, the balance of benefit and danger that parents consider when consenting to clinical trial participation is not a matter of free choice; larger frames of power and value are at work, and local actors are aware of, and part of, these. Leach and Fairhead conclude that ethical assessments of medical research must consider the context of healthcare delivery and health seeking, and draw our interest to the separations between matters of health, in its social breadth and open-endedness, and matters of life, in the sense of trial participation and mere survival; they consider how these separations are produced and negotiated around overseas research and its ethics (see also Molyneux et al. 2005a on this issue). Like Leach and Fairhead, Dilger draws on the notion of context to reflect about the methodology and ethics of his own medical-anthropological research on AIDS in Tanzania. After ‘ethical clearance’ by Tanzanian authorities, which assessed and re-shaped his study according to medical ethics standards, he went to different field sites and found that ethics, rather than being universal, as the medical ethics committee had assumed, are ‘contextual’, shaped by local ‘politics of speaking’, ‘moral constellations’ and ‘epistemologies of illness’. Dilger concludes that the ethics of medical-anthropological research ought to be guided less by medical ethics rules, but by continuous reflection about the research process, including reflexion on the fieldworker’s own position in the field. The latter point gains, he argues, particular significance where the situation under study entails extreme forms of suffering and loss – as in his own case of anthropological research on AIDS. Is ethnography in such circumstances a mere ‘bourgeois luxury’ 11

Evidence, Ethos and Experiment

or could it, as Dilger suggests, help to make better, in his words ‘locally adapted’, public health interventions, or could it even help to prise open and thus unsettle the wider structures that perpetuate the unequal distribution of suffering? The social relations that are the foundation of his ethnographic knowledge also ought to be the basis of his ethical commitment. Manton’s historical inquiry moves us into an age before ‘research ethics’ – although we can sense strong and not always aligned modes of ethos at work in his historical study of pharmacological research by mission hospitals, nationalist governments and travelling scientists and global companies in mid-century Nigeria. He examines relations and transitions across global inter-personal and institutional networks, in view of tracing different interests that went into a new drug, as well as of charting relations between local ‘field research’ and global realms of science and experimentation. Though his analysis operates on a less localised level than, for instance, Whyte’s, Manton shows how universal standards are shaped by local alliances and shape local experience – an observation that might as well be applied to research ethics as to, in his case, laboratory standards. He reveals how technical problems and tensions with people and things, and the wide and often surprising circulation of research between local instances and global negotiations, shape scientific outcomes, and even the disease in question, leprosy, itself. Just as the drug on trial – a red powder which apparently dispersed easily and stuck to anybody it got in touch with – the research that went into its clinical testing, Manton shows, produced multiple relations and transformed not only bodies, but also personal and institutional lives across continents. We end this section with Ulrich’s reflection, as an anthropologist and philosopher, upon the ongoing bioethics controversies, which oppose the advocates of global bioethics regulations to those who denounce current research ethics as hollowed-out versions of ethical reflection, or as external standards enforced by central institutions in rich countries – imperialist variations on global ‘audit cultures’. Some of these critics of formal rules propose instead to place ethics in the realm of the private, in individual decisions and morality. Ulrich shows how the latter assault against formal bioethics tends to reiterate the individualizing and privatizing tendency inherent to certain current interpretations of bioethics.6 In contrast to such privatised ethics, Ulrich insists that medical science requires ‘public ethical legitimation’. The ‘public ethics’ that he – based on Kant and on Habermas – proposes ‘positions all social actors equally and accords an equal say in the pursuit of judgement of what is right’. What is thus needed 12

Studying Trial Communities

is ‘situated accountability’ (as opposed to ‘accountancy standards’): spaces in which to account for or challenge the ethics of research. This renewed emphasis on ‘the public’ raises the question of where to locate such a ‘communicative, reciprocal’ ethics in today’s collaborative medical research settings – situations equally far removed from the discursive equality of the agora of the ancient polis, and from Habermas’ 1970s western Germany. In the old contract of ‘public health’, medical science was assumed to be directed, controlled and utilised by the nation-state as the legitimate representation of its citizens, for the benefit and welfare of the latter (see e.g. Titmus 1971). While this contractual relation has world-wide been cast into doubt by the scholarly critique of biopolitics of the 1980s, as well as by the realities of neoliberal assaults on national welfare institutions, in Africa it has been particularly brutally severed by the gradual criminalisation of the state and the hollowing of its politics of representation, and the destruction of much of its its public sector, accelerated by external economic and political pressures. Although there certainly still is a large public sector in many African countries – notably in health care and education, urban administration and the legal system – its function is at least ambiguous: both evoking projects and memories of a national collective, and embodying its destruction and absence. The primary ethical challenge today would thus be to circumscribe the lost ‘public’ to restore and safeguard the condition of a public ethics. To further this, Ulrich proposes that anthropologists explore potential ‘sites of accountability’, ranging from national and international institutions to interactions between groups and individuals involved in research, and that historians trace the genesis and demise of the ‘public’, in view of restoring this achievement of the Enlightenment. The ideal of ‘public ethics’ draws our attention to the material constitution and dissolution of the public, and the demise of the nation-state and democratic representation, and to the surge of ‘bioethics’ that has come to fill some of the spaces vacated by the democratic public sphere. In response to the present weak and undemocratic public discourse and to ‘forge reciprocal communicative contexts’, the position of state and public academic institutions in research should, according to Ulrich, be reinforced, and the ‘qualified opposition with which researchers ... are met’ by people and institutions in developing countries strengthened.7 We chose to conclude this section with Ulrich’s chapter because his reflections serve as a valuable antidote to the misunderstanding to which anthropologists are prone: to place our hope in ‘relations’. While these of 13

Evidence, Ethos and Experiment

course are our analytical target, they carry no moral value and the implication of immediacy, and its assumed moral value, that the term can carry, offers no solution to the problem of the democratic and egalitarian constitution of overseas medical research.8

Evidence The book’s second section focuses on how evidence is produced and negotiated in medical research. Participating in preparations for collaborative research in Ghana and Zambia, Feierman notices different conceptions of ‘evidence’ among clinicians working, respectively, in African government and US American university hospitals. These differences, he shows, are shaped not by ‘culture’ but by the material conditions under which doctors can put evidence into effect. Faced with the ‘normal emergency’ of government healthcare, the African doctors look for evidence fitted to the specific problems faced by their hospitals at a given moment in time – for example how to make clinical judgements without optimal diagnostics. Such limited, local, make-do evidence is problematic for the US American counterparts bent on contributing to academic research and dependent upon competitive funding. They propose instead to link global standards and local situations by research on ‘low-tech’ practices independent from latest technical devices and drugs, a proposition that will be discussed also by Ombongi’s historical paper, below. Interestingly, although Feierman does not elaborate this point, such clinical trials of ‘adapted’, low-cost interventions must adhere to the highest medical standards in order to satisfy US regulations. Thus, while the African doctors might prefer to retain the aim of state-of-art global medicine as a frame, while making do with local, temporary shortfalls in research practice, their American colleagues want to test localised forms of medicine while applying global medical standards in the trial regime. The latter approach means shifting the frame of evidence, creating another kind of universal, standardised medicine, rather than the instable, localised improvisations within one universal frame proposed by the African doctors. The former approach generalizes differences in resources and capacity: poor technologies for poor places.9 The African colleagues are not content with this because they remain committed to aspiring towards global standards, and because they do not regard local as static. For them, the lack of capacity in their hospitals constitutes not a given, but a ‘not yet’, or a ‘no longer’, with respect to the scope of medical science and the mandate of a national teaching hospital. Research on low-tech medicine means for them relinquishing the ideal of global universal science.10 14

Studying Trial Communities

Feierman concludes that African doctors – negotiating scientific aspiration, clinical practice and daily economic survival – should be critical mediators of collaborative research. This raises the question of who sets the research agenda as one of the ethical cum political-economic dimensions of collaboration that requires anthropological scrutiny. The demand, by doctors working in Africa, of medical research that inserts itself into their world, can be linked to the call for ‘pragmatic’ clinical trials that is elaborated in the subsequent chapters by Kachur and Kelly. If a large-scale trial is tailored to local technological and administrative conditions, like in Kachur’s case, is this what Feierman’s doctors have in mind when they call for local evidence? Kachur’s chapter deals also with two different conceptions of evidence around a US government research project in Tanzania: ‘experimental’, based on randomised controlled trials (RCT), which shows how well a technology works under ideal conditions, and ‘observational’ or ‘quasi-experimental’, which shows how things would work under ‘real world’ conditions.11 Within this contrast, RCTs could be said to imply transcendence by pointing at what could be done in the absence of mere political or economic constraints; observational studies remain, somewhat akin to the low-tech medicine critiqued by Feierman’s African doctors, immanent to the existing societal situation: they show what can be done within given political and economic conditions.12 Kachur describes a quasi-experiment to study drug resistance, delivery efficacy and cost involving an approved anti-malarial drug. The Tanzanian Ministry of Health agreed with the US public health organisation to introduce the new drug ‘as a matter of local policy’ rather than as a clinical trial.13 Thus, ‘national and local authorities have taken responsibility’ and the experiment does not need ‘individual informed consent’, nor specific information about the drug on trial. Yet, while the researchers overall maintain a ‘low profile’, they ‘alter or enhance’ some aspects of the delivery system to ascertain the validity – dependent upon regular drug delivery – and the ethical and legal correctness – relying upon monitoring of adverse events – of the experiment.14 This raises, as Kachur points out, the question of what variables are to be altered, compared to ‘reality’, and which ones must remain stable to maintain resemblance with normality. In other words, what is imagined as a realistic change of reality? In the given study, only the drug is shifted (apart from improved adverse event monitoring for regulatory reasons); healthcare staff levels and resources remain unchanged.15 Comparing the contrasting modes of evidence described by Feierman and Kachur, one could be inclined to see parallels between the African cli15

Evidence, Ethos and Experiment

nicians’ pragmatism and the ‘pragmatic’ research design. Both are more ‘localised’ and specific than, respectively, clinical trials of innovative lowtech interventions and the RCT, which address more generalisable levels of evidence. Yet, on another level one could argue that the clinicians’ local experimentation and the RCT share a commitment to the open-ended potentiality of science, while both low-tech and pragmatic trials share a commitment to ‘reality’ as represented by the status quo. As Kachur underlines, observational studies look for ‘real world outcomes’, not for results that reveal the potential capacity of science; similarly, low-tech interventions purposely aim below not beyond the limits of known science, to remain within the seemingly insurmountable limits of the ‘real’.16 Kelly’s chapter expands on the question of how ‘reality’ is constructed when it is modelled in trials. It also relates to Feierman’s concern with how evidence is constituted in different contexts of healthcare delivery. Again, the relationship, shaped by the political economy, between means and ends, and between evidence and ethics in clinical research is in focus. Two trials run by the Medical Research Council in Britain and The Gambia evoke both ‘pragmatic’ entanglements between everyday life and scientific evidence: scientific representations and medical realities converge in a co-productive process, albeit within two very different systems of healthcare. While the UK healthcare standards are, apart from the tested intervention, equivalent to those of the experimental context, in The Gambia, the by-and-large absence of public medical care separates the realities of research and care provision. Indeed, the distance between medical realities and trial regime is so large that participants consider the experiment a ‘healthcare system’ (see also Leach and Fairhead), and, maybe partly in response to this wide gap, the Gambian study sets itself apart as an RCT, not attempting to be ‘pragmatic’. However, as Kelly’s ethnography shows, the purity of the RCT is modified in manifold ways to accommodate the local situation, which is why Kelly recommends a greater dose of explicit ‘pragmatism’, akin to the UK pragmatic trial, testing simultaneously new technologies and modes of delivery, addressing questions of causality and effect and matters of political economy. Kelly’s chapter opens up the larger question of how ‘clinical researchers are to adopt a pragmatic position’. Engagements between science and everyday life are uncertain, and it is precisely this enduring uncertainty, incompleteness and openness that make science epistemologically as well as socially generative. Can a ‘pragmatic trial’ foster this surprising potential of the scientific endeavour? It might as well simply prove what is known already, thus ce16

Studying Trial Communities

menting ‘security’ (in Peirce’s sense of the term, opposed to the productiveness of ‘uncertainty’), which, while practically important, curbs the creative possibility of science. Is it possible to build in pragmatic or ‘real world’ elements into existing forms of clinical research, irrespective of its wider political-economic situation? If it is done, what happens to the notion of reality? Is there not a subtle difference between engaging science in (unpredictable) social and political processes, allowing for transformations of either or both, and inserting (some selected) social elements into a scientific trial regime? Feierman’s African scholars point towards a possible alternative to the accommodation of human unpredictability into trial regimes: unlike pragmatic anti-malarial trials or low-tech interventions trials, their approach is fundamentally pragmatic in that it maintains the ‘subjunctive’, trying mode of clinical engagement and pragmatic science (see Whyte 1997). Paraphrasing Peirce (quoted by Kelly): their ideas have life, generative life. The fact that it is exactly this approach to scientific inquiry that is not easily turned into large-scale collaborative trials, and cannot command funding streams – unlike pragmatic trials, quasi experiments and the like – might be indicative of this peculiar ‘insecured’ quality. How exactly to promote such a pragmatic science – which is not necessarily a simpler, lower variety of high science – remains an open question for the social study of science. Part of the answer might, again, lie in the wider public to anchor healthcare, public health and medical research (see Ulrich). If experimentation were to be part of a continuous and effective process of government, and constituted within systems of both democratic representation and control, the hiatus between science and reality might become less dramatic. Mueller-Rockstroh’s chapter on Dutch ultrasound machines in Tanzania links the problem of shifting contexts of evidence to ethics by asking how ‘good’ scientific knowledge is produced. Relating pregnant women’s, doctors’ and administrators’ understandings and uses of ultrasound, she suggests that rather than asking what is ‘good technology’, we should attend to its varied effects in different situations – intended and unintended, beneficial to some and harmful to others. The ethical question what ‘good’ technology may be is thus turned into a matter of negotiation: ‘good-ness is a choice between different worlds’; and diverse goods call forth different interests. This supports empirically Ulrich’s conclusion that ethics is, and indeed should be, a matter of contest, and that the validity of ethical judgement depends upon transparent and balanced modalities of arbitration. In an aside, Rockstroh suggests another small but important distinction: that between ‘good’ and ‘better’. While ethics discussions often understand 17

Evidence, Ethos and Experiment

the notion of ‘social good’ as analogous to ‘value’, there is an important difference between the good of a technology – the value for society or a particular group – and the prospect of betterment, ‘hope for the better’ – embodied by a technology. Rockstroh’s ethnography draws our attention to that latter, subjunctive and directional quality, which is not necessarily identical with the actual effects intended or achieved by science and technology, but which nevertheless can make people utilise technology, pursue science and change their reality. The last three chapters in this section share a more direct interest in the political context of evidence. Langwick’s chapter on government traditional medicine research in Tanzania underlines, again, that scientific proof emerges in the details of research methodologies and the structures of research institutions and collaborations, and that the conditions that allow some things to be proven and not others change over time. Her work sheds light in particular on the changes that occurred between the 1970s and today, a period during which Tanzania has taken a particularly radical turn from social democracy to radical privatization and ‘re-structuring’. For traditional medicine this implied, as Langwick shows elsewhere, a shift from herbs as resources for national health to herbs as possibilities in the globalised economy (see Langwick 2001). Langwick shows how values and practices that bolster particular forms of proof – in other words traditional medicine which is elaborated through scientific study – are sedimented in institutional structures. In her case, social studies and pharmacological and botanical studies relate in a specific way: the former covering largely what is deemed ‘unnecessary’ and ‘uncaptured’ by the latter. As a result of this order of knowing traditional medicine, Langwick shows, healers and their rituals are variously included or excluded from the development of traditional medicine. This effect of epistemological and institutional structures raises methodological, political and ethical questions, and opens the possibility of renegotiating the integration of traditional medicine and biomedicine. Geissler’s chapter on ageing Kenyan government scientists’ memories of a bygone era of nation-state public health covers a similar period as Langwick’s work – 1960s to the present – looking back from the post-neoliberal era to distant, long-lost modern times. Exploring the lives of scientific workers, the chapter discerns a particular notion of evidence linked to direct action on concrete public health problems. This notion of ‘field-work’ relied upon the integration of scientific work into the frames of the nationstate and a – however fragile and at times violent – contractual relationship 18

Studying Trial Communities

between citizen and government. This modern project claimed that it is possible to understand the world, and that it can be changed. This faith in two kinds of interdependent transformation – of knowledge and of what has been made known – permeated the men’s biographies. Their lives as quintessentially modern government scientists underline the necessity to think epistemological and political economic questions jointly, and they cast some doubt over certain anthropological and historiographic views of the state and medicine, current in the 1990s, that viewed the state as an entity vis-à-vis its people and focused analysis on its controlling and disciplining nature. Instead of the spatial imaginaries of the people as opposed to the state (or that of science juxtaposed to its subjects), which often underlies scholarly and ethical reflections on government and science, the men’s lived-in ethos of truth and transformation points to the importance of temporality to understand health science, not merely as governmental domination and discipline, but as a project. Heald’s chapter closes this section with a somewhat related reflection on ‘public health’ versus the current rights-based, ‘exceptionalist’ approach to HIV in Kuria, Kenya, an area heavily affected by AIDS. Heald links the exceptional status of HIV/AIDS among other, comparable health problems to its early association with certain rights-based interest groups, and to the fact that its outbreak coincided with the beginning of the global human rights wave that paralleled political-economic neoliberalisation. Opposition to such HIV exceptionalism has been raised by public health policy makers (e.g. Bayer 1991). Heald finds similar arguments in her Kuria informants’ commentary upon bioethical regulations around HIV research and intervention, and specifically their apparent opposition to confidentiality and voluntariness, which they argue contributes to the spread of HIV, and their calls for obligatory testing and publication of test results. In conclusion, Heald considers the idea that some of the ethical constraints of VCT might be counterproductive, and that HIV research and intervention should reconsider the emphasis on individual autonomy, rights and choice – making for an interesting convergence between anthropological and public health perspectives. While there of course are diverse opinions on the issue of confidentiality and patient rights among public health scientists and western Kenyans, the point made by Heald and the informants she quotes points back to the public health contract that the old scientific workers in Geissler’s chapter remembered from 1960s health science. The old men’s longings for authoritative, even authoritarian, and effective public health knowledge, and some Kuria people’s 19

Evidence, Ethos and Experiment

rejection of liberal and individualised approaches to HIV, seem to express a yearning for public health interventions in the interest of a wider social collective, for the (re)constitution of some sort of a ‘public’ in relation to public health. Importantly, there seems to be at present no shortcut to such larger whole, as any nostalgic attempt to return to 1960s postcolonial nation-state health would be as futile as politically dubious. Indeed, in the situation of criminal or failing states, a call for authoritarian public health policies would be dangerous, and probably detrimental to health. Thus, rather than siding with the different calls for a ‘return’ to public health that emerges in the last two chapters, anthropologists and historians ought to reflect about what happened to the ‘public’ that once held together government and citizens in one collective, which pursued – or at least claimed to do so – betterment and welfare. The next step would then be to decide what kind of collectives we might want to (re?)create to face the current epidemiological and political-economic situation.

Politics Accordingly, the book’s last section brings together chapters that occupy themselves with politics and the history of medical research and public health. Ombongi opens the section with an analysis of the shifting relation between state and biomedical policy in the colonial, postcolonial and what he calls ‘post-postcolonial’, or post-1980s era. While he stresses that colonial public health was a dimension of colonial hegemony and social control, and that science served also ‘the legitimisation of cultural superiority’, he also acknowledges that biomedicine was part of an (albeit paternalistic) state modernisation process. This modernist project fed then, in the 1960s and 1970s, into a progressive nationalist research agenda, expanding public health infrastructure and research institutions during the postcolonial era. Only by the late 1970s did it become evident that the government had become unable to deliver on its medical and scientific promise, and consequently the state lost prominence as an agenda-setting actor in public health and medical research. In its stead, new para-statal, non-state and international institutions were created while the bodies of government health science and public health decayed. Ombongi’s chapter provides us thus with a broad periodisation that can be generalised – and fruitfully compared – to other African countries. In so doing, he draws our attention, as historians and anthropologists, to the as yet largely under-researched period of the long 1980s, the era of the liberalisation of science and health, which calls for new analytic modes and approaches compared to the history of colonial medicine and research. 20

Studying Trial Communities

The following chapter by Lachenal takes on this task. Looking at the life of the Institut Pasteur in post-independence Cameroon, Lachenal works out a specifically French ethos of ‘collaboration’, which emphasises the aim to assimilate the postcolonial subject –‘Cameroonisation’ – while continuously postponing the actual transition of power and authority. The short-lived nationalisation of the Institute Pasteur in 1975 reveals the difficulties that the loss of French resources and contacts entailed at a time of economic crisis and neoliberal globalisation; nationalisation appears as a noble but doomed anachronism in the face of the political and economic restructuring of the nation and of science. Attending to the ideologies of separation supporting the continuous ‘collaboration’, Lachenal examines work arrangements and relations among staff, and traces the ‘moralisation’ of professional categories, and the culturalisation of politics, framed by old racist stereotypes. The colonial order is thereby, he argues, perpetuated and transformed within the new global political arrangement. Lachenal’s conclusion is characteristically ambivalent: while he does consider that global ethical standards could provide an opportunity to critique and address inequalities in research collaboration, he also notes with chagrin that such standardisation will entail the loss of the specifically French ethos of cooperation which, in spite of its obvious imperfections, arose in on-going face–to-face relations and thus offered possibilities of negotiation, contestation and change, which are not easily captured by formalised ethics. Schumaker’s chapter takes us back in time, to corporate malaria research and control in a colonial Northern Rhodesian copper mine. Scientists’ and industry representatives’ views of malaria control were here linked to the control of workers in a contested economic terrain. Miners and local malaria control workers, while often resisting this project, nevertheless associated scientific work with modernity’s universal promise of progress and wellbeing. Colonial antagonism did not preclude the emergence of a modernising dream that united, underneath racial repression and proto-nationalist resistance, the colonial occupants and their workers. This joint ‘research culture’ of malaria control made ‘different people and interests work together on a common project despite their often radically different understandings of the project’s meaning’. Schumaker’s study reveals the double-edged nature of modern (medical) science, noted above: disciplining in the context of existing power–knowledge alignments, on the one hand, and yet, on the other, opening horizons for hope and for transformation, which ultimately may move beyond existing arrangements. Speculating about the future rather than looking into the colonial past, Nguyen’s chapter follows Schumaker’s interest in medical charity and aid by 21

Evidence, Ethos and Experiment

discussing the impact of global flows of antiretroviral medicines (ARV), produced by US American and European companies and procured by US government and transnational charities, to Africa. He suggests parallels between the current re-formation of medical intervention in the context of HIV and the colonial history of medical intervention in Africa. In his view, ARV policies challenge national sovereignty and foreshadow a renewed ‘military therapeutic complex’, in which military might and therapeutic power, with scientific authority, are deployed for the political ordering of the world. He argues that what at present happens in Africa is indicative of ‘global futures of biomedicine’ – much in the same way in which past tropical medicine was a laboratory of colonial power and of metropolitan science – in the sense that the production and distribution of pharmaceuticals and the maintenance of bare life will be a key to the future order of government, and to the link between political and economic power and scientific knowledge. Our book ends with a final chapter by Luise White, who shares Nguyen’s concerns with ‘imperialist’ tendencies in some forms of overseas medical research and intervention in Africa. Using the case of smallpox, she traces a persistent disconnect between medical and health policy actions and the interpretations and intentions of those – African bodies, societies and nations – that have been subjected to these actions, over the course of the twentieth century. She argues that in the face of this divergence between those in power (scientifically and politically) and those whose bodies experience this power, the current utilisation of research ethics is doubtful at least, if not harmful. The question of whether a particular procedure is ‘ethical’, she states, is secondary to the question whether the whole political and epistemological process can be judged ethical, in a broader sense of the term. In view of powerful transnational interests in medical science that overshadow African national sovereignty (in her view, especially those of the USA), White insists that medical intervention and research must be viewed not primarily as an ethical issue, but as a political contest. And she concludes that: ‘the broader context of medical intervention [throughout the twentieth century] … originates so far from African concerns and African consent that it cannot be called ethical at all’, and that medical ethics ‘simply avoids the larger question’. Provocatively running counter to the entire research ethics debate, she insists that we should ‘not worry whether this or that individual consented to a particular injection’ but ‘why some injections are available and others not’, and who controls this. It must be said that White’s historiographic head-on assault on regulatory bioethics was not shared by all participants at the Kilifi conference. We nev22

Studying Trial Communities

ertheless let her chapter close this collection, giving her the last word on the matter, because it marks one endpoint of the spectrum of our fruitful discussions in and after Kilifi, and because we do think that the underlying concern with radical political economic inequality, exclusion and domination is shared widely – indeed also among some of the ethicists and scientists she critiques – and does provide an important starting point for further, joint thinking about the ethos of medical science in Africa. It is our hope that these conference proceedings, like the sustained dialogue that the meeting in Kilifi has encouraged, will contribute to the open, engaged and theoretically informed debate on the ethics and political economy of medical research in Africa and elsewhere – a discussion that is willing to include diversity of views, controversy and conflict, and more: a debate which accepts that position, interest and conflict are inevitable features and driving forces of the historical process, including the progress of medical science.

Notes 1. The papers assembled here consider themselves closely related to the latter part of the recent bioethics literature that uses ethnographic and other forms of social inquiry to open up and transform the ethics debate. We have earlier published a special issue that brought together those papers from the Kilifi conference that relate more closely to the ‘research ethics’ debate (Molyneux and Geissler 2008). The reason why we decided to publish the texts separately is not their incompatibility, but the fact that those published earlier directly insert themselves into the interdisciplinary bioethics debate, while those assembled below remain closer to their specific disciplinary origins outside the medical realm. 2. ‘Good Clinical Practice’ (GCP) is a regulatory protocol, issued by the International Conference on Harmonisation, that details practical procedures, including those pertaining to ethical matters (consent, incentives etc.) around clinical trials. It has been endorsed by the World Health Organisation and by the US Federal Drug Administration, which is responsible for licensing pharmaceuticals to the world’s largest pharmaceutical market, and it is therefore of crucial importance to the conduct of clinical trials all over the world, shaping widespread ideas about what the ethics of medical research are about (see e.g. EMEA 2002). In their GCP version, the ethical challenges of overseas research are inevitably reduced and focused on procedural matters, as exemplified by various online GCP/ethics certificates (see e.g. www.onlinegcp.org). It is this reduced version, rather than the important bioethics dialogue between philosophers, religious authorities, publics and scientists, that many of the authors in this volume react against. 23

Evidence, Ethos and Experiment

3. The Oxford English Dictionary defines ethos as ‘the characteristic spirit, prevalent tone of sentiment, of a people or community; the “genius” of an institution or system’ (although it originally has its roots in the Greek word ‘etho’ or ‘to be accustomed to’) (Oxford English Dictionary, http://www.oed.com/viewdictionaryentry/Entry/64840). 4. For an exemplary discussion of how biotechnological networks stabilise a ‘public’ see Hayden (2003). 5. The choice of trial ‘community’ was maybe, in retrospect, unfortunate, as it echoes the boundedness of the older ethnographic object and thus misrepresents the openness of the associational webs produced by overseas research – ‘network’ would have been a more contemporary formulation. Yet, to many of us, this old fashioned term also evokes one of the virtues of old ethnography, as we understand it: a methodological commitment to partake in the associational spaces that we study, including importantly the engagement with and positioning among sets of people, with their things and institutions, projects and memories. 6. Ulrich’s critique of this choice between pseudo-legal standards and personal morality echoes the observation, made by Harvey (2006), that neoliberalism erodes the democratically constituted public sphere and reduces the role of state and public institutions to legal protection of private property, leaving deontological commitments to private, often religious, morality and noeconservativism. 7. Since the Kilifi conference, several initiatives have been undertaken both by research institutions as well as activist groups to support the qualified engagement of various publics with overseas collaborative medical research (see e.g. the reflections of the activist de Cenival 2008). 8. The publication of this volume was overtaken by the second Kilifi conference on ‘The publics of public health’ (December 2009), which discussed the collectives of bioscientific work and public health in Africa (see Kelly and MacGregor, in preparation). 9. This discussion between American and African doctors recalls Ferguson’s description of ‘telos’ and ‘status’ in relation to modernity. While the African doctors retain the modern medical telos in spite of adversity, the ‘appropriate technology’ approach ascribes lastingly different (lower) status to African health systems (see Ferguson 2006). 10. The quest for low-tech solutions, advocated by doctors from high-tech institutes for the poorest people, have another problem: they tend to bypass national levels of medical decision making, reaching out to what is designated as ‘community’. Decisions are made by northern scientists, on behalf of southern patients, excluding local medical and political elites. This bypassing threatens national capacities, and may, as Feierman’s conversations with African doctors show, exclude important insights into the nature and use of medical evidence. It contributes to the erosion of the public, in Ulrich’s sense, above, and of accountability and legitimacy of research. This ‘bypassing’ is thus a particularly important ethical concern to be 24

Studying Trial Communities

11.

12.

13.

14.

15.

studied by social scientists (see also Ulrich’s observation of a Ugandan doctor’s views of the 1997 research ethics debates). This is a similar – though not identical – problematic as Feierman’s. As Kachur points out, the latter approach tries to include a clinical perspective into its field of vision, but this is not the same as seeing and acting like a clinical practitioner. When the findings of the latter are put into practice, disappointment is less likely than in the case of purified RCT results. On the other hand, the knowledge produced by such ‘realistic’ studies does not necessarily point beyond existing conditions: the pure, in a sense radical, truth of RCTs evokes political-economic demands; the realism or relative truth of quasi-experiments, by contrast, risks remaining within the seemingly given conditions of global inequality. Kachur’s trial presents us with a distinctive innovation: while the destination of the medical intervention remains the citizen’s body, the experiment pertains to the healthcare system; unit of randomisation and consent is thus not the individual but the district. While in standard experiments the body of the experimental subject is intervened into, in this case the body of the nation is modified. This policy-experimentality has some continuities with governmental experiments throughout the twentieth century (see e.g. Bonneuil 2000); yet, the contemporary experiments are carried out by a sovereign nation-state in collaboration with a foreign government. While the investigators led by Kachur have carefully considered the ethical questions arising from this new constellation, the political import of such an experimental regime would deserve further anthropological and historical reflection. Kachur describes the choice involved in the latter aspect: not to test every female patient who is prescribed ACT for pregnancy, but to monitor closely those who had inadvertently been given the drug. Since the drug had not yet been approved in pregnancy ‘to some observers … this appeared to be inconsistent’. The evidence resulting from the safety monitoring has an ambiguous status. While the primary motive of monitoring the effect of the drug on pregnant women is ethical and legal – driven by responsibility for subject health – the outcomes provide evidence of drug safety. Thus, while pregnancy exposure to the drug must not be an experiment, the effects are findings. This innovative type of trial reproduces, on an unusually large scale, the inherent tension of tropical clinical research: on the one hand, it requires that the clinical situation is different from what might be the case in industrialised countries with well-developed healthcare systems – different epidemiological profiles and different material conditions of care – and thus it has to be firmly situated in a particular place, such as a clinic. On the other hand, it aspires to universal standards of validity (and ethics) and must therefore detach itself from the place, by establishing separate, globally certified laboratories, or by transferring scientific specimens to other places, and by providing global standards of care. While simultaneous emplacement and detachment are regularly negotiated within African research hospital settings, in this case the scale of emplacement/disentanglement is larger: a district, and potentially the nation. 25

Evidence, Ethos and Experiment

16. The shift, in recent years, from research proper to ‘effectiveness studies’ or ‘evaluations’ that are more ‘real world’ and often less stringent in terms of controls and regulatory concerns/ethics is, as Kachur mentions, exacerbated by the fact that many new large-scale funders of health interventions (e.g. Pepfar) preclude the use of their funds for research in a stricter sense, and demand at the same time continuous exercises in ‘evaluation’. This conflation of experiement and intervention, could be said to be characteristic for contemporary African biopolitics.

References Angell, M. 1997. ‘The Ethics of Clinical Research in the Third World’, New England Journal of Medicine 337: 853–56. Badiou, A. 2002. Ethics: An Essay on the Understanding of Evil. New York: Verso. Bayer, R. 1991. ‘Public Health Policy and the AIDS Epidemic. An End to AIDS Exceptionalism?’, New England Journal Medicine 324: 1500–4. Benatar, S.R. and P.A. Singer. 2000. ‘A New Look at International Research Ethics’, British Medical Journal 321: 824–26. Biehl, J. 2005. Vita – Life in a Zone of Social Abandonment. Berkeley: University of California Press. Bonneuil, C. 2000. ‘Development as Experiment: Science and State Building in Late Colonial and Postcolonial Africa, 1930–1970’, Osiris 15: 258–81. CIOMS (Council for International Organisations of Medical Science). 2002. International Ethical Guidelines for Biomedical Research Involving Human Subjects. Geneva: CIOMS/WHO. de Cenival, M. 2008. L’éthique de la recherche ou la liberté d’en sortir. Bull Soc Pathol Exot 101(2): 98-101. Cooper, M. 2008. Life as Surplus: Biotechnology and Capitalism in the Neoliberal Era. Washington, DC: University of Washington Press. Emanuel, E., D. Wendler, J. Killen, C. Grady. 2004. ‘What Makes Clinical Research in Developing Countries Ethical? The Benchmarks of Ethical Research?’ Journal of Infectious Diseases 189: 930–37. Emanuel, E.J., X.E. Currie, A. Herman, Project Phidisa. 2005. ‘Undue Inducement in Clinical esearch in developing countries: is it a worry? Lancet. 366(9482): 336–40. EMEA (European Medicines Agency). 2002. ICH Topic E 6 (R1) Guideline for Good Clinical Practice. London: EMEA. Fairhead, J., M. Leach and M. Small. 2006. ‘Public Engagement with Science? Local Understandings of a Vaccine Trial in The Gambia’, Journal of Biosocial Science 38(1): 103–16. Ferguson, J. 2006. Global Shadows: Africa in the Neoliberal World Order. Durham: Duke University Press. Geissler, P.W. 2011. ‘Transport to where?’ Reflections about an awkward practice in medical research. Journal of Cultural Economy 4(1): 45–63. Gupta, A. and J. Ferguson. 1997. Anthropological Locations: Boundaries and Grounds of a Field Science. Berkeley: University of California Press. 26

Studying Trial Communities

Hayden, C. 2003. When Nature Goes Public: The Making and Unmaking of Bioprospecting in Mexico. Princeton: Princeton University Press. Harvey, D. 2006. A Brief History of Neoliberalism. Oxford: Oxford University Press. Kelly, A. and H. MacGregor. In preparation. The Publics of Public Health in Africa: Citizenship, Collectives and Social Value of Medical Research. London: Earthscan. Kelly, A. and P.W. Geissler. 2011. Introduction: The value of transnational medical research. Journal of Cultural Economy. 4(1): 1–9. Langwick, S. 2001. Devils and Development. Ph.D. dissertation. University of North Carolina at Chapel Hill. Leach, A., S. Hilton, B.M. Greenwood, E. Manneh, B. Dibba, A.Wilkins, E.K. Mulholland. 1999. ‘An Evaluation of the Informed Consent Procedure Used During a Trial of a Haemophilus Influenzae Type B Conjugate Vaccine Undertaken in The Gambia, West Africa’, Social Science and Medicine 48(2): 139–48. Lurie, P. and S.M. Wolff. 1997. ‘Unethical Trials of Interventions to Reduce Perinatal Transmission of the Human Immunodeficiency Virus in Developing Countries’, New England Journal of Medicine 337: 853–56. Molyneux, C.S. and P.W. Geissler. 2008. ‘Ethics and the Ethnography of Medical Research in Africa’, Social Science and Medicine 67(5): 685–95. Molyneux, C.S., W. Mutemi, N. Peshu, K. Marsh. 2005a. ‘Trust and Informed Consent: Insights from Community Members on the Kenyan Coast’, Social Science and Medicine 61: 1463–73. Molyneux, C.S., N. Peshu, K. Marsh. 2004. ‘Understanding of Informed Consent in a Low-income Setting: Three Case Studies from the Kenyan Coast’, Social Science and Medicine 59(12): 2547–59. Molyneux, C.S., D.R. Wassenaar, N. Peshu, K. Marsh. 2005b. ‘”Even If They Ask You to Stand by a Tree All Day, You Will Have to Do It (Laughter) … !”: Community Voices on the Notion and Practice of Informed Consent for Biomedical Research in Developing Countries’, Social Science and Medicine 61(2): 443–54. Nguyen, V.-K. 2005. ‘Antiretroviral Globalism, Biopolitics, and Therapeutic Citizenship’, in A. Ong and S.J. Collier, Global Assemblages: Technology, Politics, and Ethics as Anthropological Problems. Oxford: Blackwell, pp. 124–44. Ong, A. and S.J. Collier. 2005. Global Assemblages: Technology, Politics, and Ethics as Anthropological Problems. Oxford: Blackwell. Parker, M., T. Allen, J.Hastings. 2008. ‘Resisting Control of Neglected Tropical Diseases: Dilemmas in the Mass Treatment of Schistosomiasis and Soil-transmitted Helminths in North-western Uganda’, Journal of Biosocial Sciences 40(2): 161–81. Peterson, K. and M. Folayan. 2005. ‘The Tenofovir Pre-exposure Prophylaxis Trial: The Nigeria Debates’, NHVMAS (Nigerian HIV Vaccine and Microbicides Advocacy Group) Echoes, June. Petryna, A. 2005. ‘Ethical Variability: Drug Development and Globalizing Clinical Trials’, American Ethnologist 32(2): 183–97. ———. 2006. ‘Globalising Human Subjects Research’, in A. Petryna, A. Lakoff and A. Kleinman, Global Pharmaceuticals: Ethics, Markets, Practices. Durham: Duke University Press, pp. 33–60. ———. 2009. When Experiments Travel: Clinical Trials and the Global Search for Human Subjects. Princeton: Princeton University Press. 27

Evidence, Ethos and Experiment

Rajan, K.S. 2006. Biocapital: The Constitution of Postgenomic Life. Durham: Duke University Press. Titmus, R. 1971. ‘Why Give Blood to Strangers?’, The Lancet, 16 January: 123–25. Whyte, S.R. 1997. Questioning Misfortune. The Pragmatics of Uncertainty in Eastern Uganda. Cambridge: Cambridge University Press.

28

Chapter 1

Writing Knowledge and Acknowledgement Possibilities in Medical Research Susan Reynolds Whyte

One of the central themes in the scarce ethnographic literature on medical research in Africa is the transaction of substances. Blood is usually seen as the mediating material par excellence, weighted with significance for the researchers who remove and test it, and for the people from whose bodies it is extracted. But there is another substance that is more universal in medical research. In fact it is such an integral part of the activity of gathering data that it is hardly ever remarked upon. That substance is paper. Paper and the activities of registering and writing to which it is bound are fundamental to researchers not only at the point of constructing data but throughout the ensuing analysis, presentation of papers at conferences, and publication of manuscripts. Knowledge is objectified in written papers; its form and content are transmitted and disseminated, perhaps electronically at first, but ultimately on printed paper. For those who are the subjects of research, paper and writing have different significance. From colonial times, they were associated with relations to government and organisations based outside local communities. Although paper-writing has many other uses and meanings, it retains the capacity to mediate such relations as well as information. The point that paper is transacted in the enactment of relationships was beautifully shown by Sharon Hutchinson in her work on Nuer exchanges of blood, cattle, guns, money and paper. Concerning the two substances most relevant to the practice of medical research she says: ‘For like “blood”, “paper” is capable of spanning, whether as metaphor or medium, the experiential extremes between social intimacy and social distance as well as between human vitality and human vulnerability’ (Hutchinson 1996: 288). Both me29

Evidence, Ethos and Experiment

diate relations to inscrutable powers: the blood of sacrifices to the spirit world and paper to the forces of government. The argument I wish to pursue is that inscription, the concrete practice of medical research, opens relational possibilities for those who write and those who are written. Writing is concerned with knowledge, but knowledge can mean radically different things, as Strathern shows in her chapter. Knowing about people involves taking an epistemological perspective on them, and producing data and treatises that abstract and objectify that knowledge. Knowing people, on the other hand, has to do with being together, acknowledging one another, performing writing in the presence of the other, sharing and exchanging substances including paper, food, words and touch.1 As an ethnographer I am interested in writing and ways of knowing as practices of situated social actors, who sense possibilities on the basis of common understandings derived from the past and conditioned by the present. While my primary focus is on those who are written and those who do the field recording, I bring in the perspective of the researchers by way of contrast. Methodologically I follow the path of progressive contextualisation from a cabinet stuffed full of completed questionnaires back to the research project that produced them, on to the local world in eastern Uganda in which it took place, and beyond to the broader implications of the example.2

Research Inscriptions and Transcriptions Child Health and Development Centre, a research institute under Makerere University’s Health Sciences Faculty, has a storage problem. In the small conference room and in many of the offices are piled cardboard cartons overflowing with questionnaires from various research projects. One particular set of forms has my interest. A large grey metal cabinet is crammed with thousands of questionnaires from Dr Frank’s study on pregnancy intervals. In 1999 he followed almost 3,000 women during their pregnancies and after the birth of their children. Since the mothers were interviewed several times, a great deal of paper was generated: there were kilos and kilos of forms with information on reproductive history, mother and child health, and nutritional status. The data on the completed forms had been entered and double entered into an electronic database, analysed and written up into articles and a Ph.D. thesis (Kaharuza 2002). Shortly after he submitted it, Dr Frank was offered an excellent position as an epidemiologist in the Uganda office of the Centres for Disease Control. When time came to pack 30

Writing Knowledge and Acknowledgement

Figure 1.1: Questionnaires from Dr Frank's Project (author’s photograph).

up his office, he rediscovered another set of inscriptions from his research. They were notebooks of prose written by the fieldworkers at his behest – daily records of their activities, mothers visited, problems encountered. Rats had gotten into those records of the daily practice of field research, so he threw them away. Somehow this is paradigmatic of medical research in Africa. The medical, often quantitative, results are preserved on paper (even after being converted into electronic form) and disseminated in publications which add to the cumulative knowledge about health. The process of collecting the data, with its relationships and transactions, is lost, though in fact it may involve many people (sometimes even more than those who read the resulting articles) and have important consequences. 31

Evidence, Ethos and Experiment

Dr Frank’s knowledge project concerned risk factors for short interpregnancy intervals and the relation between short intervals and child mortality. His results showed that short intervals followed infant deaths. He documented a pattern of pro-natality/high fertility, high infant mortality, and close spacing of pregnancies. His protocol was developed on the basis of existing literature as well as an earlier survey he had done in eastern Uganda. Together with his supervisors, he had identified a set of variables upon which he needed to collect data. Analysis of these constituted the contribution to knowledge that he wrote up in his thesis. In order to obtain the data he used a methodology succinctly described in his thesis: an initial cross-sectional survey that identified pregnant women, and a cohort study with a baseline and follow-up interviews. For the first phase of the study twenty local female health workers went house to house with a pre-tested questionnaire containing questions about the current pregnancy and the outcome and date of the previous one. For the cohort study, six teams of midwives and community health workers interviewed mothers at home or at antenatal sessions about their fertility, their socioeconomic status and maternity history. Mothers were again interviewed after delivery to get information on the birth and the current status of mother and child. Mothers and children were also weighed and blood was drawn for haemoglobin, although this data was not used in the analysis. So the written material that became the published contribution to knowledge consisted primarily of mothers’ answers to questionnaires. The records that Dr Frank and his assistants produced, like so much of medical field research writing, are meant to be an accurate replication of the responses to predefined questions. Anthropologists too make fieldnote records, in addition to the narrative descriptions of participant observation that constitute their ‘fieldnotes proper’ (Sanjek 1990: 99ff.). Such records correspond to what Clifford (1990) called ‘transcription’. They require ‘a kind of copying’ (ibid.: 58), the correct recording of what the informant says, as opposed to the researcher’s fixing of an observation (‘inscription’) or her making of a more coherent ‘description’ of an observed reality (ibid.: 51). Transcription involves the discipline of reproduction, whether of a text produced by the informant (like a genealogy or a myth) or the answers to questions produced by the researcher. The form shapes the nature of the knowledge written down, far more stringently than is the case with descriptive notes. Although I wish to move beyond Clifford’s ‘graphocentric analysis’ (ibid.: 53) of fieldnotes, I believe that the notion of transcription is important for understanding what medical field researchers are doing when they write down information. 32

Writing Knowledge and Acknowledgement

Fieldworkers have tasks to do on each human subject – measuring, weighing, collecting specimens, asking questions – and they must write an accurate record before going on to the next subject. From the point of view of fieldworkers, their subjects facilitate the process to greater and lesser extents. As one fieldworker in Dr Frank’s project explained: ‘A mother who was lousy could delay you, but a mother who was bright just answered quickly. When a mother answers well, you complete soon’. From my own experience of doing household surveys in ethnographic research, I know the feeling. Some households take hours, others are speedily finished, so that you can move efficiently on to the next. Finish one question, and ask the following one. Finish a household, and proceed to do another. There is a tyranny of questionnaires that lies in the formalism of ordering: next and then next and then next. Making the record is a task to be gotten through, required by the authority of the Principal Investigator and the research plan, and accomplished more or less efficiently depending in part on the facility of the study subject. In medical field research, the Principal Investigator trains the fieldworkers not only in what they are to do – interviewing people, collecting blood, weighing and measuring – but equally important, how they are to record these actions with utmost accuracy. Labelling specimens correctly, completing questionnaires properly, keeping order in the papers are necessary because the records are supposed to be an exact representation of the reality studied. The knowledge project depends on accurate transcription. The reliable quality of the transcription must constantly be ensured by supervision, surveillance, monitoring, checking records. Yet that is not always sufficient. Failures occur because of improper understanding of the questions, because of carelessness, and sometimes because of laziness and deception. If the form is not properly followed, or if the inscription is not a faithful transcription, the content is debased and the knowledge project is undermined. The transcription forms an enduring record from which data can be selected, extracted, coded, analysed and written up for publication. In its written form knowledge is freed from the context in which it was originally inscribed. It must be abstracted so that form and content can be transmitted and accumulated and used by others.

33

Evidence, Ethos and Experiment

Being Written Let us now switch to the point of view of the people whose words and measures are written on the paper. What experience do they have of being recorded? In order to answer this question, one general point must be clear. From the perspective of people being researched, and even those local people employed to collect the data, medical research in rural Uganda is not often distinguished from other health services and projects. The fact that research produces data is not very relevant. What is more significant is that it provides jobs and often some kind of healthcare, and that it extends sociality by creating new, though sometimes brief, relationships. As Pool and Geissler (2005) point out, the context of medical research in Africa is poorly functioning national healthcare systems. In Uganda, health interventions funded and sometimes implemented by donors and NGOs bear resemblance to research projects in that they supplement the under-funded government system. Those too bring possibilities and connections and social relations mediated by paper. Certainly research subjects are used to the activity of writing as an integral part of healthcare. One could even argue that the similarity of research and healthcare is affirmed by their common performance of writing. Visiting a formal health unit – whether government or NGO – is a matter of being written. Hanne Mogensen (2005:210) describes this insightfully in her article on the practical experience of Ugandan patients. The consultation room in a Ugandan health facility is sometimes referred to as ‘where they write the paper’, and people say that ‘when she [the health worker] starts writing we know she has understood’. When people choose a government health facility rather than a private drug seller it is, among other things, ‘because there at least they do the writing’. What they write, in Ugandan government health units, is the Medical Form 5, ideally a printed form for recording the patient’s name, age, diagnosis and prescription. In fact, the paper is often a school exercise book purchased by the patient, since the requisite stationery is usually out of stock. What is the significance of that writing? It lies not in the meaning of the words for the patient, for they are not easy to read even for those of us who are literate. Like the prescriptions that doctors everywhere write, these demonstrate the superior knowledge and authority of the health worker 34

Writing Knowledge and Acknowledgement

(Whyte et al. 2002: 117–29). But writing is also simply the tangible performance of formal healthcare, the way it is supposed to be done. By practicing it, the health professional shows he is trying to help the patient. And he produces thereby the material token that he passes to a patient, again to be exchanged, hopefully, for medicine. Writing as a substantial component of healing is also familiar in Islamic practices widely known in eastern Uganda and beyond. Words are literally ingested by those taking eduwa (also called ekombe), the dissolved Koranic passages drunk as medicine. Divination using texts (lamuli from the Arabic khatt ar-raml) relies on Arabic manuscripts and involves transaction as well, in that the client must give her name, which is used as the basis for the calculations that are part of this type of divination (Whyte 1991). In both cases, it is the substance and performance of writing, as much as the content that is important as part of the personal exchange that occurs. This provides another perspective on consent procedures, in which patients perform the activity of writing as part of healthcare or research. In Uganda, consent is perhaps better known in connection with Voluntary Counselling and Testing for HIV than research, since far more people have undergone VCT. In VCT sessions, ‘clients’, as they are called, experience that inscribing a signature or thumbprint on the counselling form is an integral part of the examination. In the sessions my colleagues and I have attended, the healthworker did not actually read the consent section of the form aloud. She explained the procedure and asked several times whether the client was sure she wanted to test; she did not mention that the client’s blood might be stored for future research. But I doubt that would have made any difference to the client because putting a mark on the paper was simply a step in the process of getting a test. Recently we have been interviewing people who have started antiretroviral treatment at a project in Kampala. In describing their initiation to the regimen, they mention going to sign the paper together with the person who will help them to take the medicine, as part of getting on treatment. One man did not have a support person to sign so the nurse signed in that space. Someone had to sign; that was part of the procedure. I am not saying that signing is mechanical or meaningless, but that the performance of writing or making a thumbprint is an integrated part of the activity of getting healthcare. It has significance as part of the give and take of diagnosis and treatment, just as writing does in lamuli divination. It seems likely that the same holds for informed consent for research procedures. For many people, the signature or thumb print on paper is less a marker of rights 35

Evidence, Ethos and Experiment

than a gesture within the giving and taking of bodily substances, information and medicine. In the very different context of northern Sweden, Hoeyer (2003) has made a similar argument about the importance of seeing consent procedures as part of social, historical and moral practice, rather than as a logocentric exercise positing a rational individual. It is important to note the difference between the substantial performance of writing, with its metonymic qualities of connection and part-for-whole, and the form/content of writing with its emphasis on the record and the knowledge inscribed. The substance of writing parallels other practices of consubstantiality where substances are exchanged in shared time and space. Michael Whyte and Pabire Higenyi (1997) describe mourning practices in Bunyole, when people come to pay condolences by offering money or other material gifts. Their names are written in the school exercise books that record the contributions of clansmen, in-laws, sisters’ children, neighbours and so on, each on the appropriate page. Whyte and Higenyi suggest that the act of writing is more important than the product, the written papers, since these are neither preserved nor referred to again. Writing the name when a person comes marks physical presence (you cannot send money with someone else and have your name and contribution written in the book). The funeral books are like the ubiquitous guest books in Ugandan offices; writing in them is part of making the call. (I have often been asked to write my name even before giving my reason for coming.) There is an immediacy of experience about this substantial performance of writing, a way of being, which contrasts with the form/content of writing so important to researchers. The contrast bears similarities to Marilyn Strathern’s opposition between being and knowing: embodied relational ontology and perspectival knowledge (this volume). Yet I would not want to make the substantial as opposed to form/content contrast too absolute, for the two kinds of significance may coexist and the same people appreciate both. Writing on the Medical Form 5 is part of the personal transaction of healthcare, but it also creates a more lasting artefact, a medical record that is both a history and a time line into the future. Women know that a card showing they attended antenatal clinic may be necessary if they need to give birth in a health unit. Sick children may be refused admission to the ward if their parents cannot show a record that they have been immunised. The double significance of paper-writing was amusingly illustrated when we were sitting in the evening talking about the upcoming local council elections towards the end of 2005. One of the candidates remarked about 36

Writing Knowledge and Acknowledgement

how you can waste money trying to mobilise votes. He had bought booze for one of his neighbours, and paid a boda boda bicycle taxi to take him to the sub-county headquarters, where elections for party officials were taking place. But the neighbour wrote his own name on the ballot instead! When the candidate complained that he had wasted his vote, the neighbour only asked: ‘But will they take those names to Kampala? Then my name will be in Kampala’. The neighbour had a metonymic concern with connection; the substance of his name would be transacted to the capital. The candidate had hoped for content in the form of a vote in his favour.

Written Records and Social Relations An ethnographic approach to paper-writing as substantive performance and as form/content requires consideration of what kinds of relations it mediates and what purposes it serves. Historically in Uganda, the imported religions of the book, Christianity and Islam, brought reading (people thank those returning from a church or mosque service ‘for reading’). But the colonial government brought writing, and paper was the currency for dealing with government authorities, whether in the form of a poll tax ticket to show one had paid the annual tax, or a ‘contract’ that recorded bridewealth agreements and could be produced in court if conflicts ensued. Even today, when paper-writing is much more common, it retains the capacity to mediate relations with authority.3 This is particularly so in the field of healthcare where ‘a paper’ (olupapula) or a ‘ticket’ (etikiti) can literally be a ticket to further action. I remember my amazement one dark night in 1970 when we had been called to carry a woman having a difficult delivery to the health centre. She was in terrible pain, but when we had driven some miles on a bad road, she suddenly remembered: ‘etikiti!’ Moaning, she insisted that we return to her home to fetch the paper she had forgotten, which showed she had attended an antenatal clinic. Medical records could be weapons in domestic conflicts too. Women who had been beaten by their husbands went to the health centre to get a written report of their injuries to take to the police. They did not in fact go to the police, but let their husbands and in-laws know that they had ‘a report’ and could carry the matter to the authorities if the husband did not change his ways. In her history of birth ritual and medicalisation in the Congo, Hunt explores the role of papers and inscription in the expansion of the medical domain. Medical passports showing that a person had been examined for 37

Evidence, Ethos and Experiment

sleeping sickness were required for workers and travellers in the early 1900s (Hunt 1999: 85–86). Birth certificates were a key part of medical maternity services; hospitals wrote out birth certificates that were brought to government offices for official birth registration: ‘“living in translation” began to permeate all of colonial life in a situation where birth translated into an object of inscription – a birth certificate’ (ibid.: 264). The history of inscription as a part of healthcare remains to be studied in other African countries, but it is surely an important story and one of particular relevance for the ethnography of medical research.

Figure 1.2: The Author Writing This Paper in the Project Office at Makerere University (photograph by Michael Whyte) 38

Writing Knowledge and Acknowledgement

In her analysis of how substances bind Nuer people to each other and to distant inscrutable powers, Hutchinson suggests that paper-writing mediates relations with outsiders and with government. ‘It is better to talk to the [outsider] in his own language’, one woman explained, ‘which means you must speak to him through paper’. Wherever they went, touring local government inspectors and other prominent state officials were normally pursued by a steady swarm of little ‘papers’, each carrying a personal plea for attention and assistance. Court clerks and other literate members of the community were constantly being asked by their nonliterate associates to inscribe their requests for assistance on ‘paper’. … In brief, ‘paper’ was the principal medium through which contemporary Nuer men and women sought to tap the powers of the government. (Hutchinson 1996: 284) The substantive quality of paper/writing is clear here, but the form/content of the request was important too, for both the supplicants and the recipients of the papers. The people with whom Michael and I have stayed for many years in eastern Uganda usually hand us folded papers on which they have written their requests for help, even when we are living under the same roof and even though they also explain the problem in conversation. Hutchinson’s observation about the accepted way of relating to those with power and resources is relevant here. As people give us food, occasional gifts and friendly talk, they also interact through writing. I realise that without ever consciously deciding that those bits of paper were data, I have kept them over the years in an unorganised fashion. I come across them under other piles, in drawers and boxes, tucked into field notebooks. Are they a field record? Perhaps someday they will be empirical material by virtue of their form/content. Or do I not throw them away because they are the very substance written and handed to me by our needy friends? Probably both. The idea that paper-writing expresses or mediates hierarchies of authority and power holds for relations between fieldworkers and their subjects as well. Researchers who come to people’s homes or meet them in local venues like schools or churches are outsiders who come to write. The empirical questions for any research project are: what kinds of social relations are established with those who come to write and how do they change over the course of time? 39

Evidence, Ethos and Experiment

In southern Uganda there is a greeting convention that is usually translated as ‘well done’ or ‘thank you for what you are doing’, often specifying the activity as in ‘thanks for reading’. When we first undertook ethnographic fieldwork in 1969 to 1971, I was puzzled by the way people used to say: ‘thanks for the work of the government’ (webale ejo govamenti). I think now that it was because they always saw us writing notes. An outsider who has a connection to an area or person and comes occasionally is recognised by a special phrase. Walking along on a path, stopping here and there, we are greeted by passersby who add: webale ohutulambula, ‘thanks for checking on us’ or ‘thanks for inspecting us’. Parish priests and government officers used to do ohulambula when they made tours of their area. Nowadays government officials seldom come, but clerics make rounds and officials from NGOs come for field visits to their projects. Visitors or guests (abageni) do not just inspect. Chairs are brought to a shady place. They sit and converse, an activity appreciated by others with a greeting to the host: ‘thanks for conversing with visitors’. If possible, they should be served food and drink, though the host does not necessarily eat with them. Having visitors is not usually a private affair. All the residents of a home come to pay their greetings and sometimes friends or relatives of the host are invited to converse with the visitors and eat with them. Even though hosting visitors is demanding, they are valued – not least visitors from afar or those of higher social status. Although they are sometimes received as visitors, most researchers do not fit easily into that pattern. They must ask questions and transcribe answers and have little time to stay and converse about topics not in their protocol. Geissler and colleagues (this volume) have described the ways in which field researchers became included as quasi-kin in a malaria vaccine trial in Gambia. Embedded with families for extended periods, the fieldworkers shared food and company with the people participating in the trial. This kind of kin-like inclusion, often playful, is common in ethnographic research, but rarer in biomedical research where boundaries are sharply drawn between researcher and researched, as Geissler has also shown in describing his movement from tropical hygiene to ethnography (see Geissler and Prince 2009, 2010). The Gambia malaria vaccine trial example is not typical. Most medical field researchers are probably less like visitors or quasi-kin and more like authority figures coming to inspect or to write.

40

Writing Knowledge and Acknowledgement

Possibilities Medical field research is based on a scientific model of experiment, in which a problem (an uncertainty) is defined and possibilities are explored. The classic American pragmatists, Charles Peirce, William James and John Dewey, were inspired by the natural sciences; they emphasised practical engagement with problems and concern with the consequences of practice. Dewey underlined parallels between experiment and experience (1929: 74–107).4 Learning by doing rather than by speculative (spectator) reasoning was his watchword and radical empiricism was his method. He celebrated uncertainty because it challenges our intelligence, and he focused on possibilities to be tried and consequences to be appreciated, rather than fixed certainties. Looking to possibilities and consequences implies a temporality and a directionality that is characteristic of pragmatism (indeed some critics have asserted that the pragmatists were too future oriented). In considering medical field research it seems fitting to adopt this framework in order to focus on what those who are writing and being written are trying to accomplish, what possibilities they imagine and what consequences they realise. Introducing the notion of possibilities and consequences takes us a step beyond the contrast between knowing and knowing about, because it assumes that the ‘doing’ that characterises both may effect changes in relationships with people and the world. Temporality first. The pragmatics of those who write and those who are written are unfolded within various temporal frameworks: of the duration of projects, programmes and institutions; of the circulation, accumulation and use of scientific knowledge; and of the lives and life chances of people. A researcher may have a research programme that stretches over many years and is carried out in different localities. But from the viewpoint of any given location, the question is: how long does this project run? Is it part of an established, continuing research institution where one project follows on another? Is it affiliated to local healthcare institutions that have a continuing existence before and after the life of the research project? What opportunities are there for local people to develop relationships with the field researchers and the other experts who stand behind them? There are the long-term research programmes operating in a given area, running various research projects within a population that becomes familiar with the institution. The Medical Research Council trial sites in The Gambia have a long history. MRC has its own buildings with independent power supply (in more senses than one), its vehicles, staff and foreign vis41

Evidence, Ethos and Experiment

itors. Many Gambians know and have benefited from it. Field researchers move from one contract to another with MRC so there is continuity in that respect as well. The KEMRI Unit at Kilifi on the Kenyan coast is similarly long-established and well known. These research establishments offer very different possibilities than short-term projects like cross-sectional surveys of the type studied by Yoder and colleagues (2006). They found that the very lack of relationship to the interviewers and lab technicians encouraged people to participate in the Home-based HIV Testing and Counselling Survey. The researchers were unknown; they came briefly and left. The research had no local base. In the MRC and KEMRI cases, people benefited from access to healthcare over time. In the Home-based HIV Testing and Counselling Survey, a desired one-time service was offered. We should also remember that such high-profile, well-funded research activities are only part of the picture. At medical schools all over Africa, students like Dr Frank are carrying out Masters and Ph.D. projects that also bring them into the field, sometimes to do ‘invasive procedures’ and sometimes to count. In addition, African social scientists engage people in health research that borders on ‘medical’. Many of these projects have a relatively limited time span, but the researchers may maintain links to the locations where they did the research, or at least to some of the local leaders and field assistants with whom they worked. In any case, they usually stay in the country and are thus in a position to use what they learned in further research and teaching, and possibly to influence policy makers. As scientists, researchers adopt a time frame that has to do with the production and circulation of knowledge. They must define a clear scientific problem that has an element of uncertainty requiring a trial of some kind. The delineation of the problem and the withholding of judgement until data have been accumulated and analysed is the very essence of scientific research. The outcome of the trial is objectified in writing and disseminated to a ‘community of inquirers’ – other scientists who will weigh it and compare the findings to existing knowledge in the form of other published papers. The scientific value of the knowledge is not established immediately; it must be commoditised and evaluated first. And the significance of the knowledge may endure into the future, beyond the life of its creators. In this sense, paper is not perishable. The possibilities that researchers explore take the form of evidence – that a vaccine is effective, that high child mortality leads to close spacing of pregnancies, that bed nets reduce malaria deaths. These outcomes may have 42

Writing Knowledge and Acknowledgement

consequences for national and international health policy, and they may even have consequences for future health programmes in the very communities where the studies were undertaken. At least that is the hope and intention of many medical researchers. ‘Research to action’ is the commitment of many international health researchers, while the need for ‘evidence-based policy’ is a strong argument for more research. For the researchers there are other possibilities as well. The papers they write may be accepted for presentation at conferences and in respected journals. They will figure on their curriculum vitae and enhance their career prospects and life chances. This is so for African scholars as it is for others. While there is much attention to ‘brain drain’ in the form of emigration by qualified researchers, there is perhaps less awareness of career mobility for academics who remain in their own African countries. They too need publications in order to gain promotion. The chance to supplement a university salary with externally financed consultancies and research tasks is contingent upon having written competent reports of research accomplished. When Dr Frank completed his Ph.D. he was recruited away from the university to a more attractive position, still in Uganda, but better paid and highly regarded internationally. The pragmatism of science, with its concern for outcomes, is never the only kind of pragmatism practised by researchers. In their life projects, they may not delineate problems so explicitly, nor try solutions so systematically, but there too they look to consequences of work undertaken. While researchers look to the knowledge outcomes of their scientific projects, as well as to other possibilities, those who are written and those who do the field recording are mainly concerned with the relational possibilities offered by the process of producing the data. For relationships are possibilities – among many other things. That is particularly so for relations with people of power and authority, those who have resources, who are gatekeepers, who have connections to others with resources. Possibilities can be ominous, as rumours about nefarious intentions of researchers suggest. But far more commonly, people not only accept research projects, they form hopes and expectations about their relations to researchers. The power and resource imbalance that underlies those hopes and expectations, and the ethical issues that follow, are widely recognised in the literature (e.g. Nama and Swartz 2002). Most important in African settings are the short-term expectations of access to healthcare through a research project, which may even be perceived as one type of healthcare among others, as Leach and Fairhead show in their chapter. Molyneux and her colleagues (2005) suggest that the ‘therapeutic misconception’ – that the research intervention is for the 43

Evidence, Ethos and Experiment

benefit of the individual and has a good chance of success – may not actually be a misconception in resource-poor settings where a research project is providing healthcare or offering a specific service. The identification of research with healthcare was pronounced in the study by Dr Frank. He based his research project at Busolwe Hospital, a rural 100-bed facility, where he stayed in a staff house for a year. Busolwe is a remote posting in a poor location and has trouble attracting doctors; there are supposed to be six but often there is only one. So when Dr Frank, an experienced obstetrician from Kampala, helped out at the hospital in addition to his heavy programme of fieldwork, the staff saw the research project as a boost to their attempts to provide services. More than that, the study offered a chance for employment, new relationships and new experiences. Dr Frank needed midwives from the government health units in the county and fourteen pregnancy monitors for his field study. The six midwives were paid by Dr Frank in addition to their government salaries, for they remained on staff during the year of research. Nambozo recalls how Dr Frank had come and asked the Medical Superintendent for some midwives for research. ‘At that time the MS liked me so much, and I got that chance.’ When Nambozo remembers that chance, it is partly the change in her daily life. They stayed at a training centre in the next county. They spent two months staying together at a hotel in Busolwe. They were given bicycles and rode long distances everyday, carrying their scales and measuring boards. ‘The way we got fat from Clyde Hotel. We were eating well. On top of salary, lunch allowance. Even my hypertension came down from bicycling.’ But it was also the way the extra money enabled her to further a life project. I was struggling to educate my brother’s children [she had no living children of her own]. You don’t even use the lunch allowance to eat when you’re out. You get the money and take for the children. Some of the girls disappointed me. They dropped out of school and married. But I told them, ‘You people struggle. Me I’m a hypertensive case. I can die anytime’. At least I would leave someone … I’ve struggled for them. They’re my real children. They like me so much. Another midwife had seven children to educate. Dr Frank was understanding. ‘When someone came from home to tell one of our researchers “your child has been sent away from school because of fees”, he would just give her an advance on her salary.’ 44

Writing Knowledge and Acknowledgement

The pregnancy monitors worked under the supervision of the midwives. They were not professional health workers, though some had been trained as Traditional Birth Attendants (TBAs). Most of them had previously been employed as Safe Motherhood Assistants under a World Bank project in Bunyole. There too the relationship had been one in which written papers passed one way, and money and other material perks the other. They had been given bicycles and allowances for visiting mothers and keeping records. As they told Dr Frank, ‘We took the forms to the Safe Motherhood supervisor, but he ate our money. We hope you’re not like that man’. The recruitment of field researchers who had previously worked for another project (whether intervention or research) is probably common in medical field studies. In both cases, the fieldworkers wrote papers that they exchanged for money, but in the Safe Motherhood transactions the paperwriters remembered that the exchange had been derailed by deception. For them too, the chance to work on the research project enabled other life projects. One confided to Dr Frank that she was HIV positive – a widow who had left her husband’s home after he died. She desperately wanted to leave some money for her children. In fact, she and another of the monitors, also sickly, died after the project was completed. Others used the experience to move onto jobs in the health sector. As Mariam, a midwife, explained: ‘One pregnancy monitor later became a vaccinator. Another got a job in a drug shop. She got that chance because people were seeing her with us. They took us all to be health workers (basawo). Now she won’t be minimised. People think she’s a health worker (musawo)’. Nambozo remembers that one of the monitors she supervised was so interested in the examination of pregnant mothers. She asked questions: how do you do this and that? Nambozo taught her how to palpate. Later she joined the training for being a TBA, got a job as a vaccinator, and today is a nursing assistant at a health centre. For the fieldworkers, the connection to a research project presented possibilities for earning money and (for some) obtaining other jobs. They gained practical knowledge, and new contacts, as we shall see, but they were not concerned with the research questions themselves nor the creation of new knowledge about pregnancy intervals and the health of mother and child. When I asked Nambozo what she learned, she replied that: ‘research added on my knowledge – things I didn’t know like measuring children. It gave me experience in research. I asked Dr Frank to take me to read further. He mentioned Nsambya. But then he disappeared.’ Like Dr Frank, she learned from the research. But whereas the knowledge he ex45

Evidence, Ethos and Experiment

tracted was written and circulated, she acquired embodied skills of measuring and filling in forms properly. What Nambozo hoped for was the chance to study and move up from enrolled to registered midwife. That would have given her a promotion, a white uniform instead of pink, and a salary increase. The midwives were keen to tell how the research project had affected their relationships to village mothers; this provides another angle on the discussion of the therapeutic misconception. That needy people hope for something from a research project is nothing new. The ethnographic question is: what relationships seem to open what possibilities in a specific setting? The eastern Ugandan context is one in which less than a third of mothers deliver at a modern maternity unit and where, in general, poor service at government health facilities is expected. User fees were being charged at the time of the research project. Many people thought that the way to get decent service was either to pay health workers extra ‘for tea’ or to have a personal contact working at the health unit. ‘Do you know anyone there?’ was and is a common question when considering whether to take a patient for treatment. Even though health workers are criticised behind their backs for being rude, uncaring and greedy for ‘tea’, people are polite to their faces. You never know when you might need a health worker. As one cynic, himself a clinical officer, remarked: health workers are like little gods. In this situation, the research project brought midwives and their assistants the pregnancy monitors – sometimes even a doctor – in touch with thousands of women in their own villages and homes. ‘Those who are educated usually look down, but now they have come to us’, remarked one woman. A few suspected wicked motives in this sudden inversion. Once when Dr Frank and some of the fieldworkers approached a village, with their heads and faces covered against the dust in his old canvas-top Suzuki, people fled thinking they had come to steal children. There were a few rumours that staff from Busolwe Hospital had come to sterilise women. But midwife Mariam says the suspicion passed quickly. ‘Anyhow it was mainly men and only in some parts – those deep deep places there, not those close to the road.’ The possibilities that the arrival of the researchers brought were of two kinds: medical examinations in the village and contacts at the health units. People valued having blood tested on the spot for haemoglobin. No woman refused, even though some worried at first that their blood would be checked for Slim. They wanted do know: ‘Musawo, do I have enough blood?’ Although haemoglobin was not supposed to be done every time, 46

Writing Knowledge and Acknowledgement

some requested: ‘Ehh, musawo, you are not checking my blood.’ Dr Frank remembers that mothers loved to be weighed. They laughed at each other: ‘You’re too fat, you’re too small’. Some women complained: ‘Musawo, why are my kilos not enough?’ Nambozo recalls: ‘Even men disturbed us. They wanted to be weighed. “Why only weigh women? We hear that if you are below 50 kilos you have HIV.” Some men wanted to know their HB so we weighed and did HB on the men who asked.’ A valued service came to villages, brought by the basawo themselves. The attention to individuals’ blood and weight, the importance of writing down each one and telling them their results are not to be missed in these accounts. We had the same experience in a malaria study we once did within our research collaboration. People were overwhelmingly appreciative of having their blood tested and getting their results explained to them right in their homes. The study by Yoder and colleagues (2006) on the Home-based HIV Testing and Counselling Survey found similar positive responses: 86% of respondents gave blood and chose to receive their HIV test results. The field researchers wrote weights and haemoglobin counts on their papers. But that transaction opened the way for others later at the health facilities. Mothers were encouraged to attend antenatal regularly and to deliver at the maternity units (Dr Frank went onto the ward and paid the user fee for those in the study.) Some were referred for treatment or family planning. Ordinary medicines were not taken to the villages. But people were told to bring sick family members to the health unit and to use their new connection to get treatment: ‘When you come, ask for my name’. ‘You bring the children in the hospital please. Tomorrow by 8 before I leave so I hand them over to the doctor. You tell your problem to the musawo, she can help you.’ The midwives remembered how people sought them out at the health facilities. They come looking for you in the unit, you take them to the health worker on duty. They expected us to help everyone in the family. They even came on the weekend. We told them to come on days when we were there. They knew our names. Up to now some still come. Some we have forgotten but they remind us: ‘Don’t you remember me? You came at ours’. They want quick service or special service. She has come because of you. She has to show that she has responded to you. You counselled her and she came as you advised. To encourage her to tell others to come, especially those from far like Namulo, you give special service.

47

Evidence, Ethos and Experiment

Figure 1.3: Collecting Knowledge, Acknowledging the Other (author’s photograph).

The point here is not only that research subjects and their families saw possibilities for accessing service. They saw chances for cultivating relationships – and they were encouraged in that hope by the basawo who gave them their names. To say that rural African people are poor and powerless, and that they are mainly concerned with the health services that research might bring, goes only part of the way in understanding the social relations I am trying to get at. The possibilities of creating new connections are important to both field researchers and those they record.

Recognition The utility of research for frontline fieldworkers and their subjects is important. But we should not reduce the connections facilitated by research to their utility alone. That would not be in the spirit of pragmatism, which embraces the totality of human social existence. Relationships are what social life is about. Charles Piot (1999: 170–71) defends the tautology that 48

Writing Knowledge and Acknowledgement

people attend to relationships because they value relationships as an ‘irreducible premise’ of the social world he studied in northern Togo. If, for the capitalist, value and wealth reside in commodities – and lead in societies of the metropole to a proliferation of commodity production and consumption – for Kabre value is also measured by relations. Moreover, ... relationships are invariably mediated through exchange and embodied/externalised…. ‘When someone from the south brings a ceremony to us [in the north], or when a son buys tin for the house of his father…they show us respect’. In a similar vein, we should be asking how relationships are developed, expressed and mediated in medical research. In listening to how Dr Frank and two of the midwives recalled his study, it struck me that medical research concerns not only knowledge but also acknowledgement. Respect and recognition are being expressed and mediated, in these relationships, as paper, money, blood and treatment were exchanged. I started out by suggesting that paper-writing is a way of mediating relations between rural people and government officials or other authorities, including health professionals. That Dr Frank’s Ph.D. project was perceived as a government intervention seems very likely. Midwife Mariam recounted: ‘I used to tell people: ‘Only in Bunyole has President Museveni sent a special doctor to work on pregnant women’s health’. And they were very proud.’ To say ‘webale ejo govamenti’, thanks for the work of the government, was not wrong inasmuch as the project was based at a government hospital and involved district health staff. But in general, as I have said, there is a tendency to conflate government, NGO and research efforts that seem to emanate from Kampala or from outside. Researchers are ‘big people’, the kind who usually look down on uneducated villagers. They were, or were like, the health workers who are so often said to be harsh and uncaring. Unschooled women in particular feel the distance and discomfort in relations to health workers, as Anne Katahoire (1998) showed in her Ph.D. thesis on schooling and healthcare in eastern Uganda. Yet the distance in these relationships with the paper-writers is sometimes bridged. Paper is only one medium of transaction in hierarchical relations with researchers and health workers. Another is touch, that tangible exchange or participation in relationships to which Geissler and Prince (2010:11–14) have drawn attention. Mediation by touch is an overlooked element in rural healthcare. Whether the touch of the health worker is part of treatment or 49

Evidence, Ethos and Experiment

examination, it can convey attention and effort. Mogensen suggests that patients in eastern Uganda are aware of the unspoken practice of the healthworker even when words of explanation are not forthcoming or not registered. ‘It is noticed when the health worker touches the body or puts the ‘pen’ (i.e. the thermometer) under the armpit, looks for anaemia by examining the eyes, or checks for dehydration by pinching the skin to see how quickly it smoothes itself out’ (Mogensen 2005: 221). Certainly the touch of a health worker or researcher can be seen as rough or invasive or threatening. Examination may be interpreted as an occasion to maliciously remove bodily substance or measure people as objects not persons. But touch can also mediate concern in a relationship otherwise characterised by distance. It can express attention and recognition and sympathy. Dr Frank recalled: I had just parked that old Suzuki in a village when they came asking for help with a woman in childbirth who had failed to deliver the placenta. I went into her house, pressed the abdomen and delivered the afterbirth. The husband was grateful ‘because you touched my wife’. Later they asked for my name, they wanted to name the baby after me. And midwife Nambozo remarked: ‘When you start measuring the babies, when you touch the baby … These mothers need you to show love, to show that you feel it’s your person’.5 ‘To feel it’s your person’ may be just a figure of speech, a way of talking playfully about belonging and concern. The midwives told how they still see some of the mothers from the study, six years on: ‘When I see my mothers now, when they come to the hospital, they say: “Musawo your baby has grown” and I tell them, “You bring and I see,” and they say, “I’ll bring”.’ Maybe this light talk is really about hierarchy, trying to make the big person feel obliged. But even mere words about attachment and regard are welcome. In some cases at least, being a research subject is not experienced as subjection to the medical gaze, but as a reassuring possibility of inclusion and belonging. Sometimes, however, relationships are expressed by other media than touch or words. In a research collaboration we had with Businet+, a network of HIVpositive people in Busia District, two staff from Child Health and Development Centre were working with some informants when word came that a member of the network had died. Businet+ makes a point of attending funerals to mark their losses and publicly speak about AIDS. When I met the 50

Writing Knowledge and Acknowledgement

chairman of the network some weeks later, he made a point of telling me how Ivan and Fred from our research centre had gone with them to mourn. ‘People at the funeral saw that we have friends from Kampala.’ Although Ivan and Fred had come to Busia on a utilitarian errand, to bring them a computer and give them some training in how to use it, attending the funeral was something more. Call it an act of friendship. Call it regard. Call it recognition.6 Those whom Dr Frank’s research targeted seemed to appreciate that they were acknowledged by the basawo. But recognition was just as important for the fieldworkers themselves. One midwife explained: ‘Research is good – you meet so many people. It made me to be popular. Everyone knows me …. We created relationship … the way I was handling them. That thing helped me to be known and to be popular in Bunyole’. Another midwife expanded: I got in touch with so many mothers, became so familiar to many people. If I move anywhere I’m known …. Being known is good. It makes you to practise. Professionally it’s an advantage. I’ve ever met so many. ‘Thank you for having helped me.’ She gives you a soda when you don’t even remember her. Because you treated her well. Some come at home bringing a quarter sack of potatoes, groundnuts, oranges, some even a chicken. ‘Do you remember me? You came at home.’ Some could even welcome us as visitors. One man even slaughtered a chicken and cooked rice. Since then I’ve never forgotten him when I meet him. We reached his home. The same dynamic happens in relation to patients. Midwife Nambozo, Michael and I were walking back to her staff house at the hospital when a man wheeled up, dismounted from his bike and greeted her. ‘I’ve been looking for you musawo’, he said, handing her a couple of kilos of rice in a plastic sack. She thanked him warmly and later expounded: ‘This is what I like. People are grateful and they bring thank you. You don’t ask for money or things. Just treat them well and they give you gifts. This is my client. I counselled him and his wife; we talked long [in connection with an HIV test]. He works in the rice. Now he has come to find me as he said he would.’ The midwives felt appreciated by Dr Frank too: ‘The way he used to respect midwives. He’d say: “Those are my bosses”, or “Midwife, come and help me here. You know better, I’ve been away from this work for awhile”.’ If research projects can, among other things, facilitate relationships in which recognition is expressed, they can also reveal the temporal limita51

Evidence, Ethos and Experiment

tions of such relations. The project ends, the fieldworkers and researchers disappear. Nambozo told how the mothers chided her. ‘When it ended, eheh. Some mothers used to come here. Musawo, you decided to go forever. You could not even come to see your babies.’ But the midwives too were disappointed when the relationship fizzled out. ‘Dr Frank brought us a very good big study. His colleague Dr John promised to take us to Busia to do the same study there. But then he went for further education.’ As we parted, Nambozo gave me a message to take back to Kampala: ‘Ahh, tell Dr Frank not to forget us, to bring for us another research’. The temporality of research projects is limited. The acknowledgement that was daily expressed in seeing babies, eating and conversing together, comes to an end when that substantial interaction is no longer possible. But the temporality of scientific knowledge is not bound in the same way. The project ends but the papers continue to be published and read and referred to. The career of the researcher goes on, in other places, with other research projects, and other inscriptions. Yet we should not conclude that acknowledgement and the substantial knowing of direct interaction is what locals seek and experience, while researchers focus only on the abstracted form/content of knowledge that they offer to an impersonal ‘community of inquirers’. As my long conversation with Dr Frank drew to a close, he smiled a little sadly at the memories I had caused him to revisit. We went out in that Suzuki all the time – something I am missing now. Going out and talking to people, not just sitting in offices and going to meetings. It was my study, I was really engaged. Now I supervise others, but it’s not the same. The researcher too lives the acknowledgement of face-to-face interaction, and appreciates knowing people as well as knowing about them.

Conclusion Medical research has some resemblances to development work. (Leach and Scoones [2005] show that science and technology studies and development studies can usefully be considered together.) Their goals are different: research aims to extract data and build knowledge; development infuses and deploys resources to change a situation. But both have a kind of ‘managerialist and interventionist’ undertone, in Norman Long’s words. Powerful outsiders, often associated directly or indirectly with the state, bring re52

Writing Knowledge and Acknowledgement

sources to local people whom they see as powerless or vulnerable insiders. Long proposes the notion of ‘interface’ or better ‘interfaces’ for the ‘sites of social discontinuity, ambiguity and cultural difference’ where interactions occur. He sees a role for ethnography here: In order to get to grips with these contradictory and discontinuous processes, the practitioner or researcher needs to access and learn lessons from the ‘autonomous’ settings in which people cope with their own problems, irrespective of whether or not the foci of concern or parameters of action can be linked with outside intervention. This requires the adoption of an ethnographic stance rather than the use of experimental method. One must go where people are already engaged in interactions, problem-solving activities or routine social practices and negotiate a role or combination of roles for oneself as participant observer, active collaborator, advisor, etc. (Long 2001: 90) There are excellent examples in this volume of ethnographers who have done just this: placed themselves in settings where local people’s concerns are the focus, rather than taking a position within the medical researchers’ framing of relevance. As these illustrations show, the art of it, if we are going to have a dialogue, is not to dismiss the frames of medical researchers but to relativise them and to explore areas of overlap and incongruence between their concerns and those of other social actors. In this chapter, I have tried to explore interfaces by starting with a substance, paper, and a practice, writing, that are second nature to every researcher. The research activity of writing information and being written was ‘a site of social discontinuity’, in which differences could be described in terms of ‘knowing about’ and ‘knowing/acknowledging’. The researcher is collecting transcriptions that will allow the abstraction of data and the creation of abstract knowledge about health in Africa. He is making something that has value and can be circulated beyond the immediate time and place. The research subjects are relating to outsiders through the here and now performance of paper/writing, in a well-established pattern of interaction, particularly familiar in healthcare settings. The contrasts between knowledge and acknowledgement, knowing about and knowing, are profound. Producing knowledge is a task of forming content and commoditizing it, that is, rendering it common, transmissible and accessible to distant and future researchers. Acknowledging the other in personal interactions is to create something that is 53

Evidence, Ethos and Experiment

immediate and momentary, relatively perishable, forgettable. It might be tempting to conflate this opposition with others: us and them, researcher and researched, cosmopolitan and local, professional and patient. I believe that would be misleading for several reasons. Clearly researchers engage in both creating knowledge and sharing regard. They switch between the time of abstracting, analysing, writing, and the moments of substantial interchange with their research subjects, as well as colleagues, family, friends. Different kinds of medical research provide different possibilities for personal regard. Ethnography depends on it; in some kinds of laboratory research, the principal investigator has little direct contact with the bodies from which substances have been extracted. Yet nearly all medical field research involves some measure of time shared with research subjects, community leaders, phlebotomists, clinical staff, interviewers, field assistants, clerks, drivers. What Dr Frank missed was that substantial interaction. What about local research assistants and subjects? Are we to understand them as ‘merely’ concerned with immediate acknowledgement and material benefits? Do they also increase knowledge? They do not produce commoditised knowledge themselves, except insofar as their participation is necessary for the creation of data. But it is important to recognise that they too learn. If knowing people is not a state of being, but a practice that involves exchange and sharing, then it has the potential for changing those involved. That is why we must attend to possibilities and consequences for all those actors brought together in medical research. The new social contacts and new experiences of regard within social relations affect knowledge in the sense of knowing. It is probable that at least some of the locals involved are also interested in knowing about – in being informed about the knowledge produced by the paper-writers – although researchers seldom consider them as part of ‘the community of inquirers’ to whom they address their results. In any case, medical field research offers possibilities for both those who write and those who are written, and the task of ethnography should be to explore how we all engage in such knowledge projects in multifarious ways.

Notes 1. The argument owes much (sometimes, as in this sentence, the substance of the words themselves) to Wenzel Geissler, who saw possibilities in an earlier draft of this paper and provided inspiring comments which I have gratefully incorporated. 54

Writing Knowledge and Acknowledgement

2. The material for this article comes from a long interview with Dr Frank Kaharuza, and conversations with two of the midwife/field assistants employed on his project. His research project took place in a part of eastern Uganda where my husband Michael Whyte and I have been doing fieldwork at intervals since 1969, and I draw on our ethnographic data as well. Since 1994 we have been collaborating with Child Health and Development Centre on an Enhancement of Research Capacity programme funded by Danida. I gratefully acknowledge Dr Frank and other Ugandan colleagues at CHDC. 3. The use of written papers to mark relations with government or government-like formal organisations is so pervasive that one hardly notices it until some kind of reversal happens. In 2005 there was a bitter disagreement about where the town of Tororo will belong when the present district is divided into two. When a Commission report assigned the town to West Budama, all of the local council officials in the competing county, right down to village level, converged on Tororo, burned the report and handed over their rubber stamps declaring there was no longer any government since they were relinquishing the implements they used to stamp official letters, contracts and litigation documents. 4. ‘The conjunction of problematic and determinate characters in nature renders every existence, as well as every idea and human act, an experiment in fact, even though not in design. To be intelligently experimental is but to be conscious of this intersection of natural conditions so as to profit by it instead of being at its mercy.’ (Dewey 1925: 61) 5. In another study we did with HIV-positive people in Busia District, a positive woman who was trained as a village counsellor explained the difference between HIV positive and negative counsellors. When AIDS patients are really down and bedridden, she asserted, only the positive counsellors touch them and wash them. 6. The chapter in this volume by Gikonyo and colleagues recounts how study participants hoped for some form of recognition and acknowledgement when the study ended. They wanted a public and open expression of regard – so that others who had criticised them would hear. The fact that people with HIV/AIDS feel discounted made Ivan and Fred’s show of regard all the more important.

References Clifford, J. 1990. ‘Notes on (Field)notes’, in R. Sanjek (ed.), Fieldnotes: The Makings of Anthropology. Ithaca: Cornell University Press. Pp. 47-70. Dewey, J. 1925. Experience and Nature. Chicago: Open Court. ———. 1929. The Quest for Certainty: A Study of the Relation of Knowledge and Action. New York: Minton, Balch and Company. Geissler, P.W. and R.J. Prince. 2009. ‘Active Compounds and Atoms of Society: Plants, Bodies, Minds and Cultures in the Work of Kenyan Ethnobotanical Knowledge’, Social Studies of Science 39(4): 599–634. 55

Evidence, Ethos and Experiment

———. 2010. ‘The Land is Dying’: Creativity, Contingency and Conflict in Western Kenya. Oxford: Berghahn. Hoeyer, K. 2003. ‘“Science Is Really Needed – That’s All I Know”: Informed Consent and the Non-verbal Practices of Collecting Blood for Genetic Research in Northern Sweden’, New Genetics and Society 22(3): 229–44. Hunt, N.R. 1999. A Colonial Lexicon of Birth Ritual, Medicalization, and Mobility in the Congo. Durham: Duke University Press. Hutchinson, S. 1996. Nuer Dilemmas: Coping with Money, War and the State. Berkeley: University of California Press. Kaharuza, F.M. 2002. ‘Short Interpregnancy Intervals and Maternal and Child Health’, Ph.D. dissertation. Department of Epidemiology and Social Medicine, University of Aarhus. Katahoire, A. 1998. ‘Education for Life: Mothers’ Schooling and Children’s Survival in Eastern Uganda’, Ph.D. dissertation. Institute of Anthropology, University of Copenhagen. Leach, M. and I. Scoones. 2005. ‘Science and Citisenship in a Global Context’, in M. Leach, I. Scoones and B. Wynne (eds), Science and Citisens: Globalization and the Challenge of Engagement. London: Zed Books, pp. 15-38. Long, N. 2001. Development Sociology: Actor Perspectives. London: Routledge. Mogensen, H.O. 2005. ‘Finding a Path through the Health Unit: Practical Experience of Ugandan Patients’, Medical Anthropology 24: 209–36. Molyneux, C.S., N. Peshu and K. Marsh. 2005. ‘Trust and Informed Consent: Insights from Community Members on the Kenyan Coast’, Social Science and Medicine 61: 1463–73. Nama, N. and L. Swartz. 2002. ‘Ethical and Social Dilemmas in Community-based Controlled Trials in Situations of Poverty: A View from a South African Project’, Journal of Community and Applied Social Psychology 12: 286–97. Piot, C. 1999. Remotely Global: Village Modernity in West Africa. Chicago: University of Chicago Press. Pool, R. and W. Geissler. 2005. Medical Anthropology. Maidenhead: Open University Press. Sanjek, R. 1990. ‘A Vocabulary for Fieldnotes’, in R. Sanjek (ed.), Fieldnotes: The Makings of Anthropology. Ithaca: Cornell University Press. Whyte, M. and J.P. Higenyi. 1997. ‘Writing Mourners: Literacy and the Meaning of Mourning in Eastern Uganda’, 14th Annual Meeting of the African Studies Association, 13–16 November 1997. Columbus, OH. Whyte, S.R. 1991. ‘Knowledge and Power in Nyole Divination’, in P.M. Peek (ed.), African Divination Systems: Ways of Knowing. Bloomington: Indiana University Press, pp. 153–72. Whyte, S.R., S. van der Geest and A. Hardon. 2002. Social Lives of Medicines. Cambridge: Cambridge University Press. Yoder, P.S, A.R. Katahoire, D.Kyaddondo, Z. Akol, R. Bunnell and F. Kaharuza. 2006. Home-based HIV Testing and Counselling in a Survey Context in Uganda. Calverton, MD: ORC Macro.

56

Chapter 2

Can One Rely on Knowledge? Marilyn Strathern

How communications proceed between practitioners and subjects and leaders in medical research is necessarily the focus of much attention. Bad communication is as great a hindrance as good communication satisfies all kinds of ethical and social imperatives. Thus the Nuffield Council on Bioethics report on research relating to healthcare in developing countries states (NCOB 2002: 74) that methods must be devised for making sure that information will reach all members of a community. One obvious approach is to offer information as a fresh view on things, for out of the presentation of viewpoints some kind of compromise can be struck. In discussing how researchers try to convey information in a manner comprehensible to participants, the report thus suggests that the researcher should be able to see things from the perspective of the researched’s (local) views or ‘beliefs’. One [way] is to incorporate local belief systems into the process of providing information. For example, the researchers might say: ‘Although I as a doctor believe that the disease is caused by germs … I understand that you believe it is caused by a demon. I respect that fact that you have this belief and I should like you to try this medicine to remove the disease. Removing the disease is more important to us than whether we think it is caused by germs or a demon.’ … [I]n some circumstances it will be possible to strike a balance between such a stance and the harnessing of local beliefs in the interests of improving participants’ understanding of research (NCOB 2002: 74–75). Two views of beliefs or theories of causation are set side by side and then trumped by a further viewpoint, namely that one can suggest removing the disease without tackling the cause. My own question is to ask whether the very notion of viewpoint is not part of the problem rather than part of the solution. To get there, I must stand back a bit. 57

Evidence, Ethos and Experiment

Assumptions and Intentions This is an attempt to bring together the practices of the researchers and the researched.1 I do so not through claiming common ground in their actions and intentions – though much common ground is there – but through imposing a single analytical frame. The frame is to do with knowledge practices and with what may be taken for granted about the nature of knowledge. This chapter addresses generally shared techniques on the researchers’ part for dealing with ‘contexts’ of information. Specifically in relation to those techniques, I draw in materials from certain African ethnographies. What is presented, then, is not an open-ended set of contrasts but issues organised by my questions. What emerges is that if there are things about the researched that puzzle the researchers, then the researchers are quite as much a puzzle themselves. Differences between researchers and researched, then, and their divergent knowledge practices, are to be taken inclusively – they are related, as divisions are related, within the one frame. It is an anthropological truism that new ways of doing things, new skills, new knowledge, invariably dislodge practices and ideas already present. Even where one thinks one may be just adding, the chances are that one changes the relative weight given to existing values and perceptions. This applies no more truly than to the contexts in which people act. Shifts in habits of thinking become apparent when one realises that certain assumptions no longer hold. An obvious example is there in the move between carrying out medical field studies and the study of such studies. Becoming aware of the social processes of biomedical research will not just offer a new context for understanding such research but will alter an existing context, the background assumptions against which it is carried out. The realisation acts out some of what it means to be ‘a researcher’. Now contexts do not just happen to get displaced. Much of what I call Euro-American knowledge making exists in researchers’ (and others’) deliberate attempts to actively displace one set of assumptions by another (Strathern 1995), hence the reference to technique. Innovation or enhancement is demonstrated in the different perceptions this displacement produces. That is why the UK Nuffield Council on Bioethics (2002) made a special case of biomedical research when it is carried out in developing countries, for conducting trials in a context where research is a generally accepted practice for the advancement of medicine becomes something else when ‘the context’ is a participating population whose general expectations relate to 58

Can One Rely on Knowledge?

healthcare and wellbeing. Over time, the Council drew attention to a changing context of its own enquiry, namely the increasing push to develop international bioethical protocols.2 Here a deliberate shift of focus is revealed in the ability to switch from one context to another – a result may well be a sense of innovation that springs from showing how the fresh context alters an existing terrain. That one can shift knowingly across contexts becomes part of the knowledge that exists for dealing with them. These common techniques lie behind disquisitions on Euro-American perspectivalism.3 Whereas a paradigm may be thought of as an unacknowledged or only partly acknowledged background to an investigation (e.g. combinations of scientific and clinical modes of reasoning lying behind medical research), contexts draw attention to themselves. They offer reference points for understanding, and thus ‘make sense’ of the subject in focus. As such they constitute the focus as coming from a particular viewpoint, and thus as a perspective. Researchers are familiar with this when they grasp a perspective as the point of view of a social agent, the researcher (say) as opposed to participants in research; a person’s point of view invariably, and not always accurately, is assumed to chime with his or her interests in the matter. Such techniques for drawing information together generate specific forms of knowledge. Contexts and the way they make researchers aware of perspectives (and vice versa) offer highly significant, if thoroughly routine, conceptual tools for apprehending complex situations. Thus they enable researchers to juxtapose, almost without thinking about it, the different rationales between the distinctive aims of biomedical research with aims relating to immediately better healthcare for a particular population. A question arises. If these conceptual operations with context and perspective amount to a set of techniques for ordering and transmitting knowledge that seems integral to (Euro-American) scientific practice, among other things, does not it too have its own background? Should one not be thinking of it as embedded in a specific social and cultural context? And, if so, how much does that context embrace? Doumbo (2005) has drawn a striking parallel between the kind of care bestowed on patients by his grandfather, a ‘traditional practitioner’, and that of the French physician who was his mentor on ward rounds in the Bamako National Hospital in Mali. This does not mean the vehicle for such solicitude was the same. Verran (2001) noted a shared solicitude in teaching practices between Yoruba trainee school teachers and their mentors at Ife’s Institute of Education. In the latter case, which involved explaining 59

Evidence, Ethos and Experiment

mathematical concepts, we are explicitly told that there were subtle yet radical differences in the techniques that trainees and mentors used.4 The aims might be comparable, for both trainees and mentors aspired to teach, but the means or tools not so. Should researchers be alert, then, to differences of means in the very communication of information that is central to contemporary good practice in medical research? It would require them to use as an analytical technique the same concept of context that they are investigating, that is, to put ‘context’ into context. In being true to knowledge techniques habitual in Euro-American education and science, I suggest that it becomes possible to think of situations where the idea of context has no purchase, and where approaching diverse interests through imagining them as so many perspectives will not do. By way of explication, I draw – all too briefly – on anthropological understandings of some African materials, as well as from elsewhere. The argument is built up from much that already exists in the anthropological repertoire, and from theoretical work done on this material; it propels certain analyses of persons, powers and bodies into the centre of the debates that concern this volume. The model developed here remains necessarily generalised. But let me add that the contrast I shall set up between different ways of knowing would not be so bold and obvious if there were no puzzles still to be worked on in the interactions of researchers and researched (Yoder 1997; Molyneux et al. 2004, 2005; Geissler 2005).5

Viewpoints and Perspectives Taking different viewpoints into account is such a standard assumption in the ethics and politics of consultation it seems to flow seamlessly into equally standard assumptions about involving people in research.6 On the one hand the researcher needs to explain what he or she is about. On the other hand there must be some measure of understanding on the part of the participant, who becomes ‘informed’ about a procedure, acquiring new knowledge on the assumption that that new comprehension will displace or at least work beside old ones. For many medical researchers, I hypothecate, the issue is not just the desire to transmit knowledge but the idea that knowledge will inform practice. So they might imagine that someone who is told why a blood sample is being taken is more likely to be willing to accede than someone faced with an incomprehensible request. The reverse also holds, that knowledge is sought in order to tell one how to act. In so far as field trials are inter60

Can One Rely on Knowledge?

vention-focused (looking at treatment efficacy), other kinds of research (clinical/microbiological/epidemiological) into the relevant medical condition would have occurred at a prior stage (Helen Lambert, pers. comm.). Field trials that attempt to ascertain the effectiveness of treatment in something approaching a ‘real-life’ situation are thus already putting knowledge into action and are designed in turn to acquire knowledge about the effects of action itself (some kind of treatment intervention) so as to inform further action (a change in health policy, new clinical practice guidelines). A new perspective makes the new aim intelligible; an infinite number of such shifts may seem possible as new information becomes available. In the same way, we have seen the need to produce ethical principles or attend to internationally applicable guidelines becoming a source of new knowledge, a fresh context, for researchers. Perspectives are also created by the scope of the knowledge someone has, and we expect experts to apply their expertise. Indeed the proliferation of expertise demanded by modern biomedicine is accommodated in the view that bringing together numerous sources of expertise to solve a problem is like bringing together so many different viewpoints. However, it is highly likely that there will be more than the expertise at stake – there may well be different interests. Indeed the representation of diverse interests in reaching a decision may run parallel to diverse expertise brought to bear on a technical problem. An example from a recent consultative review in the UK makes the point: interests are here imagined as so many viewpoints or perspectives. When the UK Working Group on Human Remains in Museum Collections (DCMS 2003) was preparing its report,7 it heard commentaries both from scientific and anthropological experts, and from representatives of various ‘communities’ who wanted to or were asked to put across their view of the matter. The matter was to draw up guidelines to assist UK legislation for the tenure and repatriation of human remains currently being held in museums. Particular interest had been generated by requests from Australian Aboriginal activists. The remains ranged from whole skeletons to samples of tissue and specimens of bone and hair, in one way or another all comprising body parts. Scientific interest in such materials was justified by the medical information the remains were capable of yielding as well as by the information they held about human evolution. It was feared that knowledge would be lost with repatriation, since that might lead to their destruction or decay. Such scientific knowledge was potentially useful, and it was envisaged useful to medicine in particular, a sense much enlarged by 61

Evidence, Ethos and Experiment

the recontextualisation of interest in the materials afforded by molecular biology and developments in genetics. No-one could foresee what new contexts might, at some point in the future, render such materials valuable all over again. Statements from indigenous or first-nations people in Australia, New Zealand and North America were put side by side with the scientific view. Indeed the report was constructed in terms of a debate built up out of diverse discussions, as though there were two main ‘sides’, each with its viewpoint or perspective.8 The assumption was that either side could persuade the other to see the force of the argument being put forward. This was the report’s governing context. One or other viewpoint, or a compromise position, might come to inform the action to be taken. There was some sense that experts and representatives (in written or verbal statements) were making different kinds of knowledge claims, that scientists were largely experts representing the interests of universal knowledge about the human condition while the first-nations peoples were representatives of minority interests, expert in their own particular traditions or cultures. Nonetheless the discussion reflected in the final report ultimately framed these as perspectives that could be compared. Thus in suggesting that one way forward might be independently supervised resolution procedures, the report notes among the benefits of this proposal is that it would ‘foster a mood of understanding among parties’ (DCMS 2003: 157). Yet this is in itself to take the viewpoint of one side, the Working Group as an organ of government trying to reach some kind of agreement. Not all the first-nations representatives shared the grounding assumption here. From what they said, it appeared that the difference between the two sides could not be reduced to a difference of viewpoint or perspective. It could not be reduced to the idea that with discussion and information they would appreciate the context from which ‘the scientists’ were operating, and shift their own viewpoint accordingly. Or that informing the Working Group of the context from which they, the first nations people, were operating would offer the basis of a compromise agreement. Some of this was expressed in terms of refusal to acknowledge that their ancestors were in any sense scientific specimens. The inference I draw is that to them the difference between the two parties was not one of perspective, and was not be understood by reference to context. To take the case of some of the Australian Aboriginal representations, the point was that the Aborigines were related to their ancestors, the scientists were not. This made the Aborigines (relatives) different kinds of 62

Can One Rely on Knowledge?

people from the scientists (strangers).9 They did not inhabit a position from which an ancestor could also be a specimen. For, as different kinds of people, their relationship to the human remains gave the two parties claims of a quite different order. The one could not just put itself in the other’s shoes. To the Aborigines, there was nothing more they needed to know about themselves in order to press their claims, only the business of proving how they were related, through dance, song and other evidences of entitlement. Knowing the conduct and meaning of such performances would, in turn, only be effective when deployed by those with the right to use them. That effectiveness could not be transferred through acquiring ‘knowledge’ as such. Ultimately, it was not a matter of knowing but of being, and of being party to a particular kind of relationship. Here in microcosm is a modelling of ways in which people’s understandings or states of being bypass one another. Body or parts of the body appear more than a vehicle for the relations being claimed – by the Aboriginal representatives in their stance towards the Australian and UK governments and towards what they knew of scientific research. The apparently lifeless remains were living signs of their own past embodiments. If similar questions are raised in respect of living bodies, there may well be implications for medical research.

Sceptics and Witnesses Let me put my proposition in the strongest possible terms.10 Suppose there were people who did not invest in epistemology in the way that permeates Euro-American science and much general cultural practice besides. Suppose they treated knowledge otherwise, and not as a matter of expertise and interest that distinguishes people, because (to paraphrase Viveiros de Castro 2003) basically everyone has the same kind of knowledge. For instance, everyone learns from other people, and it is the relative power of and position of people, their relationships and state of being, that distinguishes them. Knowledge can thus become a sign of distinctiveness, as when initiates are put through a learning situation. At the same time, you cannot effectively use knowledge inappropriate to your condition, status or relations with others. You can move between different states of relationship or even anticipate the states of others in their reactions to you. But the knowledge at your disposal, or the new knowledge added to old, is the sign of that state not the substance of it. 63

Evidence, Ethos and Experiment

It may well be that knowledge attendant on status or state of being in this way is differentiated from a mundane flow of information (Kalinoe 2004), but it does not follow that information is self-validating. To the extent that people act from their positions in relation to others, knowledge is not necessarily in and of itself a direct guide to action. Certainly, if I were in Melanesia, I would recall the scepticism with which men and women receive much information and their insistence that they will not know how to act unless they have seen for themselves. This is especially true of their dealings with persons, where they rely on witnessing the outcomes of other people’s intentions rather than on what they are told. When it comes to being given information about medical conditions and their treatment, the researcher might ask, for example, about the kinds of cues people look for in those dispensing it. Or possibly find that on one occasion instruction needs to come as an order or command, that is, be presented as a matter of bodily discipline, while on another people need to be left alone to get on with things.11 The Nuffield Council on Bioethics (2002: 73) notes situations where a physician is supposed to give advice rather than information. Commentators often try to grasp the social complexity involved by observing that it may be necessary to consult not just potential participants but ‘community leaders’ (NCOB 2002: 75). Here Duombo (2005) points to the complexity that lies in the fact that individual and community consent are likely to be part of the same process. It would be surprising to find people who were not aware of innovations and new possibilities in treating disease, and ready to try anything plausible. The question is what they would take as proper witness to its effects. For medical trials or outreach programmes, what are the implications of working outside a framework where epistemology is determining? Something more is likely to be required than appealing to the participant to adopt the researcher’s viewpoint or to put ‘belief ’ into abeyance. And presumably one cannot appeal to a switch of perspectives (‘let us agree to differ and just focus on the disease’) as an impetus to action. Suppose, instead, the problems begin with the process of providing information as such. If one is outside the framework of shifting perspectives and re-making contexts, the possibility of compromise as a compromise between viewpoints may not even be conceivable. It would not be sufficient to rely on the power of knowledge, to rely on displacement by context, so that one perspective becomes replaced by another. This is not just because words lack power, but because the epistemological techniques that make another perspective persuasive are not in place. So what might be in place? 64

Can One Rely on Knowledge?

Bodies and Relations What did it mean to assert that the Australian Aborigines spokesmen regarded themselves as different kinds of people from museum scientists? Perhaps we can follow Viveiros de Castro’s extensive arguments for Amazonia (Viveiros de Castro 1998; Vilaça 2005). Here, it has been said that people are different insofar as they exist in different worlds, and the different worlds are signalled by the bodies they inhabit. Specific bodies, capable of specific operations and effects, capacities, powers, see other bodies in relation to themselves. Animals have their own embodiment and see human bodies in specific ways. Let me stress that ‘world’ is not a metaphor for context or perspective, but implies a physical habitat and in the Amazonian version the body that goes with such a habitation. Movement between such worlds is only possible under the right bodily regimen, through visions, discipline and so forth. Researchers might remind themselves, by contrast, of the Euro-American scientific view that holds that the physical world is continuous, and coterminous with the universe, the same for everyone, the only shortfall being in knowledge of it. So diverse experiences are taken as a matter of diverse perspectives on it. Mol (2002) provides a wonderful account of the ontology of this perspectivalism, in which she shows how the diverse manifestations of objects in the world are produced by constantly shifting contexts of knowledge. Take, for example, the contexts in which a disease becomes evident through the instruments and methods by which its effects are recorded. The accompanying assumption is that each such context is simply part of a larger whole, so all the perspectives brought to bear on a problem relate to one thing. People manage, she says, to create a singular world out of the diverse ones of their behaviour and experience. By ‘world’ she means a contextually produced habitat, an epistemic object. What makes the account so interesting is that she demonstrates just how the ‘one world’ (‘one nature’ [Viveiros de Castro 1998]) of Euro-American epistemology is held together. In this view, all persons – ‘human beings’ – are in physical (‘biological’) terms inhabitants of this one world, and to that extent similar to one another. Mol’s innovation is to deny both diversity and unity any primordial naturalism and to show how they are ‘performed’ in the way people put their knowledge together. The inhabitants of this Euro-American world, researchers for example, live with a unitary vision that is made to work against a natural plurality, a one world composed of many elements. As a 65

Evidence, Ethos and Experiment

philosopher and ethnographer commenting on this world, Mol (2002: 7) sees entities created by overlapping and intersecting fields (contexts) of knowledge practice that have a character she deliberately calls multiple. The term ‘multiple’ is not trivial; it implies entities conjoined and disjunct from one another in ways that cannot be reduced to the pluralities of parts gatherable into wholes. It leads her to make a radical distinction between perspective and context. This is crucial to her critique of the (pluralist) perspectivalism that implies the priority of a unitary world on which everyone has their own view. ‘Context’ by contrast carries something of the force of an analytical tool in her account. Focussing on contexts – overlapping, complex, multiple – draws attention to the way the environs (‘world’) of an object are summoned along with the coming into existence of the object. In one respect, however, her account seems congruent with the widespread social science view that regards social and cultural worlds as produced out of the contexts that people create for themselves (Dilley 1999). For, when it comes to persons, her assumption seems to be that we are dealing with human beings who, in all inhabiting this multiple world, are thus similar to one another. It remains for Law (2004) to briefly raise the question, and following Verran (1998) he does so in talking about Aboriginal land claims, about persons existing in different worlds qua persons. It is important to his rejection of the idea of ontological universalism. Nonetheless, he does not take this as far as some of his anthropological colleagues might, where the difference between perspectives and contexts pales beside the more radical disjunctions they have to make. If persons differentiate themselves from one another in varying ways, then an observer looking for a (local) counterpart to switching context might find it in the ability to switch between relations. In orienting oneself towards one set of persons rather than another, one may well become a different kind of being.12 To take an example from an area I know (Mt Hagen, Papua New Guinea), when a sister becomes a wife, or a wife thinks of herself as a sister, then a woman’s relationship to her husband, her brothers, to her father’s land, changes. Thus the way mother’s kin make bodily claims on a person can be very different from the effects that relations with father’s kin have. Being a sister or a wife is not so much a matter of context (perspective in the perspectivalist sense) as of the way in which a woman’s health and bodily strength is sustained by her relations with others. She will look to her husband for preparing the land that yields daily sustenance, in the food grown on it that she eats, and that will help form the children she bears – what nourishes her will nourish them too. She is a conduit for 66

Can One Rely on Knowledge?

the nurture than flows from the father to his child. But from the perspective (as in Amazonian perspectivism) of her natal kin what nourishes her will find its terminus in her. She cannot pass it on; her body blocks it. Rather than a conduit, for her natal kin her body is like an adornment for/extension of/a valuable from her own clan that is loaned to another for a while, but in the end must be returned to its origin. Now what the observer might refer to as relations, the actor might refer to as bodies. Any one body is enmeshed in a network of bodies, and communication is possible between them, often taking the form of flows of bodily substance or energies. What is outside the body can only come, in this sense, from other bodies. And if relations are sustained by flows and interchanges, so too are bodies. And that includes bodies in health or ill health, emanating wellbeing or affliction, and free of or prone to disease.

Communicating between Bodies How does all this bear on the social relations set up in the course of medical research in Africa? Molyneux (conference presentation) draws attention to the distance that can arise between medical researchers, who think they are doing something for the community, and the researched who interpret the interventions as something that has to be relevant to themselves as individuals. But there is another social dimension to consider, that of neither community nor individual but of persons in relations. The point is made by Reynolds Whyte (this volume), though I take rather a different route to it. Again, let me stand back a bit. The argument constructed here from science studies (Law, Mol) and from materials outside Africa (Aboriginal Australia, Amazonia, Papua New Guinea) would not be unfamiliar to certain ethnographers of sub-Saharan Africa. They underline the suggestion that when we look at relations between persons we imagine how they would translate into relations between bodies.13 Conversely, the condition that bodies are in may well imply relationships in a certain condition. Anthropologists have commented on something similar to this phenomenon for a long time, especially in respect of the way sickness is treated. A point to add is that it also has some consequence for the purchase that conveying information is going to have, and for what Euro-Americans understand as the transmission of knowledge.14 It is equally likely to have some purchase on the kinds of relations that participants, the researched, think they are establishing with researchers. 67

Evidence, Ethos and Experiment

In his studies in Kabre (Togo, West Africa; again see Reynolds Whyte, this volume), Piot (2005) describes the interdigitation of men’s and women’s bodily capacities in the household. So, for example, a man’s blood, that which allows him to cultivate and to produce children (the Kabre terms for blood and sperm are the same) comes largely, Kabre say, from the consumption of sorghum beer, and beer is the prototypical female product. Conversely, a woman’s strength and womb – which allow her to work and produce children – come from the consumption of porridge, which is produced from the prototypical male product (field crops). (2005: 69) In other words, the wife’s activities are already ‘contained’ within the husband’s, and vice versa. It is not just that the division of labour renders the tasks interdependent but the nature of the tasks (expenditure of bodily effort, ingestion of food from the soil) means that their bodies are interdependent. Indeed Kabre say that what each produces belongs to the other: the food the husband produces is ‘owned’ by the wife and the children a woman produces are ‘owned’ by her husband. He calls this a state of reciprocal incorporation, and the same form of incorporation is repeated over and over again.15 One important locus is found in the way opposite states of being incorporate (or mutually constitute) each other ‘at the level of the community’. Houses in a Kabre community are divided into one of two clans, called male and female. The clan is a largely ritual entity, responsible for age-grade ceremonies. Now amongst the male spirits of the male clan there also lives a female spirit for whom female ceremonies are performed, and among the female spirits of the female clan is a male one. Each clan and its powers thus ‘contains’ another within it. It is no surprise that the ethnographer should talk of persons having multiple identities, such that a man from the female clan will act as a female on many occasions, as male on others, while his in-married wife (who is likely to come from the male clan) can appear male. There is a hierarchical ordering here, the highest ranking ritual figure (or ‘leader’) being male, though this important man is also regarded as a ‘child’, and on occasion treated as a female, encompassing both elements (both genders; both seniority and juniority) within himself. These details show that the notion of interdependence between bodies is part of a much wider system through which social identities of all kinds participate in one another. I offer them because of this and because of the 68

Can One Rely on Knowledge?

model that the anthropologist deploys. In organising his account, Piot draws on an analogy with fractal mathematics. He is interested in the way in which forms repeat themselves throughout the social system while also being embedded in one another; thus the household with its internal division into male and female activities is contained within the community with its male and female clans. The result is a series of self-similar,16 nonlinear symbolic orders, a ‘fractal iteration’. Fractals briefly enter Law’s and Mol’s work, the science studies theorists whose arguments have been so illuminating for my account. Objects have complex relations, says Mol (2002: 149), and she means objects as they are operated on in practice. Thus, literally, in the course of an operation involving arteries, a patient’s blood vessels ordinarily ‘contained’ within the body in becoming the focus of attention become much larger than the body as a whole.17 When Law (2004: 62) wants to make clear Mol’s ensuing proposition, that the multiple body is not a fragmented (plural) one, he turns to the vocabulary of the fractional or fractal. However, the (EuroAmerican) ‘multiplicity’ of the science studies argument is not performed or enacted by the same conceptual techniques as inform gender divisions in Kabre, and is argued much more comfortably in respect of things (‘objects’) in the world rather than persons. The latter rest on the operation of division or partition between the sexes, such that one may also encompass ‘parts’ (taken) from the other. Wagner’s (1991) formula offers an abstract version of this technique, in terms of persons and relations. The fractal person, he says, is the person with relations (to others) integral to it: it exists in more than one dimension but less than two.18 Piot’s is one of two chapters on African materials in a volume on social anthropology and chaos (or complexity) theory.19 The second chapter, dealing with East Africa (Rwanda), shows how some rather similar ideas illuminate ‘communication’ between bodies (relations, persons). ‘The Rwandan person is fractal … perennially incomplete. He or she is ever involved in the process of being added to, built upon, and produced by the gifts of others’ (Taylor 2005: 144). Communication is imagined in terms of flows and blockages of substances. So, for instance, persons concerned to maintain ‘paths’ of communication between themselves enter into exchange relations, so ensuring the movement of people (e.g. women in marriage) and cattle (e.g. signifying alliances). Analogy is made with the flow of substances, especially liquids20 that have a fertilising or productive effect. Thus the principal leader (king) in pre-colonial times, who controlled the passage of fluids, kept his own body ‘open’ through the quantities of milk, honey 69

Evidence, Ethos and Experiment

and mead that he took in and excreted; the conduit between earth and sky was imprinted on his own physiology, and the health of the nation depended on him. The king could also, by withholding gifts and the like, block communication, impoverish as well as enrich his subjects. What was true for the wellbeing of the polity was also true of the individual body. Fluids, says Taylor (2005: 145) are emphasised in popular medicine. When fluids are lacking or blocked within the body, or else haemorrhage out uncontrollably, one suspects illness or sorcery. He comments on a study of healers and patients where flow/blockage symbolism suffused the narratives on both sides. This was separate from whatever went on when patients consulted, as they also did, biomedical and other practitioners who operated on a different intellectual basis. Now what was happening to bodies was also happening to relations between persons – not only by force of analogy, as the observer might have it, but by the substantial connections through which (adopting Piot’s vocabulary) the one contained or (in Wagner’s phrasing) was a dimension of the other.

Ending There is nothing new in these examples, but I hope that in good researcher mode I have contextualised them freshly. We might think of body practices as knowledge practices of a kind. And what springs to mind is the thought that in the participants’ (the researcheds’) expectations the quality of relations between them and the researcher may have significant bearing on the wellbeing that the participants experience. What is clear is that the forms of bodily communication described here do not depend on influencing people through transforming the perspectives from which they see or through transforming their understanding of contexts. These communications are ontological rather than epistemic operations, for they affect people’s conditions of being. Knowledge is a vector or aid, means rather than end. Of course one of the target audiences of the medical researcher’s efforts, the medical practitioner or clinician, is also interested in deploying knowledge as a means to the ends of effective treatment. And he or she will do so through magnifying the body as the site of intervention. But magnification (or telescoping) is itself imagined as a matter of perspective or context, diverse viewpoints on the (one) phenomenal world. The kinds of Kabre or Rwandan handlings of the body noted here instead condense or disperse the body’s elements, effects and efforts 70

Can One Rely on Knowledge?

from/to a network of persons; it is the condition of the ensuing relations that results in a person’s health and sickness, which in turn affects others. Knowledge of the state, good or bad, of those relations is crucial; yet the knowledge is no basis for action without the substantial bodily conditions that make the person into the kind of being he or she is. Let me state that this is not a theoretical account. There is therefore no need to treat it as ‘theory’ requiring translation into practical terms, or to derive from it protocols or guidance in order to turn it into knowledge from we can act. For although it draws on theory, my account is offered as an empirical resource. It is already a description of practices and assumptions (Leach 2006). Being aware of such practices and assumptions may be helpful for those moments when on the one hand researchers and fieldworkers, and on the other participants in medical research, the researched, are puzzled by what is going on, are confused when expectations are not met. An apposite example is the conclusion to a study of evaluation of health communication programmes in developing countries that indicated it was rare to find a close correspondence between changes in knowledge and changes in behaviour (Yoder 1997: 132). One cannot rely on knowledge! Perhaps one can, with more certainty, rely on relations. I hope that trying to put these within one frame will have thrown up something about the nature of the puzzles on both sides. And each should be equally puzzling, and equally interesting, to the ethnographer. This is not unimportant. The social relations that spring up around field projects may be rather more than the incidental and inevitable by-product of contacts made in the course of research. They could well be vital to it. The study of research practices is not about some peripheral part of the research process but about the enabling and disabling conditions that help and hinder the flow of communications.

Acknowledgements I am most grateful to Wenzel Geissler for the invitation to be a participant at the conference, for his and Sassy Molyneux’s inspiration, and to Kevin Marsh for the hospitality of the KEMRI Research Laboratory at Kilifi. A companion paper drawing on much the same material was given to a RAI/Leverhulme workshop in 2005, under the title ‘Using Bodies to Communicate’. (This has since been published, in Social Bodies, edited by Helen Lambert and Maryon McDonald, New York: Berghahn Books, 2009.) 71

Evidence, Ethos and Experiment

Notes 1. I use the clumsy term ‘researcher’ to cover medical research workers and fieldworkers (Mbaabu et al., conference presentation) as well as scientists or social scientists or scholars. I take research as a phenomenon peculiar to Euro-American knowledge practices. So ‘researchers’ are recipients of certain types of education and training, wherever they happen to come from as individuals. The ‘researched’ is a shorthand for subjects or participants in medical research, who may or may not share the ends of the researcher’s endeavours but ordinarily will not share the training, that is, the means. 2. The shift can be seen between the first report (NCOB 2002), addressed initially to external sponsors of medical research, and the follow-up discussion paper (NCOB 2005), which situated its review within a burgeoning international field of guidance and guidelines. On the concomitant transformation of environmental science research policy into an international arena of ‘complex processes of global-local connection’, see Fairhead and Leach (2003: 239). 3. I adopt Law’s (2004) term to distinguish it from that of ‘perspectivism’ deployed, for example by Viveiros de Castro (1998), for Amazonia. Perspectivalism creates the condition for ‘merographic relations’ (Strathern 1992); such shifts work insofar as contexts themselves are distinguishable from one another as relatively stable assemblages (Law 2004; Ong and Collier 2005). 4. In the light of what follows, it is worth noting that she characterizes ‘Yoruba logic’ as concerned with whole/part organization where English language communities (including scientific ones) emerge with an ordering logic of one/many (Verran 2001: 220). 5. One might cite the rumours about researchers stealing blood or making illicit profits mentioned by several contributors to the conference, or else the felt need to protect sponsors, researchers and participants alike through ethical procedures (cf. Molyneux et al. 2005). 6. See the advice given to Mol (2002: 11) by a social scientist anxious that she realize how many perspectives she has to take into account in her study of atherosclerosis. Fairhead and Leach (2003: 233–38) describe the ‘participatory process’ and ‘citizens’ platforms’ in environmental science policy put into practice in Guinea and Trinidad. 7. I was a member of the Working Group; an independent advisory committee under the aegis of the Department for Culture, Media and Sport; I confine my remarks here to the published report. 8. The dual structure was deliberate, and is set out at the beginning of the report (DCMS [2003: 29] talks of ‘an irreconcilable conflict between “scientists” and “indigenous people”’, and of ‘polarized views’).The last chapter before the recommendations is called ‘Resolving the conflict’. This polarization meant, among other things, that the voices of indigenous scientists were taken as some kind of compromise position between opposed camps. 72

Can One Rely on Knowledge?

9. Apropos the dual structure (above), the sentiment of irreconcilable conflict was nearer this position than polarized ‘views’. The report gives several verbatim comments on this viewpoint, such as: ‘It is our direct ancestors that are being experimented on [original emphasis]’, ‘We went to the Natural History Museum to see our ancestors and we were told that we cannot see them. For us it is like going to see somebody in hospital. To us the people in museums are not dead, they are living’ (DCMS 2003: 55). 10. I am grateful to Almut Schneider who drew me to this conclusion. I put the extreme view partly because of the profound disjunction between the ‘constructivist research paradigm’ of the anthropologist and the framework of ‘logical empiricism’ that presumes a cause and effect relationship between knowledge and behaviour (Coreil 1997: 253). See Yoder (1997: 137), citing Rubel and Hass (1990). I am grateful to Helen Lambert for these references. 11. Law (2004: 80–81) refers to a study of a particularly successful technological innovation in Zimbabwe, where a water pump manufactured as a kit in Harare is assembled and maintained by villagers; government policy demands the creation of village collectives to supervise it. However, as they have to replace worn parts villagers do not just add their own inventions, but also re-invent the social basis – they decide what local arrangements will work. The original plan disappears both technically and socially, and thus [the reason Law tells the tale] as the water pump alters shape it creates new contexts for itself. 12. I have argued elsewhere that, within a Melanesian kinship universe, acting out one set of relations means not only eclipsing another set, but altering the latter relations by the new standpoint. (When a son switches to being a sister’s son, his father becomes his mother’s brother’s in-law etc.) A new configuration of kin relations appears. 13. Law (2004: 76) has no problem with respect to ‘Euro-American’ bodies in saying that disease is located both within the body and beyond it (in the way others react to it, how it becomes an object of policy measures, etc.) – but without the sense of investment in relations with other persons that I would emphasize here. 14. In talking about transmission among persons, Leach (2006) observes that what an anthropologist might gloss as knowledge the people of the New Guinea Rai Coast translate as the substance of kinship relations. See Geissler and Prince (n.d., original emphasis) apropos western Kenya: ‘medicinal knowledge, the transformational capacity that knowledge entails, and the bodily transformation it engenders, are not so much located within entities as between them’. 15. The house itself is at once male (outside) and female (inside), each space incorporating cross-sex activity: a man’s granary is located at the heart of the inner female household while a woman’s grinding hut for preparing cereals is located in the male area outside the homestead. 16. ‘[R]ecursive self-similarity – the folding in of opposition upon opposition’ (Piot 2005: 73). 73

Evidence, Ethos and Experiment

17. Nor does the patient ‘contain’ the body, which looms larger than the person’s diverse social identities, for these can vanish or be overshadowed by (say) the anxiety of relatives as the only part of the outside world to which the patient must respond. This is not quite the same as simply saying that there are numerous perspectives that can be brought to bear. For an early exploration of these ideas that take the authors to medical practices in Africa, see Mol and Law (1994). 18. Neither singular nor plural, its dimensionality cannot be expressed in whole numbers (1991: 162). ‘A fractal person is never a unit standing in relation to an aggregate ... People exist reproductively by being “carried” as part of another’ (1991: 163). 19. Taylor cites Vincke, writing on Rwandan sorcery in 1991, who says that the [Rwandan] person has to be understood as constructed according to a fractal model (2005: 144). In relation to kingly rituals, Taylor describes what is known from nineteenth Rwandan political thought. Other Africanists interested in recursive modelling include Eglash (1997). 20. Liquids include bodily fluids such as breast milk, semen and blood, and aliments such as milk, honey and beer. ‘Liquids are employed symbolically as metonyms, parts of persons standing for and embodying the whole … or as metaphors mediating between separate semantic domains’ (Taylor 2005: 144).

References Coreil, J. 1997. ‘More Thoughts on Negotiating Relevance’, Special issue on Knowledge and Practice in International Health, Medical Anthropology Quarterly 11(N.S.): 252–55. DCMS. 2003. The Report of the Working Group on Human Remains. London: Department of Culture, Media and Sport. Dilley, R. (ed.). 1999. The Problem of Context. Oxford: Berghahn Books. Doumbo, O. 2005. ‘It Takes a Village: Medical Research and Ethics in Mali’, Science, 307(5710): 679–81. Eglash, R. 1997. ‘Bamana Sand Divination’, American Anthropologist 99: 112–22. Fairhead, J. and M. Leach. 2003. Science, Society and Power: Environmental Knowledge and Policy in West Africa and the Caribbean. Cambridge: Cambridge University Press. Geissler, W. 2005. ‘Kachinja Are Coming!’ Encounters around Medical Research Work in a Kenyan Village’, Africa 75: 173–202. Geissler, W. and R. Prince. n.d. ‘Plants, Bodies, Minds and Cultures in the Work of Ethnobotanical Knowledge: Examples from Luoland, Kenya’ (LSTMH and University of Copenhagen). Kalinoe, L. 2004. ‘Traditional Knowledge and Legal Options for the Regulation of Intellectual and Cultural Property in Papua New Guinea’, in E. Hirsch and M. Strathern (eds), Transactions and Creations: Property Debates and the Stimulus of Melanesia. Oxford: Berghahn. Law, J. 2004. After Method: Mess in Social Science Research. London: Routledge. 74

Can One Rely on Knowledge?

Leach, J. 2006. ‘Out of Proportion? Anthropological Descriptions of Power, Regeneration, and Scale on the Rai Coast of Papua New Guinea’, in P. Collins and S. Coleman (eds), Locating the Field: Space, Place and Context in Anthropology. Oxford: Berg. Mol, A. 2002. The Body Multiple: Ontology in Medical Practice. Durham and London: Duke University Press. Mol, A. and J. Law. 1994. ‘Regions, Networks and Fluids: Anaemia and Social Typology’, Social Studies of Science 24: 641–71. Molyneux, C.S., N. Peshu and K. Marsh. 2004. ‘Understanding of Informed Consent in a Low Income Setting: Three Case Studies from the Kenyan Coast’, Social Science and Medicine 59: 2547–59. ———. 2005. ‘Trust and Informed Consent: Insights from Community Members on the Kenyan Coast’, Social Science and Medicine 61: 1463–73. NCOB (Nuffield Council on Bioethics). 2002. The Ethics of Research Related to Healthcare in Developing Countries. London: Nuffield Council on Bioethics. ———. 2005. The Ethics of Research Related to Healthcare in Developing Countries: A Follow-up Discussion Paper. London: Nuffield Council on Bioethics. Ong, A. and S. Collier (eds). 2005. Global Assemblages: Technology, Politics and Ethics as Anthropological Problems. Oxford: Blackwell. Piot, C. 2005. ‘Fractal Figurations: Homologies and Hierarchies in Kabre Culture’, in M. Mosko and F. Damon (eds), On the Order of ‘Chaos’: Social Anthropology and the Science of Chaos. New York: Berghahn Books. Rubel, A. and M. Hass. 1990. ‘Ethnomedicine’, in T.M. Johnson and C.F. Sargent (eds), Medical Anthropology: Contemporary Theory and Method. New York: Praeger. Strathern, M. 1992. After Nature: English Kinship in the Late Twentieth Century. Cambridge: Cambridge University Press. Strathern, M. (ed.). 1995. Shifting Contexts: Transformations in Anthropological Knowledge [ASA 1993 Decennial conference series]. London: Routledge. Taylor, C. 2005. ‘Fluids and Fractals in Rwanda: Order and Chaos’, in M. Mosko and F. Damon (eds), On the Order of ‘Chaos’: Social Anthropology and the Science of Chaos. New York: Berghahn Books. Verran, H. 1998. ‘Re-imagining Land Ownership in Australia’, Postcolonial Studies 1: 237–54. ———. 2001. Science and an African Logic. Chicago: University of Chicago Press. Vilaça, A. 2005. ‘Chronically Unstable Bodies: Reflections on Amazonian Corporalities’, Journal of the Royal Anthropological Institute (N.S.) 11: 445–64. Viveiros de Castro, E. 1998. ‘Cosmological Deixis and Amerindian Perspectivism: A view from Amazonia’, Journal of the Royal Anthropological Institute (N.S.) 4: 469–88. ———. 2003. ‘And’. After-dinner speech given at ‘Anthropology and Science’, 5th Decennial Conference of the ASA, Manchester. Manchester Papers in Social Anthropology 7. Wagner, R. 1991. ‘The Fractal Person’, in M. Godelier and M. Strathern (eds), Big Men and Great Men: Personifications of Power in Melanesia. Cambridge: Cambridge University Press Yoder, P.S. 1997. ‘Negotiating Relevance: Belief, Knowledge, and Practice in International Health Projects’, Special Issue on Knowledge and Practice in International Health, Medical Anthropology Quarterly 11(N.S.): 131–46. 75

Chapter 3

Being ‘with the Medical Research Council’: Infant Care and the Social Meanings of Cohort Membership in Gambia’s Plural Therapeutic Landscapes Melissa Leach and James Fairhead1

Introduction Contemporary discussions of medical research frequently frame it as something distinct from ‘normal’ healthcare practice. Its infrastructural and financial trappings, involving research stations, laboratories, international staff, vehicles and funding flows set apart from and contrasting with surrounding, impoverished African people-scapes, seem to reinforce the logic of such radical distinction. So, too, do many of the institutional rules and practices that surround medical research, with its bioethical codes and studyby-study protocols, information and consent procedures for participation. Yet in settings where medical research institutions are long-established, such radical distinctions may no longer hold, especially in the perspectives and experiences of the populations who, from time to time, are invited or expected to be their study subjects. Instead, medical research institutions come to take their place amidst the multiplicity of health providers – formal and informal, biomedical and naturalistic, public and private – that characterize Africa’s therapeutic landscapes. The meanings they acquire – and how people judge their activities and whether or not to engage with them – may take on a relativistic quality. People assess and experience what might be on offer from research institutions, and the social relations and contracts entailed in 77

Evidence, Ethos and Experiment

service access, in relation to other available possibilities from other providers, as well as in the broader context of available forms of citizenship. This chapter argues that understanding these relativistic social fields of judgement will be increasingly critical to comprehending people’s engagements with medical research (and practical issues that flow from this, such as the sustainability of trial participation) as research station presence in African settings lengthens, and as health providers continue to multiply (Bloom and Standing 2001). As Nguyen (infra) suggests, research participation is now often associated with forms of therapeutic citizenship that are an emergent feature of Africa’s political landscapes. The Gambia’s peri-urban fringe provides an ideal ethnographic setting to explore these issues. A small, deeply impoverished African nation, The Gambia has long been part of global flows of people and capital through its coastal tourism, its emigration and diaspora, and its key position in medical research. The British Medical Research Council (MRC)-funded laboratories at Fajara on the coast have been in operation for more than 50 years, as the UK’s major investment in medical research in Africa. While the MRC has progressively established field stations in The Gambia’s rural Divisions from which to orchestrate studies and clinical trials, its studies have also focused on populations in the urban Western Division itself. A prime site has long been the Mandinka settlement of Sukuta in Kombo St Mary district, with MRC researchers working from within the government health centre compound. In particular, Sukuta has been the location of a series of infant cohort studies. Thus as far back as 1960–62, the Sukuta Project tracked disease patterns in ninety-five children in a longitudinal study from birth to eighteen months (Marsden 1964). From 2003 the Sukuta Birth Cohort Study (MRC 2004) began to recruit babies at birth into a cohort now numbering well over 1,000 to study how the immune system deals with early infections. This chapter explores how such infant cohort studies have been incorporated into the social worlds of Gambian child-rearing, and what they have come to mean to mothers amidst the social relations and conceptual framings of infant care and of citizenship in modern peri-urban worlds. First, it introduces contemporary Sukuta and the healthcare institutions available there. Second, it locates the procedures and practices of the current Birth Cohort Study in this context, showing the forms of integration between study recruitment and procedures, and the activities of the Sukuta health centre. A particular focus here is on the discourses and practices of MRC fieldworkers in bridging the worlds of medical research and of moth78

Being 'with Medical Research Council'

ers. The paper then turns to Sukuta mothers’ perspectives and experiences to explore how, in this multi-health-provider context, mothers find ‘joining MRC’ an effective way to address key dilemmas in keeping infants well amidst practical, social and moral hazards, as well as to find forms of solidarity and citizenship lacking elsewhere. Yet being in an MRC cohort is also perceived to entail new risks and uncertainties. In relation to these, joining involves the negotiation of maternal agency within gender, compound and peer relations. Conceptually and methodologically, this paper is part of a broader comparative anthropological research programme on public engagement with vaccination, science and medical research in West Africa and the UK (Leach and Fairhead 2007). It forms part of the Gambian component of this programme (see Fairhead et al. 2004, 2005, 2006), in which Sukuta was one of two focal sites for ethnographic research. The paper is based on ethnographic fieldwork conducted by ourselves and our Gambian research partner Mary Small during March–November 2003. This involved participant observation in the community of Sukuta and at its health centre, interviews and a variety of informal conversations with MRC fieldworkers and other health providers, and fifty detailed narrative interviews with mothers, identified during participant observation, to reflect a variety of social circumstances. These narrative interviews followed a ‘child health biography’ format, to explore, in an open-ended way, the issues, concerns and practices that had arisen in seeking to keep a particular child well. Understandings and experiences of MRC emerged as part of these conversations. Issues arising in the ethnography were subsequently explored in an interviewer-administered questionnaire survey of 800 randomly sampled mothers across Western Division (see Cassell et al. 2006). While not ethnography strictu senso, the survey responses included many narrative statements which illustrate important ethnographic themes. In drawing on these as evidence for its (ethnographically grounded) arguments, the chapter also underlines the methodological point that survey data can be a source of ethnographic evidence (see also Bledsoe 2002 for a similar argument in another Gambian health domain).

Sukuta Then and Now At the time of the first Sukuta Project in the early 1960s, Sukuta was described as a semi-rural village of 1,500 inhabitants, largely comprised of Mandinka families whose settlement went back many generations. Sur79

Evidence, Ethos and Experiment

rounded by agricultural land, the villagers were predominantly farmers, with 63 per cent of mothers in the project farming rice. A road through the countryside connected the village with Bathurst (Banjul) where some mothers travelled to sell their produce (Marsden 1964: 456). Today, Sukuta has been engulfed by the urbanization that has spread across Gambia’s Western Division. High fertility and immigration both from up-country Gambia and neighbouring countries has increased its population to nearly 15,000 people. This is despite high rates of emigration to Europe and the U.S. While it retains its village heart, divided into quarters inhabited by different founding families, it has expanded to incorporate a huge variety of immigrant populations from a variety of ethnic groups from all over The Gambia and beyond. Migrants have either purchased land, and built high-walled compounds, or rent accommodation. In the process Sukuta has transformed from a rural to urban economy, and has become an economic hub attracting commuters from surrounding villages, with a growing engagement with the coastal tourist trade. Livelihood profiles there echo those in the Western Division more generally, where in our survey only 2 per cent of women identified their husbands as farmers and only 1 per cent claimed to farm themselves. People now get by in a huge diversity of ways, including vegetable gardening for the market (6 per cent of women), trade (42 per cent women, 19 per cent of men), and many small business and craft enterprises. Remittances from diasporic members contribute significantly to most families’ livelihoods. Sukuta health centre is on the edge of the old village. It was established as part of The Gambia’s strong government primary healthcare system, with its networks of health centres and health posts, a legacy of the late colonial period and early years following independence in 1965. Since this time, the country has had major periods of relative political stability, first from 1970 under President Dawda Jawara, then briefly as part of Senegambia Confederation between 1982 and 1989, and after a military coup in 1994, under Lieutenant Yahya Jammeh, first as a military and then as a democratically elected civilian government. Economically, though, the same period has seen decline as the agricultural and groundnut export economies floundered under the effects of drought and falling world market prices, and revenues from the re-export trade and the coastal tourist trade failed to keep pace with rising costs of imports. A series of economic recovery programmes since the mid-1980s have emphasized neoliberal measures aimed at encouraging entrepreneurship and market integration, often at the expense of ordinary Gambians who have suffered from the privatization of 80

Being 'with Medical Research Council'

state services, loss of land and resource tenure, and rising consumer prices. The country’s recent Vision 2020 takes these liberal ideals to the extreme, advocating a model of private sector-led development and celebrating individual entrepreneurship. In this current political-economic context, most Gambian health centres are suffering from minimal investment, resources and infrastructure. But a few are thriving through having embraced and pursued neoliberal ideals, and Sukuta is one of these. The centre has benefited over the last decade from the unusual dedication and entrepreneurial flair of one of its midwives. Largely through her personal charisma, networking and lobbying skills, she has attracted a series of foreign donors to invest in new buildings, infrastructure and equipment. Inviting tourist groups from the nearby hotels on ‘cultural tours’ of the centre has generated further donations and links with small NGOs in tourists’ home countries. The result has been an extraordinary level of investment in men’s and women’s hospital wards, and antenatal, postnatal and delivery facilities, far exceeding the usual repertoire of a health centre and stretching the current capacity of medical staff there. A Cuban doctor posted there in the 1990s has now left, and the centre is lobbying hard for qualified medical staff. Meanwhile, an Infant Welfare clinic operates there every Tuesday and Thursday, providing routine weighing, immunizations and, if necessary, treatment to babies who, according to the schedule recommended on the Child Health Cards given at birth, should be brought every month until the age of two. The MRC has also invested in the Centre, directly and indirectly. Marsden’s 1960s project recruited newborn babies seen at the government clinic and thereafter conducted tests at a special, separate clinic run once or twice a week, timed to coincide with the government clinic and operating under a tree within the health centre compound, but managed quite separately by MRC nurses. However, given the size of the current Birth Cohort Study, and regular presence of eight fieldworkers, MRC soon established a permanent office in the health centre compound. MRC nurses would help with government health activities, such as vaccinations on the twice-weekly infant welfare clinic days. By 2005 this had been replaced by a whole new building to house the MRC study, as well as an MRC-funded laboratory on site. These developments were welcomed by the health centre as further building its profile and capacities. The laboratory regularly performs analyses for health centre patients as well as for study subjects. In infrastructure and staffing, then, there is today considerable integration between MRC and government health provision activities. At the same time though, and 81

Evidence, Ethos and Experiment

as in the 1960s, children in the cohort have all their preventative, routine, and emergency healthcare taken over by MRC. This dual aspect – of integration and separation – is critical to parental views of what it means to be in a cohort, as we shall see. The therapeutic landscape outside the health centre compound involves an array of other health institutions and providers, offering diverse forms of child protection and treatment. Private clinics – ‘family doctors’ – provide a full range of health services to those wealthy families who can afford to pay. Islamic Marabouts provide various talismans – most commonly small leather pouches containing Arabic script, to be tied round a child’s wrist, waist or neck – for protection against specific forms of physical or spiritual harm, or to promote teething and walking. Herbalists – most commonly elderly women or men renowned for their personal or inherited skills – provide a variety of plant-based cures and protective washes. Medicines are on sale in several small pharmacies in Sukuta’s main street, while table sellers in its daily market offer both biomedical and naturalistic products. Making use of both these and of home-prepared herbal remedies are the many forms of child protection and treatment that parents, grandparents, other relatives and neighbours dispense to babies and children at home.

The Birth Cohort Study – Procedures, Practices and Fieldworker Views The current Birth Cohort Study aims to understand how the immune system deals with early infections, and the legacy of infection on subsequent immunity. Its core work involves tracking immune response to three early life infections that give no symptoms in small children: placental malaria, Cytomegalovirus (CMV) and Epstein Barr virus (EBV). The study recruits babies born at Sukuta health centre, and follows them monthly for the first year, and then at three monthly intervals for subsequent years (MRC 2004). The study involves placental blood and urine sampling at birth, blood sampling from the mother at six months, and from the baby once a viral infection is found (or at four months, in the case of infection during pregnancy). Because placental blood samples needed to be analysed immediately, only babies born in the health centre between 2am and 4pm on weekdays could be recruited. Communication about the study with potential participants took a variety of forms. When MRC began their work, the study team held initial 82

Being 'with Medical Research Council'

meetings with community leaders, but relied also on the spread of informal communication about the study within the well-respected health centre where they were based, and on the individual consent process. Later in the study, to stave off what the study team saw as emergent public misunderstandings and to consolidate community support, a programme of community outreach was undertaken, centring on public meetings at the health centre. A team of MRC fieldworkers is responsible for the consent process, home visits, sample collection and monitoring. They give a brief explanation while the mother is in labour, and take samples needed at birth that will be used only if consent is given. After delivery, they explain the study more fully to the mother (and father if present), verbally translating an information sheet. A woman might sign the form at once, or if she awaits consultation with her husband, he must be present in time for the sample to be sent to the laboratory. A follow-up visit is made to the mother’s compound to take urine samples, and this is used to ensure that consent is still available. Each fieldworker remains responsible for an allocated group of babies. The MRC fieldworkers occupy a crucial communicative position at the interface between the MRC and community members, both in the consent process and their everyday work of visiting Sukuta’s compounds. Most emphasized to us that they find their work extremely rewarding. They enjoy the contact with mothers and their babies – which is evident from watching their interactions during clinic sessions – and the chance to get to know ‘their’ group well. They value the opportunity to work to improve the health of children in the Gambia, both through what the MRC study itself might eventually bring but also through more tangible, immediate, everyday interactions. These include the treatment and protection through immunizations that the MRC study clinic provides, and the advice and reminders to attend the infant welfare clinic that fieldworkers give out to study subjects – and often other mothers in their compounds and neighbourhoods – during their home visits. And they value working for MRC: both as a relatively secure form of paid employment in a Gambian context where this is so rare, and, it seems, because of the sense of belonging it brings. Fieldworkers talk of their relationship with the MRC almost as a form of citizenship that brings certain rights and entitlements, as well as duties. As ‘MRC citizens’ fieldworkers find solidarity with each other, and with an institution that connects them with values they care about. Nevertheless, fieldworkers also find their work with MRC challenging and several speak of the particular difficulties they encounter. The study is 83

Evidence, Ethos and Experiment

socially sensitive for several reasons. First, it involves asymptomatic infections and thus the study is not obviously preventative or curative in an immediate sense. Second, given widespread publicity for HIV, people tend to associate ‘viruses’ and being ‘sero-positive’ with HIV, and thus there is a problem of associated stigmatization. Third, the study involves blood taking from mothers and babies. In these circumstances, fieldworkers feel that consent and continued participation hinges heavily on their explanations and social engagement. They have developed creative ways to explain the trial, and to convey more difficult or problematic concepts. These include explaining the immune system as ‘the body’s soldiers’ (baro djato la soldaro in Mandinka). They also use analogies to explain that there are many different sorts of virus (not just HIV): ‘The virus is just like an umbrella – there are many kinds, just as the Gambia is an umbrella for many kinds of Gambians, Wolof, Fula and so on’. Despite these practices, some people have declined to participate or have dropped out during the study. Fieldworkers tend to attribute this to several issues. Some attribute refusal to wealth; to ‘those who can finance their medical care and do not need the MRC benefits’. Fieldworkers suggest that the ‘family doctors’ used by such people both provide an alternative to MRC and government healthcare, and that they sometimes advise against study participation. Fieldworkers also understand refusal as misplaced worry, whether by a woman herself, by her husband or by authorities in her compound. These worries include concern that MRC takes, steals or sells blood, which will deplete a non-replenishable life force; a widespread worry about MRC’s work in the Gambia (see Fairhead et al. 2005, 2006). In these terms, fieldworkers tend to perceive those who decline as people who have not yet understood the scientific reasons for blood samples, or the small amounts taken and lack of harm they cause. Conversations around this fade into discussion of wider problems of blood donation, and the reticence of many Gambians to do this even for close family members due to the harm it is believed to cause to the donor. Worries also encompass the idea that MRC is promoting family planning, of particular concern to some husbands; or that MRC practices include inappropriate nude examinations. Fieldworkers attribute some of these concerns to particular ethnic groups: for instance they claim that Fulas have been particularly difficult to recruit into this study, as they harbour particularly strong concerns about blood-taking. As we turn now to the perspectives of Sukuta’s mothers themselves, it will become clear that while these fieldworkers’ perceptions of reasons for trial participation and refusal are accurate to some degree, they are also 84

Being 'with Medical Research Council'

narrowly framed within the world of the study itself. They thus do not encompass the broader social contexts which, it will become clear, are important in framing mothers’ reflections on whether or not to ‘join MRC’.

Raising an Infant in an Uncertain World Mothers’ reflections on joining the MRC emerge from a social context in which giving birth to and raising an infant is considered to be fraught with hazards (see also Bledsoe 2002). Some of these hazards relate to cultural understandings and experiences of infant growth and health, and the various forces that can set it back. Others arise from the broader context of poverty and relative deprivation in which many families live. These dimensions are related, and, as we shall see in the next section, both are important in framing how people come to engage with MRC as an institution. Such engagements are, as Strathern (this volume) suggests, engagements or clashes of social worlds and ontologies – ways of being – rather than mere encounters between the ‘perspectives’ of research institutions and subjects that might be mutually accommodated through negotiation and dialogue. The hazards to infants in Sukuta families’ social worlds orientate many practices and forms of expertise and specialism towards either reducing or mitigating specific risks, or ensuring that an infant has the strength (sembo) or bodily protection (balakanta rango)2 to withstand uncertain or unpredictable events. These practices draw on multiple health providers. Parents understand each child’s particular history and health status as unique, depending on its own path through unfolding events, hazards and circumstances; a particular history that can influence how parents think about health-seeking for that child, or indeed participation in medical research. People consider these hazards to begin at the foetal stage, from the time that a woman is recognizably pregnant (harijeta), and the foetus has progressed beyond the stage of being just the man’s semen (maneo) to becoming ‘thick blood’ (yelo kunturo) at about two months, and a frog or lizard-shaped clot at three to four months. They consider that the foetus’s growth and development can be damaged not only if a mother is ill, but by jina (djinn spirits), making it inadvisable for a pregnant woman to go out in early afternoon or at dusk, when these are thought to roam freely in the winds. Sexual intercourse is thought to be beneficial, however, as it facilitates delivery and gives the baby strength, with some considering male fluid actually to add substance to the foetus. 85

Evidence, Ethos and Experiment

The period immediately after birth, these days usually (and certainly for potential study subjects) in Sukuta health centre, is considered as one of particular vulnerability. Many parents tie a piece of cloth around a newborn’s wrist as a sign that it is part of the human world; otherwise a jina might come and tie its own cloth, to claim it as one of its own. Some consider that once at home, before the eighth day, when the baby is publicly named and recognized as a human being, it should not be brought out of the house except for special washes over the spot where the placenta is buried in the front yard. People consider that neonatal deaths, and repeated neonatal deaths, can simply be the will of Allah, or they can be due to a jina or in Islamic perspective, ‘devil’ or setano in you that ‘does not like children’. If they suspect themselves of having this problem, women might go to a marabout for a special talisman (safe), this one worn on her head, to protect her future babies. Many parents also place a variety of talismans on their babies, for balakanta rango (protection). It is often a baby’s father who provides these, as part of accepted gender roles in parenting. People describe the main role of these talismans as to give protection from moo jawolu (bad people) and buwallu (‘witchcraft’), possibilities that many feel to be rising along with the individual competitiveness and mass-mixing of people in today’s crowded urban quarters and markets. Talismans are also felt to give protection in a more generalized sense, from hazards that one cannot know, including the kind that can drive a child and their family into a downward spiral of ill-health and deprivation. Those who claim not to ‘believe’ in talismans frequently put them on anyway, perhaps due to kin or peer pressure, a husband’s pressure, or simply ‘just in case’. With the nourishment of breastmilk, people consider that a baby grows and accretes strength (sembo) through a series of stages: from a scrunched up newborn to a baby that can stretch out; through a punctual moment of ‘putting on flesh’ (balo subuduno) at 2–3 months, which ‘opens up’ the body to receive and benefit from more milk, and initiate subsequent steady, gradual growth. They observe that the baby will often be restless and feverish at this time, and a first-time mother might be worried, but if she consults an elder, they will reassure her that ‘oh, it is just more flesh coming’. The later stage when a baby can hold its back and head up (‘neck straight’, kantilindo) signifies the stage at which it can be carried on its mother’s back, and also the moment when a woman might be considered available again for sexual activity (see also Bledsoe 2002). If a mother has sexual intercourse while breastfeeding, some consider that the man’s sperm enters the milk and can 86

Being 'with Medical Research Council'

lead to diarrhoea and malnourishment. The baby will lose strength (sembo doya). Generally, a baby of a good weight is an indication that a mother has abstained – even more than of her socio-economic status. If a mother falls pregnant again while her baby is still breastfeeding, people consider that the state of timparo results, causing the breastfeeding child to become sick, weak and thin, sometimes with vomiting and diarrhoea. Some interpret this as simply due to the double load of a mother attempting to nourish two children at once; others, to fluids from the developing foetus, including residual semen, entering the breast milk. Avoiding timparo is one of the reasons why women are so interested in birth spacing (and some use contraceptives to this end – see Bledsoe 2002). People suggest that a large variety of illnesses (kuran) can set back an infant’s growth and development. Within a complex set of causal fields, these are variously attributed to poor diet and care, perhaps linked to poverty; to physical happenings (e.g. falls, exposure to bad wind or damp, sometimes linked to the entering of a disease seed – kuran keso – into the body), to the agency of jina or ‘bad people’; to events which disrupt socio-ecological orders (e.g. a frog eating faeces left in a courtyard), or to Allah. The causal fields for different childhood illnesses can interact, while the state of being ill (kuranta) can be both hastened by, and further encourage, weakness (sembo doya). Maintaining health, for an infant as for an adult, is also thought to require maintenance of the quality, quantity and flow of two sorts of fluid in the body: blood (yelo) and ‘water’ (jiyo), which originates in the head, and links with the kidneys and with male and female reproductive fluids. The importance of overall blood quantity to health and wellbeing – an ‘economy of blood’ in the body (Fairhead and Leach 2006) – shapes a field of enormous worry about giving blood, whether in blood samples or in transfusions, and whether by babies and children or adults. Practices in seeking therapy for these various ailments range from visiting local herbalists, predominantly male and female elders in the village or its surroundings known for their experience in treating particular conditions; to seeking the help of a local marabout, to seeking advice and medicine at the health centre. For most ailments, the sort and sequence of treatment sought depends on how the particular course of illness interacts with practical issues such as the availability of transport, drugs and money to pay for them. These issues in turn are shaped by intra-household and gender negotiations, with mothers assuming primary responsibility for minor therapy-seeking, but often attempting to call on a father’s resources to pay for transport or more major hospital treatment. 87

Evidence, Ethos and Experiment

Routine vaccinations are now an integral part of this set of practices for promoting and maintaining infant health (see Cassell et al. 2006). Women attach particular importance to what they term the ‘three month injections’ (karo saba penko), those given at two, three and four months for DPT and Hib. These are given at bala subunduno time – complementing and boosting this phase of growth. A vaccination, like any other injection, is interpreted as a powerful poisono which goes directly into the blood, where it plays multiple roles. It is seen to have a general role in giving the baby strength (sembo). It ensures that all the illness that is, or may be, inside a baby’s body disappears. And it gives protection against further illnesses. Many mothers say that talismans and injections complement each other as means of child protection: ‘a be ka ta nyoo la’ – they both go together. One group of women in Sukuta explained how both protect a child against bad people (mo jawo) and bad wind (fonyo jawo). Moreover, they emphasise that life today is uncertain; you never know what will happen or bring a problem to your child, and both these forms offer generalized protection.

Joining the MRC – Meeting Dilemmas, Negotiating New Uncertainties Despite the complex explanations and informed consent procedures, and despite the best efforts of MRC fieldworkers to explain it in terms that mothers might understand, it seems that mothers have only a rather hazy idea of what the current Birth Cohort Study is about. However, as we show in this section, this seems less significant to them than what MRC, as a particular sort of health-providing institution, means. Signing up to join a trial is experienced as ‘joining MRC’. This brings with it rights to good quality, free treatment, which is of enormous help with the general burdens of maintaining infant health in a hazardous world. In this respect, joining MRC creates a kind of (health or therapeutic) citizenship, involving new rights and claims on resources and assistance, but further than this, distinct forms of subjectivity and solidarity centred on being part of a cohort. Nguyen (infra) has argued similarly that new and powerful forms of therapeutic citizenship are emerging amongst people infected by and affected by HIV/AIDS, through their participation in trials and roll-out of new treatments. In the Gambian case, MRC-citizenship may not be providing the kind of broader political platform that Nguyen describes. It is, however, arguably helping to fill the void left, for many, by the crumbling of other frames of belonging. Belonging to lineages and communities has frag88

Being 'with Medical Research Council'

mented, for many, through the experiences of immigration and strangerhood in the peri-urban fringe. A sense of belonging to a nation-state has been weakened by distrust, by government inability to provide healthcare alone, and by forms of migration that have scattered ‘the people of The Gambia’ across the globe. At the same time, those left behind feel excluded from those more distant frames of belonging in Euro-America, the land of dreams of great wealth, and the destination of emigrants. ‘Being with the MRC’ in part substitutes for these other, unreachable or ineffective forms of citizenship or belonging. Yet women’s narratives also reveal MRC-citizenship as ambiguous. They reveal how in this and past infant studies over the years they have weighed up the benefits of being with MRC against perceived dangers. These are dangers around blood-taking, either pressed upon them by husbands or compound elders, or that they have heard about as more widely circulating rumours. Concerns with blood-taking centre both on its bodily effects and on the wider political-economic orders of which MRC is part – and from whose benefits most Gambians are excluded. Women’s narratives begin to illustrate these points. Thus Isatou,3 an older woman, described being part of a study 17 years ago: My first child, a girl, was registered with MRC; she was booked immediately after delivery. The MRC staff were always on site to counsel women before and after delivery about the project and the need for one to join [although she could not recall the objectives of the project]. During that period everyone was registered. If you were not convinced about them they would register you and try to convince you and your husband. While some men would be convinced others would reject. During those days, when blood was taken from children, they suffered a lot. The women complained and the husbands intervened by either withdrawing their children from the study or refused children registering. MRC staff were very active and their relationship with people was very cordial. They brought incentives to attract people and people accepted. They will do home visit, meet and discuss with the couples, they will take you to the health centre when the need arose and ensured that you are brought home. People are gradually accepting MRC now, because they offer good treatment, whenever blood is taken, they ensured that enough medicines are given to replace the loss. They will never do what is bad.

89

Evidence, Ethos and Experiment

Mariama was part of MRC research in 1993/94. She recalled how to motivate people to register with them, MRC fieldworkers provided rice, sugar, oil and so on to study members. ‘We were allowed to see them regularly or whenever the child is sick we would receive free and very good medicines. They would also do follow-up visits to ensure that the child is well. They will take blood from them at least every three months and would give them medications to replace the blood loss’. She explained how, whenever MRC staff visited her or took blood, certain neighbours would advise her to withdraw from the research, commenting that the MRC steals blood and that they (mother and child) would be left short. While some withdrew, she continued as her child continued to receive prompt attention whenever they went to the main MRC laboratories in nearby Fajara. Similar balancing of the health benefits and entitlements associated with joining MRC, and the dangers, emerges in women’s narratives concerning the current Birth Cohort Study. Thus for example one mother, Nafina, perceived that ‘the study is examining children for child sickness such as malaria, and this would entail blood taking’. She remembers being counselled by MRC staff a few hours before delivery but not recalling much, and then being counselled again after delivery, with her husband, when they accepted. She feels that: Caring is much easier now I am registered with MRC, as MRC provides good medication for both child and parents. Although they have been associated with selling blood, I am convinced that the blood they take would not earn them so much money sufficient for daily living. Some have attempted discouraging me but I have not listened. Another mother, Fatmata, had her fifth child registered with MRC. She views the MRC as positive, as another institution providing good treatment. She claims not to believe rumours about selling blood: ‘They took a bit of blood from my son and I witnessed them testing it so the rumour about them selling blood was not true’. Similarly, Kadijah, a recent immigrant into Sukuta Muritani new settlement, consented to join the study when she and her husband were asked. The main message she has retained from the explanation is that the research involves taking various samples from both mother and child, including cord blood, urine, sputum, and blood samples when the baby is five and eight months old, as well as from the mother. She claims that:

90

Being 'with Medical Research Council'

Some elderly people discouraged us about the MRC programme but we did not listen. MRC has good drugs as opposed to the government clinic where you are asked to buy from a pharmacy. One good thing about registering with MRC is you are sure to have the required healthcare services, their staff will inform you when to attend clinic and you will also be given the required attention. Study subjects are given priority over others, as we are picked out from the lot, weighed, immunized and asked to go home. This saves us from the long queue as time saved will be best utilized for household chores. Whenever blood samples are taken, tablets are given to replace the loss. They also provide us with tablets for fever (if vaccinated). As this narrative illustrates, women in Sukuta also regard the MRC study as having positive interactions with routine vaccination, helping their attendance and helping them to cope with side-effects. This is a longstanding association, which goes back to earlier studies. Thus Isatou, speaking of her study involvement seventeen years ago, described how: During IWC days, they ensured that you attend or else they will pick you up from wherever you may be and will take you back after being treated. They ensured that their subjects do not default as vaccinations were always on time. MRC would provide medications especially for side effects resulting from vaccination. One unavoidable side effect was the pain leading to the persistent crying. If crying persists or the child is restless after vaccinations, the children are taken back to them and they would ensure that the mother came home satisfactorily. Thus for women in Sukuta, the MRC is seen and used primarily as part of routine health practices, and as a support to routine vaccination and clinic attendance. Indeed the supportive relationship between MRC studies and the government’s Expanded Programme on Immunization is manifested in practices at the clinic, where MRC fieldworkers have been routinely helping to administer vaccinations, thus showing themselves to all who attend to be part of ‘normal’ health practices, as well as (instead of?) part of an external ‘scientific’ organization. ‘Being with’ the MRC is, in this respect, not a substitute for using government health services, but a complement that enables one’s entitlements from (weak and unreliable) government services to be met. 91

Evidence, Ethos and Experiment

In our survey of mothers across Western Division, those who had been asked to participate in the Sukuta Birth Cohort Study made similar statements regarding the benefits of having a child registered with MRC. They emphasized the value of having one’s child routinely monitored; as a sort of safety net to guard against harm. They commented, specifically, and in a relativistic way, on the quality and availability of medicines, and the dedication of the staff, at MRC as compared with other health-providing institutions. For instance: You have good treatment as MRC has very good medicine and staff MRC has very committed staff and always plenty of medicines. The child becomes well taken care of medically, and routine follow-up is made to ensure that the child maintains a very good health. With MRC, the child can be healthy. They will check and take care of him and they will take responsibilities of every aspect of his treatment. MRC takes good care of the registered children because they make frequent visits on these particular children and they are diagnosed for any sickness. In fact my previous child was registered with MRC and they took great care of the child during the period. MRC gives us treatment; the best; whereas the Sukuta Health Centre always say ‘there are no medicines’. However, when asked in the survey whether they had heard any negative things about having a child registered with MRC, 22 per cent of the Western Division survey population had, rising to 43 per cent of those invited and accepting participation in MRC studies there, including the Sukuta Cohort Study. In all cases, these negative ideas turned on the notion that MRC takes or steals blood, or the worry that they might. For instance: I have heard that MRC steals blood. MRC takes blood from their patients and sells it. It is heard that MRC takes blood from the registered children which sometimes is too much and can cause further complications. It is heard that MRC do steal blood from some of their patients. We discuss the conceptual and experiential basis for these blood-stealing worries in the Gambia, grounded as they are in understandings of blood as a commodity in the body and in wider political-economic transactions, at length elsewhere (Fairhead et al. 2006). Here, it is sufficient to emphasize 92

Being 'with Medical Research Council'

that these anxieties are pervasive; even when mothers are unsure whether there is truth in the rumours they hear, they are faced with a field of uncertainty in which there are possibilities of harm to their infants. It is also worth noting that people’s worries are enhanced by the enormous disparities between their relative deprivation and the relative wealth of the MRC and its scientists, suggesting that the MRC must be enriching itself through an international political economy of which blood is a part. Nevertheless, for most mothers, these fears are balanced against the many benefits of joining the MRC. As one mother put this balance: ‘MRC help people in getting good health. The blood thing is the only problem’. Even those who join may hear, or themselves harbour, concerns about the MRC and bloodstealing, yet suppress these in relation to the more immediate and tangible issues of access to good care and medicines. Thus as mothers’ narratives indicate, their framing of the MRC as a routine health provision institution, yet a particularly efficient, dedicated, well-resourced and convenient one giving its members valuable entitlements – leads many to be positive about their infants joining. However, in seeking to join, they often have to negotiate with elders – and in some cases, with husbands – who, less close to the daily dilemmas of keeping infants well, place more weight on the potential dangers which joining MRC implies. Here, the degree to which mothers can exert their agency within prevailing gender and compound relations becomes critical to study participation. As discussed earlier, the formal consent process for the Birth Cohort Study initially approached women but gave them the opportunity to discuss with their husbands before signing a consent form. The mothers we interviewed varied in how far they considered themselves to have agency in the process of deciding whether or not to register. Some described the decision as having been their own. Notably, these tended to be women at a later stage in their reproductive career, with several previous pregnancies; or to be those with active careers in trade or gardening – thus exemplifying the greater decision-making power that typically accrues to Gambian women with accumulated stages in reproductive career and independent economic resources. Others – including many younger mothers – described the decision as a joint matter between themselves and their husbands, while often giving the impression that the bottom-line lay with their husband’s consent. For example: I was told about it at the health centre; I consulted my husband about it and he gave me the go ahead, that’s how it went. 93

Evidence, Ethos and Experiment

The staff asked me to join the study and my husband was there at the time and he accepted for me to join the study. When I was in labour, MRC workers were there. I was helped by them and they later asked me if I would like to join their work. I did not consent there. I had to contact my husband. My husband agreed and I joined them. My husband and I sat and discussed it. We looked at the advantages and disadvantages and finally agreed unanimously to join. In several other cases, however, women who would have liked to join could not, or did not try, to exert their view in the face of opposition from their husband. For example: The husband was not sure to allow the child to be registered with MRC. He wasn’t convinced, and had the final say to say no to the demand/register for the child to be registered. I was asked by MRC workers and there and then I made my decision not to because I knew my husband will not agree. I was asked to participate and I decided on my own to join but later my child’s father asked me not to join. Because when they started taking the child’s blood the child is always sick.

Some Conclusions The balancing of benefit and danger which frames parental decisions about participating in the Sukuta Birth Cohort Study echoes, in some ways, the balancing experienced by MRC’s own fieldworkers as they go about their work. As one put it to us: You are there, helping a mother and baby, and everything is going well. And then you have to take blood, and the mother is worried, and the child is crying, and you are going there, there …. Without blood-taking, this would be the best job in the world. But whereas this fieldworker’s statement frames the balance between bloodtaking and treatment as a procedural issue within the context of a medical encounter – a disliked procedure amidst liked treatment and care – mothers frame it more broadly, as an intrinsic aspect of engaging with the whole 94

Being 'with Medical Research Council'

MRC institution. This reflects the relative positions of fieldworkers and parents in such encounters: both are MRC citizens of sorts, but the fieldworkers are more firmly and formally ‘with’ the MRC, and speak from a more secure sense of belonging. It is such a firm sense of belonging, and of subjective identification with MRC discourses, that enables, in other contexts, fieldworkers and medical research staff to frame worries about blood-stealing as misplaced; the result of ill-founded rumours that will be vanquished by modern scientific knowledge and people’s positive engagement in medical research. They are thus seen as the last misconceptions that need to be overcome if actual and potential study subjects are to comprehend and appreciate the MRC’s work properly, and thus participate willingly in their studies. In this view, educational work through public meetings and fieldworker outreach needs continually to correct such misconceptions, while explaining what research activities are really about. It is here that the ‘popular immunology’ that fieldworkers like to draw into public meetings finds its justification. When MRC fieldworkers take it upon themselves to stand up and speak to crowds in the study’s public meetings – or to the journalists covering them for national newspapers – about the antibodies and ‘soldiers in the blood’ that MRC is trying to find, they consider themselves to be doing important outreach work that will help to build better public understanding of the Birth Cohort Study, and create the kind of modern, scientifically enlightened citizen that will engage with future MRC studies with confidence. The study’s senior scientists, meanwhile, wring their hands about the faulty, simplified picture of the study’s science that fieldworkers convey in these explanations, and lament their seeming inability to grasp the basics of immunology. Positioned at an impossibly challenging frontline between international science and publics, it seems that fieldworkers’ attempts to cross the divide creatively are not always appreciated. However, our ethnographic work suggests that such debates – focused as they are on how to improve communication about a particular study, and with the MRC as an institution that ‘does science’ – may be missing the point. As this paper has suggested, it may be misplaced to understand why people join studies like the Sukuta Birth Cohort Study in terms of their information, conceptions or misconceptions about that study, or even about medical research more broadly. First, anxieties about blood-taking persist for deep-seated reasons even amongst those joining the MRC. Second, those who join do not understand themselves to be engaging with medical research studies, but to be joining an institution that delivers routine healthcare. Third, those who join are no more ‘modern’ and ‘scientific’ in their 95

Evidence, Ethos and Experiment

outlook than anyone else. All the mothers in our ethnographic work combined a mix of biomedical, naturalistic and Islamic healing practices in seeking health for their infants. The MRC fits into this mix in particular ways for different people at different times, and brings certain advantages in meeting the dilemmas of infant care, but has become an embedded part of this plural therapeutic landscape. And fourth, being with the MRC has a place in the broader social and political landscape, linked to valued forms of citizenship in contexts in which other frames of belonging are fragmenting or under strain. This ethnography – along with similar work around other MRC studies in the Gambia (Fairhead et al. 2005, 2006) – raises some wider implications for the ways that public engagement with medical science is debated and approached. It suggests that debates in biomedical ethics which focus solely on issues of participation in particular biomedical research programmes are cast too narrowly. Rather, they need to comprehend the healthcare delivery and citizenship contexts in which research is conducted, the healthcare dilemmas which people face amidst prevailing social and cultural worlds, and the historical legacy of past interventions. Ethnographic work in the Gambia illustrates how people may be participating in research studies and clinical trials for entirely different reasons than researchers intend, while harbouring interpretations of those studies that strongly contradict research protocols (see for further examples Kelly, infra). Such examples are not restricted to settings such as those in The Gambia, and expose the naiveté of expectations of ‘informed consent’. New approaches to debating the ethics and practicalities of medical research participation are needed, which acknowledge the very different ways that researchers and populations might understand the nature of research activities, and the character and motivations of the institutions involved. In line with current approaches to ‘upstream engagement’ in public participation in science, new approaches are needed which acknowledge the very different ways that scientists and local populations live and frame the issues concerned, and which both seek negotiation between these, but acknowledge its limits (Wilsdon and Willis 2004; Leach et al. 2005; Stirling 2005). In this, ethnographic approaches become invaluable both in exposing these different framings, and in tracing the social processes and institutional imperatives which underlie them.

96

Being 'with Medical Research Council'

Notes 1. Melissa Leach and James Fairhead were joint principal investigators in the study on which this chapter draws. Mary Small participated invaluably in the ethnographic fieldwork and also oversaw the survey referred to here, while Jackie Cassell helped develop the survey and oversaw the statistical analysis of its findings in conjunction with Catherine Mercer. We thank the Medical Research Council staff and scientists who provided logistical support and useful comments at various stages of the research, especially David Miles and Marianne van der Sande. We are also grateful to the fieldworkers involved with the Sukuta Birth Cohort Study, staff of the Sukuta health centre, and the many Sukuta community authorities and members who assisted us with their welcome, time and insights. We are grateful to the joint Gambian Government/MRC Ethics Committee for granting ethical clearance for this study, and to the funding organisations which have supported it: the UK Economic and Social Research Council Science in Society research programme, and the UK Department for International Development’s Committee on Social Science Research. The views expressed in this paper are those of the authors, and not of these support or funding organisations. Any errors of fact or interpretation are ours alone. 2. Local language terms given are in Mandinka. The majority of our interviews have been with Mandinka-speakers, and the concepts and practices described thus apply only with certainty to Mandinka speakers, rather than those of other ethnic/language groups. 3. A pseudonym, as are all names in the following sections.

References Bledsoe, C. 2002. Contingent Lives: Fertility, Time and Aging in West Africa. Chicago: University of Chicago Press. Bloom, G. and H. Standing. 2001, ‘Pluralism and Marketisation in the Health Sector: Meeting Health Needs in Contexts of Social Change in Low and Middle-income Countries’, IDS Working Paper 136, Brighton: Institute of Development Studies. Cassell, J.A., M. Leach, J. Fairhead, M. Small and C.H. Mercer. 2006, ‘The Social Shaping of Childhood Vaccination Practice in Urban and Rural Gambia: A Quantitative Survey of Mothers Based on Ethnography’, Health Policy and Planning 21: 373–91. Fairhead, J., M. Leach and M. Small. 2004, ‘Childhood Vaccination and Society in The Gambia: Public Engagement with Science and Delivery’, IDS Working Paper 218, Brighton: Institute of Development Studies. ———. 2005. ‘Engaging with Science? An Ethnography of a West African Vaccine Trial’, Journal of Biosocial Science, Special Issue on Anthropology and Public Health 38: 103–16.

97

Evidence, Ethos and Experiment

———. 2006, ‘Where Techno-science Meets Poverty: Medical Research and the Economy of Blood in Medical Research in The Gambia, West Africa‘, Social Science and Medicine 63: 1109–20. Leach, M. and J. Fairhead. 2007. Vaccine Anxieties: Global Science, Child Health and Society. London: Earthscan. Leach, M., I. Scoones and B. Wynne. 2005. ‘Introduction: Science, Citizenship and Globalization’, in M. Leach, I. Scoones and B. Wynne (eds), Science and Citizens: Globalization and the Challenge of Engagement. London: Zed Press, pp. 3–14. Marsden, P.D. 1964. ‘The Sukuta Project: A Longitudinal Study of Health in Gambian Children from Birth to 18 Months of Age’, Transactions of the Royal Society of Tropical Medicine and Hygiene 58(6): 455–89. MRC. 2004. Field Protocol: Sukuta Birth Cohort Study. Version 07072004. Fajara: MRC Laboratories. Stirling, A. 2005. ‘Opening Up or Closing Down? Analysis, Participation and Power in the Social Appraisal of Technology’, in M. Leach, I. Scoones and B. Wynne (eds), Science and Citizens: Globalization and the Challenge of Engagement. London: Zed Press, pp. 218–31. Wilsdon, J. and R. Willis. 2004. See-through Science: Why Public Engagement Needs to Move Upstream. London: Demos.

98

Chapter 4

Contextualizing Ethics: Or, the Morality of Knowledge Production in Ethnographic Fieldwork on ‘the Unspeakable’ Hansjörg Dilger

There is a saying in Uganda: If a snake comes into your house do not waste time asking where the snake comes from, but kill it first and ask questions afterwards. (Noerine Kaleeba et al. 1991) Why are anthropologists doing fieldwork on HIV/AIDS in Africa? Why are they engaging in long-term research and academic debates when people are dying? In this chapter, I discuss some of the multiple challenges surrounding anthropological fieldwork on HIV/AIDS in eastern Africa with regard to its ethical and methodological implications. I argue that at a time when the boundaries between applied and basic anthropology have become less rigid,1 questions about the ethicality of ethnographic fieldwork on HIV/AIDS-related issues should focus less on why such research is being done and more on how it can be carried out in ethically, and therefore methodologically, informed ways. By drawing on fieldwork that I conducted in rural and urban Tanzania between 1999 and 2003, I will show that the planning and conducting of this research was shaped, on the one hand, by the precariousness of doing fieldwork in a context where the physical and material survival of many of my informants had to be secured on a daily basis; and where access to adequate healthcare – including confidential HIV testing and life-prolonging therapies for HIV/AIDS – was not available to the majority of the studied community (cf. Farmer 1992; Baylies 2004). In this regard, my research agenda which explored the meanings and practices surrounding 99

Evidence, Ethos and Experiment

HIV/AIDS-induced illnesses and deaths inevitably triggered questions about global inequalities and the anthropologist’s ‘right to be there and observe’ (Schramm 2005: 172).2 From the outset of my research, it was clear that my research approach would not allow for substantial improvement of my informants’ health or living conditions. In this sense, the nature of my fieldwork could critically be described as ‘bourgeois luxury’ – or, as Paul Farmer (1992: 315) has more constructively termed it, as a ‘witnessing’ of the way in which HIV-infected people live with the disease and how they die from it. In a yet more hopeful formulation, however, such fieldwork represents the view that a careful analysis of the practices and ideas that have evolved around HIV/AIDS may contribute to more differentiated knowledge about the social and cultural contexts in which the epidemic thrives, and thus to creating the basis for locally adapted strategies of prevention, care and treatment (cf. Heald 2003).3 On the other hand, my fieldwork on HIV/AIDS in Tanzania raised issues with regard to questions of confidentiality and informed consent and, more specifically, the politics of speaking about a stigmatised disease loaded with biomedical meanings. As my research evolved, I became aware that in order to establish effective communication about HIV/AIDS on the ‘local’ level, I had to learn to ‘speak properly’ – that is, in a socially and morally accepted way – about HIV/AIDS-related illnesses and deaths. Thus, while local discourse on HIV/AIDS in wide parts of eastern and southern Africa is structured essentially by concepts and practices of public health and international development, it is equally configured by silences, rumours, metaphors and ‘narratives of illness’ (Mogensen 1997) that have come to shape meanings and actions related to the epidemic in all those settings where the language of global development and biomedicine has not been pervasively established. In the following, I will draw on the various research sites in which I did fieldwork in Tanzania – kinship and village networks, Pentecostal churches, non-governmental organisations – and argue that the ethical and methodological challenges of HIV/AIDS research in eastern Africa are shaped, on the one hand, by social relationships and power hierarchies that in turn shape different modalities of speaking and non-speaking about the stigmatised disease. On the other hand, the social and moral constellations in research sites – of which the participant-observing anthropologist is likely to become an integral part – have a deep impact on shifting interpretations of illness and healing, and consequently on the way in which the ethical topoi of confidentiality and informed consent have to be dealt with by the 100

Contextualizing Ethics

researcher and his or her informants. In conclusion, I argue that the insights that can be drawn from my fieldwork in Tanzania shed light on some issues that characterise the moral and social embeddedness of ethnographic research in more general terms. These insights may therefore become relevant for the work of research commissions and ethics’ committees, which are increasingly confronted with research proposals that are crossing disciplinary as well as geographical and cultural boundaries. Before I introduce the social and moral context, which is the starting point for my further reflections on the ethics of HIV/AIDS research – as well as on the modalities of knowledge production in ethnographic fieldwork in more general terms – I will start with a description of the way I became involved in my various research sites in Tanzania through the process of applying for and obtaining an ethical clearance.

Starting Fieldwork in Tanzania4 In 1999, I submitted my application for research clearance to the Commission for Science and Technology in Tanzania (COSTECH). The proposal represented a streamlined version of my outline for Ph.D. research and carried the long-winded title ‘How people living with HIV/AIDS and their families are coping with the disease in rural and urban Tanzania’. As may be easily recognisaisable, this rather dry and formalistic heading was primarily a reflection of the way I expected the reviewers at COSTECH to react to a proposal for anthropological AIDS research that in their view might be too far removed from the fields of applied research and/or public health. After all, my proposal, which incorporated issues such as family and kinship relations, support and care within NGOs and churches, ‘traditional’ healing and the ritual cleansing of widows, was firmly situated in the tradition of basic social-anthropological research. I did not receive a response from COSTECH for several months. However, upon my arrival in Dar es Salaam I was informed that my application had been transferred to the National Institute for Medical Research (NIMR), which was to issue an ethical clearance before the actual research clearance could be processed.5 COSTECH’s decision did not come as a surprise to me at that time. Indeed, the fact that the discussion of the ethical implications of my enterprise had been neither a standardised requirement of COSTECH nor of the German Research Foundation (DFG), which had provided the grant for the project I was involved in had taken me aback, considering the intense scholarly debates surrounding the ethics of social science and anthropological 101

Evidence, Ethos and Experiment

research which have dominated discussions in Anglo-Saxon countries over the last decades (e.g. Barnes 1977, Caplan 2003, Fluehr-Lobban 2003).6 It was therefore only after I had finished my fieldwork in Tanzania – and after I had experienced how some of the methodological and ethical core assumptions of my revised proposal had proven untenable during the research process – that I started to wonder why the ethical clearance had been issued by the National Institute for Medical Research (and not COSTECH); and whether the ethical guidelines that direct the work of health researchers are appropriate to direct the work of anthropologists, too. In retrospect, COSTECH’s decision became more comprehensible to me when I realised that the processing of my research clearance through the NIMR was probably the only viable solution for COSTECH at that time. Due to the fact that my research involved interviews with ‘People infected with HIV’ and ‘People sick with AIDS’, it would have been difficult to figure out – both for the scientific officers at COSTECH and for myself – how this research could actually be carried out without being based on the same ethical assumptions guiding the work of health researchers. After all, it had initially been health researchers, development experts and AIDS activists who had developed ethical codes for the ‘personal, professional, institutional and governmental response’ to the HIV/AIDS epidemic (African Networks on Ethics, Law and HIV 1994). Health researchers were also the ones who proposed ethical guidelines for doing research with people with HIV/AIDS, especially regarding obtaining informed consent in the context of clinical trials, testing and treatment (cf. Wolf and Lo 2001; Bhutta 2002). Would it have been ethically and methodologically acceptable for COSTECH, I wondered in hindsight, not to evaluate my proposal on guidelines that were derived essentially from health research and that were, at the same time, rooted in the human rights discourse on HIV/AIDS? Would it have been ethically sound for me not to base my research proposal on these guidelines – despite the fact that I pursued a social-constructivist argument that perceived health-related behaviours as the outcome of social, cultural, economic and political processes? Would it have been possible to completely ignore the larger context of HIV/AIDS research in Africa, which is so heavily dominated by biomedically driven research and by the interests of (international) development organisaisations and funding agencies which are aimed at rapidly applicable research results? The answers that I found to these questions changed considerably over the course of my research and also over the time I analysed and processed my field data. In the following, I will show that the endeavours of finding 102

Contextualizing Ethics

answers to some of these issues – and, above all, the ethical, epistemological and methodological challenges that are implied in these answers – have to be taken back to the various social, economic and moral constellations in which I carried out fieldwork in Tanzania. I will argue that the theoretical and methodological basis of social-anthropological research on HIV/AIDS can hardly be grounded on a universally valid model of disease, health and ethics, but that the epistemologies of illness and disease, and the methodological and ethical implications that derive from them, vary with the research setting and the way in which anthropologists are involved in shifting research sites. In the next section, I turn first to the setting of NGOs and the practices of health citizenship which have shaped modalities of speaking about HIV/AIDS – as well as perceptions of ethically sound social science research on HIV/AIDS – in the context of neoliberal health policies and structural adjustment in Tanzania.

‘We Would Be Dying Like Chicken’: Notions and Practices of Health Citizenship in Dar es Salaam Starting in the early to mid-1980s, African healthcare sectors became subject to far-reaching reforms resulting primarily from the introduction of structural adjustment policies imposed by the World Bank and the International Monetary Fund (IMF) in reaction to the growing economic and political crises of African states. In wide parts of sub-Saharan Africa, these reforms caused severe cuts in governmental expenditures for social services and have led to growing privatization and transnationalization of the health field, which became ever more reliant on the funding mechanisms of international donors. The privatization and transnationalization of the health field have become most visible in the context of HIV/AIDS – whose emergence and growth conspicuously parallels, and whose course reflects on, the structural reforms in Africa’s wider healthcare sectors. In Tanzania, public health responses have been shaped largely by the growing NGO industry which has based its activities on notions of human rights, self-responsibility and empowerment and in recent years has aimed at the involvement of ‘affected communities’ into its manifold activities. Vinh-Kim Nguyen (2005: 126) has subsumed the various practices and ideas that have emerged in the fields of care, treatment and AIDS activism over the last 10 years under the concept of therapeutic citizenship – a transnationalized form of citizenship 103

Evidence, Ethos and Experiment

which makes claims on the global economic and social order based on a shared therapeutic predicament (i.e. being infected with HIV). According to Nguyen (2005: 125f.) the social and cultural practices that have evolved in the context of HIV/AIDS have been organized around a complex set of confessional technologies and processes of self-fashioning which are closely interwoven with internationally acclaimed forms of HIV/AIDS activism and essentially draw their legitimacy from the economic, political and biological inequalities existing in a globalizing world. In Tanzania, the privatization of the healthcare sector and the concurring introduction of public-private partnerships in the country’s healthcare system have also led to shifting understandings of rights and citizenship. While colonial and post-independent understandings of law in the country regarded the existence of legal frameworks predominantly as ‘constitutive of state power, not as a limitation upon it’ (Harrington 1998: 151), the transition to a multi-party system and the growing influence of the international development community from the mid-1990s onwards have defined law as a code for which the state can be made liable; thus opening a space for the perception that health is a right individuals are entitled to (ibid.). In Tanzanian society, such shifting perspectives have not resulted, however, in an open challenging of the state in relation to its alleged responsibility for people with HIV/AIDS – as for instance in South Africa where the demonstrations and activities of the Treatment Action Campaign have become symbolic for practices of health citizenship in the post-apartheid state (Robins 2004). In Dar es Salaam, it was almost exclusively on the safe ground of the NGOs that members of the HIV/AIDS support groups expressed their critique of governmental authorities which, according to them, have failed to take care of their HIV-positive and AIDS-sick citizens. As can be seen in the following statement, this critique was directed primarily at the state level, without taking account of the wider political and economic processes that from the mid-1980s onwards have forced the Tanzanian government to reduce its expenditures for social welfare and which have led to an increasing dependence of the country’s healthcare sector on external funding (Harrington 1998: 149). Zephania (male, 37), a support-group member knowingly living with HIV since 1994, said:7 Hansjörg: Has the government changed its stance on HIV/AIDS? Zephania: Yes, this is what they claim, but I haven’t seen that they have truly changed their attitudes. … So far they haven’t done anything. ... These people from the government do not come and talk 104

Contextualizing Ethics

with us – the ones who are affected by the disease. They only wait for the reports of those who are not affected. … I have been to the Muhimbili,8 I live in the quarters [of Dar es Salaam], I know the real extent of the problem. ... This question of AIDS is still far removed from them, they haven’t helped us with anything. If it wasn’t for these NGOs – if it was only for the government – we would be dying like chicken (tungekufa kama kuku). (Emphases added) By the year 1999, four of the internationally funded AIDS NGOs in Dar es Salaam had established numerous services for people with HIV/AIDS which all focused on notions of human rights and empowerment. The underlying idea of their approach being the assumption that people with HIV/AIDS – equipped with the appropriate (biomedical) information and backed up by a range of social, economic and political measures – would be able to make their own choices and decisions, thus leading a self-determined life. In addition to the provision of basic medical and counselling services, the NGOs in Dar es Salaam offered legal and economic support programmes, home-based care services for AIDS patients and also support groups for people with HIV/AIDS. These support groups comprised around 20 to 40 men and women each – the majority of them between 25 and 40 years old and with a primary or secondary school education background – who came together for group meetings on a monthly basis. Many NGO clients were members of two or more groups, which meant they could have up to four meetings per month and got to know each other fairly well. Some of the men and women I talked to had previously been interviewed by other researchers or by the media,9 and at times they played an active part in requesting an interview or asking questions about my research. All of the interviewees in the NGOs – including the NGO workers and directors –were comfortable with the way I addressed the ethical implications of my research. Following NIMR’s request, I had prepared a short research statement in Kiswahili (karatazi ya maelezo) that I read to my prospective research participants, asking them for their oral consent to participate in my research project.10 Beneath the main research question and a short description of the overall research objectives, the statement guaranteed the interviewees the right to ask questions about my research and to withdraw from the interview process at any time. Furthermore, it contained a paragraph that ensured the confidentiality of any information and names revealed during the interviews. 105

Evidence, Ethos and Experiment

This latter point of confidentiality was – due to the strong stigma attached to HIV/AIDS in Tanzania – of crucial importance for most of the men and women I interviewed. It was less important, however, for the members and leaders of the NGO SHDEPHA+ (Service Health and Development for People living with HIV/AIDS) who were all HIV-positive: according to them, HIV/AIDS had become a ‘normal disease’ (ugonjwa wa kawaida), and one of the ways they educated the public on AIDS was to talk ‘openly’ (wazi) about their HIV-infection at public events such as World AIDS Day and also in the media. For them, HIV was the acknowledged cause of their illness and the attitudes they developed towards their infection were shaped by notions of health citizenship, and more specifically by practices of ‘Living PositHIVely’ (Dilger 2001). However, even those support-group members of other NGOs who were afraid that their HIV-infection might become known to their families or neighbours, seemed to have no general problem with participating in a research project dealing with ‘AIDS’ (ukimwi). While many people with HIV/AIDS were stigmatized and discriminated against in Tanzania at that time, the premises of the NGOs offered a zone of confidentiality and protection to their clients and provided – in varying degrees – the basis for a ‘positive’ attitude towards living with their illness.11 A consequence of this general acceptance of the biomedical nature of their illness was that when the NGO clients discussed ‘witchcraft’ (uchawi) or ‘traditional healing’ (matibabu ya kienyeji) in relation to HIV/AIDS – individually or in the context of the support groups – they found these issues either irrelevant or started mocking the work of traditional healers who were often designated as ‘quacks’ or ‘liars’ (waongo).12

‘God Can Do the Impossible’: HIV/AIDS and the Neo-Pentecostal Gospel of Health and Prosperity The second setting of my research was the Full Gospel Bible Fellowship Church (FGBFC) in Dar es Salaam. I learned about this neo-Pentecostal church through two HIV-positive women in the NGO support groups who claimed that the bishop of the church, Zachary Kakobe, was healing people with HIV/AIDS. The FGBFC was founded by Kakobe in 1989 and grew rapidly during the following years. By 2000, the church claimed more than 120,000 members nation-wide and had established more than 500 regional and local sub-branches throughout the country. In Dar es Salaam alone, 106

Contextualizing Ethics

the congregation comprised roughly 30,000 members, who in 2003 could no longer gather in the church’s main building which held around 10,000 people during Sunday Services. One of the church’s main attractions were the healing prayers carried out collectively for all kinds of affliction during the Sunday services and in individual healing sessions during the week (Dilger 2005: 227–81, 2007). Additionally, between March and April 1999 the church offered healing prayers specifically for AIDS which drew around 300-400 people per week and plenty of media attention resulting in pictures of these events in the local newspapers (e.g. Daily Mail, 27 March 1999). The overwhelming interest of the public in the AIDS healings was however, also the reason that the AIDS prayers were abandoned soon after they had started. While the church members I talked to remained vague about the reasons why the AIDS healings had been discontinued, they indicated that the church leaders had underestimated the pressure that was put on HIV-infected people as a result of their exposure to public interest: Church members repeatedly told me that a person whose HIV-infection became known in his or her social and familial environment was likely to commit suicide (kujiua), since it was unbearable to live with the disgrace of this ‘shameful disease’ (ugonjwa wa aibu). The view of HIV/AIDS as a disease of sinners also shaped the atmosphere that prevailed in my interviews with church members at the home of Pastor Mbwambo, one of the church pastors. For the purpose of the interviews, the pastor had offered me a room of at his home which had earlier served as a garage. Smelling of petrol and furnished sparsely with two metal chairs, the room did not provide the ideal atmosphere for the rather intimate conversations I was aiming for. However, for the men and women I talked to, it was more important that the door to this room could be shut and that the other church members – who were waiting in the neighbouring living room for their individual healing sessions – could not overhear what we were talking about. Due to the strong stigma associated with HIV/AIDS in the FGBFC, presenting the research statement that I had used for the NGOs and that branded my fieldwork as ‘research on HIV/AIDS’ was impossible. Pastor Mbwambo therefore introduced me to the church members by stressing that I was interested in spiritual healing of all kinds of illnesses and afflictions, and by asking them to reply openly (wazi) to any questions (maswali yoyote) I might have for them. A consequence of this rather vague introduction was that over the following months I could never be certain if my 107

Evidence, Ethos and Experiment

interviewees were HIV-positive or not, and – if yes – whether they knew about their infection and also wanted to confide in me what they knew. In very few cases did my interviewees actually open up during the interviews and reveal to me that they had tested HIV-positive at a local hospital. In others, I could only suspect that they might be infected with HIV from the way in which they talked about symptoms and diseases they experienced which had often become commonly accepted metaphors for HIV/AIDS illness in Tanzania (as, for instance, ‘tuberculosis’, ‘herpes zoster’, ‘ulcers’ or ‘wounds in the stomach’). The most important mode of talking about all kinds of affliction in the church was, however by referring to malevolent spirits (pepo) and witches (wachawi) which were said to have entered or attacked the bodies of church members and to have diverted them from their state of being saved (uokovu). According to the FGBFC followers, satanic forces had taken possession of the world and were responsible for all kinds of suffering (including HIV/AIDS) plaguing the lives of men and women throughout the world. It was only through the firm belief in God and the power of the Holy Spirit that these evil forces could be exorcised from individual bodies and church followers could lead a healthy and prosperous life.

Rumours, Silences and Uncertainty: The Morality of Kinship Relations in Rural Tanzania The uncertainty about the factuality of the HIV status of my interviewees was also a central issue in my third research setting, the rural area located on the eastern shores of Lake Victoria. This region had been characterised by growing poverty and the increased outward migration of the young generation since the mid-1980s. I was familiar with the rural setting from an earlier stay in 1995–1996 when I did research on the moral discourse on AIDS and the way the epidemic had affected young people’s sexuality (cf. Dilger 1999, 2003). When I returned to the villages in 1999, however, it turned out to be rather difficult to approach people on the very sensitive issue of HIV/AIDS illness, despite the close relationships that I had established with some families and individuals during my previous stays. Shortly after my return to Mara, I presented my research plans to the director and the administrative leaders of one of the local hospitals. Without hesitation, they agreed to assist me generously with ‘any requirements’ I 108

Contextualizing Ethics

might have during my research and introduced me to one of the two hospital HIV/AIDS counsellors, asking him to put me into contact with some of his HIV-positive patients for the purpose of interviews. To the obvious irritation of the director (and of myself), however – and despite the explicit relation of dependency that shaped the hierarchy between the director and the counsellor – the latter explicitly expressed his unwillingness to collaborate on this issue and declined to put me into contact with patients he had counselled or was about to counsel.13 The second counsellor I approached turned out to be more supportive than his colleague. Through his facilitation I was able to participate in three post-test counselling sessions in which the result of an HIV test was disclosed to the hospital patient or, in one case, to his mother. In another instance, the counsellor introduced me by letter to a young couple who had tested HIV positive at the hospital and who happened to be the counsellor’s relatives. Despite these engaged efforts, however – and despite the fact that oral consent to my participation in the counselling sessions was obtained through the presentation of my research statement – interviews with the hospital patients turned out to be problematic. On the one hand, this was due to the fact that the counsellor himself insisted on being present at all of the interviews carried out in the hospital, resulting in the course of the interviews being blocked at exactly those points which I had hoped to discuss in my research (especially with regard to kinship dynamics, witchcraft or the breaking of taboos). On the other hand, even in the one case where I paid a home-visit to the young HIV-positive couple together with my research assistant we did not have the opportunity for a more elaborate conversation. After arriving at the residence of the young HIV-positive wife, we presented her husband with the letter from the counsellor in which he asked for their ‘collaboration’ (ushirikiano). However, instead of talking ‘openly’ (wazi) with us about ‘their problem’ (shida yao) the couple seemed visibly worried that the cause of their illness had been revealed and they were obviously disinclined to consent to an interview. Acknowledging the inappropriateness of approaching my interviewees through the hospital – and feeling deeply concerned about the situation I had induced for the young couple – I decided to follow the advice of my research assistant, a 40-year-old man from the area who, from the beginning of our research, had guessed that it might be difficult to approach HIVpositive men and women in rural Mara in the very straightforward manner I had intended. Over the following months we approached people who were, supposedly, infected with HIV or sick with AIDS through the kinship 109

Evidence, Ethos and Experiment

network of my research assistant, as well as through friends and other acquaintances from the rural area. This was a rather precarious enterprise as I depended on rumours and gossip that were circulating in the villages about HIV tests that had been carried out in the hospitals, as well as on information that was provided by distant relatives of HIV-infected men and women who were often more open about these issues than close family members. Following this system of snowballing and guessing, I could never be certain if the men and women of my research were ‘really’ infected with HIV or if it was only a rumour that had no ‘factual’ – i.e. biomedically confirmed basis. I often became more certain only if the person I had talked to became sick and displayed visible symptoms that could be said to be HIV/AIDS-related. In other cases, it became easier for families to talk about the cause of their relative’s illness after he or she had passed away and had been buried. Once the contact with HIV-infected persons and their families had been established, I remained vague in the ensuing conversations about the actual focus of my research by introducing the topic of my project not in relation to ‘HIV/AIDS’, but with regard to ‘chronic and deadly diseases’ (magonjwa sugu) and the situation of family care for bedridden relatives in rural areas. Having presented the objectives of my research in these very general terms, I left it up to my conversation partners how they chose to talk about their own or their family member’s illness. The consequence of this was that our conversations often addressed very different versions and interpretations of the same illness, relating its etiology to biomedical diseases such as tuberculosis and herpes zoster – sometimes also HIV/AIDS – as well as to witchcraft or the breaking of taboos, which in rural Mara is known to result in an illness called chira.14 I often felt somewhat uncomfortable with regard to the ethicality of my proceedings. However, at the same time I understood that it would have been equally critical to ask my conversation partners directly about a possible case of HIV/AIDS illness on the respective compound (about which, in most cases, I was not even certain). To confront HIV-positive men and women in rural Mara with the ‘biomedical reality’ of their illness would have been tantamount to accusing them of being immoral persons who had led a sinful life. It was therefore only after I had learned to adapt to the social and moral codes which constituted a respectful and non-offensive conversation on ‘HIV/AIDS’ that my conversation partners and I could discuss the core issues of my research project in a meaningful manner.

110

Contextualizing Ethics

From Epistemology to Methodology to Ethics: The Politics of Doing Research on ‘the Unspeakable’ According to the Ethical Code of the American Anthropological Association (AAA) ethical guidelines of anthropological research are strongly dependent on the values, codes and norms which prevail in a specific research site and can therefore not easily be transferred from one research setting to another. While, according to the AAA, ‘anthropological researchers should obtain in advance the informed consent of persons being studied … [it] is understood that the degree and breadth of informed consent required will depend on the nature of the project and may be affected by requirements of other codes, laws, and ethics of the country or community in which the research is pursued. … [It] is understood that the informed consent process is dynamic and continuous … It is the quality of the consent, not the format, that is relevant’ (American Anthropological Association 1998, emphases added). In the case I have presented on the previous pages, the research statement that I had developed in agreement with the National Institute for Medical Research in Tanzania in order to obtain the oral informed consent of my interviewees had been an important instrument for my research in the NGOs of Dar es Salaam, as well as in the hospital context of rural Mara. However, the shift to other research sites required the adaptation and switching to different social and moral constellations, expectations and values which constituted ethically acceptable and non-offensive behaviours as well as modes of speaking and doing research about AIDS in these other social contexts. The obvious differences that existed between the NGOs, the FGBFC, and the kinship networks in Mara were now, I would argue, closely related to the respective epistemologies that shape notions of illness and healing – and the concurring ethical perceptions – in these individual settings. In the NGOs, people shared a model of HIV/AIDS as it is promoted by medical researchers and by international organizations, which are the main donors of programmes on prevention, care and treatment. In the FGBFC and in rural Mara, on the other hand, epistemologies of illness and healing were shaped not solely by biomedical views of disease, but at the same time by notions of demonic possession, witchcraft and the breaking of taboos. Each of these latter epistemologies provided its own particular interpreta111

Evidence, Ethos and Experiment

tion of disease and affliction, thereby opening up moral debates and conflicts which attributed the cause of a particular illness to the moral failures of individuals, as well as to the growing deterioration of kinship networks and society at large. As HIV/AIDS is a disease that requires, at least from a ‘local’ perspective, more than the technical and medical solutions offered by governmental and non-governmental health programmes (cf. Heald 2003), anthropologists have to be sensitive not only towards the different epistemologies that govern practices and ideas around illness and healing in a specific research setting; they should also pay heightened attention to the methodological and ethical challenges that are implied in these epistemological truths, as well as to the modalities of speaking and non-speaking about AIDS they are producing. Finally, shifting epistemologies of disease and illness have varying implications for the relationships between the anthropologist and the study participants, the anthropologist’s emotional and moral involvement in the research, and the ‘production of knowledge’ in ethnographic fieldwork in more general terms. Let me illustrate these aspects with further material from my research in rural Mara.

Different Epistemologies, Shifting Moral and Methodological Agendas In Mara, the shift from one epistemology to the other challenged, on the one hand, the empirical foundations of my research. Most of the data I collected in the villages with regard to specific cases of HIV/AIDS illness were based on rumours, gossip and speculation. As mentioned above, it was often only after a person became ill or died that I could gain more certainty about the biomedical basis of my previous classifications in which I had identified my interviewees and their family members as ‘HIV-positive’, as ‘ill with AIDS’, or as the ‘relative of a person infected with HIV’. However, while in many cases doubts remained about the ‘factual’ foundations of the collected data, I want to follow Luise White (2000), who defined stories, ideas and meanings that are produced, negotiated and perpetuated in rumours as social and therefore ‘real’ and powerful facts. In rural Mara rumours about suspected HIV infections did not only play an essential role in the ways people with HIV/AIDS experienced and dealt with their own illness – whether it was ‘really’ AIDS-related or not; rumours about HIV/AIDS also determined to a large degree the way in which social and 112

Contextualizing Ethics

familial relationships for the care and support for sick relatives were constituted and the way in which social and moral obligations and priorities were (re)formulated in the course of illness, death and burials (cf. Dilger 2005, 2008). On the other hand, my adaptation to locally accepted modes of speaking about illness as well as my growing involvement in the social and moral relationships of the area challenged the methodological and ethical implications of my research on several levels. First, the strong reliance on participant observation and on less formalized conversations in the village context required a flexible use of field methods and of speaking about disease: Many of the men and women I talked to were in an advanced stadium of their illness – others were grieving for their relatives who had recently passed away – in neither case, however, would it have been possible to taperecord our conversations or to make notes about them at the time we talked. Similarly, it would have been unethical to confront the affected person and his or her family with the biomedical diagnosis ‘AIDS’. Any urge one might have felt to address the ‘real’ nature of a disease or to talk about a sick person’s approaching death gave way to the desire to give hope and reduce anxiety and fear. In many instances it was not even acceptable to talk at all as we were all overwhelmed by the situation and the helplessness we felt at being exposed to the grief and suffering of those who were about to die or had already died. On another level, reflecting on the ethical assumptions of HIV/AIDS research demands a careful reflection on how the researcher is able to protect him- or herself throughout the course of fieldwork. The immediate encounter with suffering and death in an environment in which the participant observer is gradually becoming a part requires emotional stability and the ability to keep ‘social distance’, which is normally expected from trained health workers, medical professionals or psychologists but not from social anthropologists. At times, confrontation with the suffering and death of others has a creeping effect not only on one’s emotional state but also on one’s physical condition. As Vera Kalitzkus has remarked with regard to her fieldwork on organ transplantation in Germany, an anthropologist doing research on death and suffering will at times find him- or herself struggling, in the literal sense of the word, ‘to keep the illness and death of others away from his or her own body’ (Kalitzkus 2003: 38ff., my translation). Finally, the decision to become part of the various epistemological fields from which HIV/AIDS is negotiated and acted upon implies an ethical decision by the researcher with regard to his or her own capacity to act. The 113

Evidence, Ethos and Experiment

more one becomes involved in the ‘culture of silence’ on HIV/AIDS – and the more one becomes engaged in the logic of speaking about other kinds of illnesses and disease etiologies – the more difficult it becomes to know for certain what is ethically sound behaviour and what not. Thus, once the anthropologist has gained a deeper insight into the everyday life of people in a research setting, he or she will inevitably notice that HIV-positive men and women – sometimes without their own knowledge, sometimes obviously on purpose – are infecting others with the disease. In some cases, the researcher will feel a strong impulse to address these issues openly and to prevent the person in question from infecting others. In 2000, for example, there was the case of a young man in rural Mara who had tested HIV-positive in the hospital and who, according to people in the village, had been informed by a doctor about his diagnosis. However, while people critically watched how this man continued to sleep with young girls – and allegedly infected some of them with HIV – they did not dare to confront him directly. When I asked people in the community what they could do about these events, I did not obtain a clear answer – except for the often heard statement that ‘the only possibility to stop this man is to beat him up or to kill him’. What could a non-applied researcher do in this situation? Would it have been ethically appropriate to initiate a public controversy about this issue, considering that this step might have resulted in an outburst of violence? Would it have been advisable to consult the village authorities about this case, hoping that legal action might be taken against the man in question? (Which, by the way, I did, thereby learning that the local authorities were as helpless about this issue as I myself or the other people I had talked to as they claimed to have ‘no legal evidence’ for action as test results from the hospital were confidential.) In less obvious cases deciding on ethically sound behaviours is equally difficult and triggers a slew of ethical and moral dilemmas. In principle these dilemmas are quite similar to the dilemmas also experienced by counsellors at the local hospitals in Mara – thereby rendering their sometimes unorthodox methods of counselling as described above more understandable. One of the counsellors asked me if, for instance, the widow of a man who has died of AIDS was to be cleansed by a relative of her late husband through sexual intercourse in order to avoid chira – what should he do and what were his responsibilities in this situation? Should he inform the man who was about to cleanse the widow about his late relative’s illness – thereby preventing not only the infection of this man and his wife, but possibly also of other members of his family (e.g. the couple’s unborn children)? Could 114

Contextualizing Ethics

he do this on the basis that HIV tests in the hospitals are often being carried out without the explicit consent of the patient or of his or her relatives? Should the counsellor hazard the risk that the widow – possibly together with her children – would be abandoned by her late husband’s patrilineage, thus contributing to the further weakening and disintegration of kinship networks in Mara and possibly driving the widow into ‘prostitution’? To conclude, the different modes of speaking and non-speaking about HIV/AIDS in Mara – which are organised around complex moral dilemmas as well as around different epistemologies of illness and disease – are adding further layers to the politics and ethics of the unspeakable as sketched in the previous sections. As an anthropologist in Mara, I was confronted with at least three conflicting strategies shaping the discourse and silence on HIV/AIDS illnesses and that were employed in situated – and sometimes convergent – ways by families and individuals. These discourses, silences and moralities formed an entangled web of meanings and codes that in its intertwinement with social relationships and economic realities shaped a ‘culture of silence’ around AIDS (cf. Meursing 1997; Dilger 2005: 282–330) and was essentially organised around the three following (mutually contradicting) axes: First, many people – not only those who worked in a health organization or were otherwise connected to an ‘institution of modernity’ – felt the urgent need and/or moral responsibility to control the spread of the disease by establishing pervasive biomedical knowledge about HIV and a corresponding practice of ‘openness’ and ‘disclosure’ among the population. While this practice should ideally be the result of a voluntary process that is, among others, enhanced and promoted by AIDS NGOs, some of my interviewees felt that it should, if necessary, be enforced by the Tanzanian state. This opinion was expressed most vividly by a male nurse in Mara who told me: ‘It would be good if the government tested everybody of whom they know that his or her relatives are HIV-infected. These people should then be made known to the public so that everybody knows about them. All tests should be made under the supervision of the government and the government should then announce the names of all sick people’. The second strategy is shaped by people’s desire to control the ruptures which families and communities ascribe to an increase in deaths and illnesses among young men and women as well as to the wider social and moral implications of rural-urban migration and related processes that are understood to lead to the disintegration of social relationships in the context of modernity. The idioms of illness that correspond to this strategy – 115

Evidence, Ethos and Experiment

and that challenge an over-simplified view of public health, according to which the HIV/AIDS epidemic has resulted in an all-embracing silence on cases of illness – are the idioms of witchcraft as well as possession through satanic forces or the breach of social and cultural norms (chira). All of these idioms refer cases of illness to the wider social, cultural and spiritual contexts of the affected individual and his or her families, thereby emphasizing the ‘how’ and the ‘why’ of disease as inseparable aspects of individual and collective experiences of illness and affliction (Heald 2003: 229f.). Finally, an outspoken silence around cases of HIV is triggered by sensations of danger and pain that are connected to the thoughtless and non-voluntary revealing of the ‘true cause’ of another person’s illness. Not only may an enforced disclosure raise anxieties in the affected person about his or her approaching death. It may also have serious consequences for the reputability and the social and economic standing of the affected individual and his or her family because of stigmatization and the threatened exclusion of the affected persons from kinship and community life. Finally, ‘forced disclosure’ may have an adverse impact on community life as it is feared to set in motion a cycle of mutual accusations that in the long run may endanger social relationships in unpredictable ways.

Conclusion: Knowledge, Power and the Moral Embeddedness of HIV/AIDS Research Early on in the debate on HIV/AIDS and anthropology, Paul Farmer (1992) has argued that anthropological fieldwork on HIV/AIDS implies, on the one hand, problems and dilemmas that are associated with ethnographic fieldwork in general. According to Farmer, HIV/AIDS research poses not only the question of how conflicting interests within specific research sites – represented by governments, funding agencies, NGOs, and research communities – shape the agendas of fieldwork and consequently the research outcomes. Ethical considerations in AIDS research also touch upon the relationships between anthropologists and their informants and the way in which the often unequal distribution of power in the field affects access to research data and confidential information, as well as the representation of study subjects in research reports and publications. Finally, ethnographic AIDS research addresses the way in which the source of funding affects the epistemological presuppositions and the methodologies of fieldwork (Farmer 1992). 116

Contextualizing Ethics

On the other hand, HIV/AIDS research also confronts anthropology with more specific ethical and methodological challenges. The dramatically high HIV infection rates in many African countries – and the fact that HIV-infected and AIDS-sick people in resource-poor settings have long had a low prospect of getting access to antiretroviral treatment – raise urgent questions about the applicability of research results and the ways in which informants and their communities can profit from the research in a timely manner. Certainly, the issue of applicability speaks again to more general debates on ethics and anthropology which have centred on the question for whom ethnographic knowledge is being produced: the community, the state, the funding agency or simply for the sake of ‘science’? However, while the overall relevance and accountability of ethnographic research have moved to the centre stage of debates on anthropology and ethics since the 1970s (Hymes 1972; Scheper-Hughes 1995; Caplan 2003: 5ff.), I would maintain that there is a qualitative difference between HIV/AIDS and other topics that are not immediately connected to questions of how research results can not only improve the living conditions of communities under study but actually ‘save the informants’ lives’. Finally, research on a stigmatised topic like HIV/AIDS is affected by the multiple challenges and dilemmas that result from the way in which confidentiality and access to knowledge about the HIV-positive status of individual persons is embedded in power relations that are constituted far beyond doctor–patient relationships or the local research sites. In many parts of southern and eastern Africa, people living with HIV are reluctant to disclose their illness to third parties; others do not know about their HIV-positive status, either because of the lack of access to counselling and testing services or because the information is withheld from them by the personnel of health institutions (cf. Harrington 1998: 160f.). As the confidentiality of an HIV-positive diagnosis is protected by medical laws in many African countries, the knowledge acquired by an anthropologist regarding individuals’ HIV-infection involves critical questions not only with regard to the way in which he or she obtained this information but also to whom he or she imparts this knowledge (e.g. within the research team) and for what purpose.15 In this chapter, I have argued that the various ethical implications of anthropological AIDS research – in Africa as well as in other parts of the world – cannot be dealt with as static and unchangeable entities but, have to be subject to continuous reflection and reconsideration throughout the process of doing research. In the same vein as it is impossible to offer a 117

Evidence, Ethos and Experiment

blueprint for ethically sound research in contexts of poverty and threatening violence (cf. Scheper-Hughes 1995), ethically sound codes and behaviours on confidentiality and informed consent in AIDS research can not simply be retrieved from textbooks, formalized guidelines or pre-interviews with key informants, but are gaining shape only through dialogical processes and the growing social and moral involvement of the anthropologist in his or her research setting. As a consequence, ethical codes in anthropology can not be defined with reference to the standardized procedures of health research – nor, as many AIDS activists might prefer to have it, to a universalistic discourse on human rights. While human rights values may often converge with the personal values of an anthropologist and his or her research subjects, at other times they are ‘lifted out of any particular context and raised to the level of the categorical imperative’ (Caplan 2003: 16), thereby representing a clear discontinuity with local moral worlds and becoming ethically questionable for anthropological research. I have argued that the multiple ethical dilemmas of HIV/AIDS research as described above are further complicated by shifting epistemologies of illness and healing which in turn shape different modes of speaking and non-speaking about HIV/AIDS in shifting research settings. These epistemologies open up a slew of moral and ethical dilemmas with regard to HIV/AIDS research and have a deep impact on how the ethical topoi of confidentiality, informed consent, applicability of research results etc., have to be dealt with by the non-applied working anthropologist and by his or her informants. In conclusion, I want to maintain that all of these questions are tied into the more encompassing issue of knowledge production in anthropology as opposed to knowledge production in the natural and to a large extent, also in the social sciences:16 Knowledge production in anthropology is a reflexive process that – while mostly starting out with a clearly defined research question – may lead to less predictable outcomes throughout the course of fieldwork, the analysis of field data and the writing-up of research results. Social and cultural anthropology are not based on the collection of discrete samples or of objectively quantifiable and reproducible knowledge – nor do they rest on the assumption of a morally and socially neutral researcher. As has been shown most vividly by Clifford and Marcus (1986), the subjectivity of the anthropologist and the necessity for critical reflection on the researcher’s own position in the whole course of the anthropological enterprise, challenge the ethnographic authority of the researcher and hint to the importance of making transparent the social interactions which lead to the production of knowledge in the 118

Contextualizing Ethics

course of fieldwork, data analysis and writing. For all these reasons it may seem even more necessary to reflect on some of the dilemmas addressed in this text before the research process starts. However, it is difficult (and almost impossible) to tackle all of them in a research proposal or in an application for research clearance to be evaluated predominantly by medical and health researchers whose writing and research is based on theoretical, epistemological and methodological assumptions that often significantly differ from those of anthropologists and/or the communities under study.

Acknowledgements I would like to thank Wenzel Geissler and Anita Spring for their constructive comments on earlier versions of this chapter. I am also grateful to the Commission for Science and Technology and the National Institute for Medical Research in Tanzania for their grant of a research permit. Finally, I would like to thank the German Research Foundation (DFG) and the Heinrich Böll Foundation for sponsoring my research in rural and urban Tanzania between 1999 and 2003.

Notes 1. See the discussion in Current Anthropology, 1995, 36(3). According to Antweiler (1998), ‘applied’ research does not necessarily imply the immediate translation of research results into a policy-making context. ‘Engaged anthropology’ can refer to a wide diversity of contributions to non-academic audiences through which anthropological insights are made available for public discussion. 2. Schramm’s work dealt with the ‘homecoming’ of African-Americans to Ghana and essentially involved questions about the political legitimacy of a white, European researcher doing fieldwork on this topic. However, the ethical dilemmas I experienced in encounters with death and suffering in Tanzania were similar to Schramm’s experience in that they substantially challenged the ‘mere presence’ of the anthropologist. 3. During my fieldwork in Tanzania, NGO experts and representatives of international organizations repeatedly expressed astonishment about the long-term approach of ethnographic fieldwork ‘when people were actually dying’. However, one may justifiably also ask: Have the numerous HIV/AIDS interventions that have been developed and implemented over the last 10–20 years had a significant impact on preventing all the infections and deaths that have not occurred? Is it not

119

Evidence, Ethos and Experiment

4.

5.

6.

7.

8. 9.

10.

highly relevant for the development of effective interventions to have more detailed knowledge about the social, political and economic processes that shape people’s ability and/or willingness to adapt their behaviours to the messages of local, national and international AIDS campaigns? (cf. Campbell 2003). My research in Tanzania was carried out during several stays between 1995 and 2006 (altogether 22 months). In 1995/96, the fieldwork focussed on young people’s moral perceptions of HIV/AIDS, as well as on intergenerational and gender relations within the context of modernity (see Dilger 1999, 2003). Between 1999 and 2003 – the period which is also the focus of this chapter – I studied relationships of care and support in extended kinship networks in the Mara region, as well as through observations at NGOs and in a Pentecostal church in Dar es Salaam (Dilger 2005, 2007, 2008). In 2006, I returned to Tanzania for a brief visit focusing on the introduction of antiretroviral treatment and the growing presence of NeoPentecostalism in the country. Tanzania’s National Institute for Medical Research (NIMR) was established by an Act of Parliament in 1979 and became operational in 1980. According to the NIMR website ‘the institute is responsible for carrying out, controlling, coordinating, registration, monitoring, evaluation and promoting of health research in Tanzania’. Among the research priorities of the NIMR are Malaria, HIV/AIDS and Sexually Transmitted Infections (STI). The NIMR is affiliated with the Commission for Science and Technology (COSTECH), which is the umbrella organization for research, science and technology in the country and which was established in 1986 as a successor to the Tanzania National Scientific Research Council (UTAFITI). The teaching and research position I have held at the University of Florida between 2005 and 2007 has made me even more aware of the different prevailing national standards regarding the politics and ethics of ethnographic fieldwork. At public US Universities all research proposals involving human subjects must be submitted to an internal Institutional Review Board (IRB) which confirms the ethical soundness of the project and particularly of the Informed Consent Process. According to Harrington (1998: 149), the national budget allocation for healthcare in Tanzania fell from 9.4 per cent in the 1970s to 5 per cent in 1990/1991. In the early 1990s, 80 per cent of capital expenditure for the healthcare sector was coming from external sources. Tanzania’s largest government hospital. See, for instance, the work of Janet Bujra and Scholastica Mokake on WAMATA (2000). One client of WAMATA was featured in an interview in The Express under the somewhat sensationalistic headline ‘Full-blown confessions of an HIV victim’ (The Express, Dar es Salaam: 26 November – 2 December 1998). Initially, the NIMR had requested the preparation of an Informed Consent Form. However, since it would have been paradoxical to ask prospective research participants, for whom confidentiality was one of the most pressing issues when considering an interview with me, to sign a written consent form, I made an agree120

Contextualizing Ethics

11.

12.

13.

14.

15.

16.

ment with the NIMR that a ‘research statement’ and oral consent would be the more appropriate instruments for my purposes. How fragile this basis often was became obvious when NGO workers visited clients in their homes. Thus, one of the NGOs had to park the car which was used for visits in the context of home-based care and on which the name of the NGO was written, at a distance from the respective client’s house so that his neighbours would not become suspicious of the reason for the visit and the nature of the client’s illness. A more respected form of ‘traditional’ medicine in the NGO context was Traditional Chinese Medicine. At the time of my research TCM was being offered through the Society for Women and AIDS in Tanzania (SWAA-T), whose director, a medical doctor, was involved in a research project on Chinese medicine. According to Elisabeth Hsu (2002), the popularity of Chinese Medicine among patients in Dar es Salaam is due to the ‘rapid effects’, as well as to the ‘scientific’ and ‘modern’ appearance TCM pills and treatment are said to have. It might be argued that the unwillingness of the counsellor to support my research project was based on ethical concerns about revealing the HIV-positive status of his patients to third persons. However, as I found out later, the counsellor was probably more concerned that I might gain insight into his somewhat unusual and problematic methods of counselling. While HIV-positive patients in Tanzania do have the right to be told about their diagnosis according to the national law (Mashalla 1997; Harrington 1998: 161), many counsellors do not disclose this information to all of their patients, but instead disclose it to their family members – and in some cases, not at all. For the Guiding Principle on Medical Ethics and Human Rights that has been established by the Medical Association of Tanzania and which provides legal protection for individual citizens with regard to access to medical test results as well as against the breach of confidentiality by doctors, see Harrington (1998: 161). The illness chira is said to result from the breach of norms and prescriptions concerning the regulation of sexuality and involving strict rules on how, between whom and in which periods of time sexual intercourse is allowed or even prescribed, and in which cases the neglect of these regulations may cause illness (cf. Parkin 1978: 151; Dilger 2005, 2008; Prince and Geissler 2010). According to Baylies (2004), confidentiality is particularly problematic to maintain in a context where alone the participation in a study on ‘HIV/AIDS’ may lead to the stigmatization of informants as well as of researchers and their assistants. Moreover, the disclosure of personal information, for example in a focus-group discussion may lead, upon completion of the interview, to the participants’ enhanced reflection on their own HIV-status or that of others, which implies that the mid- and long-term effects of the research cannot always be controlled. The opposition between health and social science research, on the one hand, and anthropological research, on the other, lies not so much in the fact that health researchers or social scientists would not take into account the different perspectives 121

Evidence, Ethos and Experiment

characterizing shifting research contexts. Marilyn Strathern (this volume) has argued intriguingly that health researchers are very aware of these contexts and perspectives. The opposition consists, however, in the way the researcher becomes ethically and morally involved in the values, practice and norms prevailing in the community under study.

References African Networks on Ethics, Law and HIV. 1994. The Dakar Declaration, drafted and endorsed by participants at the Intercountry Consultation of the African Networks on Ethics, Law and HIV, organised in Dakar, Senegal, from June 27 to July 1, 1994 by the UNDP HIV and Development Programme. American Anthropological Association. 1998. Code of Ethics (Approved June 1998). http://www.aaanet.org/committees/ethics/ethcode.htm (accessed 18 October 2005). Antweiler, C. 1998. ‘Ethnologie als gesellschaftlich relevante Humanwissenschaft. Eine Systematisierung praxisorientierter Richtungen und eine Position’, Zeitschrift für Ethnologie 123(2): 215–55. Barnes, J.A. 1977. The Ethics of Inquiry in Social Science. Delhi: Oxford University Press. Baylies, C. 2004. ‘Community-based Research on AIDS in the Context of Global Inequalities – Making a Virtue of Necessity?’, in E. Kalipeni, S. Craddock, J.R. Oppong and J. Ghosh (eds), HIV and AIDS in Africa: Beyond Epidemiology. Oxford: Blackwell, pp. 15–28. Bhutta, Z.A. 2002. ‘Ethics in International Health Research: A Perspective from the Developing World’, Bulletin of the World Health Organisation 80(2): 114–20. Bujra, J. and S.N. Mokake. 2000. ‘AIDS Activism in Dar es Salaam: Many Struggles; a Single Goal’, in C. Baylies and J. Bujra (eds), AIDS, Sexuality, and Gender. Collective Strategies and Struggles in Tanzania and Zambia. London: Routledge, pp. 154–74. Campbell, C. 2003. Letting Them Die: Why HIV/AIDS Prevention Programmes Fail. Bloomington: Indiana University Press. Caplan, P. 2003. ‘Anthropology and Ethics’, in P. Caplan (ed.), The Ethics of Anthropology: Debates and Dilemmas. London: Routledge, pp. 1–27. Clifford, J. and G.E. Marcus (eds). 1986. Writing Culture: The Poetics and Politics of Ethnography. Berkeley: University of California Press. Dilger, H. 1999. Besser der Vorhang im Haus als die Fahne im Wind. Geld, AIDS und Moral im ländlichen Tanzania. Münster: Lit Verlag. ———. 2001. ‘“Living PositHIVely in Tanzania”: The Global Dynamics of AIDS and the Meaning of Religion for International and Local AIDS Work’, Afrika Spectrum 36(1): 73–90. ———. 2003. ‘Sexuality, AIDS, and the Lures of Modernity: Reflexivity and Morality among Young People in Rural Tanzania’, Medical Anthropology 22(1): 23–52. ———. 2005. Leben mit Aids. Krankheit, Tod und soziale Beziehungen in Afrika. Eine Ethnographie. Frankfurt am Main: Campus Verlag.

122

Contextualizing Ethics

———. 2007. ‘Healing the Wounds of Modernity: Community, Salvation and Care in a Neo-Pentecostal Church in Dar es Salaam, Tanzania’, Journal of Religion in Africa 37(1): 59–83. ———. 2008. ‘“We Are All Going to Die”: Kinship, Belonging and the Morality of HIV/AIDS-related Illnesses and Deaths in Rural Tanzania’, Anthropological Quarterly 81(1): 207–32. Farmer, P. 1992. ‘New Disorders, Old Dilemmas: AIDS and Anthropology in Haiti’, in G. Herdt and S. Lindenbaum (eds), The Time of AIDS: Social Analysis, Theory and Method. Newbury Park, CA: Sage Publications, pp. 287–318. Fluehr-Lobban, C. 2003. Ethics and the Profession of Anthropology: Dialogue for Ethically Conscious Practice, 2nd ed. Walnut Creek, CA: Altamira Press. Harrington, J.A. 1998. ‘Privatizing Scarcity: Civil Liability and Health Care in Tanzania’, Journal of African Law 42, no. 2: 147–71. Heald, S. 2003. ‘An Absence of Anthropology: Critical Reflections on Anthropology and AIDS Policy and Practice in Africa’, in G. Ellison, M. Parker and C. Campbell (eds), Learning from HIV and AIDS. Cambridge: Cambridge University Press, pp. 191–216. Hsu, E. 2002. ‘“The Medicine from China Has Rapid Effects”: Chinese Medicine Patients in Tanzania’, Anthropology and Medicine 9(3): 291–314. Hymes, D. (ed.). 1972. Reinventing Anthropology. New York: Pantheon. Kaleeba, N., S. Ray, and B.Willmore. 1991. We Miss You All. Noerine Kaleeba: AIDS in the Family. Harare: SAfAIDS. Kalitzkus, V. 2003. Leben durch den Tod. Die zwei Seiten der Organtransplantation. Eine medizinethnologische Studie. Frankfurt am Main: Campus Verlag. Mashalla, Y.J.S. 1997. ‘Medical Ethics and Human Rights in HIV/AIDS’, in Proceedings of the National Workshop on Ethics, Law and Human Rights and HIV/AIDS in Tanzania, 13–15 January 1997, Mbagala Spiritual Centre, Dar es Salaam. Meursing, K. 1997. A World of Silence. Living with HIV in Matabeland, Zimbawe. Amsterdam: Royal Tropical Institute. Mogensen, H. 1997. ‘The Narrative of AIDS among the Tonga of Zambia’, Social Science and Medicine 44(4): 431–39. Nguyen, V.-K. 2005. ‘Antiretrovirals, Globalism, Biopolitics, and Therapeutic Citisenship’, in A. Ong and S.J. Collier (eds), Global Assemblages: Technology, Politics, and Ethics as Anthropological Problems. Oxford: Blackwell, pp. 124–45. Parkin, D. 1978. The Cultural Definition of Political Response. Lineal Destiny among the Luo of Kenya. London: Academic Press. Prince, R.J. and P.W. Geissler. 2010. ‘Purity Is Danger: Ambiguities of Touch around Sickness and Death in Western Kenya’, in H. Dilger and U. Luig (eds), Morality, Hope and Grief: Anthropologies of AIDS in Africa. Oxford: Berghahn Books pp. 240–269. Robins, S. 2004. ‘“Long Live Zackie, Long Live”: AIDS Activism, Science and Citisenship after Apartheid’, Journal of Southern African Studies 30(3): 651–72. Scheper-Hughes, N. 1995. ‘The Primacy of the Ethical: Propositions for a Militant Anthropology’, Current Anthropology 36(3): 409–40. Schramm, K. 2005. ‘“You Have Your Own History: Keep Your Hands Off Ours!” On Being Rejected in the Field’, Social Anthropology 13(2): 171–83. 123

Evidence, Ethos and Experiment

White, L. 2000. Speaking with Vampires: Rumor and History in Colonial Africa. Berkeley, CA: University of California Press. Wolf, L.E. and B. Lo. 2001. Ethical Dimensions of HIV/AIDS. http://hivinsite.ucsf.edu/InSite?page=kb-08-01-05 (accessed 18 October 2005).

124

Chapter 5

Testing a New Drug for Leprosy: Clofazimine and Its Precursors in Ireland and Nigeria, 1944–19661 John Manton

In the increasingly laboratory-centred context of mid-twentieth century pharmaceutical research, the persistent difficulties in subjecting leprosy to laboratory investigation presented a paradox, thrusting conceptions of the field, field ethos and field research conditions centre-stage in assessing the efficacy of chemotherapy in leprosy. Until 1948, with the ratification of Dapsone at the International Leprosy Congress in Havana, there was no universally recognized effective drug therapy for leprosy, and continued problems in supervising the administration of Dapsone in the following years ran alongside an increasingly active search for alternative drugs. One group of compounds with evident potential for the treatment of leprosy was developed at the laboratories of the Medical Research Council of Ireland (MRCI) on the grounds of Trinity College, Dublin (TCD), through the 1940s and 1950s, leading to the synthesis of B.663, the compound later known as clofazimine.2 This compound, subjected to clinical trials in Nigeria in the early 1960s, is still an effective component of the multidrug regimen used to treat leprosy today. The development of clofazimine, and the demonstration of its efficacy in treating leprosy, brought together chemists, biologists, clinicians and medical researchers from locations and institutions in Africa, North America and Europe over a period of over ten years, building on a complementary set of earlier networks in Irish science and medicine, and articulating a new ethos of field practice and international collaboration in leprosy research. This chapter describes the personal and institutional networks, the scientific knowledge and practice, and the medical problems which brought the MRCI research laboratory, originally conceived in order 125

Evidence, Ethos and Experiment

to investigate the chemotherapy of tuberculosis, into contact with leprosy control programmes in post-1945 Nigeria, focusing on two series of clinical trial. The first series, trialling B.283, a precursor to clofazimine developed at the MRCI laboratory in the mid-1940s, was carried out by three doctors employed by the Irish-run Roman Catholic Mission Ogoja Leprosy Scheme in 1951 and 1952. The conduct of the trial and the publication of the inconclusive and unspectacular results evoked dissent among those involved and its adherence to emerging standards for clinical chemotherapeutic trials in leprosy was questioned by participants and observers alike. However, the publication of results coincided with the opening of new MRCI laboratory buildings and the convening of a major international colloquium on the chemotherapy of tuberculosis in Dublin, and helped fuel ongoing MRCI investigations into related compounds, and interest in assaying such compounds in leprosy. This process of publication and circulation became a key component in developing relations between the MRCI research and bodies interested in the treatment of leprosy, and sustained an important avenue for the clinical investigation of MRCI-developed compounds. As a consequence of the precedent set and publicized in relation to B.283, the institutional linkages which developed between the MRCI, Geigy, the US National Institutes of Health and the Leprosy Research Board in London fostered closer relations between leprosy research agendas and the output of the MRCI. Articulating these relations, the second series of clinical trial I examine, of clofazimine itself, was carried out by Stanley Browne and Lykle Hogerzeil at the Nigeria Leprosy Service Research Unit, Uzuakoli, between 1961 and 1966. Citing the earlier trial – and the increasingly sophisticated output of the MRCI laboratory – the conduct and presentation of these trials was ostensibly much more in accord with clinical norms, gained greater currency, and produced encouraging and definite results, leading to the standardization of clofazimine for the treatment of leprosy. My investigation of the networks of chemotherapeutic research is a historical one, based largely on the recovery of archival material from a number of sites, and guided by interviews with surviving researchers. It focuses on the transnational aspects of the research process, rather than the experience of trial participants, which are largely unrecoverable. I situate my presentation amid the history of medical experimentation, leprosy history and the history of science and technology in twentieth-century Africa and post-Independence Ireland. Using the example of these trials, I exam126

Testing a New Drug for Leprosy

ine contrasts and transformations in the conduct of field research in missionary and in government clinical settings, in colonial and in post-independence Africa, and amid rapidly transforming approaches to African patients more generally, and to leprosy patients in particular, in both clinical and research contexts. The argument considers the implications of these changes for conceptions of the field in leprosy and pharmaceutical research in Africa and worldwide. It draws attention to the evolving complexity and sensitivity of clinical language and methodology as applied to rural Africa-based research in the period under consideration. It also draws attention to tensions between the exigencies of treatment and concerns with the cost of drugs, on the one hand, and the growing epistemological requirements of the clinical trials process and the business concerns of the pharmaceutical industry on the other. In developing this argument, I have been fortunate to uncover some of the notes for the trial of B.283 at the convent of the Medical Missionaries of Mary in Ogoja, Nigeria. As of the time of writing, less is readily available on the conduct of the Nigerian trials of clofazimine. Some of the more formal documentation is to be found in the Stanley Browne papers at the Wellcome Library for the History and Understanding of Medicine in London, and there is some relevant material held by Novartis in Basle: there may also be documentation at some of the smaller leprosy hospitals in Nigeria, which reputedly helped Browne in conducting his investigations. The hospital in Uzuakoli, where Browne and Hogerzeil carried out their trials, was largely destroyed in the Biafran War of 1967–1970, while the notes of Wolfgang Vischer of Geigy, and of the MRCI laboratories in Dublin have also reportedly been destroyed. In what follows I hope as much as possible to recover a history of the development of clofazimine which recognizes the diversity of influences and inspirations that led to its eventual recognition as a useful and successful treatment for leprosy.

Lichen, Tuberculosis and the Development of B.283 in Dublin The roots of the synthetic research leading to the development of clofazimine are to be found in the efforts of the post-1922 Irish State to grapple with the elevated incidence of tuberculosis in its territory. Relative to Britain and much of western Europe, where rates of infection were in decline by the last decades of the nineteenth century, tuberculosis continued to be a major 127

Evidence, Ethos and Experiment

public health problem in Ireland well into the 1940s, only gradually tailing off after the First World War, and continuing to cause concern among medical workers, policy makers and the general public (Jones 2001: ch. 1). This concern seemed all the more urgent as incidence seemed to spike in the early 1940s (ibid.: 207, n.1); in the light of this, and of developments in chemotherapy of infectious disease in the 1920s and 1930s, the State Chemist, T.S. Wheeler, and the Professor of Experimental Medicine at Trinity College, R.A.Q. O’Meara, made an approach in 1942 to the Taoiseach,3 Eamon de Valera, to set up a laboratory in Dublin to investigate the chemotherapy of tuberculosis.4 At the time, there was no specific treatment for tuberculosis. Reasoning on a calculation of the cost of tuberculosis to the Irish exchequer, the MRCI proposed a remit for a laboratory, working on: (1) Synthesis and tests of derivations of Promin which itself has given the best results in animals, but has proved too toxic for internal use with human beings. (2) The devising of therapeutic substances to break down or displace the fatty envelope of the tubercle bacillus, with a view to subsequent attack of the bacteria by known or specifically designed drugs. (3) The assessment of the effectiveness of certain aromatic derivatives in exceptionally acute forms of tuberculous infection in animals.5 In early 1944, Vincent Barry was chosen to lead the MRCI investigations, originally located at University College, Dublin (UCD). Prior to and during the early years of the Second World War, Barry had worked with a research team headed by Professors H. Ryan and T.J. Nolan, of the UCD Chemistry Department, who, along with Joseph Algar, had determined the structure of the first known naturally occurring chlorine-containing compound (Barry 1969: 154–55; Twomey 1986: 10). From this experience, Barry brought a rigorous and systematic approach to the identification, development and testing of compounds for the treatment of tuberculosis: the MRCI laboratory, in its various guises, published copiously on its syntheses and processes until its demise in 1990. His theoretical focus on interrupting the metabolism of mycobacterium tuberculosis, and his emphasis on in vitro testing of all promising compounds derived in the MRCI laboratories, streamlined the rapid development of compounds for animal and human trials. As Barry stated in a 1946 presidential address to the Irish Chemical Association: 128

Testing a New Drug for Leprosy

Many papers on the chemotherapy of tuberculosis describe as the first step the carrying out of guinea-pig protection experiments. They give no clue as to how the substance to be tested is selected, and as a careful test in animals is costly and occupies at least six months, I leave it to my audience to decide how far an investigator will get in his lifetime with this method. (Barry 1997: 45)6 Intriguingly, from the perspective of a history of leprosy treatment, Barry outlined one synthetic approach adopted by the MRCI laboratory concentrating on the properties of chaulmoogric acid (Parascandola 2003: 47–57),7 centring on chemical similarities between this acid and isolates of the lipoid fraction of m. tuberculosis. In this fashion, what has usually been represented as a project on the chemotherapy of tuberculosis, only incidentally and fortuitously germane to leprosy, can be seen to have been influenced by analysis of leprosy chemotherapeutic agents from its inception. Barry’s first major enterprise was to examine a series of branched-chain fatty acids derived from the remnants of an earlier UCD study on lichen byproducts, and by iteration determine what modifications would inhibit the growth in vitro of m. tuberculosis. This proved useful in deriving a template for the later practice of the laboratory, and its most enduring results were achieved with one of the lichen remnant compounds, diploicin. Diploicin was insoluble, and could not be investigated in vitro (Barry 1997: 49), though Barry quickly deduced that a soluble sodium salt of a carboxylic acid of diploicin inhibited tubercule growth in horse flesh broth (Barry 1969: 154). One of the main difficulties faced by the MRCI team was the maintenance of in vitro activity in serum and in an animal host. To this end, it was decided to simplify the structures being examined, so that the physical characteristics of the molecules could be altered more readily and in repeatable fashion.8 One unsuccessful compound was inadvertently not discarded, and was left in storage for a number of months.9 It demonstrated striking anti-tubercular properties upon its retrieval. It had also changed colour from near colourless to dark red, and it was surmised that the anti-tubercular activity was related to the oxidation product which had generated the pigmentation. A series of these phenazine dyes were subsequently examined by Barry and his team. Among them, one in particular maintained its effectiveness as an anti-tubercular agent both in serum and in mouse and guinea pig tuberculosis, and was as such one of the few investigated compounds to retain its activity beyond the in vitro stage (Barry 1951a: 453).10 129

Evidence, Ethos and Experiment

This compound, known by Barry’s team as B.283, was subject to encouraging animal trials (ibid.: 470), was taken orally by Barry and his associates to demonstrate its low toxicity (Lane 1951: 163)11 and was later applied in clinical assessments on human subjects in the Meath Hospital (ibid.: 155– 75) (urogenital tuberculosis) and Rialto Chest Hospital (pulmonary tuberculosis) in Dublin.

Forays into the Field: Trialling B.283 in Nigeria In July 1951, a four-day colloquium on the chemotherapy of tuberculosis was held in Dublin, under the sponsorship of the Medical Research Council of Ireland. A summary of the proceedings (Barry 1951b: 539–41), noting the international character of both the speakers and the attendance, and the patronage of the Taoiseach, Éamon de Valera, and the Irish President, Seán T. O’Kelly, was published in Nature. In the second of seven sessions Barry presented a report on the results of empirical research by the Irish team on the synthesis of drugs of potential value in tuberculosis and leprosy, focusing for the most part on trials of B.283 in Dublin, and on leprosy, in Ogoja Province, Nigeria. Since 1945, leprosy control in much of the northern half of Ogoja Province was administered by Catholic missionaries, under the medical supervision of an Irish doctor, Joseph Barnes (Manton 2005). Leprosy presented one crucial difficulty to the chemotherapeutic investigator: its causative organism, mycobacterium Leprae, could not be cultivated in vitro. Thus, the practice had arisen of assaying in leprosy any compound which had exhibited anti-tubercular activity in humans, on the grounds that the bacilli were of the same family. The state of knowledge regarding the disease and its treatment had been set out at the International Leprosy Congress in Havana, Cuba, in April 1948, which was able to present new sulphone drugs proving effective in treating leprosy. Rapid clinical improvement was followed by much slower bacteriological improvement, though marked numbers of patients exhibited sensitivity to the drugs. In order to develop a system of clinical safeguards, given that all experimentation was necessarily carried out on human subjects, minimum therapeutic requirements for clinical tests were proposed, centring on evidence of antibacterial action, absence of toxic effects or irreversible physiological changes, freedom from undue discomfort, and visible clinical and bacteriological evidence of suppression or regression of leprosy 130

Testing a New Drug for Leprosy

within twelve months (International Leprosy Congress 1949: 68–69). While such research raised hopes and promised rapid advances in leprosy chemotherapy, the pronouncements of the Congress on classification of leprosy disappointed Nigerian representatives. Drawing on the Pan-American classification system agreed in Rio de Janeiro in 1946, which reconciled previous systems based on heterogeneous combinations of clinico-anatomical, histopathological and immunological factors into a single continuum ‘believed to be based on a biologic interpretation of the clinical facts’ (ibid.: 73), the Classification Committee of the Congress proposed a polar opposition of recognizably lepromatous (malignant) and tuberculoid (benign) leprosy, each internally differentiated according to measure of severity, bridged by a class of indeterminate (undifferentiated) cases. The clinical subdivision would be organized according to class, degree of severity, manner of evolution, localization (on the body), morphology and clinical form, thus marrying classical descriptions of symptoms with prescriptive aids to therapeutic options. The Nigerian representative objected to the confusing subdivisions, noting that in Nigeria ‘the indeterminate group is not merely a group but covers a considerable proportion of all cases, and embraces types not described in the proposals’ (ibid.: 75). The notion that tuberculoid and indeterminate cases were non-infectious was vigorously contested, and proved a controversial issue in Nigerian leprosy research over the following years, especially in the light of the Congress’ recommendation that these cases should not be included in the evaluation of chemotherapeutic trial results (ibid.: 69), a recommendation which would cast doubt on the conduct of the trial of B.283 carried out at Ogoja. The roots of this trial lie in Barnes’ European visit of 1948–1949, between tours in Ogoja, during which he married Elizabeth Allday, an English doctor working in the new British National Health Service. His return to Ireland had also introduced him to Barry’s work on the chemotherapy of tuberculosis.12 Barnes and his wife travelled to Ogoja in July 1949, and set about trialling new chemotherapeutic agents in the leprosy settlements there. Joseph Barnes was unconvinced by the use of Dapsone in treating leprosy, as he objected to what he saw as its excess toxicity in the large doses administered in the late 1940s. Prior to its standardization, and its adoption as the chemotherapeutic agent of choice by the Nigerian Leprosy Service, Barnes and Allday obtained sanction to use sulphetrone, a more expensive but ostensibly less toxic derivative of Dapsone, in its place. The year 1948 was the first full year of research at the British Empire Leprosy Relief As131

Evidence, Ethos and Experiment

sociation (BELRA) Leprosy Research Unit at Uzuakoli, with promising new results reported in the administration of sulphetrone to patients with tuberculoid leprosy, held in this account to constitute the majority of Nigerian cases (BELRA 1948: 5–6). In the case of B.283, however, the experimental work carried out in Ogoja was both novel and somewhat exotic. The initial trial, begun in January 1951, was with ten patients and, consistent with the recommendations of the Havana Congress, all ten had been diagnosed with lepromatous leprosy (Allday and Barnes 1952: 422). No reports were yet available on the success of the drug in treatment of tuberculosis in Ireland, although concurrent human trials were in progress. Alongside the trial in lepromatous leprosy, a report on which was published in the Irish Journal of Medical Science, ten patients with tuberculoid leprosy were also treated with B.283.13 This dual trial, in contradiction to the recommendations of the Havana Congress, reflected new research trends occurring in Eastern Nigeria at the time, where the predominance of tuberculoid and ‘indeterminate’ over lepromatous leprosy (International Leprosy Congress 1949: 75) seemed to render internationally conceived and agreed experimental strictures worthless. An analysis of the clinical management of the first trial of B.283 at Ogoja, transferred from Dr Barnes to Dr Denis Freeman midway through the trial period, offers a unique insight into the social, infrastructural and informational contexts of and constraints upon methods then developing in clinical experimental conduct and procedure. The correspondence between the two men, and between them and the various parties to the broader experimental context of post-1945 pharmacology, is peppered with observations on the conduct of healthcare in Nigeria, Britain and Ireland, on the appropriate equipment and services required to successfully document clinical outcomes related to drug trials, and on the mechanics and expertise evolving around diagnostic procedure, as bacteriology and pharmacology converged in the clinical nexus. Scholarly literature on the evolution of experimental methods in biomedicine is largely inattentive to more marginal or less expressly successful trials: in this regard, the Ogoja example offers a worthwhile and telling corrective to dominant historical models of clinical experiment.14 In comparison with other indications of interest in B.283 for the treatment of leprosy, such as the prospective trials at Makogai, Fiji and Sokoto, Northern Nigeria, indicated in Barnes’ correspondence,15 extensive documentation exists in Ogoja of the experimental work carried out there. This is not the case for all compounds examined in Ogoja: indeed, it is the trans132

Testing a New Drug for Leprosy

fer of responsibility from Barnes to Freeman, far from ideal as this was felt to be, which enables us to examine the progress of this trial. In contrast, the experience of Barnes, Allday, Chambers, Feeney and Freeman with drugs such as Dapsone, sulphetrone, chloromycetin, calciferol and myvizone register but little on the archival and published record. One strand in this correspondence concerns the conceptual and philosophical outlines of the trial process in which the two clinicians were engaged. Much of this material is abstracted from reflection on the practicalities and logistics of carrying out research in such a remote location, and thus equips us with a sense of the infrastructural demands of laboratory and clinical work in ostensibly marginal colonial settings. Barnes conceded that his interim report on the B.283 trial had been inclined towards optimism, in contrast to the caution which had characterized his first spell in Ogoja, but the ensuing discussion between the two doctors, based on the individual cases, raises significant questions about the adequacy of diagnosis as practised in Ogoja and of the presentation of clinical information to a broader scientific and lay audience.16 Robert Cochrane, among the most eminent working leprologists, based by this time in Britain, had visited Freeman in Ogoja, and had queried some of Barnes’ diagnoses on which case selection for the B.283 trial was based. Commenting on the resulting disagreements, Barnes wrote that they may have arisen: because you had only an abbreviated summary history of the cases and none of the original photographs ... from these photoes [sic] and descriptions you may classify for yourself these cases, but it is not likely many will disagree for as you know there is a variety of classifications for leprosy. It must have been difficult for Cochrane who never saw these cases if he had not all the data I have now given you, as well as a missing lepromin reaction and tissue section which from your letter I understand he requires for the complete diagnosis of such cases.17 Barnes’ queried the accuracy of the term ‘atypical’, applied to the diagnosis of a presentation of leprosy he had classed as ‘pre-lepromatous’, noting infiltrations on the nose and ears, giving positive smears, as evidence of the case having crossed a border to lepromatous. Coupled with diagnostic difficulties were a series of technical issues which obviated the validity of some of the eventual conclusions of the trial. The nearest pathology lab was in Lagos, and at the time of the jointly con133

Evidence, Ethos and Experiment

ducted trial on B.283 Barnes and Freeman relied on black and white photography to document their diagnoses. Neither doctor was confident in their documentary abilities with a camera: thus the advice sent by Barnes to Freeman 1. white background such as a wall or sheet. 2. Lesions so adjusted to catch the rays of the sun on the slant. 3. three feet distance from lesion to lens. 4. timer 1/100 sec. 5. aperture vary according to lighting – if in doubt – 3 apertures f.5, f.7, f.9 for 3 separate snaps. 6. panchromatic film.18 The striking dissociation of disease and symptom from patient, so often noted as a hallmark of laboratory pathology, is very much to the forefront in these technical considerations, nowhere more so than in the characterization of lens-object distance with regard to the ‘lesion’. In contrast, the patient notes and published report on the trial were not anonymized; Barnes enquired on the physical and material progress of a number of patients in his correspondence with Freeman,19 and his notes contain remarks such as ‘Wona’s physique is, like his character, tough and lean’, ‘Dick presents a most striking picture in his red jersey, cast-off battle dress and navy shorts’. Indeed for Barnes, the distinction between trial and treatment, key to the specifications outlined in Havana, was at best distracting, and possibly spurious, as evidenced by his report that ‘we have been unable to employ an untreated control group as our patients would not remain isolated without the inducement of treatment’ (Allday and Barnes 1952: 422). Alongside updates on equipment needed in Ogoja and advice on the conduct of the first B.283 trial, Barnes’ letters from Ireland reported medical and personal news relating to novel chemotherapies and social and professional events in Dublin. Amongst the personal vignettes was included repeated commentary on the doings of Freeman’s father, Edward, a prominent Dublin physician, whose pronouncement that ‘to learn a technique is easy, what is difficult is judgment of its use’ had particularly impressed itself on Barnes.20 Such comments are a reminder of the professional and international context in which the somewhat disputed reports on the efficacy of B.283 in leprosy were promulgated. Indeed, the setting for this material is especially complex, as it draws on, and illustrates, an international milieu which throws the Ogoja material into relief. The changing state of knowledge with regard to treating tuberculoid leprosy, signalled in the recommendation of the Havana Congress that clinical trials be carried 134

Testing a New Drug for Leprosy

out on lepromatous cases, impacted on the validity of the results of many trials being carried out in Nigeria in the early 1950s. Freeman’s second trial on B.283 in tuberculoid leprosy seems, for instance, to have disappeared without a trace. At the same time, Cochrane and others continued to assent to the utility of B.283 in the treatment of both tuberculoid and lepromatous leprosy, even as it became clear that other compounds were both cheaper and more effective.21 A positive result of the Ogoja trial, and one which perpetuated its reputation over the next decade, was that Barry was able to demonstrate some success in the treatment both of tuberculosis and of leprosy by the time of the Dublin colloquium, thanks in part to a series of photographs exhibited by Barnes, relating to promising interim results of the first Ogoja trial of B.283 (Barry 1951b: 540). In a testament to the complementarity of clinical and laboratory-based studies at this particular conjuncture in the history of chemotherapy, the scientific team based around Barry saw Barnes’ desire to trial B.283 in the treatment of lepromatous leprosy as promising ‘fresh clues to guide synthetic work’, and it was noted that ‘the activity in vivo of these basic fatty substances is of especial interest, as it provides some justification for the long preoccupation of workers in the field with compounds of a fatty character’ (Barry 1951a: 473).22 Meanwhile, for Barry and his team, the seeds of a long association with the treatment of leprosy had been definitively sown.

‘The Strange Story of Dr Barry’s Leprosy Cure’: The Construction of Clofazimine The Dublin colloquium underlined the place of the MRCI lab at the heart of de Valera’s conception of Irish scientific excellence. However, the pharmaceutical industry remained undeveloped in mid-twentieth century Ireland. Consequently, the synthesis, evaluation and production of B.663 (clofazimine) was the result of a growing collaboration between the MRCI laboratory and the Swiss drug company Geigy. The compound was originally synthesized at the MRCI laboratory in 1954 (Yawalkar 2002: 121). While less active in vitro than the family of compounds from which B.283 had emerged, it proved prodigiously active in murine (mouse) tuberculosis. The exceptional difficulties of synthesizing many of the compounds developed by Barry’s team on site in Dublin, B.663 among them, lent great importance to the collaboration with Geigy, who could provide facilities 135

Evidence, Ethos and Experiment

for production, toxicology and co-ordination of animal and human trials on a scale to which the MRCI could never aspire. Some idea of the scale of the investigations mounted by Geigy is given in a document prepared for Stanley Browne, detailing the solubility and stability of the compound, its toxicity in various preparations, its absorption into fatty tissues, its side-effects, its therapeutic effects in vivo both on its own and in combination with other tuberculostatic drugs, and its clinical effects in human tuberculosis as examined at Davos, Paris, Illiers, Vienna, and Borstel in the late 1950s. The introduction to this document cites the earlier trial of B.283 in Ogoja, and notes a shared side-effect; the red pigmentation of skin, hair, and urine resulting from continued use. This side-effect mirrored the manufacturing process, of which Wolfgang Vischer noted ‘Almost everyone who came into contact with the substance had a problem shaking it off… clofazimine is a red powder that will fill the whole room at the slightest sneeze. The development chemists were not very taken with it!’ (Novartis Foundation for Sustainable Development) Preliminary investigations into human tuberculosis had not been encouraging, and further assessment was almost abandoned (Yawalkar 2002: 121), but interim results in murine leprosy23 trials conducted by Y.T. Chang, a previous collaborator of Barry’s at the National Institutes of Health, USA (Barry 1969: 159), evidently persuaded Geigy to reconsider, and at a 1960 London meeting between Barry, Vischer and R.G. Cochrane, by now Director of the Leprosy Research Unit in London, it was suggested that Browne trial B.663 in leprosy at Uzuakoli, Nigeria (Yawalkar 2002: 121). Having been appointed Director of the Leprosy Research Unit at Uzuakoli in 1959, Stanley Browne returned there to begin the trial in September 196024 in collaboration with Lykle Hogerzeil. By the time of his arrival in Nigeria, Browne had earned a notable reputation as a leprologist as a result of work carried out in the Belgian Congo, where he had begun as a surgeon in 1936 (Hunt 1999: 209, 228). His structured way of working, and of organizing, reporting and publicizing his investigations25 underlined the contrast between his trials in B.663 and the earlier trials of B.283 carried out in Ogoja. Already at the time of Barnes’ trials, the Unit at Uzuakoli was at the forefront of systematic evaluation of chemotherapy in leprosy, with the scale and facilities to carry out complex series of investigations such as those needed to standardize the dosage of Dapsone in the treatment of leprosy, research carried out at Uzuakoli in the early 1950s. By the time of Nigerian Independence in October 1960, it had a worldwide reputation for the quality of its research. 136

Testing a New Drug for Leprosy

However, the pilot trial of B.663 was very small in scale, three patients receiving B.663 in combination with Dapsone, and three receiving B.663 alone.26 All had lepromatous leprosy, of comparable severity. There seem to have been four other patients receiving B.663, in an exercise which was not recorded in the published results of the trial (Browne and Hogerzeil 1962: 6–10). The trial determined that B.663 improved the clinical state of the patient with lepromatous leprosy, leading to a fall in the bacterial index, an effect which was enhanced in combination with Dapsone. A supplementary report noted a form of drug resistance, and discounted B.663 as a serious rival to Dapsone, due to its high cost (ibid.: 182–84). Further studies, by Browne, and by F. Imkamp at Liteta Leprosarium, Zambia, demonstrated anti-inflammatory properties of B.663, which lessened the dependence of an important cohort of leprosy patients on cortico-steroids, and proved a crucial boost to the reputation of the drug, which was introduced under the trademark Lamprene in 1969. This process led to considerable tensions in the relationship between the MRCI laboratory, five of whose scientists had lodged a Swiss Patent Application in 1956,27 and Geigy. Any evaluation of this collaboration must take account of the pressures resulting from changes in the structure of pharmaceutical research funding, which altered the relations between Geigy and the MRCI laboratory even as clofazimine was beginning to demonstrate its impressive promise in human clinical trials, and which has impacted on published pronouncements on the development of the rimino-phenazines. Much of the eventual strain between the two bodies resulted from technical problems involved in synthesizing clofazimine on a large scale, due to starting and intermediate materials being difficult to obtain, and poor yields. Barry evidently felt that a drug which had been shown as successful in the treatment of leprosy should be available to those most in need, and that the prohibitive cost of Lamprene undermined the purposes of his research (Bannon 1974: 9). While the agreement between Geigy and the MRCI laboratory meant that the research in Dublin could continue to benefit from royalties accruing from Geigy’s marketing of Lamprene, Barry’s team continued to work on more easily synthesized compounds with analogous action, creating a rift between the erstwhile collaborators. Meanwhile, a series of trials through the 1970s encouraged the World Health Organization to examine the potential of a multidrug therapeutic (MDT) approach to leprosy, culminating in the adoption of an MDT regimen from 1982. Clofazimine was a recommended component of this regimen, and has since been available without charge to leprosy patients in treatment. 137

Evidence, Ethos and Experiment

Conclusion Stanley Browne derived significant acclamation from his role as Director of the Leprosy Research Unit, first at Uzuakoli and later in London, and from his early success with B.663. This acclamation enabled him to act as arbiter with respect to later trials of the compound and signalled a sea-change in the relation of the research institute to the treatment of leprosy. Much of this change was due to the existence of successful chemotherapeutic agents, starting with Dapsone in the late 1940s, but developments in the conduct, reporting, and reception of clinical trials in the post-war decades are clearly discernible in the contrast between the trials of B.283 in Ogoja, and B.663 in Uzuakoli. Whatever the validity of Barnes’ methods and diagnoses, its publication in The Irish Journal of Medical Science gave it a currency, and gave the relation of Barry’s work on tuberculosis chemotherapy to leprosy a continuing identity, which emerged as vital in the development of clofazimine, from Barry’s work, and in the Nigerian research centre at Uzuakoli, in the 1960s. More to the point, at a time when the balance in medical research was shifting ever more decisively from the clinical to the laboratory-based, and the clinical trial took its place, as an ancillary of healing, among the range of technical operations performed in expanding colonial healthcare institutions, the trial of B.283 at Ogoja provided an especially eloquent index of the plans, aspirations and capacities of even the most remote of medical enterprises.28 Indeed, the less than ideal physical and geographical circumstances of most leprosy research, the impossibility of cultivating leprosy outside a living human host, and the persistent privileging of local clinical knowledge – at times promoted in the most wilful terms – all help us understand the experimental clinical setting of Ogoja as characteristic rather than anomalous. With the burden of leprosy in Nigeria so heavily skewed towards the tuberculoid, the strictures elaborated at Havana in 1948 seemed to impoverish the context for leprosy research in Nigeria. The technical capacity to sustain a successful clinical trial and to firmly ground the results of a clinical research intervention did not exist in Ogoja in the early 1950s: Barnes’ location of counter-arguments to objections to the conduct of the trial of B.283 in lepromatous leprosy in notions of personal practice and individual clinical expertise exposes a moment of transition in clinical research in leprosy. Over the next decade, leprosy work in Nigeria contributed to an enriching of the classificatory and diagnostic contexts elaborated in Havana, enabling 138

Testing a New Drug for Leprosy

a more robust circulation between the local instance and the global phenomenon of leprosy and its control, and providing a more systematic grounding for the appreciation of research results in clinical leprosy trials. Indeed, the intervening years wrought infrastructural and capacity changes so great that it is difficult to consider the trial of B.663 at Uzuakoli as linked in any more than a circumstantial way to the Ogoja trials. Technical and methodological advances had transformed the ethos of leprosy research and conceptions of the field in leprosy and had enmeshed the local in a series of more closely aligned and clearly articulated transnational concerns and networks. The startling success, from a scientific point of view, of a trial hardly larger or more ambitious than that carried out at Ogoja, testifies to a shift in attitudes to local clinical knowledge even regarding a condition as experimentally intractable as leprosy. The methodological developments separating the trials in Ogoja and Uzuakoli, contingent as much on infrastructural capacity as on growth in knowledge, affected the work of the MRCI laboratory in other ways. As the main link between the two trials, and the foremost locus of Irish medical research from the 1940s through the 1980s, the fate of the MRCI laboratory reflected the development of experimental chemotherapy in the twentieth century.29 Increasingly unable to replicate the success of large pharmaceutical companies, the laboratory was seen to represent a diminishing return on investment: the MRCI was folded into the new Health Research Board in 1986, and the laboratory followed its parent organization into oblivion, much to the consternation of the international leprosy medical fraternity, in 1990. While it has not yet been possible to contrast the approach to the patient in the trials under consideration, some preliminary remarks can be made about the construction of the field in leprosy. Leprosy, as an object of study, remains to this day firmly anchored in its human host. No conclusive results can be demonstrated without recourse to an infected individual. At the same time, developments in classification, the more widespread dissemination of diagnostic techniques and equipment, and the availability of a recognized and standardized treatment, none of which were available in Ogoja in 1951, exacerbated and generalized the dissociations between disease and patient already experienced by Barnes and Freeman in their ‘lesion’ photography. Strikingly, Allday and Barnes’ contribution to the Irish Journal of Medical Science names the patients who took part in the trial, a personalization which seemed alien, if not intrusive, ten short years later (Allday and 139

Evidence, Ethos and Experiment

Barnes 1952: 421–25). Indeed the intimacy of Barnes, Allday and Freeman’s engagement with trial participants foreshadows contemporary concerns, where the recognition of the human face, the identity and the name of historical sufferers of leprosy is now seen by leprosy activists as part of the process of promoting a positive, non-discriminatory image of leprosy and its sufferers. It should also be noted that the clinical literature regarding B.283 and clofazimine makes no distinction based on race between European tuberculosis patients and African leprosy patients, nor does the treatment of trial subjects appear to differ in any way. More investigation of the conduct of the trials, the nature of the contract between Geigy and the MRCI laboratory, and the modelling of the networks that maintained the links between the MRCI investigation into the chemotherapy of tuberculosis and the treatment of leprosy will need to be carried out before a definitive account can be written of this fascinating current in the histories of leprosy control, clinical experimentation, Irish science and medicine in rural Africa.

Notes 1. I gratefully recognize the impact of my discussions on the history of leprosy treatment with Jo Robertson on the conception and development of the argument presented here. I have also benefited from discussions of Clofazimine, its synthesis, and its action with Colin McDougall, Stanley McElhinney and Raphael Darcy. The support of a Wellcome Trust Fellowship has made this research possible. Any errors or omissions in the resulting text are, of course, my own. 2. I routinely use the standard name clofazimine to refer to the drug synthesized by the MRCI as B.663 (Dublin), produced by Geigy as G 30 320 (Basle), and marketed as lamprene. In the main set of clinical trials to which I refer, the compound was known as B.663, and I will adopt this usage in reference to these trials and to its original synthesis in Dublin. 3. Taoiseach is the official designation for the Irish prime minister. 4. National Archives, Ireland (NAI). TAOIS/S 13025 A. Medical Research. T.S. Wheeler and R.A.Q. O’Meara, Memorandum for An Taoiseach in regard to the organization in Eire of research on the chemotherapy of tuberculosis. Dated 12th November 1942. 5. NAI, TAOIS/S 13025 A. Letter, dated 9 April 1943, from J.M. O’Connor, Honorary Secretary, Medical Research Council of Ireland to the Secretary, Department of Local Government and Public Health. 140

Testing a New Drug for Leprosy

6. This is a reprint of a 1946 address to the Irish Chemical Association. 7. Chaulmoogra oil was derived from an Indian remedy for leprosy, and was in vogue in biomedical circles from the early twentieth century. 8. Barry (1997: 47) notes that previous inquiries had posited that a combination of physical – rather than specific chemical – properties were at the basis of bactericidal action of branched-chain fatty acids. 9. Twomey (1986: 11) likens this inadvertent process to Alexander Fleming’s serendipitous results with penicillin. 10. Intriguingly, of the compounds listed by Barry which retained anti-tubercular activity in the guinea pig, the three which had some measure of success in treatment of leprosy – Dapsone (diaminodiphenylsulphone), conteben (thiosemicarbazone), and B.283 (2-anilino-3-imino-5-phenyl-phenazine) – were noted as having ‘one property in common: they all contain a basic nitrogenous group’, while seemingly sharing no other obvious relationship. 11. Lane’s note is the only reference I have seen to self-administration of compounds at the MRCI laboratory; Barry (1969: 156) later claimed that B.283 ‘toxicity was considerable’. 12. Author’s interview with Joe and Betty Barnes, 27 March 2000. 13. Ogoja Convent Files. Letter dated 29 September 1951 from J. Barnes (Dublin) to D. Freeman (Ogoja) mentions the two sets of cases. 14. See Lederer (1995) and Marks (1997) for an overview of approaches to the history of the clinical trial in the twentieth century. 15. Ogoja Convent Files. Letter from Barnes to Freeman, dated [28] September 1952. 16. Ogoja Convent Files. Letter from Barnes to Freeman, dated 5 May, 1952. 17. Ibid. 18. Ogoja Convent Files. Letter from Barnes to Freeman, dated 2 April 1952. 19. Ogoja Convent Files. Letters from Barnes to Freeman, dated ‘last Sunday in Sept.’ [1951] and 2 April 1952. 20. Ogoja Convent Files. Letter from Barnes to Freeman, dated 14 June 1952, reporting the impact of his plenary address to the joint British Medical Association/Irish Medical Association meeting in Dublin in 1952 on Freeman’s father’s growing reputation among Irish physicians. 21. Ogoja Convent Files. Letter from Barnes to Freeman, postmarked 14 June 1952. This letter mentions a trial carried out in England by Cochrane on a number of patients with tuberculoid leprosy. 22. Twomey (1986: 10) comments on the unwitting and reluctant participation of the Irish researchers in a then-fashionable field in biochemical enquiry! 23. References to murine leprosy in the medical literature acknowledge the development by the 1960s of an experimental model for assaying toxicity in chemotherapeutic agents for leprosy on the mouse footpad successfully inoculated with m. leprae. 24. Wellcome Library Archives. WTI/SGB/C.1/4/1 – B.663 in leprosy. This date is given in a handwritten note dated August 1960. 141

Evidence, Ethos and Experiment

25. Wellcome Library Archives. WTI/SGB/C.1/4/1 – B.663 in leprosy. Browne’s notes for F. Inkamp’s trial of B.663 in cortico-steroid dependent patients give some idea of the rigour of his observations regarding patient history, type and degree of sideeffects observed, degree of neuritis and lepra reaction observed, psychological effects of treatment, method of drug administration, and toxic effects observed. 26. Wellcome Library Archives. WTI/SGB/C.1/4/1 – B.663 in leprosy. Note headed ‘B 663 – Uzuakoli patients’, detailing the timing and labelling of biopsies taken from the patients in the trial subject group. 27. Dr Raphael Darcy, Department of Chemistry, UCD, has kindly made available to me a number of documents referring to the agreement between the MRCI and Geigy. Among these is a note, dated 4 April 1957, in which J.G. Belton, M.L. Conalty, J.F. O’Sullivan and D. Twomey, co-inventors with Barry of B.663, agree that Barry shall act in their interests regarding the Patent Application, no. 1048/76, dated 4 October 1956. 28. The chapters by Ann Kelly, Patrick Kachur, and Steven Feierman in this volume also concern themselves with aspects of this tension between clinical and laboratory-based medicine. 29. Personal correspondence with Stanley McElhinney, Trinity College, Dublin.

References Allday, E.J. and J. Barnes. 1952. ‘Treatment of Leprosy with B.283’, Irish Journal of Medical Science 6(322): 421–25. Bannon, R. 1974. ‘The Strange Story of Dr Barry’s Leprosy Cure’, Hibernia 29 March: 9. Barnes, J. 1951. ‘Diaminodiphenylsulphone in Leprosy’, The Lancet 29 September: 595. Barry, V.C. 1951a. ‘An Organic Chemists’ Approach to the Chemotherapy of Tuberculosis’, Irish Journal of Medical Science 6(310): 453–73. ———. 1951b. ‘Chemotherapy of Tuberculosis’, Nature 168(4274): 539–41. Barry, V.C. 1969. ‘Boyle Medal Lecture: Synthetic Phenazine Derivatives and Mycobacterial Disease: A Twenty Year Investigation’, Scientific Proceedings of the Royal Dublin Society Series A, 3(16): 153–71. Barry, V.C. 1997. ‘Antitubercular Compounds: Presidential Address’, Irish Chemical News Winter: 44–50. British Empire Leprosy Relief Association. 1948. Annual Report. Browne, S.G. and L.M. Hogerzeil. 1962. ‘B.663 in the Treatment of Leprosy’, Leprosy Review 33: 6–10. Hunt, N.R. 1999. A Colonial Lexicon of Birth Ritual, Medicalization, and Mobility in the Congo. Durham, NC: Duke University Press. International Leprosy Congress. 1949. Memoria del V Congreso Internacional de la Lepra : celebrado en La Habana, Cuba del 3 al 11 de Abril de 1948 / organizado por el Gobierno de la República de Cuba con la colaboración de la Asociación Internacional de la Lepra. Havana: Editorial Cenit. Jones, G. 2001. ‘Captain of All These Men of Death’: The History of Tuberculosis in Nineteenth and Twentieth Century Ireland. Amsterdam: Editions Rodopi. 142

Testing a New Drug for Leprosy

Lane, T.D.J. 1951. ‘Chemotherapy in Urinary Tuberculosis’, in Medical Research Council of Ireland, Proceedings of the Colloquium on the Chemotherapy of Tuberculosis. Dublin: MRC Ireland, pp. 155–75. Lederer, S.E. 1995. Subjected to Science: Human Experimentation in America Before the Second World War. Baltimore: Johns Hopkins University Press. Manton, J. 2005. ‘The Roman Catholic Mission and Leprosy Control in Colonial Ogoja Province, Nigeria, 1936–1960’, unpublished D.Phil. dissertation, Oxford: University of Oxford. Marks, H.M. 1997. The Progress of Experiment: Science and Therapeutic Reform in the United States, 1900–1990. Cambridge: Cambridge University Press. Novartis Foundation for Sustainable Development. n.d. The Story of Multiple Drug Therapy with Dapsone, Lamprene® and Rimactane®. http://www.novartisfoundation.com/ en/projects/access_health/leprosy/multidrug_therapy/history_mdt.htm (accessed 31 October 2006). Parascandola, J. 2003. ‘Chaulmoogra Oil and the Treatment of Leprosy’, Pharmaceutical Historian 45(2): 47–57. Twomey, D. 1986. ‘An Irish Solution for a Non-Irish Problem’, Irish Chemical News Spring: 10–12. Yawalkar, S.J. 2002. Leprosy for Medical Practitioners and Paramedical Workers, 7th ed. Basle: Novartis Foundation.

143

Chapter 6

Elucidating Ethics in Practice: Focus on Accountability George Ulrich1

The theme of accountability has not been well illuminated in the current debate about medical research ethics.2 The predominant tendency has been to focus on the definition and enforcement of a relatively narrow range of rigidly defined standards, usually through coercive means. Ethical standards thus come to assume a quasi-legal character and the notion of accountability is implicitly modelled on a juridical paradigm. A related salient tendency is to assimilate ethical review and accountability procedures to models derived from the financial sphere, which, too, imply an external enforcement of compliance with established standards. This is often perceived negatively by practitioners. A recent interpretation, which will mark the point of departure for the present analysis, regards ethics reviews as an instrument of an increasingly dominant ‘audit culture’ that imposes bureaucratic control on, and thereby distorts, essential professional practice and scientific research. Clearly, this does not ensure genuine ethical compliance. However, neither does the most common counter reaction, which consists in a (partial or wholesale) dismissal of formalized ethical standards and review mechanisms and retreat to a purely private sense of ethical obligation. In contrast to both of these tendencies, the aim of the present chapter is to place renewed focus on the theme of accountability in a distinct ethical sense. This should not to be regarded as running contrary to the ongoing codification of authoritative national and international standards of medical ethics (and professional ethics more generally), but rather as a means of properly contextualizing such standards and in a wider sense reinforcing ethical awareness and compliance within the contexts in which research and other professional activity unfolds. This is particularly important when it comes to areas of professional activity involving extreme inequities and 145

Evidence, Ethos and Experiment

acute power imbalances – such as is the case with externally sponsored medical research in low-income countries or communities. What is required in order to develop a viable theory of ethical accountability is a combination of conceptual analysis and empirical elucidation of how ethics mechanisms function, or fail to function, in practice. One needs to look not only to philosophical sources but also to concrete research settings: what are the implicitly operative norms and what expectations of accountability are being asserted or indirectly indicated (e.g. through expressions of frustration) in connection with research? How are diverging ethical interpretations negotiated in real-life situations, and how is professional conduct affected by the formalization of ethical standards and institutional review procedures? This field of inquiry, which I have elsewhere associated with a notion of ‘emergent ethics’ (Ulrich 1999), involves a significant social science component.3 Anthropology in particular has an important role to play in view of its rich tradition of qualitative analysis of health-seeking behaviour and the experiential dimension related to medical practice. As the ethical implications of this analytical perspective remain largely undeveloped, a primary objective in the present context will be to identify key questions for further investigation in what may be foreseen to be a joint effort involving medical practitioners, social scientists and ethics specialists. The exposition that follows will fall into two main parts. The first part is predominantly conceptual. It begins with a brief examination of the critique of contemporary audit cultures, as developed by Strathern and others, as this may be regarded as symptomatic of a more widespread scepticism about formalized ethics procedures. Following this, I seek to reconstruct a Kantian/Habermasian notion of ethical accountability, as distinct from idioms of accountability pertaining to financial administration and the legal sphere. I argue that all professional activity, including research, is inherently public in nature and hence subject to a requirement for public ethical legitimation of actions taken. This gives rise to a variety of conceptual and practical challenges having to do with the underlying notion of the public sphere and the distinction between public and private ethical commitments. Another key issue in need of conceptual clarification has to do with how to establish valid, operational normative references, in particular in settings where the actors involved cannot be assumed to share a common moral or religious outlook. While not the primary focus of the present paper, this question is addressed in general terms. I acknowledge that formalized ethics codes have an important role to play but suggest that a less 146

Elucidating Ethics in Practice

rigidly defined categorization of pertinent ethical considerations may be more conducive to framing concrete discourses of ethical accountability. In view of the proposed concept of ethical accountability, the second part of the chapter is devoted to the examination of concrete factors that inhibit its realization in practice, notably in relation to externally sponsored research in developing countries. These are grouped under three main headings, namely the problem of weak accountability structures, ethics as a language of power and the disconnect between science and social process. In all three regards one finds that the discourse of ethics is structurally unhinged, largely on account of familiar global imbalances in power and access to resources. Despite such complications it is suggested, in conclusion, that a range of concrete measures can be taken to reinforce ethical accountability in practice. I regard this as imperative above all because it is a precondition for ensuring dignified relations of interaction in settings that otherwise militate against human dignity.

The Case against Audit Culture [T]he market model of audit and accountability, which treats people like commodities, has inevitable damaging consequences when applied to the public sector. … The time has come to hold audit itself to account so that we may realize the true extent of the disastrous social costs of this coercive new form of governance. (Shore and Wright 2000: 85) ‘As an instrument of accountability,’ Marilyn Strathern observes, ‘audit is almost impossible to criticize in principle – after all, it advances values that academics generally hold dear, such as responsibility, openness about outcomes and widening of access’ (Strathern 2000: 3). Nevertheless, this is precisely what is being done by the majority of authors contributing to the influential volume edited by Strathern titled Audit Cultures, which draws on extensive comparative empirical analysis to develop a critical social science perspective on a set of presuppositions that are so ingrained in contemporary administrative practice as to appear self evident and of obvious utility and legitimacy. The background to the investigation is that audit procedures give rise to widespread misgivings among the practitioners affected and in some cases contribute to a certain culture of scepticism, even cynicism about the relevance of public accountability. This may extend to professional ethics as well,4 in so far as this domain is seen to revolve around externally imposed procedures of monitoring and control. Several contributors to the Strath147

Evidence, Ethos and Experiment

ern volume implicitly or explicitly interpret the contemporary paradigm of professional ethics as an extension of the predominant audit culture. The notion of audit in current discourse is modelled on a financial accounting idiom and is semantically linked with a broad cluster of related terms, which aside from accountability include notions such as good transparency, efficiency, quality control, quality assurance, accreditation, responsibility, and performance (Shore and Wright 2000: 60). Together this constellation of values has been asserted with remarkable force in public affairs far beyond the realm from which they originally derive: ‘The concept of audit in turn has broken loose from its moorings in finance and accounting: its own expanded presence gives it the power of a descriptor seemingly applicable to all kinds of reckonings, evaluations and measurements’ (Strathern 2000: 2). ‘The past two decades have witnessed a striking proliferation in the use of the term audit and its extension into contexts where it was seldom previously used’ (Shore and Wright 2000: 58). While seemingly benign, the predominant idiom of audited accountability constrains behaviour by establishing a context within which ‘only certain operations will count’ (Strathern 2000: 2). The ‘audit explosion’ (Power 1994) at one and the same time creates a powerbase for a new class of experts and leads to a widespread internalization of norms reinforcing external control over qualitative scientific and professional work. Peter Pels extends this analysis to professional ethics as such, which when ‘understood in terms of a set of quasi-legal rules, [forms] part of a specific technology of the (professional) self ’ (Pels 2000: 136). Chris Shore and Susan Wright, similarly, analyse audit in terms of the Foucauldian notion of normalization power. Audits, they suggest, ‘rely on techniques of the self which render political subjects governable by requiring that individuals behave as responsible, self-activating, free agents who have internalized the new normative framework. … thus, squaring the circle of efficiency, economy and arms-length control’ (Shore and Wright 2000: 71). A particularly worrisome feature of audit is the way in which it corrupts reciprocal inter-subjective relations. To be accountable, according to the operative paradigm, means to be obliged to answer to others but not necessarily to be able to pose questions in turn. ‘Audit is essentially a relation of power between scrutinizer and observed: the latter are rendered objects of information, never subjects in communication’ (ibid.: 59). The rise of audit cultures, by the same token, implies a fundamental deterioration of trust and is antagonistic to any sense of community. This, indeed, makes for a peculiar contextualization of ethics. Virtually all contributors to the 148

Elucidating Ethics in Practice

Strathern volume view this combination of managerial control and undermining of reciprocity, community and basic interpersonal trust as a direct function of the global rise of neoliberalism. In the words of Pels, ‘auditing and accounting have become the operational signs of the global spread of neo-liberal values’ (Pels 2000: 135). The prevailing critical conclusion, then, is that ‘audit technologies are not simply innocuously neutral, legal-rational practices: rather, they are instruments for new forms of governance and power. They embody a new rationality and morality and are designed to engender … new norms of conduct and professional behaviour. In short, they are agents for the creation of new kinds of subjectivity: self-managing individuals who render themselves auditable’ (Shore and Wright 2000: 57). This is perceived to have detrimental consequences for conscientious, socially committed professional practice. Such a critique of audit culture may well be justified when it comes to the consequences of imposing accountancy standards – in an ethical guise – on a broader realm of academic and professional practice. But the response to this cannot be to jeopardize the very notion of ethical responsibility. As aptly pointed out by Ananta Giri in his contribution to the Strathern anthology, ‘a Foucauldian critique of audit as a contemporary governmentality is not enough … There is need for a post-Foucauldian response in terms of subjective preparation and the development of moral commitment, and here Kant provides much needed reflections for self-criticism’ (Giri 2000: 188). In seeking to retrieve a sense of what ethics is fundamentally about, Giri strongly emphasizes an aspect of personal responsibility and self-improvement which, in addition to Kant, in large measure is inspired by a Gandhian notion of Swaraj. ‘Accountability’, he notes, ‘has multiple meanings …. It refers not only to being accountable for what one is expected to do or perform but to one’s responsibility beyond legal minimalism, to the growth of oneself and the other and thus to the creation of dignified relationships in society’ (ibid.: 173f.). Interestingly, the corrective proposed by Giri to the excessive formalization of audit culture tends to point in the direction of another extreme, which is to constitute ethics as a personal calling and hence predominantly a private matter. This resonates in various ways with practices commonly encountered among researchers in the field. One seemingly widespread reaction to frustrations with formalized ethics consists in establishing one’s own conscience and a sense of personal integrity as a valid alternative to the ostensibly flawed public accountability procedures. Another characteristic 149

Evidence, Ethos and Experiment

reaction consists in seeking to construe the interaction between professional and client, or researcher and researched, as a privileged ‘covenantal’ relation that is immune to external scrutiny. Both of these reactions have it in common with Giri’s construal of ethics as a matter of personal calling that they essentially privatize ethics, that is, remove the locus of ethical judgement from public view. This, in my view, is fundamentally problematic. There is no doubt that private ethical commitments can serve as an invaluable impetus for ethical conduct in one’s professional capacity, but it cannot serve as a substitute for public justification. Scientists and medical professionals cannot, in other words, tacitly grant themselves a license to pursue a given course of conduct, no matter how strong the conviction about its correct character, if the decision cannot be justified in a suitable public context. Aside from being inherently paternalistic, the problem with appeals to ethical immediacy is that they provide no safeguard against cutting corners and developing attitudes of arrogance or simple bad habits, be this in the form of unintentionally exposing research participants to unwarranted risk, neglecting to maintain proper standards of respectful communication, launching initiatives that may be of questionable relevance and thus possibly wasteful of scarce resources, or reinforcing inequities and disrespectful relations of collaboration between international and local actors. The continuous low-level erosion of ethical standards of conduct, in my view, is a real scandal that needs to be seriously addressed. I therefore propose that the appropriate response to the critique of audit culture lies in retrieving a concept of genuine ethical accountability in the public sphere. This may well involve an aspect of self-improvement, but beyond this it also involves restoring a notion of public responsibility and hence accountability in a sense not co-opted by the prevailing legal/financial idiom. Giri, in fact, expresses a similar aspiration when advocating a sense of ‘responsibility beyond legal minimalism’. I shall in the next section follow Giri’s lead and look to the moral philosophy of Kant for an account of the defining characteristics of ethical conduct.

Retrieving a Concept of Ethical Accountability In all the world and in all of life there is nothing more important to determine than what is right. (Lewis 1955: 27) To achieve a distinct concept of ethical accountability, it is necessary to set the discourse of ethics apart not only from concepts and procedures de150

Elucidating Ethics in Practice

riving from the financial realm but also from parallel legal notions and practices. Legal accountability is a highly powerful paradigm that in many respects requires attention and reinforcement (as, for example, in campaigns against impunity), yet is fundamentally different from ethical accountability and sometimes gets in the way of the latter. Conceptually, the defining feature of ethics is that it is a mode of judgement which, based on a notion of right action, stands apart from calculations of expediency. Ethics is not about achieving set goals, maximizing benefits or even complying with rules, law or local custom; it is about right action. This is the central premise of Kantian ethics. The adequate form of the ethical imperative, according to Kant, is therefore not hypothetical (‘if you want to achieve such and such an outcome, then do this’) but rather categorical (‘do this, because it is right’). In response to the question why one should comply with a rule or law, Kant recognizes two possible answers: due to fear of punishment or out of respect for the law – that is, out of recognition that it embodies a principle of right action. The former reaction reflects the coercive aspect of law and involves a calculation of expediency: does the risk of punishment, for example, outweigh the possible gains from breaking the law? The latter mode of conduct is inherently ethical and is subject to a principle of legitimacy of action (Habermas 1997: 28ff.; Alexy 2005). It is closely linked with a notion of autonomy, which on Kant’s understanding implies recognizing the maxim or principle of what is right and acting in accordance with it – in other words, imposing upon oneself the principle of right action rather than being heteronomously determined in one’s behaviour by external authority or arbitrary whim (Kant 1991 [1785]). The discourse of audited accountability, while seemingly appropriating the language of ethics, essentially revolves around questions of expediency and coercive mechanisms of rule compliance. It is therefore not genuinely ethical in nature. The intuitive recognition of this, as we have seen, often leads practitioners to abandon any commitment to public accountability and instead construe their ethical commitment as something purely private – a matter of conscience and personal integrity. This reaction arguably has some justification in Kantian thinking in so far as this is typical of what can be called a monological approach to ethics – an approach that views the ethical evaluation as an internal matter. Jürgen Habermas, while in many other regards remaining loyal to the key premises of Kantian ethics, presents an alternative, dialogical approach to ethics, which instead of viewing ethical judgement as something that unfolds in the individual mind situ151

Evidence, Ethos and Experiment

ates it in a public context. This has come to be known as the theory of discourse ethics. Ethics, to Habermas, is essentially a matter of broaching normative evaluations in a communicative setting among parties implicated or affected by the relevant decisions. It hence comes to revolve around publicly accounting for the legitimacy of one’s action with reference to a more or less well defined set of shared normative references and with due consideration to how the issue at hand is perceived by and affects other partners in discourse. In my view this establishes a suitable framework within which to understand the notion of ethical accountability. What it implies, first and foremost, is that the ethics of medical research, or any other branch of professional ethics, cannot be construed merely a matter of exercising private conscience. When it comes to difficult decisions, deliberations must as a general rule be presented in a relevant public forum – what I have elsewhere styled a site of accountability (Ulrich 2001: 99) – where they are open to scrutiny and can be subjected to criticism and contradiction. This may lead to disagreement about how a given dilemma should be resolved, but such disagreement should in fact be recognized as an integral component of healthy ethical discourse. It is fundamentally different from charges of unethical behaviour in a sinister sense, where the basic character of the person being criticized is called into question. Failure to make this distinction typically places professionals in a defensive position vis-à-vis ethics and thereby stifles the very type of candid ethical discourse that is so important to any branch of professional conduct, not least in contexts where significant differentials of power and competence are at play. It is obviously difficult to comply with the requirement of situated ethical accountability in cases where the research methodology involves an element of secrecy or duplicity, or where the very fact of conducting research is kept out of public view, but this is precisely an added reason why covert research is fundamentally contentious and should be approached with great caution.5 The primary reason for insisting so strongly on the open-ended communicative aspect of ethical accountability is that the exercise of any professional competence is an inherently public activity utilizing resources which have been conferred upon the professional through education, professional licensing, institutional affiliations, allocation of funds and general recognition of competence and authority, and so on. Science, too, in the form of accumulated knowledge and access to sensitive data is a precious public resource intrinsically linked to a universalist notion of equal entitlements and commitment to the common good. This confers upon 152

Elucidating Ethics in Practice

scientists, and health professionals generally, an important mandate which must be managed in a responsible fashion and upon which the general public has a claim. It is therefore reasonable to insist that the primordial moment of ethical significance lies in the public articulation of ethical deliberations justifying one’s decisions.6 Ethical accountability, then, means to act out of concern for what is right and assume responsibility for explaining the reasoning underlying one’s actions in suitable public contexts. Ethics is not primarily about the enforcement of norms through coercive means. This may come in addition – as an aspect of legal accountability being superimposed on the ethical – but ethical accountability proper is about being able to pose questions and being obliged to provide responses concerning the legitimacy of one’s acts. Ethical accountability thus has a communicative, reciprocal character. This stands in stark contrast to the notion of accountability adopted from the financial realm which, as has been shown by several contributors to the Strathern volume examined above, is premised on an asymmetry of authority that undermines communicative reciprocity and trust. Whereas audited accountability always unfolds in predefined relations of authority and obligation (who has to answer to whom and who is entitled to control what, etc.), the ethical judgement will never recognize an assertion of social position or status as a valid ground for refusing to be held to account. It must be emphasized that the notion of ethical accountability presented here is not a peculiar theory-based invention. Rather, it is an implicitly operative expectation in real-life communicative interaction, even in contexts marked by disagreement about the applicable standards of right action. And yet it is an expectation that is regularly disappointed. What thwarts the realization of ethical accountability in practice is a lack of functional accountability structures and the ongoing scandalous misrecognition of discursive competence of many social actors. Both factors will be further examined below as factors contributing to an ‘unhinging’ ethics in practice. It may be argued on theoretical grounds that the very notion of the public sphere as outlined above is fundamentally problematic outside a few ‘Western’ liberal democracies and hence a questionable premise for a theory of ethics of global reach. Phrased differently, it may be argued that a Habermasian public sphere in which social actors, irrespective of their particular identities, are committed to reciprocal communicative relations and the pursuit of a common good has never firmly been consolidated in fragile emerging democracies (postcolonial or otherwise) and, moreover, that 153

Evidence, Ethos and Experiment

what has been established is currently in the process of being dismantled by the neoliberal thrust to privatize public functions and conflate the distinction between common good and private (commercial) interest. I generally agree with this analysis but do not believe that it compromises or renders unrealistic a commitment to public ethical accountability. To the contrary, the prevailing weakness of public institutions confronts us with a profound need to clarify what type of public context could underpin a theory of situated accountability. It is in this spirit that I have previously flagged the identification of relevant sites of ethical accountability as a key area for empirical elucidation. This is a necessary but all too often neglected complement to the quest to articulate relevant ethical standards, which will be briefly reviewed in the following section.

Question of Normative References A central question in any reflection on professional ethics concerns the normative standards by which actors shall be held to account. Some approaches to ethical theory focus virtually exclusive attention on this question. The adoption of ethics codes for professions and institutions has been a momentous accomplishment, notably in the medical field, but a drawback of this approach to ethics is that it risks assuming a formulaic character that is contrary to real-life ethical complexity. Whether contextualized as pseudo-legal norms or as a standard of reference for administrative control in the manner of audited accountability, professional codes may have the effect of reducing the exigency of ethical compliance to a vacuous ‘checklist approach’ whereby serious ethical concerns are given mere cursory, pro forma attention. There is an inherent risk, in other words, that the formalization of professional ethics removes the dynamic aspect of ethical discourse. This has, for example, been a primary reason for the strong opposition of European anthropologists towards the adoption of a code of professional ethics. Peter Pels, cited above, has been a vocal proponent of this stance (Pels 1999). Against this sceptical inclination, it could be argued, however, that the role of ethics codes when properly construed is not to impose a layer of external control, premised on mistrust, upon the practitioners of a given profession, but rather to inspire and frame ongoing attention to genuine ethical complexities on the part of the various actors involved in and implicated by a given realm of professional activity. This, indeed, is how the American Anthropologist Association has sought to 154

Elucidating Ethics in Practice

frame its code of professional ethics (Fluehr-Lobban 2003). A well-articulated ethics code, accordingly, can be understood as a public statement by members of a given profession about the standards by which they expect to be held accountable by the general public, communities and individuals facilitating and/or affected by their work.7 I would be inclined to support the latter interpretation of the role of ethics codes (cf. Ulrich 1999, 2009). My general concern as articulated in various writings (Ulrich 2000, 2001, 2007b) has nevertheless been to place a stronger focus on the communicative of ethical accountability by seeking to identify and describe the key areas of ethical concern that need to be addressed in any discourse of professional ethics. I have in this connection proposed a taxonomy consisting of five main categories of ethical issues, namely: 1. ethical issues having to do with protection of research participants against harm (direct or indirect; physical, psychological or material; active or passive); 2. ethical issues having to do with reciprocity and respect; not what one does to and for others, but the quality of how one interacts; I have sometimes described this class of ethical issues as communication issues; 3. ethical issues having to do with relevance and justice, advancing the general human and social good; 4. ethical issues having to do with fairness in collaboration (which in effect means negotiating vested interests at the individual and institutional levels); 5. ethical issues having to do with professional/scientific honesty and validity. The advantage of this approach, in my view, is that an enhanced awareness and conceptual clarity about the nature of the ethical issues confronting professionals in their work will help not only in fostering compliance with key principles but also in identifying and analysing possible conflicts between different levels of ethical consideration in complex situations. The approach captures a much broader range of ethical concerns than formalized codes, which typically are limited to a summary recapitulation of essential minimum standards. When it comes to reinforcing public ethical accountability, it may moreover be found that the quest to foster a shared understanding of, and capacity to analyse, the pertinent issues is more ef155

Evidence, Ethos and Experiment

fective than the articulation of an authoritative list of positive norms. This is especially true for research and other professional activity conducted in a cross-cultural context, as experience shows that authoritative norms and prescriptions for correct conduct often have a divisive effect and are suspected of surreptitiously imposing particularist ‘Western’ values, whereas the attempt to attain a shared diagnosis of the relevant issues serves to position the protagonists in discourse in a common moral and communicative space within which informed evaluations can be effectively asserted and the options for resolving genuine ethical dilemmas can be explored, negotiated and pursued. Either way, it is essential to recognize that we are talking about ethical obligations pertaining to the public persona. This is by definition limited in comparison with the ethical commitments of any given actor in private. The ethics that govern interaction in the public sphere, including any branch of professional ethics, do not, in other words, comprise the full range of values and concerns that govern our individual or communal pursuit of a good life; they concern only the issues that we need to resolve in common in order to ensure the possibility for all, to the best of their ability, to pursue their own life aspirations – on the condition that this does not infringe on the ability of others to do the same. Such a delimitation serves in essence to accommodate normative pluralism, the key premise being that divergent ethical outlooks and valuations, often based on religion or culture, can coexist in the public sphere as long as a set of minimum standards and premises are shared by all partners in discourse and matters of common concern addressed in a normatively inclusive manner. Such agreed minimum standards will predominantly be of a procedural nature (ensuring the ability of everyone to take part in decisions affecting them, protecting the basic equality and integrity of all participants in discourse, ensuring fair outcomes of processes involving the utilization of public resources, etc.) but may also in some respects go beyond this to commonly agreed substantive objectives (e.g. alleviating suffering, doing good within one’s area of competence, promoting social or global justice, the pursuit of truth). Medical science and professional practice is clearly an area of conduct that involves actors with divergent moral outlooks who need to come to a common basic understanding about the way in which the profession is practiced. Medical ethics is thus subject to exactly the delimitations described above with all the advantages and complications that this entails. Two premises of public ethics are particularly contentious in many social/cultural contexts and give rise to widespread resistance. One has to do 156

Elucidating Ethics in Practice

with the underlying public–private divide. While this is essential to ensuring an element of inclusiveness and respect for the moral valuations of others, it runs counter to most people’s immediate moral intuitions, especially in settings characterized by a weakly constituted public sphere. A second contentious aspect of the delimitation of public ethics is the concomitant disqualification of authority based on substantive morality, thus also religious authority, with regard to determining standards of professional conduct. Public ethics is, to borrow a phrase from Habermas, inherently post-metaphysical.8 In order to ensure applicability in communal contexts in which religion-based worldviews nevertheless do hold sway, as can be found in all parts of the world, the proponents of a unified set of global ethical standards typically seek to demonstrate that such standards are at least consistent with the predominant values in all major cultures and religions. This idea is obviously appealing but not convincing in every regard, and it may be anticipated that residual fundamental disagreements about values and standards will act as a stumbling block to the attempt to assert established international ethical standards in contexts of cross-cultural or inter-religious interaction. It remains uncertain whether this problem can be alleviated by the proposed shift of focus from positive ethical standards to the underlying issues of common concern. If, however, some normative differences do prove to be irreducible in cross-cultural professional interaction, then it is all the more important to forge reciprocal communicative contexts in which they can come to the fore and be recognized for what they are.

Ethics Unhinged It is one thing to articulate an ideal notion of ethical interaction, quite another to see it realized in practice. So it is with ethical accountability: while advocating the concept, it must be candidly acknowledged that there are powerful factors that inhibit its realization. This is notably true in skewed contexts, such as pertain to development cooperation and externally funded research, that are characterized by extreme differentials in access to knowledge, resources and power. In such settings the discourse of ethics is unhinged on account of at least three important factors: dysfunctional structures of accountability, the misappropriation of the language of ethics as a language of dominance and a disconnect between science and social process. In reviewing each of these factors, my concern in the following will be to elucidate the underlying problem and indicate questions for further empirical investigation. 157

Evidence, Ethos and Experiment

The Problem of Weak Accountability Structures A direct consequence of the public character of ethical accountability is that it unfolds in concrete contexts. The ethics of scientific practice must thus be conceived as a situated ethics embedded within determinate structures, which I have proposed to characterize as sites of accountability. Ethics review boards both at national and institutional levels naturally fulfil this function, but so do a range of other entities such as funding agencies, publishing houses, the public media and self-appointed watchdog agencies. It must be noted, furthermore, that a certain discourse of accountability continually unfolds in relations among peers and in the interaction between scientists and research participants or between scientists and the local communities hosting research. While less structured and formally regulated, such mechanisms too are highly important for the realization of ethical accountability. They presuppose a basic degree of empowerment of the various parties involved in the process of scientific research, as will be further examined in the next section. A central premise of the present approach to professional ethics is that any improvement of ethical standards in practice depends on the functioning of concrete structures of accountability of the nature described above. This, in fact, is tacitly taken for granted in developed societies where scientists and other professionals are (for the most part) under continuous public scrutiny with regard to their professional conduct. However, when it comes to research conducted in developing societies, one invariably finds that accountability mechanisms are weak and not conducive to embedding genuine ethical discourse within the contexts in which professional practice unfolds. This, as we have seen, may on the one hnad lead to the rise of externally contituted audit structures of perceived questionable legitimacy or, on the other hand, to a defensive retreat into an excessive reliance on private judgement and a sense of personal integrity. Paradoxically, the two ethical aberrations complement and reinforce each other. A prime example of weak accountability structures is found in the way in which the press in many developing countries either fails to report effectively on contentious foreign-sponsored research initiatives or do so in an exaggerated fashion that feeds a general sense of suspicion and disempowerment rather than informed monitoring and public discussion. As a case in point, Kenya’s leading daily The Nation on 28 April 1997 carried a back-page feature proclaiming in large script that ‘AIDS studies “kill infants”’. The underlying issue had nothing to do with direct hazards arising 158

Elucidating Ethics in Practice

from the research in question, but rather with the difficult question of what minimal standard of care to adopt for research participants in clinical trials in a context in which the prevailing standard of care in the surrounding society is virtually nil. Clearly this is a highly pertinent ethical issue, but The Nation’s alarmist report did not in fact facilitate the articulation of a constructive/critical Kenyan perspective on the issue. A related illustration of malfunctioning accountability at the local level is found in the way in which misgivings about community-based research are often articulated in the form of ‘rumours’, such as about blood and organ stealing, deliberate spreading of disease, surreptitious fertility reduction, and so on. In an insightful recent article, Geissler and Pool make a convincing case for treating such persistent rumours associated with externally sponsored research in sub-Saharan Africa as critical local commentaries on issues of medical research ethics. ‘Given local unfamiliarity with the conceptual models and terminology of the international medical ethics debate and the lack of direct engagement of study populations in discussions about ethical concerns, local communities make use of their own models and terminologies to express and debate their concerns’ (Geissler and Pool 2006: 980). Rumours, on this analysis, ‘give local populations a way of commenting on medical research as part of a wider system of exploitative appropriation of value’ (ibid.). The drawback, however, of articulating ethical misgivings in such a manner is that it will have only limited impact on the way in which research is conducted and will not contribute to establishing an environment of reciprocal ethical interaction. Like exaggerated reports in the press, the idiom of rumour is effectively a symptom of disempowerment. So while agreeing with the joint authors that ‘in order to improve the ethics and the practice of medical field research in Africa, we have to recognize its study subjects as interlocutors in ongoing global ethics debates, not merely as objects of ethical responsibility’ (ibid.: 981), I would insist that this can only be accomplished by reinforcing structures and procedures of situated accountability. It must be acknowledged that this entire area is poorly defined and stands in need of elucidation, both through further conceptual analysis and through ethnographic and historically reflexive investigation. A primary issue in need of empirical clarification has to do with identifying relevant sites and mechanisms of public accountability at the local, national and international levels. How do accountability procedures function in practice, and what are the main blocks and obstacles encountered? It would in this connection also be helpful to examine the relative merits of informal and 159

Evidence, Ethos and Experiment

formalized ethics procedures. How do they each affect professional practice, relations among practitioners, and relations between practitioners and the public they serve or engage in their work? Furthermore, what is the relation between review and accountability procedures in different contexts or at different levels? Can national ethical review procedures, for example, be substituted by ethical review by an international body, or alternatively by less formal local procedures, in cases where the competence and integrity of the national bodies may be subject to doubts? To what extent are researchers empowered to publicly challenge the authority of formal or informal monitoring bodies? Are local and overseas visiting researchers differently positioned in this regard? What other means of recourse do researchers have, and is it ever ethically justified to disguise or misrepresent the objective and/or methodology of research? These are questions where anthropological research has an important role to play in enhancing our understanding of the experiential dimension of real-life ethical interaction.

Ethics as a Language of Power Closely related to the issue of how accountability is situated in concrete institutional or communal structures is the question of who has competence to pronounce ethical judgement. Ethical discourse ought, according to its nature, to be reciprocal and non-coercive. Being essentially non-instrumental, it does not permit uses of language that are either commanding or manipulative and does not recognize status differences when it comes to pronouncing ethical judgement. To the contrary, the discourse of public ethics positions social actors on the same level and accords all an equal say in pursuit of judgement about what is right. This, obviously, may also entail exposing and contesting existing injustices and structures of domination. Ethics, in other words, is premised on a notion of what Habermas has described as ‘domination-free communication’, that is, communication that is oriented towards reaching a common understanding (see, for example, Habermas 1990: 133ff.). In reality, however, the language of ethics often functions as a language of power and as a way of asserting authority and disqualifying the voice or experience of others. This marks another key example of context distortion. Rather than ‘level the playing field’, the introduction of formalized standards of ethics in some cases reinforces existing power imbalances in donor-sponsored research and other areas of professional collaboration. The ability to command the language of ethics reflects on professional competence, and one could argue that without this it is not possible to attain 160

Elucidating Ethics in Practice

full professional status. This connection probably holds true for all professions, but applies in particular to the area of medicine due both to the sensitive nature of the work performed and to the way in which the pledge of an ethical oath has been established as an age-old prerequisite for entering the profession. When it comes to medical science, the ability to pronounce on ethical issues is an indispensable precondition for taking part in decisions about research methodology and design, for establishing priorities, and for managing and monitoring research activity as it unfolds. In effect, professional ethics serves as a standard by which to both legitimize and de-legitimize scientific practices. It is thus also a powerful factor in intercollegial relations, as is concretely felt in conjunction with collaborative research where ethical references are often advanced to assert authority, and where local actors are sometimes denied recognition of the competence to pronounce ethical judgement on account of their lack of command of the prevailing idiom. Ethics debates tend for this reason to be highly sensitive. Such a dynamic clearly came to the fore in the well-known disputes of the 1990s concerning research in Africa and Asia into the effect of anti-retrovirals on perinatal transmission of HIV. When the American watchdog organization Public Citizen, supported by an editor of the New England Journal of Medicine, Marcia Angell, sharply condemned trials conducted in Africa and Asia as unethical, the chair of the Ugandan AIDS Research Committee, Edward Mbidde, responded in equally strong terms, insisting on a connection between local autonomy of research and local competence for ethical evaluation: ‘There is a mix up of issues here which needs to be clarified. It is not NIH [the National Institute of Health] conducting the studies in Uganda but Ugandans conducting the study on their people for the good of their people. If this is not acceptable ... then this is tantamount to ethical imperialism’.9 What Mbidde is rightly protesting against is the concentration of ethical competence within global centres where research is being initiated, organized and monitored (Geneva, Atlanta, Washington DC, London, etc.). Like the inclination to privatize ethics, this too effectively implies a removal of accountability from the contexts in which the research is conducted. External monitoring and ethical review is obviously warranted in many cases, yet the inclination to simply bypass or disregard local review structures and local expertise undermines the very foundation of situated professional ethics and should be resisted. There rests, accordingly, a heavy onus on visiting researchers in developing countries – and in particular on large institutions sponsoring such research – to recognize and respect local competences to pass judgement 161

Evidence, Ethos and Experiment

on ethical issues and, going beyond this, to engage constructively with local review mechanisms, even when such mechanisms are structurally weak. Rather than lament the apparent contradictions in ethics discourse, I would suggest that the connections between ethics and power should be openly recognized and mobilized in conjunction with ethics education and awareness raising. In view of the personal and institutional dynamics related to actual ethical practice, ethics education should aim progressively to foster: 1. familiarity with established ethical standards and/or pertinent levels of ethical consideration; 2. the ability to adapt and apply standards in complex situations; 3. the ability to negotiate terms and priorities of research initiatives, and in so doing ensure a higher degree of reciprocity in research collaboration; 4. the ability to communicate and teach ethics; and 5. the self-confidence required to participate in international standard setting. The implicit thrust of such a curriculum is, needless to say, not only to disseminate knowledge about established ethical standards but also to empower individual scientists and practitioners through a reinforcement of ethical ‘voice’. Empirical science studies can contribute to such a programme of ethical capacity building by better illuminating the concrete inter-subjective dynamics associated with ethical evaluations and by presenting examples of genuinely reciprocal and constructive collaboration in relation to ethical challenges and uncertainties.

Disconnect between Science and Social Process The very concept of medical research ethics presupposes a certain degree of integration of science and social process. Going beyond the question of structures designed to reinforce accountability, narrowly speaking, this is the general context in which ethical standards can meaningfully be realized. In practice, however, one often finds that the required degree of integration is lacking. Research initiatives in low-income societies are all too often externally imposed and not adequately tailored to local needs, local conditions and local perceptions of relevance. This marks a further example of context distortion with far-reaching ramifications for the realization of ethical accountability. 162

Elucidating Ethics in Practice

The social context in which research is conducted may directly affect the application of ethical standards, such as, for example, when the element of choice that is necessary for research participants to give voluntary informed consent is simply absent, namely when the only choice is between experimental treatment or no treatment at all; when impoverishment is so widespread that any reward or reimbursement for research participation assumes a coercive character; or when the essential premises of a given research protocol for technical or other reasons cannot be meaningfully explained in the context in which the research will be carried out. These are familiar difficulties that are recognized and to some degree addressed by conscientious practitioners. However, the underlying issue that has to do with a lack of integration of science and social practice tends to be given less attention. In order to bring this out, it is necessary to adopt a broad interpretation of the nature of science. Viewed in a comprehensive perspective as a social activity that is integral to modern society, one finds that science is never only about advancing our knowledge of a specific subject matter, never merely about procuring a set of data that may be useful for a particular purpose. It is always also about enhancing the conditions for further scientific activity within the same general area, be this through intellectual capacity building, through building a resource base in the form of libraries and institutional infrastructure, or through publications that identify questions in need of further examination. The very fact of accommodating and facilitating scientific activity, moreover, shapes the public’s view of the world. Modern society is structurally dependent on such a continuous production of knowledge, not only in a narrow instrumental sense but also for the basic manner in which we interpret our natural and social environment and possibilities of exerting agency. It is, by the same token, a central feature of social modernization in Western society that the production of knowledge and public administration practices are intimately related through a process of what Anthony Giddens (1991) has called ‘looping effects’.10 Scientific activity by its very nature thus always reaches beyond itself, and it is in this broad perspective that the question of the relevance of research has to stand its test. Historically, foreign-sponsored research in Africa and Asia has not performed well on this count. Third-world modernization processes during colonialism and well into the postcolonial era have not involved an effective integration of scientific input, and the preponderance of externally generated research initiatives would appear to have perpetuated a structural disconnect between science and social 163

Evidence, Ethos and Experiment

process. To be sure, this is not the full picture. Several contributions to the present volume document how national and regional centres of medical research during the late colonial and early postcolonial era have fostered cultures of public service which clearly were perceived as part of a greater project of nation building.11 Such historical documentation is highly valuable for the understanding of how research is locally contextualized and how it has been perceived as integral to a notion of public good. But the predominant picture is nevertheless that the ability to frame medical research as a genuine public good has been frail in comparison with an overall pattern of subordinating research to selective or particular interests, often alien to the interests of the local communities in which research is in fact conducted. The latter tendency would seem to have become all the more predominant in the current neoliberal era, where powerful international actors once again determine the flow of knowledge and goods, rarely in the best interest of marginalized low-income populations and without leaving much space for local initiative and self reliance. This is deeply problematic from an ethical point of view. If structurally, as individual or as community, one is consistently being excluded from benefiting from public goods, then it becomes fundamentally ambiguous why one should contribute to science as an ostensible public good. Without a basic sense of shared purpose overriding particular interests, the very context of ethical valuation and accountability becomes fundamentally distorted. It is for this reason, ultimately, that externally sponsored research in low-income societies should without exception involve a firm commitment to embedding research in structures of genuine collaboration so as to reinforce local initiative and local relevance. One might further draw the implication that externally sponsored medical research and health interventions in developing countries ought, as far as possible, to be channelled through existing state structures and local academic institutions, even if such institutional structures may be found in the short terms to be less efficient than freshly created, externally controlled parallel structures. The predominant tendency, however, is exactly the opposite, as is evident by the propensity of major international actors such as the philanthropic Global Fund or the US-supported Centres for Disease Control (CDC) to create parallel structures in support of research and large-scale health interventions. Generally, it is my impression that even when a certain obligation to facilitate local capacity building is acknowledged by large private or public entities – be they multinational pharmaceutical companies, international research conglomerates, or bilateral or multilateral donors – 164

Elucidating Ethics in Practice

their actual success in facilitating effective local contextualization of research remains contentious, at best.12 Despite the emerging focus on capacity building and local relevance, there is thus still a predominant tendency for externally sponsored research in developing countries to remain disconnected from ongoing social and political processes and to not genuinely reinforce local autonomous research and policymaking capacities. Instead, development processes continue to unfold with a strong economic, scientific and policy-related impetus from outside parties, without solid internal institutional checks and balances and without being embedded in a broader civil society context of the type that in European modernization has been instrumental in facilitating a continuous monitoring and re-appropriation of developments at the systemic level. This general situation has a profound impact on virtually every level of consideration related to professional ethics. It affects, for example, the standard of what counts as good science, the manner in which one weighs risks and potential benefits, the dynamics of trust and mistrust between researchers and lay research participants, lacunae of communication and relations between collaborators in research. But beyond this it also fundamentally affects the underlying ethical context in terms of how obligations and sites of ethical accountability are constituted and who has voice and competence to pronounce ethical judgement. Concrete science studies are needed to illuminate the ways in which research activity is embedded in or disconnected from a broader social context and how this affects ethical relations. This will add an important dimension to the evaluation of ethical issues at all levels and will further help in establishing priorities for research and institutional capacity building.

Conclusion It has been a central premise of the present argument that anthropological research has a major role to play in elucidating how ethical discourse unfolds in practice. Actual findings are in many regards discouraging. On all three counts examined above, it has been found that there are powerful forces working against the forging of genuine ethical accountability – in particular in the very contexts where this is most needed, namely where research (and development cooperation more generally) is predominantly planned and initiated by external actors and where relations of cooperation are characterized by debilitating asymmetries in status, authority and access to vital resources. Against prevailing tendencies to either retreat to 165

Evidence, Ethos and Experiment

a personalized ethical calling or leave the field to authoritative monitoring bodies, I have argued that it is necessary to develop a situated approach to ethical accountability. I have argued, specifically, that this entails a combination of conceptual and practical measures at different levels, including: 1. to define the space of encounter between protagonists in research as a public sphere in which all actors are subject to a strong requirement to publicly articulate ethical deliberations; 2. to identify the main parameters of ethical consideration of relevance in this sphere; 3. to confront donors and other powerful protagonists in the organization of research with an obligation to recognize and respect the ethical competence of local counterparts; 4. to develop models for ethics education and awareness raising that are designed to strengthen such competence; 5. to identify potentially relevant sites of accountability and investigate how such structures can be rendered more functional; 6. to reinforce, when possible and relevant, the national structures in developing countries capable of supporting public ethical accountability; and 7. to consistently focus on the local relevance of research. Against the paternalistic thrust of either privatizing ethics or usurping ethical competence within global centres of excellence, this means, in effect, to strengthen the qualified opposition with which researchers and other professionals are met on the ground when working in low-income and relatively disadvantaged societies or communities. To do so one needs squarely to confront the concrete factors that derail ethics in practice, and by the same token denounce any misuse of office as well as exaggerated or misrepresented allegations of ethical misconduct which, too, undercut reciprocal ethical discourse. To be sure, it is not easy to maintain or rebuild an aspect of trust in public institutions that in a context neoliberalist globalization appear to be perpetually in jeopardy. But it is also possible to overstate this point. It may be argued that the agenda presented here is itself an expression of liberal values that are inherently Western in origin and alien to many of the contexts in which they are being applied. It is not. Rather, it is about the respect being owed to individuals and communities being engaged in or otherwise affected by externally funded medical practice and research to be 166

Elucidating Ethics in Practice

treated with dignity as authors of their own essential commitments, even under highly adverse social and economic circumstances, and fully capable of forming valid ethical judgements when structurally empowered to do so. This is not an alien expectation in any professional setting, cross-cultural or otherwise. The plea for reinforcing a culture of ethical accountability, ultimately, is about giving definition and voice to existing moral intuitions. To be sure, their force is weak in comparison with concomitant impetuses that overshadow or even obliterate the ethical, but hence all the more reason to reclaim a viable and exacting concept of ethical accountability.

Notes 1. As inspiration for this paper, the author is grateful to the organizers and participants in the conference Locating the Field: The Ethnography of Medical Research In Africa, Kilifi, Kenya, 5–9 December 2005, and in particular to Wenzel Geissler for ongoing stimulating discussions and intellectual challenges. 2. It is thus telling that the notion of accountability only receives one cursory reference in the influential and voluminous Principles of Biomedical Ethics; Beauchamp and Childress (2001: 149). Emanuel et al. (2000), similarly, only touch upon the notion of accountability twice in ‘What makes Clinical Research Ethical?’, JAMA 283: 2701–11, and on both occasions this happens with reference to the rather narrow context of independent ethics review. 3. Compare Habermas (1990: 45ff.) on ‘the phenomenology of the moral’. 4. See, for example, Patricia Cohen, ‘As Ethics Panels Expand Grip, No Field Is Off Limits’, New York Times, 28 February 2007: ‘[t]o many faculty and graduate students, review boards are like a blister that gets worse with every step. … the system’s “mission creep” is having a pernicious and widespread effect on humanities and social science research’. 5. I do not wish to state categorically that research conducted without full openness or without approval from the relevant public authorities is in every case unacceptable – it may in some cases be the only way to subject powerful actors or perpetrators of illegal or unethical acts to scientific study – but the lack of public scrutiny in such cases significantly raises the threshold for ethical justification and imposes an added burden on the researcher and hosting institution to define and respect a reasonable procedure of ethical accountability. 6. It must be emphasized that this is not intended to establish a different standard of ethical accountability for health professionals acting within a commercially privatized context. Quite the contrary: privately owned corporations also utilize public resources when engaging in any form of research and for this reason too need to be

167

Evidence, Ethos and Experiment

held publicly accountable for their actions precisely in the manner outlined above. 7. Compare Art. 9 of ‘The Code of Conduct for the International Red Cross and Red Crescent Movement and NGOs in Disaster Relief ’: ‘We hold ourselves accountable to both those we seek to assist and those from whom we accept resources’ (The Sphere Project, Oxfam Publishing, 2000). 8. Public authority cannot, in other words, be based on a comprehensive moral cosmology, that is a system of objective right, which legitimizes a hierarchical social order, prescribes behavioural norms for different social groups and extensive obligations of the individual towards the community and state, and establishes meaning and purpose in life by embodying a vision of the good life which individual society members may seek to realize in accordance with their own capabilities and social position. Such visions of the good life may of course be otherworldly in their orientation, as was the case with the medieval Christian appropriation of Aristotelian eudemonistic ethics. In what Jürgen Habermas has styled post-metaphysical society, the presence of a shared moral cosmology can no longer be taken for granted, and what does remain of substantive morality is relegated to the private sphere, where it becomes a matter of personal commitment made in accordance with individual beliefs and preferences as well as communal affiliations. 9. Letter to the NIH dated 8 May 1997; quoted in part by Varmus and Satcher 1997: 1003–5. 10. Compare Ian Hacking’s notion of the looping effect of human kinds: ‘that is, the interaction between people, on the one hand, and ways of classifying people and their behaviour on the other. Being seen to be a certain kind of person, or to do a certain kind of act, may affect someone. A new or modified mode of classification may systematically affect the people who are so classified, or the people themselves may rebel against the knowers, the classifiers, the science that classifies them. Such interactions may lead to changes in the people who are classified, and hence in what is known about them’ (Hacking 1995: 239). A related observation, albeit with a rather different valuation, lies behind Michel Foucault’s concept of discourse power. 11. See in particular the chapters by P. Wenzel Geissler and by Guillaume Lachenal in the present volume. 12. The chapter by Kenneth Ombongi in the present volume documents how the colonial and post-independence Kenyan state, in different ways, attributed great significance to medical research and public health infrastructure and services. However, ‘[f]rom the 1980s onwards, the role of the state in biomedical development dwindled. To fill up the emerging “void” non-state and international organizations strengthened their role in Kenya’s disease and research policy’ (Ombongi, present volume). The author implicitly makes an argument for once again reinforcing the role of the state so as to foster and promote ‘tropical medicine for the South by the South’ (ibid.). I consider this to be consistent with the argument developed in the present pages; however, with the proviso that locally initiated re168

Elucidating Ethics in Practice

search and public health campaigns must anyhow accommodate international cooperation, and that dysfunctional public structures, if such should persist, must remain open to candid criticism.

References Alexy, R. 2005. ‘The Nature of Legal Philosophy’, in S. Coyle and G. Pavlakos (eds), Jurisprudence or Legal Science? A Debate about the Nature of Legal Theory. Oxford: Hart Publishing. Beauchamp, T.L. and J.F. Childress. 2001. Principles of Biomedical Ethics. Oxford: Oxford University Press. Emanuel, E.J., D. Wendler and C. Grady. 2000. ‘What Makes Clinical Research Ethical?’, JAMA 283: 2701–11. Fluehr-Lobban, C. (ed.). 2003. Ethics and the Profession of Anthropology: Dialogue for Ethically Conscious Practice, 2nd ed. Altamira Press. Geissler, P.W. and R. Pool. 2006. ‘Popular Concerns about Medical Research Projects in sub-Saharan Africa – A Critical Voice in Debates about Medical Research Ethics’, Tropical Medicine and International Health 11(7): 975–82. Giddens, A. 1991. The Consequences of Modernity. Stanford, CA: Stanford University Press. Giri, A. 2000. ‘Audited Accountability and the Imperative of Responsibility: Beyond the Primary of the Political’, in M. Strathern (ed.), Audit Cultures: Anthropological Studies in Accountability, Ethics and the Academy. London: Routledge, pp. 173–95. Habermas, J. 1990. Moral Consciousness and Communicative Action. Cambridge: Polity Press. ———. 1994. Justification and Application: Remarks on Discourse Ethics. Cambridge, MA: MIT Press. ———. 1997. Between Facts and Norms. Cambridge: Polity Press. ———. 1998. ‘Remarks on Legitimation through Human Rights’, Philosophy and Social Criticism 24(2/3): 157–71. Hacking, I. 1995. Rewriting the Soul. Princeton: Princeton University Press. ICRC. 2000. ‘The Code of Conduct for the International Red Cross and Red Crescent Movement and NGOs in Disaster Relief ’, The Sphere Project. Oxford: Oxfam Publishing. Kant, I. 1991 [1785]. Groundwork of the Metaphysics of Morals, trans. H.J. Paton. London: Routledge. Lewis, C.I. 1995. The Ground and Nature of Right. Columbia University Press. Pels, P. 1999. ‘Professions of Duplexity: A Prehistory of Ethical Codes in Anthropology’, Current Anthropology 40(2): 126–27. ———. 2000. ‘The Trickster’s Dilemma: Ethics and the Technologies of the Anthropological Self ’, in M. Strathern (ed.), Audit Cultures: Anthropological Studies in Accountability, Ethics and the Academy. London: Routledge, pp. 135–72. Pojman, L.P. 2001. Ethics: Discovering Right and Wrong, 4th ed. Wadsworth Publishing. Power, M. 1994. The Audit Explosion. London: Demos.

169

Evidence, Ethos and Experiment

Shore, C. and S. Wright. 2000. ‘Coercive Accountability: The Rise of Audit Culture in Higher Education’, in M. Strathern (ed.), Audit Cultures: Anthropological Studies in Accountability, Ethics and the Academy, London: Routledge, pp. 57–89. Strathern, M. (ed.). 2000. Audit Cultures: Anthropological Studies in Accountability, Ethics and the Academy. London: Routledge. Ulrich, G. 1999. ‘Comment to Peter Pels “Professions of Duplexity”’, Current Anthropology 40(2): 126–27. ———. 2000. ‘What We Are Talking about When We Are Talking about Ethics: Towards an Integrated Approach to the Ethics and Politics of Development and Development Research’, The Politics and Ethics of North-South Relations – Contributions to the Development Research Agenda of the New Millennium. Copenhagen: FAU. ———. 2001. ‘Optimum Ethical Standards’, in ‘Proceedings Seminar on Health Research Ethics in Africa’, Acta Tropica 78(Suppl. 1). ———. 2003. ‘Charges and Counter-charges of Ethical Imperialism: Towards a Situated Approach to Development Ethics’, in P.Q. van Ufford and A.K. Giri (eds), A Moral Critique of Development: In Search of Global Responsibilities. London: Routledge. ———. 2007a. ‘Towards an Ethical Base for Human Rights Field Operations’, in M. O’Flaherty (ed.), The Human Rights Field Operation: Law, Theory and Practice. Aldershot: Ashgate, pp. 69–85. ———. 2007b. ‘Towards a Theory of Global Ethics in Support of Human Rights’, in W. Benedek, K. de Feyter and F. Marrella (eds), Economic Globalisation and Human Rights. Cambridge: Cambridge University Press, pp. 39-66. ———. 2009. ‘The Statement of Ethical Commitments of Human Rights Professionals: A Commentary’, in M. O’Flaherty and G. Ulrich (eds). The Professional Identity of the Human Rights Field Officer. Aldershot: Ashgate, pp. 49–82. Varmus, H. and D. Satcher. 1997. ‘Ethical Complexities of Conducting Research in Developing Countries’, New England Journal of Medicine 337: 1003–5.

170

Chapter 7

When Physicians Meet: Local Medical Knowledge and Global Public Goods Steven Feierman

In the US, as in Britain, as in the whole of the industrial world, medical researchers work to develop scientific evidence that can be applied systematically to clinical practice.1 When must a patient be given a specific antibiotic, and for how long? What is the relative efficacy of different antibiotics for each particular disease-causing organism? Which of many potential screening tests is preferred? How long should the course of treatment continue, and at what dosage? How can the course of treatment be structured so as to decrease the probability that resistant bacteria will emerge? A body of careful research-based knowledge addresses each question, and informs each therapeutic practice. While the system of evidence-based medicine has great strengths, it is important not to romanticise it. Critiques point out that prescriptive conclusions may serve bureaucratic purposes as well as therapeutic ones (resolving disputes between medical specialties over spheres of competence, or serving the interests of private health insurers). Evidence-based prescriptions may also draw attention away from ambiguities and zones of ignorance – creating the illusion, sometimes, that one can eliminate the fuzziness of difficult clinical judgements. Even the best and most authoritative medical practice is not always rigorously evidence-based: some legitimate clinical practices are not (and probably never will be) supported by evidence from randomised controlled trials.2 Nevertheless, rich sets of research studies do help well-informed physicians to make good clinical choices and to predict probable outcomes. All the details of evidence-based medicine assume, as a given, the underlying existence of a set of necessary resources – a body of medical technologies – many of which are unavailable in African settings subject to health crises. 171

Evidence, Ethos and Experiment

What happens to the idea of evidence-based practice when medical resources are not available? Does each medical toolkit, each locally available group of technologies, have its own separate evidence-based medicine, tailored (sometimes tragically) to the particular set of tests, drugs, staffing patterns, hospital supplies and transport realities that exist on the ground? Or does northern country practice remain as an unattainable ideal? Elite African physicians, who read standard medical journals (BMJ, New England Journal of Medicine, and others), are aware of industrial country protocols but rarely have the tools to follow them. If it is not possible to adhere to standards that claim to be international, then physicians often craft something else for themselves – a set of improvised, untested, evidencepoor compromises. A very different approach calls for resource-poor countries to adopt the low-tech, evidence-based practices promoted by international donors. Under conditions of poverty, donors want, quite reasonably, to find the practices that achieve the greatest health benefit for the least money (Jamison 2006b).3 But this in turn presents problems. It means that well educated, developing-country physicians must often adopt clinical practices they know to be inadequate according to general international standards. African physicians are being asked to adopt not best practice but what, in the most favourable light, might be called good-enough practice (or at least good enough for the poor). In addition, this set of internationally approved practices, formulated in Geneva, Washington, London, Seattle or Paris, is meant to be applicable in every developing country, and therefore cannot easily account for local differences. Even once the best globally targeted research is done, there is a need for evidence-based medicine that is tailored to disease conditions and available technologies in each place. In industrial countries, information flows continuously between everyday clinical practice and biomedical research, which then helps to address local problems: many physicians contribute at the soft and permeable boundaries of the research establishment. In resource-poor settings, the possibility for clinical research by local physicians is much more limited. If skilled physicians employed by national governments were to rethink evidence-based medicine, they would be doing the equivalent of patching an airplane in mid-flight and redesigning it at the same time. Their situation brings to mind the words of Walter Benjamin, who wrote, ‘the state of emergency in which we live is not the exception but the rule’ (Taussig 1987: 466). African physicians, under conditions of normal emergency, rarely have the time and material resources to construct local or regional versions of evi172

When Physicians Meet

dence-based practice, except for those rare occasions when, in their work for donor agencies, international and local agendas coincide. Lower-level health workers, who have intimate knowledge of medicine on the front lines, are even less likely to have their feedback heard by international experts. These issues presented themselves to me with special intensity during a research-planning trip to Zimbabwe and Ghana in 1998 with seven other representatives of the University of Pennsylvania, including three faculty members in paediatrics and one in neurology. In both countries we met with physicians and other researchers with the intention of forming collaborating research groups. All of us together discussed research and clinical interventions that might ultimately lead to reductions in the prevalence of childhood neurological disabilities. Early in those conversations it became clear that the different sets of physicians (distinguished not by country of origin but by location of practice) had radically different research priorities – even different definitions of the content and distinguishing characteristics of particular diseases. The divergences were so interesting – seemed to me potentially so important – that I began to take careful notes of our conversations. My colleagues in Philadelphia later reviewed and commented on my notes. One of the most striking things was the frustration expressed by the physicians based in Zimbabwe and Ghana about the extent to which they had to fly blind – to address common problems in everyday medical practice without basic epidemiological and clinical knowledge. They felt (although this is not how they put it) that they were practicing medicine under difficult conditions with little knowledge of the particular variety of best practice their local situation would call for. They were workers in a knowledge-based field; they had the most detailed and realistic understanding of local medical problems; but they did not have the resources to create a local form of evidence-based practice or to modify a global one in addressing those problems. When I discussed these issues more recently with two academic physicians in Philadelphia, both of them argued that American physicians also feel insecure in their clinical judgements. One sent me an email saying: I heard something very familiar in the complaints and anxieties of the African physicians you quote. Many colleagues, especially those practicing in community settings, work under conditions far removed from the world of clinical trials and continuously feel like they are ‘winging it’ (i.e. improvised evidence-poor compromises) even when 173

Evidence, Ethos and Experiment

there is some data to help them make decisions .... There is something similar (and yet very different) in their fears about a hollowing out, a moral crisis, even when the situation is objectively (resources, knowledge) so different. It is true that some physicians in the US learn from direct experience that poverty and violence disturb the process of treatment leading, for example, to the emergence of multidrug-resistant tuberculosis in the 1980s and 1990s (Farmer 1997: 347). Other sources of American medical insecurity are quite different from African ones. American physicians know that rigorous medical evidence applies only imperfectly to their own patients, that randomised controlled trials rarely account for the fact that patients often suffer from several diseases at once (and not only the one disease that is addressed in a trial), and that treatment effects are very different in community hospitals than they are in teaching hospitals where trials are carried out. Physicians in tropical Africa, by contrast, must often do entirely without the benefit of clinical trials. Yet another set of issues in the US concerns individual variation in the presentation of symptoms and in the effects of treatments. Some of these are grounded in subtle shadings of personality; others can be measured biologically. Researchers are in the process of finding genetic markers that make it possible to identify in advance the small number of people who will not respond to a particular drug (Rosser 1999: 662–64; Kravitz, et al. 2004: 661–87). This is very different from the problem in many African teaching hospitals where, as we will see, the lack of simple testing materials leaves doctors completely in the dark about which antibiotic to use. Our group’s discussions in Zimbabwe and Ghana showed how physicians from two continents, who shared with one another a respect for medical evidence and for patients’ needs, thought about the practical possibilities in radically different ways. The differences between them were especially clear because the issues they were talking about – on the causes of childhood neurological disability – had not yet become the focus of an internationally accepted policy language. In the absence of this sort of globalising and hybridising language, Zimbabwean and Ghanaian physicians talked about the everyday problems they were dealing with, and the knowledge that they would need to deal with them more effectively. This knowledge was often locally specific, as we will see – specific to countries and to regions within countries. And the needs for knowledge, in situations where physicians knew they did not have basic technologies, were 174

When Physicians Meet

often quite modest. A typical request asked for information about the percentage of children in a particular patient pool who responded to one drug as opposed to another; the physicians knew it was unrealistic to ask for equipment with which to make a precise determination. Using local epidemiological information in place of targeted tests was the equivalent of making slight improvements in aim when throwing darts at a board. The modesty of their goals emerged from their own intimate knowledge of medical conditions in their own country. The American team, meanwhile, talked about how start-of-the-art techniques could be brought to bear on local problems, about epidemiological research that might affect policy and about possibilities for joint research that would address Zimbabwean or Ghanaian concerns at a standard that might be funded in the US. Local medical knowledge and local leadership deserve special attention at this moment, when so much of the research informing African practice is funded from outside the continent. Most of the research institutions that make key decisions (whether funded by northern-country governments, international organisations or NGOs) are global in scope. In most cases, they do not begin from a small-scale medical predicament on the ground, and then move up from there. They are looking to do good on a grand scale. They seek out projects that will contribute a public good – preferably a global public good – something that can, potentially, benefit all the world’s people and, in the process, solve problems in a number of places (Das 1999; Kaul et al. 1999a). It is as though they are flying above the earth in a spacecraft, surveying the whole globe, and looking for knowledge, or actions, that will have a positive effect on the greatest number of people, wherever they may be. And of course the universalising tendencies of medical best-practice are well suited to the process by which institutions create large-scale interventions. It would be wrong to say that local factors are ignored in this process. Internationally based research agencies devote great effort to studying disease conditions and treatments that are important in poor countries, from drug regimens for multidrug-resistant tuberculosis to the life cycle of parasitic worms.4 In addition, all practitioners of international health acknowledge the importance of studying epidemiological patterns. They know that distributions of disease vary from one social setting to another, and these settings must be understood if interventions are to be successful. Local factors are taken into account also in studies of the behavioural peculiarities of vulnerable populations. These are seen as relevant in understanding the risks people take and the ways they respond to educa175

Evidence, Ethos and Experiment

tional programmes. These ways of studying the local rarely, however, focus on locally or regionally specific forms of medical practice and reasoning (including those of the international agencies themselves) as important subjects of inquiry. The need for locally specific knowledge emerged with particular clarity in our group’s discussions of the causes of cerebral palsy, and more generally, in our discussions of brain damage among newborn infants caused by kernicterus. This latter condition is the consequence of neonatal jaundice in its dangerous form. Kernicterus is the staining, and therefore poisoning, of some brain cells by bilirubin, which is a by-product of the destruction of red blood cells. The causes of kernicterus include a number of conditions in which blood cells are destroyed. It can happen if the newborn has been deprived of oxygen in the birthing process, or as a result of infection, or of an immune response. For a long time, physicians thought that moderately high levels of bilirubin among newborns were likely to cause kernicterus, and so they treated elevated levels (called hyperbilirubinemia) with either phototherapy (exposure to light), which is safe, or in more serious cases with exchange transfusion, which is dangerous. In this latter procedure, the infant’s blood is withdrawn in small increments and replaced, in the same increments, with donor blood. In the US extensive research on the dangers of hyperbilirubinemia came only after the introduction of Neonatal Intensive Care Units (NICUs) in the late 1950s and early 1960s. These succeeded in saving the lives of low birth weight infants who, in the days of earlier technologies, would have died. Among other things, the NICUs introduced the use of ventilators to help the newborns breathe, along with ways of monitoring blood oxygen. Babies who would have died in previous years were now surviving, and among these kernicterus emerged as a significant threat (Brett and Niermeyer 1998: 111–29). We can see, then, that intense concern with kernicterus in industrial countries emerged from the introduction of important new technologies. Inevitably, the general availability of new technologies leads physicians to redefine research agendas. For this reason, the varying availability of the whole array of medical technologies in resource-poor countries has implications for deciding what research is useful. If babies do not survive at very low birth weights, as they would in NICUs, then kernicterus might not have the same causes as in the US or Western Europe. The causes of kernicterus vary from one time or place to another with changes in disease conditions and technologies. It is important, however, when thinking about technologies, to think not only about machines 176

When Physicians Meet

(ventilators, or umbilical catheters for monitoring blood oxygen), but to broaden the focus to the whole of sociotechnical practice (Hughes 1991) – all the factors that shape care, including staffing patterns, the frequency with which operating rooms are available, the measures taken to deal with the HIV+ status of many of the jaundiced babies, and others. The care of newborns at Harare Central Hospital, and the incidence of kernicterus and other causes of neurological damage, was shaped first of all by the fact that the hospital was extremely understaffed by developed country standards, relative to the number of births. The hospital was the site of about 21,000 births in 1998, with many of the newborns HIV positive, and therefore in need of special care. It managed with a paediatrics staff of 12 faculty and 16 residents. The University of Pennsylvania team, by contrast, came from two adjacent hospitals with about 2,000 birth per year, and more than 200 faculty and 90 residents in paediatrics. A rough calculation, not accounting for a number of factors,5 shows that while there were about 7 births at the Hospital of the University of Pennsylvania for each paediatrician at the adjacent Children’s Hospital, there were 750 births per paediatrician at Harare Central Hospital, and since Harare Central was a tertiary care hospital, in the middle of an AIDS epidemic, many of the births were high risk and many of the infants very sick.6 These numbers have profound implications for the use of technology and for the division of labour in both places. Nurses at Harare would have had to do tasks that, at Children Hospital of Philadelphia (CHOP), would have been reserved for physicians. And, other factors altogether to one side, there would have been a labour-time limitation at Harare on the use of complex and time-consuming technologies. It would be wrong to make a blanket generalisation about the presence or absence of contemporary medical technologies in Harare Central Hospital, as it was then. (Conditions now are very much worse.) The hospital had some of the technologies of NICUs, but not others, and so it was highly improbable that babies at the very low birth weights would do as well as in fully equipped units. It had ventilators for low birth weight infants, but not foetal monitors for the birthing process. Nor did it have the capacity to administer surfactant, needed (along with ventilators) for infants with respiratory distress syndrome (Kambarami et al. 1999). It had shunts to drain off excess cerebrospinal fluid, and thus prevent the neurological damage of hydrocephaly, but there was not enough operating room time to insert the shunts. It would be wrong, then, to say that they needed research on medicine that was consistently low-tech, but just as wrong to assume 177

Evidence, Ethos and Experiment

the availability of coherent packages of more advanced technologies. Health workers needed the knowledge for dealing with the particular anomalous hybrid that existed at that time and place. Their situation was transitory and locally specific, not generalisable from one sub-region of the country to the other, nor even from one moment in time to another. The technical possibilities would change whenever a new machine was donated or (if spare parts were unavailable) whenever a machine broke. Ghanaian health workers dealt with a no less heterogeneous and unpredictable set of technologies. What both sets of physicians were calling for, in each case, was the systematic knowledge to create their own evidence-based medicine for the unsystematised and unpredictable technologies available to them. One of the key questions was about the birth weights at which the staff needed to intervene vigorously to guarantee survival. The fact that the hospital had been given a number of ventilators made it possible to sustain at least some of the low birth weight babies, but then some members of the staff were concerned about the implications of saving the lives of marginally viable infants. Would these infants suffer from cerebral palsy? Would their needs for care have a negative impact on other children in their families, or on the earning capacity of their parents? The medical staff adopted a rough guideline for resuscitating newborns: the provisional dividing line was a birth weight of 1,000 grams. They withheld resuscitation if babies fell below that line, administered it if they rose above. This rule was, however, not grounded solidly in any body of evidence. The research knowledge on which to build best practice did not exist. A pamphlet on these issues, circulated by the World Health Organization (WHO) for local health workers, explained, ‘Viability of the newborn in terms of gestational age [for which birth weight is, in part, a proxy] may differ according to local circumstances. Local data on survival by gestational age are very useful in determining the policy on the gestational age at which resuscitation should be initiated’ (World Health Organization 1997). The international advisers, in other words, could give little help because they recognised that the relevant data must be gathered and analysed at the local level. The physician in Zimbabwe who discussed the 1,000 gram policy worried that it might have been ensuring the survival of severely disabled babies. It is clear, from a range of evidence, that caring for the severely disabled under conditions of poverty puts pressure on kin-based networks of care, with extremely damaging effects on health and welfare.7i The advisory pamphlet, distributed by the WHO, claimed no knowledge of these networks and therefore left space for local judgement: ‘A determinant 178

When Physicians Meet

of when to start or to stop resuscitation is the capacity of the institution/country for the management and further care of malformed, very preterm or severely damaged newborns’. One Zimbabwean physician, in a discussion of his priorities for needed research, asked for studies of health outcomes of resuscitation at a number of different birth weights, so as to be able to make rational decisions about heroic life-saving measures in the context of local outcomes and their implications for people’s lives. The research in question was not purely biomedical. It was deeply social, in the sense that it focused on those medical facts most needed for making responsible decisions in the context of Harare’s everyday living conditions for families – decisions on which babies to save and which to let go. The subject of these decisions sounds brutal, but it is, in fact, a set of issues health workers have to face everywhere. In many industrial country hospitals, bioethicists participate in the decision-making process, but physicians in Harare had to make their decisions without the availability of consulting philosophers. In order to face these decisions responsibly, they needed knowledge about the probable outcomes of one or another course of action. They asked that research to provide this knowledge be treated as a priority. Physicians in developed countries, working at home, have done just these kinds of studies about the outcomes of saving babies at varying birth weights, of varying gestational ages, and in varying conditions. The relevant articles prescribe measures to be taken under a whole range of conditions. They offer charts that predict survival at varying birth weights between 250 and 3,000 grams. And they identify conditions (like a mother’s fever), that increase the probability the child will suffer from cerebral palsy (Draper et al. 1999; MacDonald 2002; American Heart Association 2005). The results of all these studies, however, are valid only for the settings where they are done, because the outcomes depend so profoundly on the availability of a whole array of medical technologies. They depend, also, on continuous attention to the neonate, over a period of time, by a team of physicians and nurses – attention of a kind that was clearly not available at Harare Central Hospital, given the levels of staffing that we have already seen. The presence in Harare of so many HIV-positive babies also affected outcomes. It is important to note just how locally specific the Harare research needed to be if it were to predict outcomes for newborn infants at risk of kernicterus. The problem would not have taken the same form in rural Zimbabwe, where ventilators might not have been available, where some 179

Evidence, Ethos and Experiment

births were attended by traditional birth attendants, and where women in childbirth often found it (and still find it) impossible to get to a hospital in time for the birth. In some rural districts, neonatal mortality rates at that time were as high as 67 per 1,000 live births (Kambarami et al. 1997). In Ghana, a senior physician was concerned with this issue, but not in exactly the same form. He was concerned with the need to reduce the number of infants treated with dangerous exchange transfusions, when they might simply be taken outside in the mid-day sun, for phototherapy, instead. When a severely jaundiced infant was slow in getting to the hospital – perhaps because a traditional birth attendant did not immediately identify the problem – the infant’s condition worsened, and the time for phototherapy passed, leaving exchange transfusion as the only choice. Here, too, as with the Zimbabwean decision on whether to ventilate babies, the Ghanaians expressed a need for research on the later effects of life-saving procedures. One senior Ghanaian physician asked for studies of the impact of the resulting disabilities on families. In Ghana, we held our discussions in Accra, in the south. The issue does not exist in the same form in the north of the country. We learned that, while there were twelve paediatricians in the whole of Ghana, none of them was located anywhere north of Kumasi, in a part of the country as large as Austria and Switzerland combined. A representative of the Ghanaian Department of Social Welfare explained that in northern Ghana, and at times in the south, traditional practitioners ‘put out’ severely disabled babies, who then did not survive. It is difficult to know exactly what this meant, in light of the fact that not all disabilities are visible in infancy, and that northern Ghanaian practices were sometimes looked down on by officials in the south. In any event, hospital-based decisions to save severely impaired babies had different social consequences in each region: they were not the same in northern Ghana and in the south, in the country and in the city, nor were they the same in Accra and Harare. It is hard to imagine that research priorities for northern Ghana would focus on the outcomes of the most heroic measures for extremely low birth weight infants. Mechanical ventilators were not available, and those babies were unlikely to survive. Some medical research results, even ones created in a high-tech northern hemisphere context, transferred easily to a range of very different settings. This was the case with knowledge about levels at which elevated bilirubin becomes dangerous. In the United States, much of the concern with high levels of bilirubin came, as we have seen, after the creation of NICUs. For a period, neonatologists thought that bilirubin levels above 180

When Physicians Meet

twenty milligrams per decilitre were dangerous, and required treatment with exchange transfusion. Then Jon Watchko and Frank Oski published an article (1983) whose tone was heavily ironic (rare for medical journals), on vigintiphobia – fear of the number twenty. They showed that there was little evidence that bilirubin above 20 mg/dL was, by itself, dangerous. The danger came from particular disease processes which might lead, in some cases, to elevated bilirubin. A Zimbabwean paediatrician explained that after this article appeared in the US, exchange transfusions in Harare went down 50 per cent. And a senior Ghanaian paediatrician also expressed a strong preference for phototherapy over exchange transfusion, presumably in response to the same finding. In this case, the medical research from the US was immediately usable in both African countries. In other cases, it was more difficult to assimilate northern country research conclusions, because they were based on data that could not easily be compared with bodies of African data. Zimbabwean physicians requested help in bringing their own database on the causes of childhood disability into line with the standards of the Centers for Disease Control in the US. Only then would they be able to contribute data that might shape international standards, and only then would those standards be translatable into conclusions about local practice. Even then, the textured complexities of local practice would need to be captured in constructing best practice. The presence or absence of ventilators, the likelihood that a newborn will be attended by a traditional birth attendant, and the practice of ‘putting out’ disabled infants all shape paediatric practice. Within these varying contexts, physicians were asking important questions about managing the boundary-line between life and death for infants who were severely compromised. The answers to these questions needed to be based on research, and they needed to be locally specific. The implications of an infant’s survival, in the middle of an AIDS epidemic in Zimbabwe, where there is a severe shortage of adults to care for healthy children, are different from the implications of that child’s survival in the US or Western Europe. Everywhere, including northern countries, that boundary is a contested one. In each place local values and practices, along with medical technologies, interact. And in each case, systematic research is necessary if we are to understand the outcomes of different kinds of policies and different kinds of medical practices. The important thing to remember here is that it would be profoundly wrong to see practice in the US or UK as universal and at Harare Central 181

Evidence, Ethos and Experiment

Hospital as local. American practice on infant resuscitation today is specific to the US (and also varies regionally within the US). To take only one example, the practice of resuscitation in the US today is shaped profoundly by the country’s religious politics. The subject is too complex for a full exposition here but, for example, in 2005 the US Department of Health and Social Services, influenced by the politics of abortion, affirmed ‘the legal principle that infants who are born alive at any stage of development are persons entitled to the protection of the law’ (Wall and Partridge 1997: 64– 70; American Academy of Pediatrics Neonatal Resuscitation Program Steering Committee 2003; Sayeed 2005). This ruling interacted with two laws – one that attempts to guarantee that, at a time when many hospitals work on a for-profit basis, patients in need of emergency care will not be turned away, regardless of their ability to pay; the second sets standards for states in the prevention of child abuse. The question raised by the ruling of 2005 (and by The Born-Alive Infants Protection Act of 2001) was whether a non-viable infant, or an infant whose viability was ambiguous, was now protected by a law requiring that he, or she, be given emergency medical care. A decision not to resuscitate could, imaginably, be treated as a violation of the law. American physicians still withhold resuscitation today, but the way they practice cannot help but be shaped to some extent by the legal environment. For example, one set of practices urged by an article in the journal Pediatrics describes a scheme in Portland, Oregon, in which a whole team meets together to decide on guidelines for the viability of preterm infants (Kaempf et al. 2006). The team includes maternal-foetal medicine specialists, neonatologists, obstetricians, neonatal nurse practitioners and nurses from labour and delivery units and the NICU. These guidelines then shape medical practice, and are also the basis of counselling for mothers of preterm babies. One way to see this scheme is as an enlightened consultative approach to a difficult problem. But there is another possible interpretation: the scheme is one that protects hospital staff from legal vulnerability in a setting where any one physician, or nurse or lay person can charge the hospital with violation of the Emergency Medical Treatment and Labor Act. In this particular local case, one could argue that an entire structure of medical consultation and care has grown up to deal with the religious politics of the United States at this moment. The article about it reports a piece of medical research into best practice (on how mothers who receive counselling perceive their own levels of satisfaction with the hospital’s efforts), but its approach to neonatal resuscitation is as locally 182

When Physicians Meet

specific, as heterogeneous and hybrid, as resuscitation in Accra’s hospitals, or Harare’s. In the Harare and Accra discussions, fascinating contrasts occasionally emerged in the ways the different sets of physicians used basic medical terms. Each local usage worked in its local context: it defined knowledge that was useful, given the particular technologies that were available. At the same time, research physicians were poignantly aware of what might be possible, given more expensive technologies, or if there were a wider range of detailed local knowledge, and so local usages were in tension with northern country ones. In Harare, for example, some physicians, but not all, used the term ‘birth asphyxia’ as the standard cause for neurological damage among newborn infants. One physician explained that ‘birth asphyxia is a problem because tertiary hospitals are overwhelmed by caesarean sections, which are then performed late’.8 We will see that American physicians objected to this blanket use of the term. In Zimbabwe, however, the term had such great currency that it became a part of the language of mothers of disabled children. We saw this when the American group visited a day-care centre for disabled children in a Harare suburb. The centre was organised by community mothers for their own children. The mothers had created an educational play that they sometimes performed for other mothers, and they now performed for us. The dialogue was in Shona, but I stood with a woman who participated in the centre’s work, who interpreted for me. Early in the play, a pregnant woman goes to the hospital. The nurses say, ‘Don’t push, it’s not your time yet’. The interpreter explained to me, ‘If the mother doesn’t push, the baby is slow in coming out and there’s a problem’. Back in the play, with the woman still in labour, the nurse goes off for a tea break. Later, after the baby has been born, and once it is clear that the baby has a problem, mother and child visit a male physician (played by one of the mothers), who speaks the only two sentences in the whole play that are in English: ‘Your baby was slow to come, the problem is birth asphyxia’. Local medical usage had crossed over: the term ‘birth asphyxia’ had been adopted by at least some members of the community of patients’ families. One of the American physicians objected vigorously to the use of the term ‘birth asphyxia’ by his Harare colleagues. (The Zimbabwean usage would later be repeated, in important respects, in Ghana.) ‘It’s meaningless’, he said ‘to make a diagnosis of birth asphyxia without a foetal monitor’. He pointed out that asphyxia was not the only possible cause of the disabilities we were seeing. Some might in fact have been caused by congenital 183

Evidence, Ethos and Experiment

infections, others by metabolic disorders. He explained that the diagnostic process moves to consider these latter possibilities when, through monitoring, ‘we have a record to show that there had not been a problem in the period of childbirth’. Since the Zimbabwean physicians had neither foetal monitors nor technologies to test for metabolic disorders, their use of the term ‘birth asphyxia’ was a practical one, if not always descriptively accurate. Whether the neurological damage had resulted from a compressed umbilical cord during labour, or from a metabolic disorder, the possibilities for medical intervention with the disabled child were the same. Under these conditions, the term ‘birth asphyxia’ was a practical disease entity – its content determined by the technologies available both for monitoring the birthing process and for intervening in its outcome. In fact, the disagreement over birth asphyxia was not merely idiosyncratic and local. The term was a contested and problematic one in the medical literature, concerning the relationship between observable events (the fact that a newborn does not breathe, or has a poor heart rate, muscle tone and colour) and cellular processes in the brain. One article suggested that the gold standard in identifying birth asphyxia was some combination of phosphorous magnetic resonance spectrography, Doppler cranial ultrasound and near-infrared spectrography (Blair 1993; see also Hull and Dodd 1991). These were not easily available even in northern countries. At the other extreme was the definition in the pamphlet distributed by the WHO, which defined birth asphyxia much more simply as ‘the failure to initiate and sustain breathing at birth’ (World Health Organization 1997). In other words, in the international medical literature, as in our discussions, ‘birth asphyxia’ had different meanings depending on the technologies available in the different settings. The American physicians made it clear, in their remarks to me, that disagreements over the diagnosis of birth asphyxia were not grounded in lack of regard for the training and intellect of their counterparts in Harare and Accra. The physician who insisted on the use of a birth monitor said, ‘The individual [physicians] there [in Harare Hospital] are doing their best in a situation of crisis’. Another Philadelphia physician, in the same conversation, said, ‘Their medical training is impressive …. They do things that are well documented in the literature’. If the American physicians had not had access to the technologies of Children’s Hospital, they might well have fallen back on the simpler definition of birth asphyxia. The university physicians in both Ghana and Zimbabwe often spoke about the implications of doing their work in the absence of the relevant 184

When Physicians Meet

tools, and about the need for context-specific knowledge. One Zimbabwean physician said, ‘We need basic knowledge. For birth asphyxia we do the Apgar score. In many cases these are just functional diagnoses. The real diagnoses are not known. As many as thirty-eight per cent of our patients have no known diagnosis.’ He went on to say that ‘Among the disabled there is a need for screening. What is the prevalence of the G6PD trait in this country?’ Notice that the physician began by acknowledging just what his American counterpart had said: he was diagnosing birth asphyxia in the absence of valid medical knowledge. The Apgar score is a rough but important way of making a quantitative estimate of the condition of the infant, at one and five minutes after birth. It assigns points to the quality of the heart rate, respiration, muscle tone, reflex irritability, and colour of the body. He and his colleagues were relying on this rough score, based on observation with the naked eye, because the appropriate technologies were not available. As a result, he was dealing with just functional diagnoses’ rather than diagnoses based on detailed knowledge of organic processes. He was saying that it would have been far better to establish the causes with precision. But this could not be done. The Zimbabwean physician did not ask for us to bring all the machines and diagnostic tools that would be available at CHOP, in Philadelphia. He knew that in a situation of normal emergency this would not happen. Instead, he requested locally specific knowledge, so that he and others could make intelligent decisions based on well-grounded judgements of probability. As an example, he wanted to know the proportion of the overall patient population that suffers from G6PD (Glucose-6-phosphate dehydrogenase) deficiency. Deficiency in this enzyme can occur in foetuses, and can be a cause of haemolysis – the destruction of red blood cells. The deficiency could therefore explain a proportion of the large number of cases of kernicterus that were being lumped together under the term ‘birth asphyxia’. He was saying that one way of beginning to take this gross category apart was to gather information about the prevalence of G6PD deficiency. Knowing these numbers would then open up thinking about the most costeffective ways to reduce the number of neurologically damaged infants. When physicians in Zimbabwe and Ghana thought aloud about what they saw as useful research, one of the most common requests was for the kind of epidemiological information that could inform day-to-day clinical decisionmaking, especially for those kinds of clinical decisions that needed to be made in the absence of laboratory tests. The doctors were trying to learn how to approximate best practice if most of the instruments for it were missing. 185

Evidence, Ethos and Experiment

One of their answers was to improve care by learning about the general distribution of disease patterns in the population from which patients were drawn. This knowledge would permit them to define best practice as the practice most probable to succeed given that distribution of disease. The availability of local epidemiological knowledge would, in addition, help to shape future investments in equipment, medical education and research, because it would make sense to spend the small amount of available money on technologies and training for the most common diseases. It is important to emphasise that this knowledge would have been useful not only in tertiary care settings but at every level of the medical system. With pneumonia, as with G6PD deficiency, a better picture of prevalence in the patient population would enable practitioners to practice more effectively, even in the absence of basic tests. One physician in Zimbabwe told us that she faced an especially difficult problem, at Harare Central Hospital, with paediatric pneumonia. The majority of her patients were infected with HIV; their pneumonia infections were not, therefore, responding to antibiotics in the way that might otherwise have been expected. The appropriate way to deal with this situation would be to culture the organism for each patient, then resistance-test it to see what drugs would control it, and then prescribe the correct antibiotic. This could rarely be done, however, because of a shortage in culture material, as this physician explained with considerable bitterness. Once again, the solution seemed to lie with local-level epidemiological studies. The physician asked for research on drug resistance within the wider population from which her patients were drawn. If she knew this, then she would use the particular drug most probable to be effective, even in the absence of testing. In fact, she later did some research of this kind, herself. Over and over, we heard senior physicians say that they did not have the resources to know crucial facts about local disease conditions. Work had to begin at the ground level; they could not be concerned, immediately, with global solutions, when they did not know with precision what the local problems were. One senior Ghanaian paediatrician said, ‘There is little information about the circumstances in which babies are born ... the highest number of deaths is in the immediate post-partum period, but few post-mortems are done and little blood is taken’. The same physician explained, later, that he and his colleagues had no idea about the actual causes of the hypoxic ischemic encephalopathy – that is, dysfunction of the brain as a result of shortage of oxygen and inadequate flow of blood. Once again, we find that those who were actually dealing, 186

When Physicians Meet

among infants, with the causes of neurological damage, expressed an urgent need for basic information about the causes of disease. These sophisticated African physicians asked, in the context of many different conversations, for the most basic information about the situation in which they themselves were living and working – for example, information about the effects of the drug regimens in common use. In Ghana, where malaria is a major problem, and where many children are treated with chloroquine, a Ghanaian physician pointed out that ‘we have no knowledge of how much damage is done [to vision] as a result of the longterm use of chloroquine’. He also said, ‘We treat all out born babies who come to the hospital [that is, all the many babies born outside the formal medical system] with high-powered antibiotics, because we assume that there were not sterile conditions in childbirth’. The effects of this broad use of antibiotics on newborns were also, of course, unknown. This was still more basic knowledge that was needed before a Ghanaian evidence-based medicine could be constructed. The questions about which drugs to use, and about what their effects were, were shaped not only by the level of resources available to the teaching hospital, but also by those in use at lower levels of the health system. Once again, this was expressed as a failure of knowledge, but in this case without a request for research. In Harare, it was extremely difficult to treat young children who came to the hospital with meningitis. The disease could be caused by any one of several organisms. If patients arrived without prior treatment, the hospital laboratory at that time could culture, stain and identify particular organisms under the microscope. In most cases, however, the patient had been to a primary care clinic where laboratory studies had not been done. Now that a lab was available, the pathogens were no longer visible, since they would have been suppressed by the earlier round of antibiotics. The health workers with whom we met recognised the need for systemic change in healthcare in their own countries. One of the uses of research, as they saw it, was to do exactly this – to bring about change in national priorities in the allocation of resources. In Zimbabwe, health workers said that they could not succeed in winning rehabilitation services for disabled children unless they could find some substitute for the system of DALYs. These are Disability Adjusted Life Years – a measure favoured by the World Bank for measuring the relative losses that people suffered as a result of different disease conditions. There have been great debates over the distortions created by this particular method of measuring disease burdens (Lopez et al. 2006). 187

Evidence, Ethos and Experiment

In Zimbabwe, in 1998, professionals who worked to rehabilitate the disabled felt that the DALY system undercounted the larger social effects of disability, and they wanted better measures, so that they could prove to the Ministry of Health that rehabilitation was an important priority. Ghanaian health workers also wanted broad measurements of disability and its effects. In their case, they wanted to overcome a division between the Ministry of Health (which concerned itself with issues like the birthing process) and the Ministry of Social Welfare, which was in charge of rehabilitation. Here, as in Zimbabwe, correct research on the burden of disease could possibly change national priorities. One could imagine that even with better measures of the burden of disability, rehabilitation services might have proven too expensive. But it is also possible to imagine that these data would have made an even more compelling case for research on the prenatal and perinatal causes of disability. Zimbabweans and Ghanaians were especially interested in what they could learn from each other. One member of the Pennsylvania team had been born in Ghana, and returned to work there from his base in Philadelphia. Another Ghanaian paediatrician had spent many years in Zimbabwe. And one of the Zimbabwean physicians came along with us to Accra. Ghanaians were especially interested, for example, in the Zimbabwean approach to training rehabilitation technicians. These were generalists who learned some of the skills of each of a number of rehabilitation specialists: they learned some of the work of physical therapists, occupational therapists, speech therapists and so on. This had multiple advantages: they provided services at the district level, across the country, in places where there was no hope of getting multiple rehabilitation specialists. In addition, rehabilitation technicians held qualifications that were not easily transferable to wealthier countries: they were less likely than their more specialised counterparts to contribute to the brain drain. The local research interests of the Zimbabwean and Ghanaian physicians, as presented to us, can be seen as a complement to international research seen as a global public good. A locally usable texture of research knowledge must exist alongside globally usable knowledge. The idea of a global public good has been important as a way of explaining to political leaders and publics in the industrial world that wellbeing in their own nations is served if they provide support for health improvements in poor countries – that the parts of the world are intertwined in such a way that even the rich reap a benefit from certain kinds of health provision for the poor. A global public good must, however, meet a criterion of universal usefulness. According to an authoritative statement: 188

When Physicians Meet

Their benefits are quasi universal in terms of countries (covering more than one group of countries), people (accruing to several, preferably all, population groups), and generations (extending to both current and future generations.) This property makes humanity as a whole the publicum, or beneficiary of global public goods. (Kaul et al. 1999b) This defines a very different set of priorities than the ones discussed by physicians in Harare and Accra. They were interested in locally specific knowledge that would permit them to develop a form of evidence-based medicine that would work under conditions they faced, given local disease conditions, equipment, bureaucratic structures, patients’ resort to traditional healers and so on. The criterion of the global public good, by contrast, required that solutions be useful everywhere in the world, in every local context. This makes sense as a way of prying resources out of donors, but leaves national physicians without much power, unless they leave their national contexts and work for international donors. Debates about inequalities in international health are usually cast as contests over resources. Is enough money being given? If donor countries gave larger sums, would the money do any good? Should pharmaceutical companies be able to keep the rewards for their intellectual property, even if the high cost of drugs leads to a reduction in the number of people who can be treated? These debates are rarely cast in terms of the locus of knowledge, the nature of biomedical knowledge (universal or local), or the relative authority of health workers employed within the country or answerable to external donor agencies. Many issues are therefore not addressed because the people who would address them are effectively silenced. Donor agencies working in Africa have concentrated on AIDS, and then after AIDS, on other infectious diseases, but rarely on non-communicable diseases. The most recent edition of the World Bank’s authoritative Disease Control Priorities in the Developing World points out that ‘Although 50 percent of deaths ... of children under age five occur at ages younger than 28 days, relatively little attention has been paid to this age group’ (Jamison 2006a: 4). We have seen that eight years earlier, Ghanaian and Zimbabwean physicians were expressing frustration about this problem, but did not have the power to define priorities separate from those of international donors. The points of greatest contention on health in poor countries come where authorities must decide how much of acceptable medical practice 189

Evidence, Ethos and Experiment

needs to be jettisoned to improve the health of the poor, given an inadequate level of funding. The logic of finding simple, effective interventions is clear and (in one sense) admirable. Small measures do help: iron and folate supplements and tetanus injections for pregnant women, hygienic care of the cord after birth, and other simple measures, do have a significant impact, even though they are defined on a world scale to be used in thousands of varying local contexts (Lawn et al. 2006). The search for simple solutions is also, in many cases, a move to circumvent the need for local medical expertise. Many international planners argue that in the absence of institutional capacity in poor countries, interventions need to be made so simple that the lowest level of health workers can undertake them (Gericke et al. 2005). The picture we get is of a radically simplified form of medical practice that, for most patients, can be carried out by health workers with low levels of education, using inexpensive drugs and equipment, according to plans drawn up by international experts. The logic of cheap measures does break down at moments to reveal the brutality of the political logic underlying it. What is troubling, however, in light of our discussions in Ghana and Zimbabwe, is that the breakdowns inevitably emerge as debates among authorities in rich countries about what is to be done for (or to) the poor. There is an assumption that the capacity to make medical decisions at a national or regional level is irrelevant. This is a self-fulfilling diagnosis, because it is part of a process by which national capacities are hollowed out. We can see the marginal position of national research physicians in many of the crucial debates over appropriate research and care in the poorest countries. In 1997, for example, physicians in the US and around the world engaged in a bitter and intense discussion about poverty and research ethics. The research, meant to find a cheaper way to control the transmission of HIV from mother to child, used a shorter course of a drug already in use in the US (AZT), and also gave women in the control group placebos, even though a known drug was effective. The principal participants in the debate were, on the one side, the editor of the New England Journal of Medicine and on the other, the Directors of the US National Institutes of Health and Centers for Disease Control. Physicians from countries where the trials were being conducted contributed only marginally. The Ugandan physicians who made public comments supported the trials because the treatment used in the US at that time cost $800, more than eighty times the government per capita health expenditure (Angell 1997; Bagenda and Mudido 1997; Varmus and Satcher 1997; Mbidde 1998). The trial provided 190

When Physicians Meet

some women with the drug: without the drugs provided for research, there would have been none at all. Seen from the point of view of Uganda’s research physicians, they could accept the AZT trials, contribute to making them effective, or reject them. They would have found it difficult, however, to initiate an entirely different line of research, especially if it had nothing to do with HIV. Elite African physicians who are engaged in both clinical practice and research occupy a strategic position that is not duplicated anywhere else in the structure of international health. They come face to face with patients in clinical practice, they have a rich texture of knowledge about the actual practice of healthcare in their own countries, and they know about international priorities. Most important of all, they are concerned with a full range of health problems, and not only those defined as international priorities. It is interesting to think about how the course of medical research in their countries would change if they were playing a fuller role in defining priorities. In the absence of this sort of knowledge, best practice defined by even the best, most skilled and best-informed international agencies can never be good enough. To say that it can never be good enough is not, of course, to say that the world would be better off without it. If African children lacked the fruits of vaccine initiatives, or of research on micronutrients, they would be worse off. Even when international researchers make mistakes, the world of international medical assistance is so varied and diverse that the errors can be corrected. Multidrug-resistant tuberculosis, for example, appeared at first to be too expensive to treat in resource-poor settings, but then critics of two-tiered medicine (one for the rich, the other for the poor) showed the dangers of leaving this disease untreated (Farmer and Nardell1998; Farmer 2001). Work presented in this volume shows that international researchers can be enormously imaginative about the strange technological hybrid that is medicine. The project described by Patrick Kachur, for southern Tanzania, gives serious attention not only to basic research on the effects of antimalarials but also to the design of age-specific dosing envelopes and the organisation of drug distribution outlets. The problem, of course, as contributions to this volume on malaria vaccine trials in Gambia show, is that at a certain moment the researchers will go home, and who will be in charge of medical care at that moment? What resources will be available to them? And what research will they be able to do? International funding and control over research, combined with radically simplified forms of good-enough medicine, presents an additional 191

Evidence, Ethos and Experiment

problem for the reproduction of the medical profession within the continent. Medical students everywhere need to be taught about practice as a form of intellectual inquiry. In a context where research is organised from the outside, and where efficacious drugs are not available, teachers and students can find that all avenues for inquiry are cut off, that the realities of their everyday situation reduce them to the level of unskilled local practitioners. Dr Erastus Amayo, of the Department of Medicine at the University of Nairobi, is quoted on this subject in an article on clinical realities in eastern Africa: [It] makes you feel you may lose in your area of interest – Your proficiency, your particular area, and even your particular [ability at] solving diseases .... The basic thing in teaching ... you want to have something that is challenging the students and you to eventually reach a diagnosis and that was what medicine was supposed to be – to tax your mind, with a differential diagnosis, and do the investigation and come up with the diagnosis .... Students fear ... their biggest worry is that they will not be recognized [as competent physicians with requisite skills acquired in patient care] (Good et al. 1999). Compare this sense of futility, and of having reached an intellectual dead end, with the excitement expressed by a physician working at an American teaching hospital during the early years of the epidemic: Because I like to solve mysteries, I found it very intellectually stimulating because, again, of the multipresentations. You could see children with autoimmune disease. You could see children with low platelet counts ... and children with interstitial pneumonitis. So it was an extremely varied presentation, and intellectually it was very interesting and fascinating. (Bayer and Oppenheimer 2000: 35) To explore the contrast between these two physicians, at two moments, is not to say that intellectual inquiry occurs only in industrial countries, or only by international physicians. John Iliffe has described the great part African physicians played in uncovering the AIDS epidemic’s existence, and also some of its important characteristics (Iliffe 1998). Nevertheless, the conditions of clinical practice at most tropical African university hospitals present difficult problems for physicians who wish to continue to teach clinical practice as a form of inquiry. What is at stake here, when we look at physicians at different national and institutional locations around the world, cannot be reduced to a differ192

When Physicians Meet

ence in point of view. Marilyn Strathern, writing in this volume about communication between practitioners and subjects, asks us to ‘Suppose that there were people who did not invest in epistemology in the same way that permeates Euro-American science’. When we are talking about physicians, however, what is both fascinating and crucial is the degree to which physicians in Africa, who invest in epistemology in exactly the same way as their European or American counterparts, come to a radically different yet extremely fruitful understanding of the knowledge that is needed to improve healthcare. They are embedded in the world of sick people in a very different way from their international counterparts. They must gain bodies of knowledge that their counterparts cannot fully imagine from a distance. They have their own very different embodied, and moral and scientific understanding of their work. And they do this under conditions of normal emergency.

Notes 1. I am grateful for the comments of Wenzel Geissler, Robert Aronowitz, Elise Carpenter, Julie Livingston, Charles Rosenberg and Donald Silberberg. I am grateful also for the help of Elizabeth Alpern, Dennis Durbin and Kwaku Ohene-Frempong. 2. Randomised controlled trials are the gold standard of biomedical research. Walter W. Rosser points out that ‘an adequate RCT (randomised controlled trial) to prove that papanicolau smears reduce mortality from cancer of the cervix would require more than a million women allocated randomly to receive or not receive smears with followup for at least 30 years for credible results’ (Rosser 1999: 661–63). For the more general points on the uses and weaknesses of disease specificity as supported by clinical research, see Rosenberg (2002); see also Aronowitz (1998). 3. This view has, of course, been challenged by Paul Farmer and Partners in Health. A book about Farmer’s career and approach quotes one of his associates as saying, ‘do you know what appropriate technology means? It means good things for rich people and shit for the poor’ (Kidder 2004: 90). 4. Wellcome Trust, Successful Applications, 2000. Strategies for the management of multidrug-resistant tuberculosis in Kampala, Uganda, http://www.wellcome.ac.uk/ doc_wtd003303.html. New insights in the epidemiology and control of Taenia solium, taeniasis/cysticercosis. http://www.wellcome.ac.uk/doc_wtd003304.html, accessed 8 June 2006. 5. Many of the physicians at Children’s Hospital of Philadelphia (CHOP) were engaged in research, and not only clinical care. In addition, some children at CHOP were brought from more distant obstetrical units, and not only the one at the Hospital of the University of Pennsylvania. 193

Evidence, Ethos and Experiment

6. The estimated number of births for CHOP was provided by Dr Elizabeth Alpern and the estimated number of physicians by Dr Dennis Durbin. The CHOP Neonatal Intensive Care Unit, at that time, admitted about 600 children per year. The staff at Harare Central reported that they averaged about 2,000 births per month. Kambarami et al. (1999) reported that Harare Central Hospital’s Neonatal Unit admitted 5,397 patients in 1998, and that this was 25% of Harare Hospital deliveries for the year (for a total of more than 21,000). Of infants admitted to the NNU, 19.3 per cent died. It seems clear that many of these admissions would have qualified for NICU care at CHOP. Other researchers report that for a three-month period between March and June 1999, Harare Central Maternity Hospital (the maternity unit of Harare Central) had 3,722 women giving birth (Feresu et al. 2004). 7. My own detailed research on caregiving networks in Tanzania shows that there is a balance between the labour, money and other resources that caregivers within a network can provide, and the needs of sick and disabled people, or young children, within that network. If there are too many demands for care relative to the overall capacity of the caregivers, the entire network can tip over into helplessness. One way to deal with these imbalances is to send a child to live with a relative, but here, too, there are limits to resources. In my own understanding, visible AIDS orphans, without support, appear when extended caregiving networks can no longer cope. A child with cerebral palsy makes very large demands for care, and so affects an entire network. On migration within networks, see Ansell and van Blerk (2004). 8. Conversations were not tape-recorded. All quotations are taken from my own notes. I took great care with them, but the word order is not precise, and there are, inevitably, elisions.

References American Academy of Pediatrics, Neonatal Resuscitation Program Steering Committee. 2003. ‘Born-alive Infants Protection Act of 2001, Public Law No. 107–207’, Pediatrics 111(3): 680–81. American Heart Association. 2005. ‘Neonatal Resuscitation Guidelines’, Circulation 112(Suppl. 1, Part 13): 188–295. Angell, M. 1997. ‘The Ethics of Clinical Research in the Third World’ (editorial), New England Journal of Medicine 337(12): 847–849. Ansell, N. and L. van Blerk. 2004. ‘Children’s Migration as a Household/Family Strategy: Coping with AIDS in Lesotho and Malawi’, Journal of Southern African Studies 30(3): 673–90. Aronowitz, R.A. 1998. Making Sense of Illness: Science, Society and Disease. Cambridge: Cambridge University Press. Bagenda, D. and P.M. Mudido. 1997. ‘We’re Trying to Help Our Sickest People, Not Exploit Them’, Washington Post September 28, 1997. Bayer, R. and G.M. Oppenheimer. 2000. AIDS Doctors: Voices from the Epidemic. New York: Oxford University Press. 194

When Physicians Meet

Blair, E. 1993. ‘A Research Definition for “Birth Asphyxia”?’ Developmental Medicine and Child Neurology 35: 449–55. Brett, J.A. and S. Niermeyer. 1998. ‘Neonatal Jaundice: The Cultural History of the Creation and Maintenance of a “Disease” of Newborns’, in N. Scheper-Hughes and C. Sargent (eds), Small Wars: The Cultural Politics of Childhood. Berkeley, pp. 111– 29: University of California Press. Das, V. 1999. ‘Public Good, Ethics, and Everyday Life: Beyond the Boundaries of Bioethics’, Daedalus 128(4): 99–133. Draper, E., B. Manktelow, D.J. Field and D. James. 1999. ‘Prediction of Survival for Preterm Births by Weight and Gestational Age: Retrospective Population Based Study’, BMJ 319: 1093–97. Farmer, P. 1997. ‘Social Scientists and the New Tuberculosis’, Social Science and Medicine 44(3): 347–58. ———. 2001. ‘The Major Infectious Diseases of the World – to Treat or Not to Treat?’ New England Journal of Medicine 345(3): 208–10. Farmer, P. and E. Nardell. 1998. ‘Nihilism and Pragmatism in Tuberculosis Control’, American Journal of Public Health 88(7): 1014–15. Feresu, S.A., S.D. Harlow and G.B. Woelk. 2004. ‘Risk Factors for Prematurity at Harare Maternity Hospital, Zimbabwe’, International Journal of Epidemiology 33: 1194– 201. Gericke, C.A., C. Kurowski and A. Mills. 2005. ‘Intervention Complexity – A Conceptual Framework to Inform Priority-setting in Health’, Bulletin of the World Health Organization 83(4): 285–93. Good, M.-J. DelVecchio, E. Mwaikombo, E.M. Amayo and J. M’Imunya. 1999. ‘Clinical Realities and Moral Dilemmas: Contrasting Perspectives from Academic Medicine in Kenya, Tanzania, and America’, Daedalus 128: 172–73. Hull, J. and K. Dodd. 1991. ‘What Is Birth Asphyxia?’ British Journal of Obstetrics and Gynaecology 98: 953–55. Hughes, T.P. 1991. ‘From Deterministic Dynamos to Seamless-Web Systems’, in H.E. Sladovich (ed.), Engineering as a Social Enterprise. Washington, DC.: National Academy Press, pp. 7–25. Iliffe, J. 1998. East African Doctors: A History of the Modern Profession. Cambridge: Cambridge University Press. Jamison, D.T. 2006a. ‘Investing in Health’, in D.T. Jamison, J. Breman, A. Measham, G. Alleyne, M. Claeson, D. Evans, P. Jha, A. Mills and P. Musgrove (eds), Disease Control Priorities in Developing Countries, 2nd ed. New York.: Oxford University Press and The World Bank. Jamison, D.T., , J. Breman, A. Measham, G. Alleyne, M. Claeson, D. Evans, P. Jha, A. Mills and P. Musgrove. 2006b. Disease Control Priorities in Developing Countries, 2nd edn. New York: Oxford University Press and The World Bank. Kaempf, J.W., M. Tomlinson, C. Arduzza, S. Anderson, B. Campbell, L. Ferguson, M. Zabari, and V. Stewart. 2006. ‘Medical Staff Guidelines for Previability Counseling and Medical Treatment of Extremely Premature Infants’, Pediatrics 117(1): 22–29. Kambarami, R.A., M.Z. Chirenje and S. Rusakaniko. 1997. ‘Perinatal Mortality and Associated Socio-demographic Factors in Two Rural Districts in Zimbabwe’, The Central African Journal of Medicine 43(6): 158–62. 195

Evidence, Ethos and Experiment

Kambarami, R.A., P. Matibe and D. Pirie. 1999. ‘Risk Factors for Neonatal Mortality: Harare Central Hospital Neonatal Unit – Zimbabwe’, Central African Journal of Medicine 45(7): 169–73. Kaul, I., I. Grunberg and M.A. Stern (eds). 1999a. Global Public Goods: International Cooperation in the 21st Century. New York: Oxford University Press for The United Nations Development Programme. ———. 1999b. ‘Defining Global Public Goods’, in I. Kaul, I. Grunberg and M. Stern (eds), Global Public Goods: International Cooperation in the 21st Century. New York, pp. 2–19: Oxford University Press for the United Nations Development Programme. Kidder, T. 2004. Mountains Beyond Mountains. New York: Random House. Kravitz, R., N. Duan and J. Braslow. 2004. ‘Evidence-based Medicine, Heterogeneity of Treatment Effects, and the Trouble with Averages’, The Milbank Quarterly 82: 661– 87. Lawn, J., J. Zupan, G. Begkoyian and R. Knippenberg. 2006. ‘Newborn Survival’, in D.T. Jamison, J. Breman, A. Measham, G. Alleyne, M. Claeson, D. Evans, P. Jha, A. Mills and P. Musgrove(eds), Disease Control Priorities in Developing Countries, 2nd ed. New York, pp. 531–50.: Oxford University Press and The World Bank. Lopez, A.D., C. Mathers, M. Ezzati, D. Jamison and C. Murray. 2006. Global Burden of Disease and Risk Factors. New York.: Oxford University Press and The World Bank. MacDonald, H. 2002. ‘Perinatal Care on the Threshold of Viabililty’, Pediatrics 110(5): 1024–27. Mbidde, E. 1998. ‘Bioethics and Local Circumstances’ (editorial), Science 279(9 January): 155. Rosenberg, C.E. 2002. ‘The Tyranny of Diagnosis: Specific Entities and Individual Experience’, The Milbank Quarterly 80(2): 237–60. Rosser, W.W. 1999. ‘Application of Evidence from Randomised Controlled Trials to General Practice’, Lancet 353(20 February): 661–63. Sayeed, S.A. 2005. ‘Baby Doe Redux? The Department of Health and Human Services and the Born-alive Infants Protection Act of 2002: A Cautionary Note on Normative Neonatal Practice’, Pediatrics 116(4): e576–e585. Taussig, M. 1987. Shamanism, Colonialism, and the Wild Man: A Study in Terror and Healing. Chicago: University of Chicago Press. Varmus, H. and D. Satcher. 1997. ‘Ethical Complexities of Conducting Research in Developing Countries’, New England Journal of Medicine 337(14): 1003–5. Wall, S.N. and J.C. Partridge. 1997. ‘Death in the Intensive Care Nursery: Physician Practice of Withdrawing and Withholding Life Support’, Pediatrics 99(1): 64–70. Watchko, J.F. and F.A. Oski. 1983. ‘Bilirubin 20 mg/dL = Vigintiphobia’, Pediatrics 71: 660–63. World Health Organization. 1997. Basic Newborn Resuscitation: A Practical Guide. Maternal and Newborn Health/Safe Motherhood Unit, Division of Reproductive Health (Technical Support). WHO/RHT/MSM/98.1. Geneva: World Health Organization.

196

Chapter 8

The Plausibility Design, Quasiexperiments and Real-world Research: A Case Study of Antimalarial Combination Treatment in Tanzania S. Patrick Kachur

Experimental and Observational Designs: A Primer Since the late 1980s, the concept of evidence-based medicine has increasingly shaped clinical practice and health policy – initially in northern countries (Sackett et al. 1996) – but eventually expanding globally along with related political and economic forces and the agendas that drive them. The evidence-based medicine movement was appealing to policy makers in the context of broader neoliberal reforms in the public health sector and beyond (Berguer 2004). Principles of evidence-based medicine were rapidly generalised from academic biomedical research settings to clinical practice with individual patients and onto more general recommendations for whole populations (Eriksson 2000; Macintyre et al. 2001). Ideally, the evidence-based approach considers the entire body of biomedical scientific findings related to a given treatment or preventive strategy. It favours interventions that can improve survival, health outcomes, quality of life or cost-effectiveness. In the process, some forms of evidence are privileged over others. Experimental evidence, particularly evidence from randomised controlled trials (RCT), is regarded as the most compelling (Guyatt et al. 1993, 1994). That is because the experimental approach is believed least prone to bias. If you want to know how well a new vaccine works for preventing 197

Evidence, Ethos and Experiment

pneumonia in children, you could design a simple experiment. You could choose a sample of children known to be at risk for pneumonia and randomly assign them to receive the new vaccine or not, then track the occurrence of pneumonia in the vaccinated and unvaccinated groups. You could make certain to treat both groups identically in all other respects and carefully monitor their health outcomes over time. Such an RCT design could provide evidence of the new vaccine’s efficacy – how well it can work under ideal conditions (Last 1995). As an experimental researcher, you could isolate the biomedical effect of the vaccine from its context by giving the unvaccinated children an inactive, dummy or placebo vaccine which the children, parents and study staff could not distinguish from the genuine article. You would also be aware that other factors, such as mother’s education or household socioeconomic status, might contribute to this risk. The random allocation of children to the treatment or comparison group is a simple way to isolate the impact of the vaccine. If the randomisation is conducted as planned, then all other factors should be equal between the two groups and you can confidently attribute any differences in outcome that you observe to the vaccine. This gives the RCT design a high level of internal validity, which appeals to proponents of evidence-based policy. Experimental designs are particularly well-suited to establishing the efficacy of straightforward biomedical interventions that affect measurable outcomes within a short period of time. Your experimental study of the pneumonia vaccine would be feasible if the incidence of childhood pneumonia was expected to be high and occurred within a relatively short follow-up period. It gets more difficult if the researchers have to continue assessing children for months or years, during which time some might move away, if the vaccine involves multiple booster doses that some children miss, or if other services available in the community also help prevent pneumonia or mitigate its impact. Experimental researchers often go to extraordinary lengths to minimise these potential threats to their design: providing incentives, restricting enrolment, censuring follow-up and so on. So much so that the academic biomedical experimental setting often becomes so altered it no longer resembles practice – the real-world context in which the intervention ultimately must be delivered. That is, the experimental design often has limited external validity or generalisability (Charlton 1995). For public health researchers generalisability is key. We are, of course, keenly interested in experimental studies that establish the biomedical efficacy of interventions delivered and evaluated under ideal conditions. This 198

The Plausibility Design, Quasi-experiments, and Real-world Research

is particularly so when we are considering newly developed interventions without an existing body of evidence to compare them with the standard of care. But once a certain level of efficacy has been established, we are responsible for devising workable delivery strategies with a high level of effectiveness. That is, delivering interventions that achieve the promised public health benefit when deployed under real-world conditions (Last 1995). There are several ways of approaching effectiveness studies. Sometimes these are referred to as quasi-experiments, or natural experiments, to distinguish them from the highly valued standard for true experiments (Robson 2002). You could assess the effectiveness of the vaccine by observing rates of pneumonia before and after it was introduced, or by comparing pneumonia incidence in one community where the vaccine is used and another where it is not. Either approach would allow you to observe what happens when people are free to accept an offered service or not, or when strict criteria for inclusion do not apply, as well as how the existing public health system can be adapted to incorporate the new intervention. It might also require less intense follow-up and intervention on the part of the research team. However, observational designs including quasi-experiments are prone to biases that the experimental researcher would have eliminated (or at least minimised). In place of the satisfyingly self-evident result provided by a RCT (e.g. ‘The new vaccine works better than a placebo’), at the end of a quasi-experiment, investigators are left with a qualified set of conditional outcome statements (e.g. ‘The new vaccine can work if health workers are able to provide the recommended number of doses .... The new vaccine prevented X cases of severe pneumonia for every $100 spent …. The new vaccine might be more effective if successful interventions improved demand and uptake ...’) and obliged to consider critically what other factors might have influenced it. Was there a change in the overall prevalence or virulence of pneumonia? Was this same trend present in the comparison area where vaccine was not available? Did enough people take advantage of the vaccine for it to have a measurable effect? Was it delivered in the same manner at every site where it was available? Did factors other than the vaccine get introduced during the observation period? How did the research team distort the usual healthcare delivery system, and how might this have affected the outcomes? These questions demand that observational studies be considered from a much more situated perspective than experimental studies.

199

Evidence, Ethos and Experiment

A Polarised Debate This tension between almost purely positivist and more relativist perspectives, as well as between deductive and inductive approaches to creating medical and public health knowledge is at the core of an ongoing debate over the relative importance of experimental and observational studies (Banta and European Advisory Committee on Health Research 2003). Academic biomedical proponents of the evidence-based movement maintain that the RCT approach is the ‘definitive expression of late twentieth century scientific philosophy in clinical medicine’ (Holmberg and Baum 1995) and lament how little clinical practice is actually informed by defensible science (Callaham 1997). They are quick to cite examples where RCT findings reversed recommendations based on rather convincing – but ultimately biased – results from non-experimental studies (Rossouw et al. 2002). At the same time, they tend to overlook the larger number of instances where RCTs and observational studies produce nearly identical results (McKee et al. 1999; Benson and Hartz 2000; Ioannidis et al. 2001; Concato 2004). They prize the internal validity afforded by the randomisation process and are critical of the statistical methods that non-experimental researchers have devised to explain away bias and statistical confounding after the fact (Vandenbroucke 2004). They view clinical evidence as a hierarchical organisation of knowledge, at the top of which rest findings from large-scale RCTs and meta-analyses of such trials. One mark of the success of the evidence-based movement is that its hierarchical perspective has become widely institutionalized over the past three decades. Some version of this hierarchy of evidence has been adopted by national and international bodies including the Canadian Task Force on the Periodic Health Examination, the Cochrane Collaboration, the North of England Evidence Based Guideline Development Project, the Revised Scottish Intercollegiate Guidelines Network, US Agency for Health Care Policy and Research, and US Office of Disease Prevention and Health Promotion and its Preventive Services Task Force, among others (Kroke et al. 2003). The apparent hegemony of the evidence-based movement and the RCT design may be illusory, however. RCT design studies are outnumbered by observational studies published in the most respected academic journals by a ratio of three or four to one (Ray 2002). Yates maintains that the evidence debate overstates the role of a relative few zealous, but highly prolific, academic biomedical proponents (Yates 1997). Since the start of the evidence-based movement, there have been equally vocal scholars who 200

The Plausibility Design, Quasi-experiments, and Real-world Research

warned about overemphasising RCT findings (Black 1996, 2001; Petrie 1998), and ignoring results based on observational approaches situated in specific healthcare delivery systems. Concerns about overly relying on experimental evidence have been voiced by experts in fields as diverse as alternative medicine (Verhoef et al. 2002), medical informatics (Harris et al. 2006), perinatology (Lorenz and Paneth 2003), physical medicine and rehabilitation (Horn et al. 2005), and even equine veterinary medicine (Muir 2003). Smith and Pell made a dramatic point by publishing a metaanalysis of all RCT studies demonstrating the efficacy of the parachute in the British Medical Journal. They found none and concluded that the seemingly obvious benefit of the parachute has yet to be proven through a scientific RCT (Smith and Pell 2003). Smith and Pell’s work is a deliberately provocative example. Even so, it helps to demonstrate that the body of published opinion on the evidence question over the past decades would benefit from careful discourse analysis and political-economic examination – tasks that are beyond the limits of this paper and its author. The polarised nature of the discussion may be rhetorical, at least in part. Both forms of research seek to measure similar outcomes and both are analysed based on statistical theories of probability. Proponents of both views have tended to overstate their differences and overlook commonalities. For example, most of the seminal descriptions of evidence-based medicine do acknowledge a role for observational evidence and clinical judgement – even as they privilege large-scale experimental findings (Guyatt et al. 1993, 1994; Sackett et al. 1996). More regularly, the discussion underscores a longer-standing divide: that between the academic medical centre and the individual clinical practitioner. Berguer links the evidence-based movement to an emergent academic-industrial complex – one that ensures its own continued funding by trumpeting the need for ever more experimental studies (Berguer 2004). Other scholars maintain that it is irresponsible to apply RCT evidence averaged over many patients to the individual decisions that clinicians and their patients face daily (Charlton 1995), and that practicing clinicians are too busy to differentiate between findings that are statistically significant and those that have relevance for their daily practice (Muir 2003). By overemphasizing RCT approaches and their associated statistical techniques, others have expressed concern that clinical practice researchers will risk overlooking important inputs they can make towards understanding the pathophysiology of health conditions and distort the dialectic on which medical education is based (Petrie 1998). 201

Evidence, Ethos and Experiment

Moving Beyond the Debate It is not surprising that proponents of experimental and observational approaches have engaged in a highly polarised debate. Even the historical origins of the two approaches are distinct. Experimental research can be traced back to Mendelian genetics and agricultural research (Yates 1997) while observational studies were developed to assess interventions addressing education and complex social issues (Prideaux 2002). As early as forty years ago, clinical practice researchers began suggesting other ways forward based on hybrid research designs including the pragmatic clinical trial (Schwartz and Lellouch 1967). Rather than meeting the explanatory objectives of a RCT, pragmatic clinical trials aim to collect practical information to guide routine clinical practice. Pragmatic trials are still frequently randomised, but acknowledge that at least some of the complex characteristics of clinician and patient that defy randomization can be incorporated into a trial and its assessment (Godwin et al. 2003). The relative advantages of pragmatic clinical trials were immediately evident for interventions for which randomisation and blinding would be impossible or unethical, such as surgical techniques. However, a number of clinicians were quick to apply the model to less invasive medical interventions (Helms 2002). In an extensive review of published literature, Carles Vallvé identified only ninety-five trials that claimed to follow a pragmatic design, of which only four met his strict definition of a pragmatic clinical trial (Vallve 2003). However, Charlton maintains that nearly all contemporary multi-site mega-trials are already more pragmatic than explanatory (Charlton 1995). Another variation on the same theme, intended to incorporate the range of complex clinical situations in which an intervention might be evaluated, has been proposed in the large, simple trial design (Sandercock 1998). Epidemiology has its roots in observational and descriptive methodology (Last 1995). But its close link to academic biomedical research traditions and institutions has resulted in the application of evidence-based medicine to public health policy. Like their counterparts engaged in clinical practice research, some public health professionals have sought to steer a course of “subtle realism” between extremes (Banta and European Advisory Committee on Health Research 2003). Some have suggested ways to minimise bias and enhance the credibility of observational methods (Radford and Foody 2001). Others have devised frameworks for designing and evaluating public health interventions research (Campbell et al. 2000). These works tend to emphasise that some interventions come into common usage without the benefit of RCT evidence, either because they are 202

The Plausibility Design, Quasi-experiments, and Real-world Research

intrinsically sensible, because sufficient observational evidence has already accumulated in their favour, because random assignment would be impractical or unethical, because strong advocacy has brought them into common practice, or because anxious policy makers are often disinclined to wait for experimental conclusions. Habicht and others have suggested a much more important reason to reconsider the hierarchy of evidence for population-level or community-based studies. They suggest that public health interventions should not be judged by the standards used for clinical research (Habicht et al. 1999). They argue that to demonstrate effectiveness, researchers need not prove that the intervention caused an observed public health gain. Rather, if they can demonstrate that the intervention was delivered adequately and a public health benefit was documented, the onus should be on demonstrating that it is plausible the two are linked. The RCT approach uses randomisation to arrive at conclusions about biomedical causality, linking intervention to outcome through probabilistic theory. Plausibility evidence, on the other hand, is assembled from multiple additional indicators along the way to support the credible conclusion that the intervention was delivered sufficiently and could reasonably be interpreted to have caused or contributed to the observed impact. One example comes from the evaluation of a 1992 California state statute requiring motorcycle drivers and their passengers to wear helmets (Kraus et al. 1994). The authors measured a reduced rate of severe and fatal injuries related to motorcycle crashes after the law was introduced. But the findings were purely observational and the only link between the law and the outcome was ecological – they occurred in approximately the place and over the same period of time. The plausibility that one followed from the other could be enhanced by presenting information that: the law was, in fact, enforced; motorcycle use continued following the law; helmet use increased (Kraus et al. 1995); the rate of motorcycle crashes was unchanged (or increased); and emergency response or access to trauma care remained constant (or decreased).

IMPACT-Tanzania: A Case Study in Real-world Plausibility Evaluation The US Centers for Disease Control and Prevention worked with the Ifakara Health Institute and other partners to conduct a similar observational plausibility evaluation in rural Tanzania. The rest of this paper will describe that project and illustrate some of the deliberate choices we made 203

Evidence, Ethos and Experiment

in order to balance concerns about scientific credibility and relevance to the real-world healthcare delivery system. In recent years, malaria control programmes in Africa have struggled to confront rapidly evolving resistance to the drugs most widely used to treat this common and potentially fatal illness. Many malaria researchers have suggested that new strategies based on artemisinin-containing antimalarial combination treatments (ACTs) should be deployed as soon as possible (Warhurst and Duraisingh 2001; Whitty et al. 2004). ACT combines widely used synthetic antimalarial drugs with compounds derived from a 2000year-old Chinese herbal treatment, quinghaosu, or Artemesia annua. Proponents of ACT strategies believe that this approach can forestall the progression of drug resistance (Price et al. 1998) and reduce malaria transmission (Nosten et al. 2000). The evidence base for these claims is entirely observational and comes from areas where malaria transmission is very low and the resources to deal with it are far better than elsewhere in the malaria-endemic world (Barnes et al. 2005; Bhattarai et al. 2007). Smallscale drug efficacy studies, often conducted as RCTs, have shown that ACT can be more efficacious than common antimalarial monotherapies under experimental conditions (Adjuik et al. 2004). We felt there was a need for real-world plausibility studies to demonstrate whether this strategy was feasible to deliver through existing health systems in a typical African setting of high malaria transmission and whether it would be associated with any measurable effect on malaria transmission or drug resistance (Kachur et al. 2001).

Figure 8.1: Map of IMPACT Tz Study Districts in Southern Tanzania. 204

The Plausibility Design, Quasi-experiments, and Real-world Research

The Interdisciplinary Monitoring Project for Antimalarial Combination Therapy in Tanzania (IMPACT-Tz)1 was a multiyear evaluation, comparing experiences delivering ACT for routine treatment of malaria in Rufiji District with the delivery of the nationally recommended monotherapy in three adjacent districts (Figure 8.1). The project was covered under seven distinct research protocols, all of which were independently reviewed and approved by the institutional review boards of the Centers for Disease Control and Prevention, the Ifakara Health Institute, and the Medical Research Coordinating Committee of the Tanzania Commission on Science and Technology. The project’s primary outcomes were related to biomedical markers of antimalarial drug resistance in the parasite’s genetic code. But the more immediately accessible lessons continue to come from other observations. Exploring how to deliver a combination therapy regimen within existing health systems was paramount to the project and would later inform the adoption of another ACT nationwide. Details on how the ACT was delivered, used and perceived by the community, and how this affected household-level decision-making were collected as well. These findings were useful for building a plausibility case, but also generated new knowledge on their own. Because we wanted to assess the intervention in as close to a realworld setting as possible, the project maintained a low profile, at least insofar as delivering the medicines and promoting their use. Data collection activities, however, were often large-scale affairs that were highly visible. Global Factors Health System Factors

Provider and Community

Household and Consumer

Pharmaceutical

Figure 8.2: A Strategic Framework for Considering Interventions to Optimise Antimalarial Combination Treatment in Tanzania (adapted from Marsh, Kachur and Mehra, 2002). 205

Evidence, Ethos and Experiment

Residents of Rufiji and the comparison districts often did not link these large-scale surveys and other studies to the ACT intervention. Balancing between delivering the ACT in a real-world context and the obligations one accepts as a researcher led us to make some very conscious decisions about when and where to alter existing health delivery systems. This paper will address three particular areas where these decisions came into play. The first is in describing existing drug use patterns and alternative health seeking for malaria in order to help researchers and public health authorities develop the intervention approach. This step involved employing a strategic framework based on previous experience with malaria treatment in East Africa and generating local data on each aspect. The second area of concentration was the delivery of the ACT drugs themselves, where we favoured the real world – with some enhancements. And the third had to do with monitoring the safety of the treatment – where we made a substantial effort that would not have been possible without the additional resources that the research project provided.

A Strategic Framework for Delivering Malaria Treatment We considered interventions to optimise the use of ACT in the intervention district based on a strategic framework originally developed to consider possible interventions for malaria home care (Marsh et al. 2002). The framework identified five broad groups of factors that influence optimal drug use and represent opportunities for interventions to improve it: global factors, health system factors, provider and community factors, household and consumer factors and pharmaceutical properties (Figure 8.2). These influences are complexly interrelated and can be highly situational. At the start of the project, it was important to describe all these elements thoroughly – first, to characterise the context in which the observational study was being conducted and, second, as a basis for negotiating and planning the intervention and key portions of the study design. Prior to devising interventions to implement and promote ACT for first line treatment in the intervention district, researchers and programme officials reviewed data from a variety of sources to contribute to a situation analysis based on this strategic framework, including: 1. Ethnomedical and care-seeking studies already completed in the study areas. 206

The Plausibility Design, Quasi-experiments, and Real-world Research

2. Data from Health Information Systems, Demographic Surveillance Systems and National Census. 3. Data from the National Malaria Control Programme on implementation of sulfadoxine-pyrimethamine (SP) as the nationally recommended first-line treatment. 4. Legal and regulatory standards and policy documents including standard treatment guidelines. In addition, between 2000 and 2002, we completed a series of baseline studies to measure biological and other outcomes prior to the national implementation of SP as first-line treatment (September 2001) and the introduction of ACT with SP plus artesunate (SP+AS) in Rufiji District (February 2003). Several of these contributed to the situation analysis. These included: 1. In vivo drug efficacy studies of SP and SP+AS (2000, 2001, 2002). 2. Community-wide surveys of malaria infection, drug resistance markers, and care-seeking behaviour (2000, 2001, 2002). 3. Inventories of outlets where pharmaceutical medicines were available (2000, 2001, 2002). 4. Longitudinal interviews with health officials and other stakeholders at national and district levels (2000 to 2002). 5. Rapid ethnographic assessments of care-seeking behaviour and existing attitudes towards commonly available pharmaceutical medicines (2001, 2003). 6. A health facility survey to establish trends in malaria infection among patients and health worker attitudes and practices related to malaria treatment (2002). 7. Surveillance systems to monitor the occurrence of adverse events associated with antimalarial therapy in all study districts (2002 to present). 8. A review of the drugs management infrastructure in Rufiji District (2002). Each of these elements contributed to the understanding of access and utilisation of antimalarial medicines in the study districts. Table 1 presents some of the more relevant observations that we expected to influence access and utilisation of the ACT even before the interventions were developed or deployed and as a starting point for negotiating the research design with local 207

Evidence, Ethos and Experiment

Table 1: Examples of Factors Likely to Influence Access and Utilisation of Sulfadoxine/Pyrimethamine (SP) Plus Artesunate (AS) in Rufiji District, Presented in the Context of a Strategic Framework for Intervention Specific Factors Identified

Sources of Information

GLOBAL FACTORS •

• • • •

Both SP and AS are registered for use in Tanzania, but no co-packaged product has been registered Multiple local and multinational SP products are available AS is only available at Part I pharmacies in urban areas and at extremely high prices A wide range of pharmaceutical medicines is available throughout the country Drug-resistance tests have shown some foci of SP resistance in other parts of the country and subregion

• • • •

Tanzanian Ministry of Health Essential Drugs Lists Pharmacy Board’s roster of registered products Inventory of outlets stocking pharmaceutical medicines in study districts National Malaria Control Programme’s sentinel surveillance system for antimalarial drug efficacy testing and East African Network for Monitoring Antimalarial Treatment

HEALTH-SYSTEM FACTORS •

•

• • •

•

•

SP is deregulated for over-the-counter use; AS and CQ are restricted to prescription only Malaria treatment guidelines were revised and deployed in 2001 although implementation of SP as first-line treatment was patchy in the first six months Quality of some locally available SP products is poor Adequacy of management systems for drugs supply in Rufiji is weak Supplies of SP through essential drugs kits have been more than adequate throughout Rufiji District since late 2001 A majority of residents of Rufiji District live within 5 km of a formal sector health facility Health systems in Rufiji District have been strengthened, especially for treatment of childhood illness through IMCI. However, health facility utilization rates do not appear to have increased.

• •

• • •

208

Pharmacy Board’s roster of registered products Interviews with policy makers and assessment of health workers attitudes and practices (Goodman et al. 2007) Quality testing of SP products obtained from inventory of drug outlets Rapid Assessment of Drugs Management Systems in Rufiji District Findings from Tanzania Essential Health Interventions Project (Finlay et al. 1995) and IMCI Multicountry Evaluation (Armstrong Schellenberg et al. 2004)

The Plausibility Design, Quasi-experiments, and Real-world Research

COMMUNITY AND PROVIDER FACTORS •

•

•

•

•

•

•

Retail providers for pharmaceutical medicines outnumber formal health facilities in Rufiji District more than 12-fold SP product availability, however, is mostly limited to formal sector health facilities and Part II drug shops Retail providers are more accessible, flexible and convenient than formal sector health facilities Providers’ individual knowledge of recommended treatments and dosages for SP has improved but is not perfect Mass media coverage emphasized risks associated with SP adverse drug reactions rather than its efficacy advantages over CQ Malaria transmission patterns and crosssectional prevalence of malaria appear to be decreasing over time in study districts Fewer than half of patients treated for malaria at health facilities actually had malaria parasitemia at the time

• • • • • •

Inventory of drug outlets (Goodman et al. 2004) Health-worker assessments Media review Malaria transmission studies in study districts (Abdulla et al. 2007) Cross-sectional surveys of malaria parasitemia and care-seeking behaviour Health Facility Survey

HOUSEHOLD AND CONSUMER FACTORS •

•

•

• • •

• • •

Pluralistic beliefs about illness causation and preferred treatment for malaria-like illnesses are common Social norms suggest a preference for traditional remedies when symptoms of cerebral malaria are present Pharmaceutical medicine use is common for fever; 60 per cent of all age groups access fever medicines through retail sector sources Children under five are more likely to utilize formal sector health facilities Drug resistance may cause some patients to reinterpret the cause of illness Community members recognized that CQ resistance was becoming a problem, but few prefer SP drugs Literacy among adults in written Kiswahili is more than 50 per cent Individual knowledge of recommended dosages is poor Health seeking from formal or retail sector sources appears relatively stable among households from all socioeconomic strata

• • • • • •

209

Prior research on care-seeking in study districts (de Savigny et al. 2004) Rapid Ethnographic Assessments in 2001 and 2003 Cross-sectional surveys of parasitemia and care-seeking behaviour Prior research on treatment preferences in the Kilombero Valley (Muela et al. 2002) Data from National Census 2002 Analysis of household economic data from 2001 cross-sectional survey of careseeking behaviour (Njau et al. 2006).

Evidence, Ethos and Experiment

PHARMACEUTICAL FACTORS • •

•

•

•

• •

•

Biological efficacy of SP monotherapy and SP+AS remains high in Rufiji District Molecular markers suggesting the potential for rapid spread of resistance were present in Rufiji as early as 2000 SP and AS differ in their dosing schedules, elimination half-life and modes of action: therefore the two medicines must be coadministered according to differing schedules Many consumers perceive SP to be slow acting or exceptionally risky and would have preferred to continue using CQ Prevalence of confirmed or likely severe adverse drug reactions to SP and related drugs was low in the first 12 months of first-line treatment Consumers express preference for bittertasting medicines to treat malaria Few locally available products feature adequate labelling, locally relevant instructions or images or unit doses Consumers readily appreciate advantages to unit dose packaging and blister packs, although they are relatively unfamiliar with these features

• • • • •

In vivo drug efficacy studies in Rufiji, Ulanga and Morogoro Drug packaging and registration information for SP and AS products Rapid ethnographic assessments in 2001 and 2003 Passive surveillance for adverse drug reactions implemented in study districts Inventory of drug outlets and sales-level surveys in selected outlets in study districts

and national leaders. The research team presented this strategic framework and situation analysis to local and national authorities including the National Malaria Control Program, the Pharmacy Board (now Tanzania Food and Drugs Authority), the Medical Stores Department, and the District Health Officers and Council Health Management teams from the study areas. Even though the project had received ethical and procedural review from a range of Tanzanian and international authorities, negotiating the details required that the researchers actively engage with these relevant political entities at subnational and local level as well.

210

The Plausibility Design, Quasi-experiments, and Real-world Research

Introduction of the New Treatment in an Existing Health System Deciding where, when and how to introduce an ACT on a pilot basis in a setting where careful evaluation would be possible required compromise between the Ministry of Health and the evaluation team. Even so, the study sites and specific ACT to be evaluated were agreed upon relatively easily – both because the local research partner and Ministry of Health had successfully collaborated on the evaluation of insecticide-treated nets and Integrated Management of Childhood Illness in these same districts, and because these sites and the proposed treatment drugs lent themselves to an evaluation design that minimised a number of factors that would have threatened the generalisability of a more randomly selected design. The intervention and comparison areas, for example, were geographically adjacent and formed part of a single ecosystem, but human populations were separated by a large uninhabited game reserve. These features satisfied the researchers’ concerns about controlling for ecological conditions while keeping the intervention and comparison populations relatively uncontaminated. For public health planners, on the other hand, it was key that the ACT be implemented and evaluated at district level, since the district is the most peripheral unit for planning and implementation of health policy in Tanzania. Randomising health facilities or sub-district areas might have resulted in a more robust experimental design for answering some particular research questions, but would not have provided the same realistic basis for lessons about how existing health system structures could deliver ACT. District and national health officials were particularly reluctant to consider randomly assigning health facilities or sub-district areas because they perceived that the random selection would violate their obligation for providing services equitably To routinise the delivery of the new treatment, the District Medical Officer for Rufiji District, and the Manager of the National Malaria Control Programme officially designated ACT as the first-line treatment for uncomplicated malaria in the district. Doing so involved recommending the simultaneous use of two drugs already registered, licensed and approved in a combination that had been shown in RCT-design studies across the continent to be more efficacious and equally safe. In addition, this particular ACT was already authorised and pre-approved by the World Health Organization (WHO). Deploying the combination was consistent with the national malaria treatment policy, but off-label for the local approval of the 211

Evidence, Ethos and Experiment

two individual drugs. The Ministry of Health was interested in gaining experience with a combination therapy approach because it felt pressure from the World Health Organization and the agitation of highly vocal international health advocates (Attaran et al. 2004) who publicly called their interim policy decision ‘medical malpractice’. National malaria control officials also wanted to gain experience with an ACT, because they anticipated they would recommend this type of treatment in the near future (Tanzania officially adopted a fixed combination ACT in 2004 and implemented this change in 2007). However, without the resources provided by the project, it is unlikely that the Ministry would have had the resources necessary for a wide-scale multiyear evaluation comparing ACT with conventional malaria treatment as early as was possible. The project allowed the Ministry to respond to its critics and ACT advocates without undertaking a new nationwide policy change for which it was under-resourced. This decision also appealed to the research team’s interest in working through the existing health system without creating an artificial experimental setting. Negotiating the study design and intervention elements involved continuous engagement between the research team and local and national – level policy makers for nearly two years before the first routine treatment dose of ACT was delivered. National and district-level health officials almost certainly perceived the benefit of obtaining an innovative new treatment at no cost. Through the project, funding was secured up front to guarantee a sufficient quantity of ACT for the study district for a minimum of four years or until a new ACT was available nationwide. Although CDC was relatively unknown in Tanzania at the time of the study, national partners in the Ministry of Health had worked with a number of the investigators. The reputation of Ifakara Health Institute (IHI), the local research partner, was probably more influential. The Institute is well known throughout the country – even among informed lay persons – as a leading source of interventions about malaria prevention and treatment. IHI had conducted the first studies of a candidate malaria vaccine in East Africa, established the effectiveness of insecticide-treated bed nets and evaluated the impact of integrated management of childhood illness. All of these studies were conducted in partnership with national and local health officials in the IMPACT-Tz districts. In addition, each of these projects had prioritised sharing findings and considered their implications by convening national policy makers, local political leaders and community members. As a result, IHI scientists were already well known to health workers and district officials. In addition, the investigators had no commercial interest associ212

The Plausibility Design, Quasi-experiments, and Real-world Research

ated with the ACT being promoted. In fact, the drugs provided in the context of the IMPACT-Tz product were all generic products that could not be patented by any single company and were procured under the authority of the World Health Organization. This, too, may have encouraged national and district health officials that they would not be unduly criticised for collaborating with the project. (Interestingly, when the Ministry finally recommended a proprietary ACT regimen for nationwide use, the decision attracted public and media suspicion that government officials had a financial interest in the decision.) Introducing the ACT through a district-level policy decision with support from the National Malaria Control Programme and the drugs regulatory authority, provided some objective distance between the evaluation team and the national and local authorities responsible for formulating and implementing treatment policy. That the ACT was given as a routine measure also had other important implications for the researchers, providers and consumers. Most notably, it allowed the ACT to be given without individual informed consent. In that respect it appeared just like any other treatment and was not presented as something extra or experimental. Health workers were expected to explain the treatment regimen and to exclude certain patients, but in ways they do for virtually every other medicine they use. Introducing the ACT through existing health systems was more or less successful and resulted in delivering more than 1,500,000 treatment doses to the 200,000 residents of Rufiji District over a period of four consecutive years. The implementation team actively chose to maintain a separate delivery system for the ACT medicines and supporting materials that operated independent of the national system for managing other essential drugs. Quality-assured medicines for ACT were only available through the World Health Organization with whom the project had negotiated subsidised procurement. Project staff then distributed the drugs to the district health supervisors outside the routine national-level drugs procurement, distribution and management systems. If these adaptations had not been made, the health facilities would have faced regular stock-outs, like they did with many other routine medications. In all health facilities, ACT was provided by the existing health workers with training and supervision from the local authorities. The project conducted random audits, as a disincentive for leakage of the medicines, and as a tool to assist supervisors in overseeing the delivery and management of ACT supplies. Project staff also worked with local health planners to implement several modest interventions intended to improve community awareness and demand for the ACT. 213

Evidence, Ethos and Experiment

Developing guidelines for the introduction, use and management of the ACT was also achieved through close collaboration and negotiation between the study team and the official health system. Researchers worked closely with local and national public health authorities to develop tools for health worker training and information, education and communication to consumers. These developmental steps were reviewed and agreed upon by local and national public health authorities before being deployed. Consensus was achieved through a series of workshops where researchers presented findings from baseline evaluation and formative research activities to public health authorities, collectively negotiated the implementation plan and jointly developed supportive materials for health worker training and supervision and to enhance consumers’ likelihood of taking the complete three-day treatment (Figure 8.3). Researchers had initially planned to develop a comprehensive social marketing strategy around a pre-packaged ACT product. In the end, however, the ACT was promoted only through registered health facilities and clearly identified with the generic names of the two component drugs.

Figure 8.3: Participatory Methods Were Used to Conduct Formative Research to Develop Information, Education, and Communication Materials in Support of ACT Implementation in Rufiji District. 214

The Plausibility Design, Quasi-experiments, and Real-world Research

There was early but convincing evidence that the consistent supply of ACT at health facilities coincided with an increase in the proportion of people who sought treatment for potential malaria illness at health facilities, rather than retail outlets (Kachur et al. 2006). This increase in health facility utilisation was more pronounced in the ACT intervention district than in comparison districts where the nationally recommended monotherapy was maintained. Because of the additional inputs into drug supply and management, it will be difficult for researchers to discern to what extent this change in behaviour was related to the perceived advantages of the ACT itself, or to other factors related to the study itself, its interface with the healthcare delivery system, or simply the reliable availability of malaria treatment more generally. To overcome the lack of a pre-packaged ACT product, researchers and district health officials developed colour-coded dispensing envelopes for each of the four recommended age-specific doses (Figure 8.4). In addition, the ACT ultimately expected to be ready for first-line deployment would be packaged similarly, in four unit dose regimens (IMPACT-Tz researchers participated in pilot testing these packages and supporting health infor-

Figure 8.4: Colour-Coded Dispensing Envelopes Were Developed for Deployment of ACT. 215

Evidence, Ethos and Experiment

mation, education and communication materials for two of the three likely candidate ACT regimens). Formative research suggested that the dispensing envelopes could be helpful for improving both health workers’ ability to prescribe the age-appropriate dose and consumers’ ability to complete the recommended three days of treatment. This innovation, which was intended to be temporary, remained in place throughout the four year evaluation period. Ultimately, the dispensing envelopes became one of the few easily discernable features that distinguished the ACT from other pharmaceutical treatments available at health facilities in the intervention district. The dispensing envelopes were printed with the IMPACT-Tz evaluation named on the reverse side, but the role of the Rufiji District Council Health Management Team appeared even more prominently. In followup assessments, researchers found the dispensing envelopes had become the most visible aspect of the project for consumers and that they were regarded as a key feature of the ACT malaria treatment. For instance, popular references to the dosing envelopes or the number of tablets to be swallowed on the initial visit (seven for an adult patient) replaced the clumsy generic names (sulfadoxine/pyrimethamine and artesunate) of the two ACT component medicines in focus-group discussions and interviews conducted with consumers.

Monitoring Safety of the Newly Deployed ACT In general, the provision of medicines for ACT involved some reasonable compromises and some relatively straightforward alterations to the realworld setting. For the most part, these remained largely invisible to consumers. Conversely, dealing with the risk of adverse reactions or inadvertent exposures to ACT required a more deliberate approach where researchers and district health workers undertook a level of activity that would not have been likely outside the context of an internationally supported evaluation. Although the ACT regimen deployed in Tanzania was recommended by the World Health Organization as safe and effective, it had not been widely used elsewhere. Most medicines, when introduced in developing countries, have already undergone extensive monitoring for potential side effects or risks associated with taking them. Many have also accumulated data concerning their safety during pregnancy (Couzin 2005). This was not the case for the ACTs which are being newly introduced on a large scale for the very first time in African countries. 216

The Plausibility Design, Quasi-experiments, and Real-world Research

In a classical RCT design, individually randomised patients are informed of the potential for adverse reactions to experimental drugs and followed closely to identify if any symptoms or conditions occur more predictably in the treatment vs control groups. This process is often overseen by the sponsoring institutional review board, which may convene an independent data safety and monitoring board to review the safety and efficacy data at one or more points before the end of the trial (Bankert and Amdur 2000). Because the ACT procured by the investigators and delivered by the local health authority had already been subjected to RCT studies outside the country and was endorsed by the national health authorities even prior to the start of the project, this step was not necessary as part of the IMPACT-Tz evaluation. Even so, investigators and public health officials were also aware that although the regimen had been approved and recommended, it was seldom available and too costly to have been used widely. As one of the first settings where an ACT became routinely available, IMPACT-Tz researchers and our colleagues in national and district-level authorities for public health felt a responsibility to incorporate safety monitoring into the evaluation. In northern countries, this sort of post-marketing surveillance – monitoring the safety profile of approved treatment regimens – rests with the drugs regulatory authority and is usually supported by the pharmaceutical companies that own the patented products (Steenburg 2006). ACT drugs are natural plant extracts and, as such, no single company owns the technology for their production, and no international pharmaceutical company maintains an infrastructure that would allow post-marketing surveillance in tropical Africa. Furthermore, the drugs regulatory authority in Tanzania operates only a rudimentary system for monitoring drug safety. For these reasons, the research team made a conscious decision, in partnership with national and local authorities, to take on additional responsibility for these tasks. Prior to the project, passive systems existed for collecting information about potential drug exposures during pregnancy and for reporting suspected adverse drug reactions in Tanzania. These systems were poorly functional. Initially, the evaluation team took responsibility for enhancing the reporting systems in the intervention and comparison districts. This improved the situation greatly, so much so that more than eighty per cent of the adverse reactions reported to the Tanzania Food and Drugs Authority in 2004 and 2005 came from the four IMPACT-Tz districts and another five that were supported by the project’s local research partner – indicating that the reporting system in the country’s remaining 112 districts 217

Evidence, Ethos and Experiment

remained essentially non-functional. But still, this enhanced system did not work well enough to confidently identify previously undescribed adverse reactions nor the potential risks that might be associated with taking the ACT during the first few months of pregnancy. Data from Asia suggested that ACTs were far safer than inadequately treated malaria in pregnancy, but the number of women who had received the treatment was too limited to establish conclusive evidence for safety during pregnancy (McGready et al. 2001). This sort of knowledge gap is the case, and most new drugs are licensed with a precaution about using them, particularly in the early months of pregnancy (Couzin 2005). We trained local providers to exclude potentially pregnant women from the ACT, but realised early on that this still resulted in substantial numbers of women who received ACT or standard antimalarial treatment before they recognised they might be pregnant. In the end, we adapted our approach to look actively for these women among a population of 135,000 people who participated in a demographic surveillance system. These persons were visited three times a year and asked to provide information on births, deaths, pregnancies and migrations into or out of the household. We used the system to track ACT and other antimalarial drug exposures among pregnant women, and introduced a mechanism to revisit these women and assess the outcomes of their pregnancies. For those women who had taken the malaria treatments while pregnant, researchers revisited them and their children up to three times during the child’s first twelve months of age. The research team deliberately enhanced the existing system for reporting adverse drug reactions, but went beyond this and established an active screening system independent of the official health sector. However, the evaluation team carefully shared findings related to drug safety with the national drugs regulatory authority and local health officials as well as the pharmaceutical suppliers.

Implications for Validity, Generalisability and Protection of Human Subjects The decision to depart from a more traditional experimental approach in favour of a more situated plausibility evaluation has important implications for sharing our findings and conceiving future public health evaluations. While we endeavoured to keep the ACT intervention grounded in realworld systems, we intervened to ensure the integrity of the evaluation. This 218

The Plausibility Design, Quasi-experiments, and Real-world Research

will require careful explanation when we report our findings. It undoubtedly limits the generalisability and external validity of our findings, but in ways that may be relatively easy to qualify. For instance, one might present as given the fact that weak drugs delivery and stocks tracking systems will have to be enhanced everywhere ACTs are introduced. If that is accepted, our experience is not only generalisable, but also an important source of evidence on how this can be accomplished. The plausibility of such a claim can also be strengthened with evidence from settings where such enhancements have not been made. For example, Zurovac and colleagues evaluated ACT coverage after it was introduced in Zambia without any measures to enhance delivery and tracking. They demonstrated substantial challenges resulting in stockouts, leakage, rationing and poor coverage, even where the new treatment was in stock (Zurovac et al. 2005). Because the evaluation was not based on principles of random assignment, it will also be important for the researchers to present our findings in light of the known factors that could confound conclusions about how the intervention might have affected the observed outcomes. For this reason we also recorded known potential confounders. These included coverage of insecticide-treated mosquito nets, availability and use of antimalarial drugs obtained from sources outside formal health facilities, use of monotherapy relative to ACT, and relative estimates of the malaria transmission intensity in intervention and comparison communities. Presenting these observations alongside the findings of the primary evaluation will be essential. To some extent, it may be possible to employ specific statistical techniques that can quantify or even minimise the impact of known confounding variables. However, it will still be possible that some additional factor which we did not anticipate nor measure may have influenced the evaluation outcomes. For this reason, we as researchers must present all of our findings in the context of an accurate description of the evaluation design and acknowledge the potential limitations of our quasiexperimental approach. As noted earlier, the entire body of peer-reviewed evidence to support the claims that ACTs will reduce malaria transmission or forestall drug resistance, comes from similar observational settings, but without the benefit of any contemporaneous comparison. As a result, the IMPACT-Tz evaluation, despite its limitations and potential for unanticipated and unmeasured bias, will represent a more objective opportunity for examining these claims than has been possible previously. In addition, both the evaluators and the public health authorities in Tanzania agreed at the outset that the benefits of evaluating an ACT intervention under con219

Evidence, Ethos and Experiment

ditions realistically achievable through existing public health delivery systems would produce more immediately useful data than a carefully controlled artificial experiment. This expectation has been at least partially realized, as lessons learned and materials developed for the IMPACT-Tz evaluation helped inform the delivery of ACT in Zanzibar, Burundi and Sudan as well as the ultimate introduction of a related ACT for first-line treatment in mainland Tanzania in 2007. Our decision to forego the real-world approach in favour of actively identifying potential adverse drug reactions and following women inadvertently treated with ACT in the early months of pregnancy was another important departure from traditional biomedical approaches for generating knowledge. Providing a novel treatment, even one that will be life-saving, carries with it a responsibility for assessing the potential for rare adverse events and reproductive health risks. In northern countries, this responsibility falls squarely to the multinational pharmaceutical corporations that manufacture and promote the drugs. Out of commercial interest, these multinationals often go to great lengths to track reproductive outcomes associated with their drug, both to identify potential harmful effects, and to guard against litigation for unsubstantiated claims. The same is essentially true in Tanzania, but there is no means to enforce this requirement and the commercial and legal motivation for drug producers is altogether absent. Moreover, the approaches that the pharmaceutical industry uses to register possible cases in the north would be impractical in African settings. As a result we felt an obligation to collect more systematic data, and had a far more visible presence in the lives of women who received malaria treatment during pregnancy and their families. In some respects, we might have liked to obtain more positivistic evidence about drug safety in pregnancy, but enrolling women intentionally in an experimental design would have been irresponsible. Moreover, international standards for review and approval of experimental research provide an exemption for the evaluation of safety and effectiveness of licensed registered products once they enter widespread utilisation. Moving from experimental designs to more situated observational approaches also has implications for the way scientists identify themselves and their agendas to participants, local officials and community members. Most of the conventions for engaging human subjects in biomedical research are derived from experimental models. There is considerable debate about how these can be adapted to research settings in other cultural contexts (Angell 1988, 1997). It is by no means evident that these conventions consistently improve participants’ informed understanding of the research setting and 220

The Plausibility Design, Quasi-experiments, and Real-world Research

their rights, even in Euro-American settings. The move towards real-world research, and implementation evaluation, also incorporates a subtle, but important, shift in these conventions. For example, in the IMPACT-Tz evaluation, ACT was introduced as a matter of local policy. As a result, the research team was not obliged to obtain individual informed consent each time someone was treated for malaria. Perhaps it is a reasonable compromise, then, that we obtained consent whenever we enrolled participants in one of the data-collection activities that were part of the multidisciplinary evaluation. But the rationale behind when to require consent was not always apparent to the investigators, local public health officials, nor even to our institutional review bodies – much less to the participants. International standards recognise some exceptions to the requirements for individual informed consent, particularly when data collection is intended to monitor ongoing public health programmes or to obtain evidence on the safety and efficacy of approved medicines in wide-scale use. IMPACT-Tz is an example of just such an evaluation – sometimes referred to as a Phase IV trial (Steenburg 2006; Glasser et al. 2007). Even so, we felt an obligation to collect systematic data when women were inadvertently exposed to ACT during their pregnancies. This led some of the research team and health workers to question whether we should do more to prevent that exposure. We operated under the expectation that the known risks of malaria in pregnancy would outweigh the potential risk of reproductive toxicity, based on relatively limited experience with ACTs elsewhere. While we discouraged ACT treatment during pregnancy, we also actively explored the cases in which it was used inadvertently. This balance has arisen as ACTs are becoming more widely deployed for routine treatment in growing numbers of malaria-endemic countries. In nearly every setting, policy makers have reached similar decisions about discouraging ACT use in early pregnancy, but refrained from actively testing for pregnancy prior to treatment. The IMPACT-Tz evaluation has provided a model for openly considering and addressing safety concerns in an ongoing dialogue between research institutions and the public health officials responsible for recommending new treatments.

Conclusion The IMPACT-Tz evaluation experience helps illustrate how contemporary public health questions may demand alternative approaches that can accommodate complex political and social contexts in ways that classical experimental research deliberately avoids. Approaches such as observa221

Evidence, Ethos and Experiment

tional studies require that the researchers actively engage with other policymaking and delivery actors to negotiate issues of study design, intervention implementation and interpretation of findings throughout the lifetime of the research project, and often beyond. In addition, observational researchers and those who interpret their findings, are obliged to consider the contextual factors that have influenced the study and its outcomes, in ways that experimental research often minimizes, or randomises away. Observational research is not new to public health, nor is it incompatible with more privileged RCT designs. In many ways it accommodates the positivist biomedical tradition very well. Yet by encouraging a more deliberately situated set of observations and conclusions, it opens the controlled laboratory research setting to some of the noise of the real world. By trading off some scientific rigor for greater generalisability, observational research studies offer an opportunity for public health scientists and their counterparts to explore issues that are impossible to experimentally control, or where experimental control interferes with the social and political process critical to making an intervention work. In some ways, observational research also leaves the roles of researcher, regulator, policy maker and participant less clearly defined than experimental research. That, too, may present an opportunity as well as a limitation. Engaging with local and national public health officials before and throughout the evaluation period allowed IMPACT-Tz researchers to influence new policy and implementation priorities more directly and more immediately than was previously common. Finally, the conventions that we as researchers have inherited for representing our intentions and ensuring that participants take part in an informed and just manner, may not be well-suited to non-experimental approaches. When and how to obtain individual and/or community consent for participation in observational studies are not as clearly definable as they are for experimental research. Certainly, excluding observational studies from internationally respected guidelines for the ethical conduct of research involving human subjects seems absurd. But it is not without risk of occurring. In particular, new initiatives in global public health including the US-government support for HIV/AIDS and malaria – specifically prohibited or otherwise limited the amount of funds that could be spent on research, while at the same time urging that newly developed treatments and prevention technologies be deployed with minimal delay. Policy makers may feel compelled to roll out new treatments and prevention interventions as soon as they are approved, and constrained from using resources to evaluate as222

The Plausibility Design, Quasi-experiments, and Real-world Research

sociated risk or attributable benefit. The potential for these efforts to sidestep both conventional means (such as post-marketing surveillance for safety) as well as innovative approaches for the protection of research participants is high. This possibility may ultimately be more important than the methodological tradeoffs in internal validity and generalisability.

Acknowledgements I would like to thank the other IMPACT-Tz principal investigators Salim Abdulla, Peter Bloland and all of the project co-investigators and collaborators. An earlier version of this paper was presented at the Conference on Ethnography of Medical Research in East Africa, in Kilifi, Kenya. I am especially grateful to the organisers, Wenzel Geissler and Sassy Molyneux for inviting me to prepare a paper, and to all of the delegates who contributed valuable comments that have been incorporated into this revision, particularly Wenzel Geissler and Ann Kelly. Vicki Marsh and Christian Lengeler contributed to the evolution of the strategic framework described here.

Notes 1. IMPACT-Tz is a multiyear implementation research evaluation project that rests on a collaborative platform comprising the US Centers for Disease Control and Prevention (CDC), Ifakara Health Institute, the National Institute for Medical Research, Muhimbili University College of Health Sciences, the London School of Hygiene and Tropical Medicine (UK), and the Tanzanian Ministry of Health, including its National Malaria Control Programme, the Tanzania Essential Health Interventions Project, and the Council Health Management Teams of Kilombero, Morogoro, Mvomeru, Rufiji and Ulanga Districts. IMPACT-Tz is primarily supported by funding from CDC, the United States Agency for International Development and Wellcome Trust.

References Abdulla, S., K. Adazu, H. Masanja, D. Diallo, A. Hodgson, E. Ilboudo-Sanogo, A. Nhacolo, S. Owusu-Agyei, R. Thompson, T. Smith and F.N. Binka. 2007. ‘Patterns of Age-specific Mortality in Children in Endemic Areas of Sub-Saharan Africa’, American Journal of Tropical Medicine and Hygiene 77: 99–105. Adjuik, M., A. Babiker, P. Garner, P. Olliaro, W. Taylor and N. White. 2004. ‘Artesunate Combinations for Treatment of Malaria: Meta-analysis’, Lancet 363: 9–17.

223

Evidence, Ethos and Experiment

Angell, M. 1988. ‘Ethical Imperialism? Ethics in International Collaborative Clinical Research’, New England Journal of Medicine 319: 1081–83. ———. 1997. ‘The Ethics of Clinical Research in the Third World’, New England Journal of Medicine 337: 847–49. Armstrong Schellenberg, J., J. Bryce, D. de Savigny, T. Lambrechts, C. Mbuya, L. Mgalula and K. Wilczynska. 2004. ‘The Effect of Integrated Management of Childhood Illness on Observed Quality of Care of Under-fives in Rural Tanzania’, Health Policy Plan 19: 1–10. Attaran, A., K.I. Barnes, C. Curtis, U. D’alessandro, C. Fanello, M.R. Galinski, G. Kokwaro, S. Looareesuwan, M. Makanga, T.K. Mutabingwa, A. Talisuna, J.F. Trape and W.M. Watkins. 2004. ‘WHO, the Global Fund, and Medical Malpractice in Malaria Treatment’, Lancet 363: 237–41. Bankert, E. and R. Amdur. 2000. ‘The IRB Is Not a Data and Safety Monitoring Board‘, Institutional Review Board 22: 9–11. Banta, H.D. and European Advisory Committee on Health Research. 2003. ‘Considerations in Defining Evidence for Public Health’, International Journal of Technology Assessment in Health Care 19: 559–72. Barnes, K.I., D.N. Durrheim, F. Little, A. Jackson, U. Mehta, E. Allen, S.S. Dlamini, J. Tsoka, B. Bredenkamp, D.J. Mthembu, N.J. White and B.L. Sharp. 2005. ‘Effect of Artemether-lumefantrine Policy and Improved Vector Control on Malaria Burden in KwaZulu-Natal, South Africa’, PLoS Medicine 2: e330. Benson, K. and A.J. Hartz. 2000. ‘A Comparison of Observational Studies and Randomized, Controlled Trials’, New England Journal of Medicine 342: 1878–86. Berguer, R. 2004. ‘The Evidence Thing’, Annals of Vascular Surgery 18: 265–70. Bhattarai, A., A.S. Ali, S.P. Kachur, A. Martensson, A.K. Abbas, R. Khatib, A.W. AlMafazy, M. Ramsan, G. Rotllant, J.F. Gerstenmaier, F. Molteni, S. Abdulla, S.M. Montgomery, A. Kaneko and A. Bjorkman. 2007. ‘Impact of Artemisinin-based Combination Therapy and Insecticide-treated Nets on Malaria Burden in Zanzibar’, PLoS Medicine 4: e309. Black, N. 1996. ‘Why We Need Observational Studies to Evaluate the Effectiveness of Health Care’, British Medical Journal 312: 1215–18. ———. 2001. ‘Evidence Based Policy: Proceed with Care’, British Medical Journal 323: 275–79. Callaham, M. 1997. ‘Quantifying the Scanty Science of Prehospital Emergency Care’, Annals of Emergency Medicine 30: 785–90. Campbell, M., R. Fitzpatrick, A. Haines, A.L. Kinmonth, P. Sandercock, D. Spiegelhalter and P. Tyrer. 2000. ‘Framework for Design and Evaluation of Complex Interventions to Improve Health’, British Medical Journal 321: 694–96. Charlton, B.G. 1995. ‘Randomized Clinical Trials: The Worst Kind of Epidemiology?’, Nature Medicine 1: 1101–2. Concato, J. 2004. ‘Observational Versus Experimental Studies: What’s the Evidence for a Hierarchy?’ NeuroRx 1: 341–47. Couzin, J. 2005. ‘Drug Safety: Gaps in the Safety Net’, Science 307: 196–98. De Savigny, D., C. Mayombana, E. Mwageni, H. Masanja, A. Minhaj, Y. Mkilindi, C. Mbuya, H. Kasale and G. Reid. 2004. ‘Care-seeking Patterns for Fatal Malaria in Tanzania’, Malar Journal 3: 27.

224

The Plausibility Design, Quasi-experiments, and Real-world Research

Eriksson, C. 2000. ‘Learning and Knowledge-production for Public Health: A Review of Approaches to Evidence-based Public Health’, Scandinavian Journal of Public Health 28: 298–308. Finlay, J.F., M.M. Law, L.J. Gelmon and D. de Savigny. 1995. ‘A New Canadian Health Care Initiative in Tanzania’. CMAJ 153: 1081–85. Glasser, S.P., M. Salas and E. Delzell. 2007. ‘Importance and Challenges of Studying Marketed Drugs: What Is a Phase IV Study? Common Clinical Research Designs, Registries, and Self-reporting Systems’, Journal of Clinical Pharmacology 47: 1074–86. Godwin, M., L. Ruhland, I. Casson, S. Macdonald, D. Delva, R. Birtwhistle, M. Lam and R. Seguin. 2003. ‘Pragmatic Controlled Clinical Trials in Primary Care: The Struggle between External and Internal Validity‘, BMC Medical Research Methodology 3: 1–7. Goodman, C., S.P. Kachur, S. Abdulla, P. Bloland and A. Mills. 2007. ‘Drug Shop Regulation and Malaria Treatment in Tanzania: Why Do Shops Break the Rules, and Does It Matter?’, Health Policy Plan 22: 393–403. Goodman, C., S.P. Kachur, S. Abdulla, E. Mwageni, J. Nyoni, J.A. Schellenberg, A. Mills and P. Bloland. 2004. ‘Retail Supply of Malaria-related Drugs in Rural Tanzania: Risks and Opportunities’, Tropical Medicine and International Health 9: 655–63. Guyatt, G.H., D.L. Sackett and D.J. Cook. 1993. ‘Users’ Guides to the Medical Literature. II. How to Use an Article about Therapy or Prevention. A. Are the Results of the Study Valid?’, Evidence-based Medicine Working Group, JAMA 270: 2598–601. ———. 1994. ‘Users’ Guides to the Medical Literature. II. How to Use an Article about Therapy or Prevention. B. What were the Results and Will They Help Me in Caring for My Patients?’, Evidence-based Medicine Working Group. JAMA 271: 59–63. Habicht, J.P., C.G. Victora and J.P. Vaughan. 1999. ‘Evaluation Designs for Adequacy, Plausibility and Probability of Public Health Programme Performance and Impact’, International Journal of Epidemiology 28, 10–18. Harris, A.D., J.C. McGregor, E.N. Perencevich, J.P. Furuno, J. Zhu, D.E. Peterson and J. Finkelstein. 2006. ‘The Use and Interpretation of Quasi-experimental Studies in Medical informatics’, Journal of the American Medical Informatics Association 13: 16–23. Helms, P.J. 2002. ‘Real World’ Pragmatic Clinical Trials: What Are They and What Do They Tell Us? Pediatric Allergy and Immunology 13: 4–9. Holmberg, L. and M. Baum. 1995. ‘Can Results from Clinical Trials Bbe Generalized?’, Nature Medicine 1: 734–36. Horn, S.D., G. Dejong, D.K. Ryser, P.J. Veazie and J. Teraoka. 2005. ‘Another Look at Observational Studies in Rehabilitation Research: Going Beyond the Holy Grail of the Randomized Controlled Trial’, Archives of Physical Medicine and Rehabilitation 86: S8–S15. Ioannidis, J.P., A.B. Haidich, M. Pappa, N. Pantazis, S.I. Kokori, M.G. Tektonidou, D.G. Contopoulos-Ioannidis and J. Lau. 2001. ‘Comparison of Evidence of Treatment Rffects in Randomized and Nonrandomized Studies’, JAMA 286: 821–30. Kachur, S.P., S. Abdulla, K. Barnes, H. Mshinda, D. Durrheim, A. Kitua and P. Bloland. 2001. ‘Re.: Complex, and large, trials of pragmatic malaria interventions’, Tropical Medicine and International Health 6: 324–25.

225

Evidence, Ethos and Experiment

Kachur, S.P., J. Schulden, C.A. Goodman, H. Kassala, B.F. Elling, R.A. Khatib, L.M. Causer, S. Mkikima, S. Abdulla and P.B. Bloland. 2006. ‘Prevalence of Malaria Parasitemia among Clients Seeking Treatment for Fever or Malaria at Drug Stores in Rural Tanzania 2004’, Tropical Medicine and International Health 11: 441–51. Kraus, J.F., C. Peek, D.L. Mcarthur and A. Williams. 1994. ‘The Effect of the 1992 California Motorcycle Helmet Use Law on Motorcycle Crash Fatalities and Injuries’, JAMA 272: 1506–11. Kraus, J.F., C. Peek and A. Williams. 1995. ‘Compliance with the 1992 California Motorcycle Helmet Use Law’, American Journal of Public Health 85, 96–99. Kroke, A., H. Boeing, K. Rossnagel and S.N. Willich. 2003. ‘History of the Concept of “Levels of Evidence” and Their Current Status in Relation to Primary Prevention through Lifestyle Interventions’, Public Health Nutrition 7: 279–84. Last, J.M. (ed.) 1995. A Dictionary of Epidemiology. New York: Oxford University Press. Lorenz, J.M. and N. Paneth. 2003. ‘When Are Observational Studies Adequate to Assess the Efficacy of Therapeutic Interventions?’, Clinics in Perinatology 30: 269–83. Macintyre, S., I. Chalmers, R. Horton and R. Smith. 2001. ‘Using Evidence to Inform Health Policy: Case Study’, British Medical Journal 322: 222–25. Marsh, V., S.P. Kachur and S. Mehra. 2002. Home-based Management of Malaria. A Policy to Strategy and Implementation Brief. London: Malaria Consortium. McGready, R., T. Cho, N.K. Keo, K.L. Thwai, L. Villegas, S. Looareesuwan, N.J. White and F. Nosten. 2001. ‘Artemisinin Antimalarials in Pregnancy: A Prospective Treatment Study of 539 Episodes of Multidrug-Resistant Plasmodium Falciparum’, Clinical Infectious Diseases 33: 2009–16. Mckee, M., A. Britton, N. Black, K. Mcpherson, C. Sanderson and C. Bain. 1999. ‘Methods in Health Services Research: Interpreting the Evidence: Choosing between Randomised and Non-randomised Studies’, British Medical Journal 319: 312–15. Muela, S.H., J.M. Ribera, A.K. Mushi and M. Tanner. 2002. ‘Medical Syncretism with Reference to Malaria in a Tanzanian Community’, Social Science and Medicine 55: 403–13. Muir, W.W. 2003. ‘Is Evidence-based Medicine Our Only Choice?’, Equine Veterinary Journal 35: 337–38. Njau, J.D., C. Goodman, S.P. Kachur, N. Palmer, R.A. Khatib, S. Abdulla, A. Mills and P. Bloland. 2006. ‘Fever Treatment and Household Wealth: The Challenge Posed for Rolling Out Combination Therapy for Malaria’, Tropical Medicine and International Health 11: 299–313. Nosten, F., M. van Vugt, R. Price, C. Luxemburger, K.L. Thway, A. Brockman, R. Mcgready, F. Ter Kuile, S. Looareesuwan and N.J. White. 2000. ‘Effects of Artesunate-mefloquine Combination on Incidence of Plasmodium Falciparum Malaria and Mefloquine Resistance in Western Thailand: A Prospective Study’, Lancet 356: 297–302. Petrie, D. 1998. ‘RCTs on Trial’, Annals of Emergency Medicine 32: 633–34. Price, R., C. Luxemburger, M. van Vugt, F. Nosten, A. Kham, J. Simpson, S. Looareesuwan, T. Chongsuphajaisiddhi and N.J. White. 1998. ‘Artesunate and Mefloquine in the Treatment of Uncomplicated Multidrug-resistant Hyperparasitaemic Falciparum Malaria’, Transactions of the Royal Society of Tropical Medicine and Hygiene 92: 207–11.

226

The Plausibility Design, Quasi-experiments, and Real-world Research

Prideaux, D. 2002. ‘Researching the Outcomes of Educational Interventions: A Matter of Design. RTCs Have Important Limitations in Evaluating Educational Interventions’, British Medical Journal 324: 126–27. Radford, M.J. and J.M. Foody. 2001. ‘How Do Observational Studies Expand the Evidence Base for Therapy?’, JAMA 286: 1228–30. Ray, J.G. 2002. ‘Evidence in Upheaval: Incorporating Observational Data into Clinical Practice’, Archives of Internal Medicine 162: 249–54. Robson, C. 2002. Real World Research, 2nd ed. Malden: Blackwell. Rossouw, J.E., G.L. Anderson, R.L. Prentice, A.Z. Lacroix, C. Kooperberg, M.L. Stefanick, R.D. Jackson, S.A. Beresford, B.V. Howard, K.C. Johnson, J.M. Kotchen and J. Ockene. 2002. ‘Risks and Benefits of Estrogen Plus Progestin in Healthy Postmenopausal Women: Principal Results from the Women’s Health Initiative Randomized Controlled Trial’, JAMA 288: 321–33. Sackett, D.L., W.M. Rosenberg, J.A. Gray, R.B. Haynes and W.S. Richardson. 1996. ‘Evidence Based Medicine: What It Is and What It Isn’t’, British Medical Journal 312: 71–72. Sandercock, P. 1998. ‘What Questions Can Large, Simple Trials Answer?’, Cerebrovascular Diseases 8: 30–36. Schwartz, D. and J. Lellouch. 1967. ‘Explanatory and Pragmatic Attitudes in Therapeutical Trials’, Journal of Chronic Diseases 20: 637–48. Smith, G.C. and J.P. Pell. 2003. ‘Parachute Use to Prevent Death and Major Trauma Related to Gravitational Challenge: Systematic Review of Randomised Controlled Trials’, British Medical Journal 327: 1459–61. Steenburg, C. 2006. ‘The Food and Drug Administration’s Use of Postmarketing (Phase IV) Study Requirements: Exception to the Rule?’, Food and Drug Law Journal 61: 295–383. Vallvé, C. 2003. ‘A Critical Reveiw of the Pragmatic Clinical Trial’, Medicina Clinica (Barc) 121: 384–88. Vandenbroucke, J.P. 2004. ‘When Are Observational Studies as Credible as Randomised Trials?’, Lancet 363: 1728–31. Verhoef, M.J., A.L. Casebeer and R.J. Hilsden. 2002. ‘Assessing Efficacy of Complementary Medicine: Adding Qualitative Research Methods to the “Gold Standard”’, Journal of Alternative and Complementary Medicine 8: 275–81. Warhurst, D.C. and M.T. Duraisingh. 2001. ‘Rational Use of Drugs against Plasmodium Falciparum’, Transactions of the Royal Society of Tropical Medicine and Hygiene 95: 345–46. Whitty, C.J., R. Allan, V. Wiseman, S. Ochola, M.V. Nakyanzi-Mugisha, B. Vonhm, M. Mwita, C. Miaka, A. Oloo, Z. Premji, C. Burgess and T.K. Mutabingwa. 2004. ‘Averting a Malaria Disaster in Africa – Where Does the Buck Stop?’, Bulletin of the World Health Organization 82: 381–84. Yates, D. 1997. ‘Randomized Control Trials and Evidence-based Medicine – What’s in a Name?’ European Journal of Emergency Medicine 4: 123–24. Zurovac, D., M. Ndhlovu, A.K. Rowe, D.H. Hamer, D.M. Thea and R.W. Snow. 2005. ‘Treatment of Paediatric Malaria During a Period of Drug Transition to Artemether-lumefantrine in Zambia: Cross Sectional Study’, British Medical Journal 331: 734. 227

Chapter 9

Remember Bambali: Evidence, Ethics and the Co-production of Truth Ann Kelly

And yes, I was happy to participate and again, yes. MRC has been very open with us and they show us how it was done. They come to our village, they stay for a week until one am in discussions and we appreciate this. Establishing a rapport has changed things very much. You coming here is very important, because things must be done on village level. It is the moral way of doing things. I only hope MRC continues with this relationship, they have been building between us. You remember Bambali, yes? (Alaklo of Bambali, 16 February 2005)1 ‘And remember this village was the most difficult to recruit’, Sheriff says with a grin. Arriving in Bambali, children surround the Medical Research Council (MRC) Landrover, making it almost impossible to open the doors. ‘Dr Vasee thought we would have to pull out entirely – and Bambali’s a big village. We would have had to take on at least two or three more villages to the west to get the numbers needed.’ Sheriff Jobe was the senior field worker on the Malaria Vaccine Trial (MVT) and cohort studies conducted from 2002 to 2005.2 The MVT was the most intensive investigation to be undertaken by the British Medical Research Council (MRC) at their unit in Farafenni, Central Gambia, involving constant contact and surveillance. Fluent in Woloff, Mandika and Fula, Sheriff was responsible for recruitment of the volunteers and the day-to-day management of the ‘field’ aspects of the research. He is recognized by villagers wherever we go. As we walked towards the bantaba, the Alkalo3 and a group of men whom Sheriff refers to as ‘the prominent opinion leaders’ rise to welcome us. This was the village where, three years ago, Sheriff spent every evening for a fortnight, sometimes until one or two o’clock in the morning, explaining the purpose and the procedures of the trial, and trying to convince 229

Evidence, Ethos and Experiment

the sceptics that the MRC had the community’s interests in mind. The Alkalo invites us to sit and drink freshly brewed Attaya,4 but Sheriff is eager to show me the rest of the village. So, accompanied by a few of these men, we stroll through Bambali, stop to greet people and occasionally sit with them in their houses, eat groundnuts and get out of the sun. As he asks after the family and apologizes for not having visited in such a long time, it strikes me that Sheriff seems to know everyone in the village personally. One stop elicits an invitation to return the following day for a proper meal. Sheriff immediately agrees. Though he has worked for the MRC for almost twenty years, most of which he has spent at the coast, Sheriff tells me that he ‘came from a village not unlike this one’. He suggests that in addition to a meal with his friend, we should really come back for an evening, to see ‘Bambali in the moonlight’. Where the houses grow farther apart, we follow two men lumbering under the weight of large stones. The most remote of the trial villages, Bambali is only about 400 metres away from the river Gambia. As the saying goes, ‘from Bambali to the river, from the river to the sea’ (or as an entomologist working at the Farafenni station rephrased this geographic isolation: ‘from Bambali to the swamp, from the swamp, to the mosquitoes’). As we approach, we see a group of men rebuilding the bridge that had been washed out this past September, during a particularly violent rainy season. Amongst these men, the vast majority of whom are volunteers, Sheriff is at his most gregarious. He calls out to one young man in a fluorescent mesh shirt and, after embracing, introduces us in English: ‘and here, is number 57! So how are you Mr 57? Meet Anna. Mr 57 is a lazy man’. They both laugh and mock wrestle each other. After this interchange, Sheriff greets all the men this way; recalling how this one was difficult to recruit, how another had continually missed the blood sampling, and that this one preferred having his finger prick taken from the thumb on his left hand. For everyone, Sheriff recalled a number: the badge that identified him as a past participant in the trial and had anonymized his blood samples to the research team. This moment of recognition will provide the starting point for an inquiry into the social relationships formed within and around a clinical trial in the context of developing world research. Over the course of the MVT study, through conversations with trial-volunteers, non-participants and field staff, Wenzel Geissler, Robert Pool and I investigated the processes of mediation and material exchange between populations and scientific protocol. It was clear from those discussions that the success of the MVT trial depended upon the ability of fieldworkers and researchers to deviate from 230

Remember Bambali

the static rules of research practice and negotiate ethics ‘on the ground’ (Geissler et al. 2008). Unlike trials that the MRC had conducted earlier in the area, the MVT trial placed a nurse and a fieldworker in each village, providing support and free healthcare to the entire community for the duration of the study. This inclusive and resolutely pragmatic approach recast the participant in terms of an ongoing ethical relation between a research institution and the community as a whole. Our contention was that during the MVT trial, the practice of research, which transforms a person into a sample, was itself transformed from a clinical experiment to a ‘trial community’. The designation ‘57’ carries with it more than the man’s place within the experimental aggregate, but rather a marker of shared memory. The striking effect of hearing a number deployed as a form of address, points to the ill-fit between the epistemological framework of scientific experimentation in the developing world and the particular kinds of blood relations generated within it. This paper hopes to develop this observation by following what happens to Mr 57 after the trial ends. Where experiment and everyday life are inextricably linked, the formal precepts of ethics protocols fall short. But beyond a problem for clinical research ethics, I would like to suggest that the very concept of the ‘end of a trial’ invokes an epistemological position at odds with the ways of knowing that characterized the continuous engagement between the trial’s fieldworkers and its villagers. The goal here is to rethink the social relations developed through the practice of clinical research in Africa in terms of the scientific integrity – rather than the ethical padding – they provide to experimental practice. To some extent, this position is common sense: the equitable treatment of trial participants describes both the quality of their consent and the knowledge generated through their participation. The problem is that securing these qualities in an African research context are understood as distinct problems. My argument is that regarding Sheriff ’s memory of Bambali as an aspect of ‘good community relations’ short-circuits how we imagine a more equitable research. Instead, I would like to consider greeting ‘Mr 57’ as an act of the same order as collecting of his blood. Holding these activities alongside each other, I believe, has considerable consequences for the social benefit that scientific activity in this setting can generate. But first, let us return to the historical record. A brief reminder of how the epistemic and ethical commitments of science took shape will reveal the ways in which social value and scientific integrity of clinical research practice are mutually constitutive. 231

Evidence, Ethos and Experiment

Blind Benefit In 1946, Anthony Bradford Hill, a professor of epidemiology and medical statistics at the London School of Hygiene and Tropical Medicine, randomly allocated in sealed envelopes the treatments of either streptomycin and bedrest or bed-rest alone to patients suffering from Tuberculosis at six different centres across the UK.5 In each case, two radiologists and a clinician – unaware of the treatment the patient had undergone – looked at a patient’s x-ray films and discovered that in those cases in which streptomycin had been administered along with bed-rest, the patient’s condition improved dramatically. What set Bradford Hill’s experiment apart from the standard methods of clinical evaluation of the time was its reliance on statistics. The careful planning, execution, and reporting that characterized the streptomycin trial followed precisely the recommendations outlined by Ronald Aylmer Fisher, in his book, The Design of Experiments (1935). Fisher’s interest was on the effects of fertilizer on grain production. Suppose, Fisher argued, an experimenter wished to know which of two varieties of fertilizer was better. According to the conventions of experimentation, one would select a piece of land and subdivide it into narrow strips. He would then organize the strips into two groups – one treated with fertilizer A, the other with fertilizer B – ensuring that each group had an equal exposure to sunlight, access to water, or soil fertility. The reliability of the experiment depended, first, upon the experimenter’s awareness of factors that might influence his results and, second, on his ability to fairly distribute them between the two groups. But what if there had been some factor that the experimenter had overlooked? It is possible, for instance, that the size of grain was not uniform. How, then, Fisher asked, was one to know whether or not a 10 per cent difference in yield between the two groups reflected a true difference in the fertilizers or merely marked an incidental difference in the plots? Isolating and expunging all confounding factors so as to make the plots identical was an impractical strategy; some unforeseeable specificity was almost certain to escape the experimenter’s notice. Fisher’s innovation was to build the capriciousness of nature into his experimental design. Using a coin to decide which plots would receive treatment would help ensure that differences were distributed equally through the sample and thus release the experimenter from the burden of creating comparable groups.6 In so doing, randomization creates a framework by which the experimentally defined relationship between variables can be interpreted and their contingent (or confounding) interrelationships can be dissolved (Fisher 1966: 2, 670–77).7 232

Remember Bambali

In the context of research on human subjects, whereby radical variability exists across an experimental sample, random assignment offers considerable advantages. Contemporaries of Bradford Hill had acknowledged that single-patient case studies were insufficient to assess the merits of new therapies. One approach offered was to administer treatment to a large group of patients and then compare their survival rates to a group of patients of equivalent size that had not been treated. However, Bradford Hill believed that controlling for unknown factors only solved part of the problem. An investigator’s preference, he argued, could also influence therapeutic outcome; his inclination to give the sickest patients treatment rather than placebos drastically undermines an experiment’s results. Here, Fisher’s technique of assignment proved invaluable. Treating participating patients based on a random allocation schedule ensured ‘that neither our personal idiosyncrasies, consciously or unconsciously applied, nor our lack of judgment have entered into the construction of the two (or more) treatment groups and thus biased them in anyway’ (Hill 1953: 132). The impact of Fisher’s ideas on statistical thought and clinical practice has received a great deal of attention.8 Here, I would like to emphasize the political significance of randomized design. It is important to remember that Fisher was a contemporary of Lysenko, an infamous and politically connected Lamarkian scientist, under whose direction leading Soviet geneticists were exiled or sent to labour camps. Understandably, then, Fisher, along with many of his generational cohort, spoke out against putting science at the service of the state. I have suggested that Fisher advanced randomization as a framework of interpretation. Its ‘rigorously specified uncertainty’ relieved the experimenter ‘from the anxiety of considering and estimating the magnitude of the innumerable causes by which his data may be disturbed’ (Fisher 1966 [1935]: 68). But a chance arrangement not only released the experimenter from the burden of creating comparable groups; randomization also liberated the outcomes of the experiment from authoritative judgement. According to Fisher, ‘any thinking man’ could tell whether or not fertilizer A was better than fertilizer B if he was presented with consistent and generalizable findings (Marks 2003: 934). The value of randomization, then, was that it preserved the difference between drawing inferences and making decisions based on those inferences. Statistical analysis, Fisher insisted, had nothing to do with utility: We aim, in fact, at methods of inference, which should be equally convincing to all rational minds, irrespective of any intentions they may 233

Evidence, Ethos and Experiment

have in utilizing the knowledge inferred … There is something to be gained by at least being able to think of our scientific problems in a language distinct from that of technological efficiency. (1966 [1935]: 55) In short, according to Fisher, the autonomy of research not only ensures the accuracy of its outcome, but also the liberal orientation of the information it produces. By immunizing the experiment from the influence of politics, the randomized design places research at the service of society writ large. Though Bradford Hill’s use of randomization relied more heavily on common sense than on statistical theory, his rationale for the methods echoed Fisher’s efforts to shift the burden of proof from the experience of the doctor to the rigour of the experimental design: “Once it has been decided that a patient is of the right type to be brought into the trial, the random method of allocation removes all responsibility from the clinical observer.” (Bradford Hill 1966: 130) However, in the context of the clinic, the effort to suspend expert bias carries more complicated moral consequences. Because doctors are believed to be acting for the patients’ benefit, the idea that treatments should be decided by the flip of a coin was – and still is – a rather difficult notion to accept. Although he saw the potential application of Fisher’s design to clinical research about a decade earlier, it was not until 1946 that Bradford Hill found an opportunity to introduce randomization to the clinic. Tests conducted in America in vitro and on animals strongly suggested the efficacy of streptomycin in humans. However, in the wake of the Second World War, the British Government did not have the means to finance a widescale purchase of the product. Randomization, then, both guaranteed experimental accuracy and ensured the fair allocation of medicine that was in short supply. Under these circumstances, Bradford Hill argued that ‘it would have been unethical not to have seized the opportunity to design a strictly controlled trial which could speedily and effectively reveal the value of the treatment’ (Bradford Hill in Mainland 1954: 157). Thus, carried into medical research, Fisher’s randomized design not only generates independent facts to serve the interests of the greater good, but also anonymizes individual patients to ensure the fair allocation of scarce resources. Bradford Hill’s Streptomycin trial marked a transformation in the requirements of medical evidence. Grounded in statistical concepts, clinical experiments re-conceptualized the difference between what is known and what is unknown. But 1946 was a landmark year in the history of UK clinical research 234

Remember Bambali

in other respects: the introduction of the National Health Service Act created an entirely new environment within which research took place.9 The Act brought clinical facilities together under a free, population-wide, comprehensive healthcare system. Though the history of the NHS is clearly beyond the scope of this paper, what I would like to suggest here is that the dominance of the RCT as the gold standard of evidence is closely linked to ideas about the redistributive ethics of the welfare state. Under a unified system supported by limited public funds, intervention must be proven to be a justifiable expense; questions of effectiveness, in other words, involve ones of efficiency.10 Designed to respond to and accommodate clinical realties, medical experiments involve a methodological balancing act between the requirements of rigor and relevance (Will 2007). As an instrument of both empirical truth and institutional accountability, the RCT situates scientific evidence within the structures of policy. To summarize: the RCT is founded upon diverging and perhaps even conflicting ideas about what constitutes ‘good evidence’. On the one hand, set against the backdrop of communist science, the systematic use of randomization disentangles evidence from the influence of politics and the state. The value of Fisher’s design inheres in the liberal orientation of the information it produces. On the other hand, oriented to the demands of the national healthcare system, the RCT scales scientific truth to the demands of a particular society. This tension between bias and interest, knowledge and value, universality and context-specificity undergirds the second stipulation of the 1947 Nuremburg code of ethics: “The experiment should be such as to yield fruitful results for the good of society, unprocurable by other methods or means of study, and not random or unnecessary in nature.”11 Framed ethically, ‘fruitful’ research refers first to its internal coherence: well-designed trials justify subjecting a patient to experimental conditions because meaningful knowledge will be gained through the process. Second, ‘fruitful’ results speak to the utility of research or, rather, its necessity – qualities which can only be judged by the socio-political context in which research takes place. In short, ethical debates and discussions over trial design come to the same point: generating social benefit means squaring not entirely collateral ideas about the value of scientific research. In the context of research in The Gambia, the convoluted relationship between good research and the goods it produces gains considerable amplitude. Where public infrastructure is absent, the expected benefits that accrue from research are immediate, tangible and inevitably short-term. The social value of science, therefore, gains a temporal dimension distinct from one associated either 235

Evidence, Ethos and Experiment

with a knowledge-economy (in the form of research papers, project grants, industry collaborations), or with the possibility of novel interventions to improve health. It is at this juncture between research practice and the boundaries of the experiment that we find Mr 57.

A Social Contract The Malaria Vaccine Trials (MVT) and cohort studies were initially met with scepticism. By and large, villagers refused to participate; many argued that the blood-taking regimen demanded by the investigators12 was too strenuous for those with such limited financial and physical resources (Geissler et al. 2008.13 As so many of the chapters in this volume demonstrate, resistance to medical research is growing across Africa: in recent years, scientists have been accused of Satanism, remote blood sucking, and the theft and subsequent resale of African blood (Geissler 2005). The participants approached by the MVT researchers were well-acquainted with western research practice. The first Medical Research Council (MRC) unit was established by the UK government in The Gambia in the 1950s and several of the young men involved in this study had been born into earlier MRC cohorts. Their uneasiness with the MVT trial was the product of a longstanding distrust of the MRC’s practices and weariness with medical research that has offered little in the way of tangible benefits. In addition to the week of lengthy discussions Sheriff held with village leaders, to allay villagers’ suspicions about the research, the MVT research team extended their promise of free healthcare for volunteers to the entire community. With nurses and fieldworkers stationed in the villages for the duration of the trial, almost none of the participants dropped out or refused treatment. In the end, the vaccine proved ineffective. But the trials themselves, conducted according to high ethical and medical standards, were deemed a great success. Indeed, during discussions held at the trial’s conclusion, villagers reflected on the experience of research as being an overwhelmingly positive one: MRC’s intervention gave us a stable service of healthcare, which is a remarkable improvement for our economy. (Focus-group Discussions, Palen Fulla, 20 February 2005)

236

Remember Bambali

Joining the MRC trials was to give us good health and to help us with healthcare, because here is very far away from the health facilities, especially in Farafenni. (FGD, Bombali, 18 February 2005) On average, the study villages are 20–40 km away from the closest hospital. Thus, the presence of a nurse and the availability of emergency transport profoundly altered the villagers’ access to healthcare. Whilst the trial’s efficient management had impressed researchers, it was its medical efficacy that participants praised. Regardless of the vaccine’s statistical ineffectiveness, according to the villagers and the nurses who lived with them, the treatment was a complete success. Participants, who usually contracted malaria five or six times during a year, swore that they had been malariafree since the vaccination. Now that the trial had ended, participants argued that they were, once again, falling sick. As one participant commented: I am very happy with MRC. MRC has been very good for the life of this village. In fact, we are all crying that MRC has gone … continued care would be a great help. Healthcare is always a problem. (Interview, Yallal Ba, 31 January 2005) As Leach and Fairhead argue in their aptly named chapter, ‘Being with the MRC’, volunteers experience MRC research projects as health-service institutions rather than as scientific interventions (Leach and Fairhead, this volume). Indeed, when asked, ‘Why did you participate in the MVT?’ respondents unanimously cited the trial’s immediate and tangible benefits: Joining the MRC trials was to give us good health and to help us with healthcare, because here is very far away from the health facilities especially in Farafenni … A trial team came here and told us about a vaccine on malaria. Because malaria was a major cause of illness that effects our communities and it even kills, they have a vaccine that they are making a trial to prevent this common illness, malaria in our society … it is good, research is very good because the healthcare benefits we gain and also improve the health status of the community. (FGD, Dibba Kunda Woolof, 29 Jaunuary 2005, my emphasis) For these participants, malaria is both an individual and a communal problem. It is a disease that threatens the village, the community and the greater Gambia; it is confronted on a daily basis and situated within general wellbeing. Regardless of the vaccine’s efficacy, the experience of malaria had been significantly 237

Evidence, Ethos and Experiment

relieved by the presence of the trial in the volunteers’ villages. For the staff and the villagers, the MVT – as a project and an intervention into the community – had worked; they were bewildered that the scientists could not see this. However, the trial’s closure posed serious problems for the trial villages: as one participant explained, ‘I woke up sick last week and now there is no one here to give me treatment’. But beyond their disappointment in the discontinuation of care, for many of the villagers, the termination of the MVT suggested that the MRC’s relationship to the village had been strictly instrumental. In the words of the Alkalo (political leader) of Pallen Fulla: Yes, we would be happy to participate in another study, but fundamentally, we need to be given a good incentive to continue in future programmes. This means creating a good relationship with the community and the programmes, not stopping services anytime they want. If it continues in this way, in the long run, when they need the community they will not be welcomed. (Interviews, Pallen Fulla, 12 February 2005, my emphasis) The Alkalo expresses an important issue about the place the MVT had acquired within the community. During the trial, the villagers, whether or not they had volunteered, welcomed the MVT researchers into their community and supported those who chose to participate. Because the village had cooperated, the MRC had benefited. The MRC, in turn, had helped Pallen Fulla by providing access to healthcare. This mutual dependence constitutes a ‘good relationship’; as long as the trial continues, the grounds of benefit and interest are equal and expectations are met. But the moment services are withdrawn, a power dynamic re-enters the equation, and the foundations of rapport are disturbed. By ‘stopping services any time they want’ the MRC had failed to live up to its responsibility to protect the welfare of the community. Gambian participants justified their involvement in terms of the everyday clinical care that they and their families received. The trials’ closure effectively undid the social relationships and health security that made research acceptable in the first place. The critical point is that the ethics of clinical research is experienced in terms of a working relationship between the institution and the community. What became clear by the conclusion of the trial was that the scientific knowledge generated through that very particular blood relation did not meet the participants’ expectations of social benefit. For the MRC, the post-trial situation in Bambali is hardly exceptional. As Molyneux and co-workers (2005) observe, the confusion between treat238

Remember Bambali

ment and research is endemic to clinical trials conducted in the developing world. Though a clinic might be good for the village, such an undertaking is clearly the responsibility of the government and not within the remit of a research facility.14 The participants are no more naïve to the force of this logic than the researchers: The commitment with MRC is on contract. But though we have the interest and the willingness to participate with MRC, let us pray that there will be a continual extension of this contract. (Interview, Imam (religious leader), Sara Kunda, 18 January 2005) As the MRC is not going anywhere, it is highly likely that ‘the contract’ to which the Alkalo refers will be extended by other research teams in the future. But other than hoping for an uninterrupted flow of clinical trials, might there be a way to imagine a good clinical research that would be compatible with a Gambian notion of public good?

Still Part of That Outcome Consider, for the sake of argument, what might have happened if the vaccine had been successful. Participants, community members, The Gambia and the world would have received immediate benefit.xv It is this hypothetical promise of fruitful results that permits the transformation of human beings into research subjects. As it happened, the trial’s results were negative. The villagers’ status as participants evaporated: no vaccine, no nurse and no transport to the hospital. For the MRC and for the scientists working under its institutional auspices, this was not the case. Though admittedly not always granted the same degree of attention, negative and positive results are equally valid forms of information (e.g. Horton 1999; Tallon, Chard & Dieppe 2000; at the trial’s conclusion the MVT results were transformed into published papers, scientific leads and future grants. This discrepancy between the benefits experienced by researchers and participants was eloquently articulated by one of the volunteers: I have not been to school but from what I understand MRC does research in laboratories and needs to make outcomes. We are part of that outcome; it is because of us it is made. Then they take this outcome and share it with others. They transfer it around, so even if the trial is over, we are still part of it, and they are still benefiting. We 239

Evidence, Ethos and Experiment

should also [continue to benefit]. We need a clinic built here. One that stays. So then we would keep benefiting. (FGD, Bambali, 26 January 2005) The transference of MVT outcomes – the transformation of capital into work and from information into knowledge – casts the bifocal character of clinical research in sharp relief. As an anonymised constituent of the experimental sample, the individual embodies the wellbeing of the population at large. But by participating in a trial as a patient-citizen, he represents a public that exercises jurisdiction over healthcare interventions. Social benefit is, therefore, the product of aggregation, because for a knowledge-based society, unbiased information is an unmediated good. It is also a public investment: the value of research arises through the collective endeavour of its participants. In the case of the MVT trial, the flow between outcome and benefit moves only in one direction despite being ‘part of that outcome’; in this relation, the participants do not belong to the same socio-political universe as the people who will directly benefit from the increase in knowledge. Their connection to MVT’s scientific activity terminates in its experimental practice. (Strathern 1996) I would suggest that the model of kinship generated between the MVT staff and its participants suggests a way that the length of the link might be extended. On the day of the trial’s formal closure, Edrissa, a nurse who had been stationed in the trial village, Bambali, described his experience. He had spent almost three years living in Bambali and had grown very close to the volunteers and their families. The conclusion of the trial had left him feeling that he had let them down: Almost three years we have been living with them and now it is stopping. I don’t know how they are going to tell them but it is going to be a big problem. For people like us we are in the same country, we are the same community, we will meet each other. We will see each other at festivals and that relationship will still be there … We are brothers, and so when I see them, and I will feel badly that I had to leave them. It is not good how it is ending. We are giving them the vaccine and then we are leaving. It is not right. (Staff Interviews, 26 January 2005) The relationship he had built with the villagers, living with them, sharing meals and caring for the sick, had come with responsibilities, which Edrissa believed could not be abandoned so quickly. Edrissa had not just been a 240

Remember Bambali

nurse but ‘their nurse’, just as Bambali had become ‘his village’. The MVT had exploited both Edrissa’s technical skills and Gambian heritage, deploying his relationship to the people of Bambali in a particular way. As a nurse, Edrissa’s responsibility towards these people might end with the trial but, as a brother, his commitment to the villagers endured. Edrissa’s familial bond with Bambali is coextensive with both the continual project of wellbeing and with the demands of research. For without his attachment to the villagers, the MVT team would not have been able to recruit or to maintain the number of participants needed for the trial. Quite literally, the MVT was a coordinated project whose success depended not only on the techniques of statistical analysis but also on the fieldworker’s memory of which finger a participant wanted to have pricked.

Conclusion That good community relations are essential to medical research in Africa is hardly a radical claim. However, for those involved in that practice, these relations are not only enabling research to be conducted but are, ultimately, the greatest source of benefit. To take seriously the twinned aim of experimental medicine – to produce knowledge in the interests of social wellbeing – requires consideration of what constitutes benefit for the people under study. As the MVT makes clear, the notion of public good espoused by the MVT participants and fieldworkers describes the production of vital and prospective connections between the researchers and the researched; ’it is good, research is very good because the healthcare benefits we gain and also improve the health status of the community’ (see above). What if we were to understand the intimate practices of knowing that bind the fieldworkers and the villagers as being on a par with those connecting the scientists to the number of parasites found in a blood sample? Framed as part of the experimental work, the recognition of Mr 57 becomes a critical component of a ‘fruitful’ outcome, rather than merely a bi-product of its experimental practice. Contrary to the ethical parameters built around a sample, Edrissa’s and Sherriff ’s memory of Bambali and its volunteers provides an extended framework within which to continue rapport, as the Alkalo suggests, ‘in the long run’. As Susan Reynolds Whyte points out in her account of dealing with misfortune in Uganda, the realistic appreciation of uncertainty and the hope for a better future provide a basis for the improvement of life (Whyte 1997: 232). So too with clinical research in The Gambia: the scientific integrity and social value of ideas 241

Evidence, Ethos and Experiment

resides in the memory they preserve and the futures they suggest. The challenge is how to build those futures into the protocols of international scientific research and hold them stable so that they might serve as an engine for social progress.

Acknowledgements This paper draws its insights from a collaborative research project conducted by Wenzel Geissler, Babtunde Imoukheuede and Robert Pool. Further fieldwork was conducted with the generous support of a Wellcome Trust Bioethics Grant (#2173). Sincerest thanks to Wenzel for his engaging commentary on the text and for continually providing inspiration for my work. This research would not have possible without the reflections of the trial participants and MRC fieldworkers. I am grateful to them for their openness, generosity and care. In particular, Sherrif Jobe’s guidance has been invaluable.

Notes 1. Quotations in this paper are taken from a series of interviews and focus-group discussions conducted a month after the Malaria Vaccine Trial’s (MVT) activities were brought to a close. Focus-group discussions, lasting roughly an hour each, were held in five villages of varying size, distance from Farafenni fieldstation and ethnic composition: Bambali, Sara Kunda, Dibba Kunda Wollof, Yallal Ba, and Pallen Fulla. Focus-group discussions were open to all men in the village and included those who volunteered, those excluded from the study, and those who had declined participation. Interviews were conducted with village political and religious leaders and with outspoken critics to medical research. Interviews focused on their experiences of the MVT and their options of the Medical Research Council’s (MRC) work in general. 2. Recruitment began in July 2002 and the first disease events were not until October. The trial was terminated in December when interim results proved the vaccine ineffective. An intensive 28-day follow-up cohort study concluded the MVT investigations in early January 2005. 3. The Alkalo is the political head of the village. This was, traditionally, a hereditary position but now is often elected or appointed by the government. The bantaba, a wood platform with a thatch roof, is the political and communal hub of the village. 4. A cheap Chinese powdered green tea, Attaya is drunk by Gambians throughout the day and offered as a gesture of hospitality. 5. Alder Hey, Great Ormond Street, Hammersmith, Glasgow, and Highgate. 242

Remember Bambali

6. That randomization renders groups comparable is a misunderstanding propagated by Bradford Hill, who, as Marks argues, emphasised the reliance of randomization at the expense of Fisher’s more subtle ideas about a framework of uncertainty (see Marks 1997: 147). 7. For a more thorough discussion of Fisher’s statistical rationale in setting up these experiments see Marks (1997: 140–45) and Senn (2003: 52). 8. E.g. Hotelling (1951: 35–46); Yates (1951: 19–34); Savage (1976: 441–500); Gigerenzer et al. (1989: 117–20). 9. Aneurin Bevan, the Secretary of State for Health under Clement Attlee’s Labour government, introduced the Act in 1946. 10. Archie Cochrane, regarded as the godfather of Evidence-based Medicine (EBM), emphasises the located – or ‘ready-made’ nature of medical research. Commissioned by the Nuffield Provincial Hospitals in 1971, Cochrane’s monograph, Effectiveness and Efficiency, opens with a comprehensive and – quite damning – statistical survey comparing the number of medical personnel and the percentage of GNP spent on the NHS to the overall mortality and morbidity rates in the UK. For Cochrane, the reason for the disparity between health expenditure and health benefits is plain. The NHS, he argues, lacks a formalised method of evaluating the service it is providing. The solution he offers is the Randomised Controlled Trial (RCT) (Cochrane 1972). 11. This is the second stipulation following from the need to procure informed consent (the Nuremberg Code 1947 in Beaucamp and Childress 1979: 45). 12. This included a continuous twenty eight-day follow-up, during which, more than one hundred male volunteers had to receive finger pricks every day. 13. The villagers often explained their resistance to blood sampling in terms of their relative poverty and consequent lack of blood. This vulnerability reframes the presence of the MRC as both benefactor and as benefiting from the participants’ powerlessness. As a participant from Bambali put it: The MRC are mostly here trying to have controlled areas to study treatments and coming to Africa for the reason of helping those who need it most (Interview, 25 February 2005). 14. It is not my intention to ventriloquise the MRC, rather to suggest that the relationship between international research and the state is complex. As one nurse stationed in Bambali explained: MRC is not trying to run a clinic, because that would be in competition with the hospital. We help them with transport, we keep their vaccine for them and we work hand-in-hand with the hospital staff. We take care of some of the people who they can’t because we’ve got the best medicine and medical staff here (Interview, 28 January 2005). 15. In November of 2000, clauses 19 and 30 were added to the Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects, stipulating that all populations must benefit from the results of research by being assured of access to the best proven prophylactic, diagnostic and therapeutic methods identified by a study. 243

Evidence, Ethos and Experiment

References Beauchamp, T.L. and J.F. Childress. 1979. Principles of Biomedical Ethics. New York: Oxford University Press. Cochrane, A. 1972. Effectiveness and Efficiency: Random Reflections on Health Services. London: Nuffield Provincial Hospitals Trust. Fisher, R.A. 1966. [1935]. The Design of Experiments. Edinburgh: Oliver and Boyd. Geissler, W. 2005. Kachinja are coming! Encounters around a medical research project in a Kenyan village. Africa 75: 173–202. Geissler, W., A. Kelly, B. Imoukhuede and R. Pool. 2008. ‘He Is Now Like a Brother, I Can Even Give Him Some Blood’ – Relational Ethics and Material Exchanges in a Malaria Vaccine ‘Trial Community’ in The Gambia’, Social Science and Medicine, in press. Gigerenzer, G., Z. Swijtink, T. Porter, L. Daston, J. Beatty and L. Krüger. 1989. The Empire of Chance. Cambridge: Cambridge University Press. Hill, A.B. 1953. Assessment of therapeutic trials. Transactions of the Medical Society of London, 68: 128–47. Horton, R. 2001. ‘The Clinical Trial: Deceitful, Disputable, Unbelievable, Unhelpful and Shameful – What Next?’, Controlled Clinical Trials 22: 593–604. Hotelling, H. 1951. ‘The Impact of R.A. Fisher on Statistics’, Journal of American Statistical Association 46: 35–46. Marks, H.M. 1997. The Progress of Experiment. Cambridge: Cambridge University Press. Marks, H. M. 2003. Rigorous Uncertainty: Why RA Fisher Is Important. International Journal of Epidemiology. 32 (6): 932–37. Molyneux, C.S., D.R. Wassenaar, N. Peshu and K. Marsh. 2005. ‘Even If They Ask You to Stand by a Tree All Day, You Will Have to Do It (Laughter) ... !’: Community Voices on the Notion and Practice of Informed Consent for Biomedical Research in Developing Countries’, Social Science and Medicine 61(2): 443–54. Savage, L. 1976. ‘On Rereading R.A. Fisher’, Annals of Statistics 4: 441–500. Senn, S. 2003. Dicing with Death. Cambridge: Cambridge University Press. Strathern, M. 1996. ‘Cutting the Network’, The Journal of the Royal Anthropological Institute 2(3): 517–35. Tallon, D., J. Chard and P. Dieppe. 2000. ‘Relation between Agenda of the Research Community and the Research Consumer’, Lancet, 355(9220): 2037–40. Whyte, S.R. 1997. Questioning Misfortune: The Pragmatics of Uncertainty in Eastern Uganda. Cambridge: Cambridge University Press. Will, C. 2007. ‘The Alchemy of Clinical Trials’, Biosocieties 2(1): 85–99. Yates, F. 1951. ‘The Influence of Statistical Methods for Research on the Development of the Science of Statistics’, Journal of the American Statistical Association 46: 19–34.

244

Chapter 10

Foetuses, Facts and Frictions: Insights from Ultrasound Research in Tanzania Babette Müller-Rockstroh1

Introducing a Research Object Dating from 25 January 2004, the letter by the National Institute for Medical Research in Dar es Salaam provides ethical clearance for my study on ‘Ultrasound in Tanzania’ with the following words: ‘Herewith you are permitted to conduct the above named study. We expect sound insights into a good technology that has not been researched in our country. We are looking forward to your reports biannually.’ Since US President Harry Truman’s inaugural address announcing a ‘fair deal’ that initiated the beginning of a flow of goods from high-income to low-income countries, medical experts in offices in Geneva, in New York, in Washington, have attempted to define ‘appropriate technology’ as that technology eligible for travel long distances. How much may a technology cost to be affordable for low-income countries? How long does it need to run, and how many patients does it need to meet in order to be considered efficient? Is it of use, and will it be used in a new context? Small, portable, comparatively cheap in relation to other diagnostic imaging technology such as X-ray, MRI or CT scans, with a wide range of medical applicability yet modest in its infrastructural requirements, ultrasound technology has been promoted as ‘appropriate’ for travelling the world by the World Health Organization since the mid-1980s (WHO 1985). Common accounts of technology transfer in the wake of Truman’s promotion of ‘science and technology’ for the world thereby seem to take for 245

Evidence, Ethos and Experiment

granted that technology naturally does (and should do) what it is supposed to do. This is the departure point for my research on ultrasound transfer to Tanzania: the nature of technology. I draw on Medical Anthropology and Science and Technology Studies that have repeatedly shown that technology upon leaving its country of origin may do quite different things and may have unintended consequences. Alongside its travel, as these studies have shown, technology gets appropriated and transformed as it is put to new uses by new users. Yet it may partly also remain the same and may thus change the places it travels to (van der Geest and Whyte 1988; Akrich 1992, 1993; De Laet and Mol 2000; Whyte et al. 2002). Following ultrasound technology to Tanzania, I introduce technography as heuristic device to study this mutual shaping of technology and society. Doing technography, an ethnography of a technology in transit, allows us to understand evidence – facts or in my context also foetuses – as being done in practice (cf. Mol 2002). It thus provides us with questions that frame the work in which context, technology and its depicted objects get co-constructed. In short, I claim technography to be a critical-constructive mode and valuable contribution to current medical research and research on medicine in Africa. In Tanzania ultrasound is still a fairly new technology. Whereas Teaching Hospitals and some private healthcare facilities in urban settings have used ultrasound since the beginning and mid-1990s, many of the rural district hospitals have only since recently been provided with ultrasound technology within a Dutch development project transferring ultrasound machines to a range of sub-Saharan African countries, among them Tanzania. The following accounts of ultrasound in Tanzania are based on a longer ethnographic study conducted in 2004 in which I followed these machines from their places of manufacture to the immediate buyers of the technology, the Tanzanian Ministry of Health, to their new workplaces, Tanzanian healthcare facilities, and into the lives and bodies of women in northwest Tanzania. Two questions guided this pursuit. The first was a question of ontology – what does ultrasound become in different contexts? The second was an epistemological question – how can we know and study this becoming?

First Context: A Dutch Multinational Company In the manufacturing hall of Philips, a Netherlands based multinational company, ultrasound is a device to reduce maternal mortality statistics. It is simultaneously also a device for development, as the manager of the ultrasound-transfer project in Tanzania explained to me: not only are there 246

Foetuses, Facts and Frictions

masses of women in Africa who are ‘either pregnant, or sick, or both’, women in Africa, when given the chance, are also known to be more productive than men, more reliable in paying off micro-credits and the ones responsible for childcare – thus for raising and educating future generations. Helping women out of disease, so Philips claimed, meant taking care of the continent as a whole – the perfect remedy being ultrasound technology. Ultrasound’s capacity to alleviate the lot of Africa and African women lies in the technology’s hardware, consisting of a transducer, a central processing unit (CPU), a monitor and a keyboard. The machine emits and receives ultrasound waves after they are deflected by body tissue; these signals, once electronically enhanced, can be displayed on the monitor in shades of grey according to the density of the scanned tissue. In developmental terms it is ultrasound’s ability to visualize the (mal)formation and (mal)presentation of foetuses that counts as well as the possibility to measure them and thereby establish gestation age. These measurements are the effect of the software inbuilt in ultrasound machines. Embodied in ultrasound are so-called foetal growth charts. Upon pressing the right key, any new foetus is compared to the growth and associated age of these ‘stored foetuses’ and thus provided with the proper reference, for example thirtysix weeks, three days. Among other factors that count, it is this technologically generated pregnancy time that is used for medical decision making: is the pregnancy to be terminated or continued? And which delivery mode is to be chosen to ensure good health for mother and child? Africa (and likewise African women) in many ways provided a ‘blind spot’ for the company, as the project manager explained. In particular, it was the lack of technical maintenance capacity and operating knowledge that proved to be a constraint for entering Africa as a market. In order to ensure the good working of their machines, the company engaged in capacity building, in line with contemporary development policy. For five years after sale, Tanzanian technicians trained by the company were to visit the workplaces of the technology regularly and ‘repair’ machines by replacing parts known to break down easily – a task later to be taken over by their business partner, the Tanzanian Ministry of Health. The second capacity to be built consisted of the professional bodies operating the machines. While the company left this choice to the Tanzanian partner, it set a precondition: in line with the internationally pursued decentralization policy, the majority of future operators should work in rural areas. In order to be able to teach the basics of ultrasound operation within the twoweek course that the company had developed in terms of balancing costs 247

Evidence, Ethos and Experiment

and output, ultrasound operators first studied the theory and scanned each other in a ‘lab’ on the premises of the company. In the second week operators were then moved to the radiology department of the big governmental Teaching Hospital in the capital in order to learn scanning ‘under real African conditions’, as the company put it. Following ultrasound machines in the context of a multinational company demonstrates that technology transfers work in a particular ‘space of thought and action’ (Escobar 1995: 42), in which ‘Africa’ is depicted as both the essential other and essentially the same. It is the different mortality statistics for mothers with the same rights and the same bodies as their ‘white’ sisters that call for ultrasound technology to travel and for it to be maintained in the same way; preventing the breakdown of machines, following the company’s regime of repair, prevents the subsequent breakdown of bodies. Likewise, professional bodies (e.g. healthcare professionals) are both the same and different – in their simple lack of knowledge. And while district hospitals are a different type of hospital to teaching hospitals, learning in one facility in the eyes of the company can be easily extrapolated to the other type: after all, it is all Africa! In providing Africa with imaging machines, colonial images – such as ‘the white man’s burden’ of the continent and of women – are reproduced. These images mingle with neoliberal imaginations of an economic development (see Langwick, this volume). Hence, in the context of a multinational company, ultrasound rather than being a simple diagnostic imaging technology becomes a mobile actor in humanitarian ‘traid’ (trade/aid). What happens to ultrasound technology, though, once manufactured and released into the real world?

Second Context: Tanzanian State Agencies For the Tanzanian Ministry of Health, who bought the Dutch ultrasound technology, however, ultrasound was primarily seen as a device for professionalization. Concerned with the slow degradation of the discipline of radiology, meant to propel the country into modernity after independence, buying into the Dutch ultrasound technology seemed to open up a new era. Former president Julius Nyerere’s pursuit of his own, socialist-inspired politics had resulted in the withdrawal of many international companies and their services; X-ray machinery had slowly gotten older and obsolete, as had radiographers and radiologists. Even before the Dutch proposed to loan/sell ultrasound technology, radiographers (who increasingly travelled to international conferences after the country was opened to the liberal248

Foetuses, Facts and Frictions

ization politics guided by the International Monetary Fund and the World Bank) noticed that their colleagues elsewhere in the world had extended the profession to include sonography into the paramedical technical production of diagnostic images. Buying into ultrasound technology hence promised to serve several goals: to brush up imaging standards in the country, thereby attracting more radiologists and radiographers, to obtain new rather than secondhand technology as is often the case with donations, and to execute decentralization policy by serving rural over urban areas. As it turned out, however, there were not enough radiographers in rural areas who could operate both X-ray and the new ultrasound machines. So the Tanzanian state agency turned to midwives as ultrasound operators at the district level. Born out of necessity, this decision had several advantages too: midwives were considered to be closer to the women, an asset in regions such as in the northwest of Tanzania where people still believed in witchcraft and might associate the new technology with such kind of practices, according to the Head of the Imaging Unit. Furthermore, choosing midwives as second-order operators allowed the state agency to buy a few more sophisticated ultrasound machines for the radiographers and a large amount of cheaper basis machines, thereby demonstrating a desired imaging hierarchy and reinforcing the traditional referral system. Following ultrasound to the level of Tanzanian state agencies illustrates that the state apparatus in many ways did work: professionals were recruited and trained, most of them worked in rural areas, new professional boundaries were drawn in that professional conducts obtained a passus about ultrasound usage and the state representative obtained a seat in the international steering committee overseeing and evaluating future technology transfer projects to the country. Decentralizing ultrasound thus served the state agency to reinstall itself as central and responsible manager of healthcare services. At the same time, however, radiographers and radiologists still remained less educated than their fellows ‘in the West’ trained in PET, MRI and CT scans. Moreover, seeing themselves as technicians working for but not as medical personnel, some of the radiographers I met treated the foetus like any other organ without taking into consideration its specificity of developing, moving and growing. ‘Twins. Twins are the same. That is what is meant by twins’, one of the radiographers told me when I asked him why he would not measure both twins in order to exclude certain pathology, such as the twin-to-twin transfusion syndrome. At the same time, nurse-midwives obtained their triple 249

Evidence, Ethos and Experiment

job, that of an ultrasonographer but without official acknowledgement or incorporation in reproductive health programmes. Trained according to ‘Western’ obstetrics, nurse-midwife-ultrasonographers, in contrast to their radiographer colleagues, could use their new skills to move into private facilities. This transformation of ultrasound into a device for professional politics did not hinder the machines travelling to their future work places. Pregnancy thus came to meet with ultrasound accidentally rather that systematically. In several of the healthcare facilities equipped with ultrasound, the technology did not catch on despite the seemingly similarly trained professional bodies and high number of women with pregnancies and pregnancies ‘at risk’. Elsewhere I have in detail described how a pregnancy has to be made fit for ultrasound, that is, made into a pregnancy scan-able and eligible for imaging technology (Müller-Rockstroh 2007). In the following, however, I concentrate on two healthcare facilities where ultrasound has found its way in clinical practices.

Third Context: Healthcare Facilities For all healthcare facilities under study in the research, ultrasound turned into a device to ‘run the hospital’. For the intern doctors in the Teaching Hospital, the introduction of ultrasound was highly appreciated as it allowed doctors to do medicine ‘by the book’ – modern obstetrics. Referring to the replacement of traditional clinical diagnosis of ectopic pregnancies by ultrasound, one of the interns said: What the old ones did, someone such as this old mzee, this doctors you met at H2 [the emergency unit, BMR], for example, they simply took a syringe, injected it and drew blood. If the blood clotted, most certainly it was an ectopic. So they need not do ultrasound. But you see these were the old days. We cannot do that anymore, simply because it is not practiced anymore, it is not scientific. If you open a textbook, it will say: for diagnosis of ectopic use ultrasound. Ultrasound in the Teaching Hospital for some conditions thus has become a routine examination – at least in the normal working hours of ultrasound. At night, however, what counted as emergency for ultrasound, changed. During the day, Paulina Mkumbi,2 for example, would have immediately been referred to ultrasound: thirty-three years old, she had just one living child aged nine, and had already lost one ovarian tube due to a previous ec250

Foetuses, Facts and Frictions

topic pregnancy. During a nightshift with the above intern, Paulina Mkumbi reported with severe abdominal pain, spotted bleeding, three months menstrual cessation and a positive urinary pregnancy test. According to the knowledge of the intern, all clinical signs pointed to either an ectopic pregnancy or a missed abortion. In daytime, he said, his doubts would have made him send her straight to ultrasound. Her arrival at night, however, changed ‘everything’, he remarked. For him as an intern, getting the ultrasonographer on call from home implied getting in contact with the supervising nurses in order to make a telephone call first, then calling his immediate superior doctor and convincing him of the urgency of an ultrasound, waking a driver and sending a car to fetch the ultrasonographer and boss. Hence, night-time ultrasound required a careful balance between ‘knowing enough’ to convince one’s superior and ‘knowing too little’ to legitimize the mobilization of these extra resources. In the case of Paulina Mkumbi, the intern decided against an ultrasound at night and wrote a referral for the following morning. By the time the ultrasound was finally done on the following day, Paulina Mkumbi’s second ovarian tube had already ruptured, resulting in high blood loss and subsequent infertility. In the small district hospital, situated about 35 kilometres away from the Teaching Hospital, the new ultrasound machine enabled a greater capacity to deal with the masses of pregnant women reporting every day for antenatal care or for delivery. As the facility, a former mission hospital had had an ultrasound machine before, this old machine was put in the theatre to be used for pelvic, abdominal and pregnancy scans. The new machine, in contrast, was put in the maternity ward to be used for pregnancy and, surprisingly, for the postpartum period. In particular, the young nurse-midwifery students who worked in the antenatal care room profited from the new availability of scans. For them ultrasound turned into a knowledge device and midwifery teacher. Nursemidwives often complained that women were ‘ignorant’ of their last menstruation, making it necessary to establish gestation age as a reference for a palpated fundal height.3 Students frequently sent women for ultrasound to ‘fine-tune’ their, the students’ hands, also in preparation of future work in places without this technological possibility. The new ultrasound at the maternity ward, in contrast, turned into an effective executor of national health regulations, on the one hand, and a compensator for national health regulations on the other hand – not for pregnancy but as an after-birth scan. Following governmental policy promoting hospital rather than home birth, for example, all women who had 251

Evidence, Ethos and Experiment

given birth before arrival in the hospital were sent to ultrasound to scan for possible leftover placenta tissue. This dangerous condition occurred frequently when women delivered without ‘proper assistance’, as the midwives explained. They added that the situation had become worse after the government had cut down the teaching programmes for traditional birth attendants who had formerly been taught by hospital personnel how to manage the placenta stage correctly. As syphilis was endemic in the region, the district hospital was moreover pressured to do anything to fence in the disease. A major cause for intrauterine foetal death and prematurity, syphilis thereby indirectly increased the necessity for scanning after the birth of such children as both might result in incomplete delivery of placenta and foetal tissue. Using ultrasound after birth, the nurse-midwives at the maternity told me, had significantly decreased the return of women with puerperal sepsis, and thus decreased maternal morbidity and mortality. Given this decrease to be one of the main goals of the Tanzanian health policy and its sponsors, the midwife-ultrasonographer told me she was more than surprised that this after-birth scan did not meet with the approval of the Dutch ultrasound teacher at the refresher course, who emphasized that she should rather ‘work on properly scanning the foetus’. For the matron of the maternity ward, however, ultrasound mainly turned into a device that delayed the discharge of women from the ward. In her evaluation she stressed the need to allocate beds to women, to sterilize specula for uterine evacuation, to schedule the duty hours of the doctor according to ultrasound results and to keep track of the clock in order to discharge women at a time that allowed them to reach home in daylight hours. This task was made more difficult by women’s ‘ignorance’ in believing diagnoses, she said. While ultrasound was meant to speed things up, women often requested a second examination to confirm the diagnosis of intrauterine foetal death before consenting into induced labour. In order to facilitate understanding, she had thus advised her staff to use ultrasound to stress the importance of syphilis treatment. Consequently, in the antenatal care ward, I heard several of the nurses explaining to women that the kioo [Kiswahili: mirror] had seen the white stripes, the wadudu, and that these wadudu were responsible for their dead child (see also Gammeltoft 2007 on the local usage of ultrasound/pictures in Vietnam with regard to both regional and individual history). Following ultrasound to the hospital level illustrates that ultrasound does not easily embed in clinical practices. Schedules, working hours, time preferences, material resources, knowledge, hospital architecture and ge252

Foetuses, Facts and Frictions

ography shape and are shaped by ultrasound technology with the objective to effectively pass patients on through clinical space. Yet, how are patients moved, in the sense of being affected, by ultrasound?

Fourth Context: Women’s Homes, Women’s Bodies For all women in northwest Tanzania that I spoke to ultrasound became a device to support the ability to carry children. For Paulina Mkumbi, a former teacher living in the regional capital, it was out of the question to not have an ultrasound. Not only her pregnancy but also Tanzania as a country were after all modern, she said: ‘we women who live in town, we are modern now. We see with open eyes. So if you are pregnant, what we do is like any other women in your [Germany, BMR] country, we have an ultrasound to test our wombs’. Unlike Paulina, Janice Mtutu had no prior knowledge of the new technology. Like her, however, Janice also had her womb tested: after having observed a disrupted bleeding pattern for three months, along with nausea (for which she ate earth in order to strengthen her womb and a possible child), and wearing a necklace with traditional medicine against witchcraft afflictions, she convinced herself her pregnancy should be examined at the hospital when she got concerned with the lack of child movements at around 37 weeks of gestation. I met Janice again after her second ultrasound examination, whereupon she told me the following story: At the first mirror, I saw it moving. She [the nurse-midwife, BMR] told me that the child had left the stomach [but] I saw it moving there at the glass. I felt it moving then. They want me to do what? Take the pipe to remove the dirt. […] They brought me to have the mirror again. This time I did not see it moving. I saw dirt only, ashes. The child had truly left the stomach. […] I saw some white things too. The nurses they tell you these are wadudu, […] the nzoka, the nzoka that has bitten my child. In this region in northwest Tanzania (Usukuma), nzoka is one of the key agents of affliction. It thereby comprises a range of complaints: nzoka could startle a person to the extent that she or he has a fit, it can cause stabbing pain in the genital region, it can cover a woman’s eggs with pus thereby dis253

Evidence, Ethos and Experiment

abling her to conceive, and it can ‘bite’ unborn children so that these will die. Despite these clear-cut effects on reproductive health, the etiology of nzoka at first seems contradictory. While some people told me nzoka was inherent to being Msukuma, thus something you are born with, others explained nzoka was acquired by eating earth. In particular, older people also told me that afflictions by nzoka had increased since the big trees in the region, seats of all kind of spirits, had been cut down thereby forcing these spirits to inhabit new seats, people’s heads and bellies. What condenses in the term nzoka, in my opinion, is a range of invisible agents that have one thing in common: they perform and display vertical disorders in the social fabric. It is the practice of eating earth that thereby links the seemingly contradiction in etiology. As anthropologist Wenzel Geissler (Geissler 1998, 2000; Geissler et al. 1999) has shown for the Luo in Kenya, earth in agricultural societies is both the ground for production, but it is also the place where ancestors are buried. By eating earth, people thus literally incorporate their lineage. It is here where nzoka also becomes gendered although nzoka can in principal afflict both sexes. Yet earth is considered to fill the unborn with strength, and so women become more vulnerable to this agent. In particular, women who had taken ‘white medicine’ before getting pregnant, a term used to refer to the three monthly contraceptive injection, for example, were obliged to eat earth in order to compensate for the ‘lack of blood’ caused by the drug and to prevent possible ‘bleeding problems’ as a result of the injection. These problems consisted of dizziness, lack of blood in pregnancy but also of child mal-positioning in the womb. While not all women I spoke to were certain that ultrasound could see nzoka, the particular configuration of ultrasound in the hospital makes clear why women may remain thinking they carry nzoka rather than a child or a foetus even though they have been scanned by ultrasound. There is, first, the fact that the nurse-midwife-ultrasonographer in accordance with her professional ethics (see above) do not show the ‘baby’s first picture’ (Mitchell 2001); second, that the monitor is placed in such a way that women see moving pictures which, in the absence of explanations, are often taken for child movements (and movements signify the presence of a child); and third, that wadudu inhabit the world around ultrasound in the antenatal ward. While wadudu in the antenatal ward refer to syphilis, in the ultrasound room they rather refer to spirit affliction. Despite all the things lost in translation, Janice Mtutu in the end followed the biomedical trajectory. Other women I talked to after having ultrasound scans, however, took 254

Foetuses, Facts and Frictions

the new information home and chose to deliver their babies at places and with people whom they thought were better able to deal with the diagnosed risks than the hospital. Following ultrasound into women’s lives illustrates a body formed within socio-technical relations with kin, ill-wishing neighbours, healthcare professionals, antenatal records, hospital beds and monitors. All women referred to ultrasound as a machine that could ‘see each and everything’. Some thereby repeated the scanning order: ‘a head’, ‘a belly’, ‘a leg’. Professional diagnoses may thus provide temporary certainty that there is a child and that it is alive. Leaving the hospital grounds, though, exposes women again to affliction of witches or of spirits. Ultrasound in this context loses its ‘natural’ connection with establishing concrete facts and foetuses. With the loss of the foetus and return of the mother as end user, ultrasound has reached the end of its travel – for the time being. What then has technography yielded and what questions has it raised?

Doing Technography: Questioning Technology in Contexts Doing technography depicts technology in contexts. What is evident – what ultrasound, a foetus, a pregnancy, a child is – thereby becomes contextual matter, as I have shown above. In this last section I will use the empirical insights to pose some of the questions that the ultrasound research as I have presented raises for the study of (medical) research on biomedicine in Africa.

What Are the Effects of Technology? Common medical technology assessment deals with the question whether a technology is a good means to a pre-determined end. A good ultrasound machine is certainly a machine that works – that does not break down and that makes bodily (mal)formations visible. In Dr Dodoma’s statement at the beginning of this chapter, another ‘good’ seems to resonate: a good ultrasound machine here is a machine that responsibly deals with existing human and material resources, for example. Talking about ‘good’ technology can also encompass a much larger set of ethical, aesthetical and political questions, such as whether the introduction of ultrasound in Tanzania results in doing away with skills such as hand-knowledge or re-skills professionals. Or, whether the channelling of ultrasound technology to Tanzania merely disposes of obsolete technology, or indeed opens a mar255

Evidence, Ethos and Experiment

ket? And if it does open a market, who is to pay for these machines in the future? These effects also have to be made visible. Thus, I argue it is important to explore not only how technology works but also how it works out for the network of people and other things that it becomes part of. What are the repercussions of ultrasound for old, young, urban, rural women, for hospital architecture, documentation systems, for traditional healers, for foetuses and for nzoka? Preventive maintenance, for example, dismisses African reparability, the capacity of innovative tinkering with machines and human bodies. This new regime thus sets new rules, distributes blame in particular ways, and closes debates on how maternal mortality could have been tackled otherwise. We may thereby consider the performance of the same measures for men and machines by the Dutch company as business strategy, and yet have to ask ourselves where we want tinkering to stop and standardization of care to take over. In short, instead of regarding ultrasound technology merely as a cultural product that travels, it has to be studied as a producer of culture, or as the philosopher Nelson Goodman has put it, as ‘world making’ (Goodman 1978).

Which Reality is Most Important? One way to broaden the question what makes a ‘good’ technology is thus to study the implications of the technology for those touching and being touched by it. A diagnostic imaging technology by definition is meant to depict what lies beneath the skin. The question of ‘good’ here is simple enough: a ‘good’ technology is a technology that displays bodily reality as precisely as possible. Yet, as shown, ultrasound has unexpected effects on the broad network of humans, things, attitudes, knowledges, conventions and identities of which it is part of. Concerned with the communication between practitioners and their (research) subjects, Marilyn Strathern (this volume) calls us to ‘suppose there were people who did not invest in epistemology in the way that permeates Euro-American science’. Her concern resonates with the classic focus of medical anthropological research: patient worlds. In answer to her call, Steven Feierman, writing in this volume about how physicians of different cultural contexts meet, urges us to focus our attention to those that ‘invest in epistemology in exactly the same way as their European or American counterparts’, and yet come to very different conclusions with regard to the needs and demands of African healthcare. I agree with Feierman that it is necessary to shift away from studying patients as the essential other, and include those that appear so deceptively familiar in our search for better healthcare. In a context 256

Foetuses, Facts and Frictions

where distance to healthcare facilities is not an issue during the day or at night, where there are mobile phones, ambulances, post-partum care, neonatal intensive care units and blood banks, the display of a foetus may be the most important insight into a womb. But where there is syphilis, where there are scarce resources to mobilize transport, where women need to pay for uterine evacuation unless they have it done immediately after delivery, where bed economy in the hospital is a tricky task, and where people do not travel at night due to witchcraft fears, the most important reality to be scanned for may be placentas. And then those who speak for placentas and after-birth scans need to be listened to, instead of their reality being brushed away. While, like Feierman, I am concerned with acknowledging multiple medical realities, I disagree with one important implication of his paper, that is, that translations of medical technology are but effects of their faultiness and mal-adaptation to the contexts they are supposed to serve. Better technology, he seems to suggest, would not need such appropriation. Here Feierman’s argument sounds strangely similar to that of the public health experts in Geneva, Washington and New York that he criticizes. Based on my experiences with ultrasound in Tanzania, it seems more probable to me that any technology inevitably changes as it embeds in practices already in place. While Bruno Latour (1988) makes it appear as if Louis Pasteur generated a new world, doing ethnography on medical research in Africa we cannot help notice that biomedicine can only go where there are already some rails, to use Latour’s image. Worlds are thus never generated from a zero level but are built upon something already in place. How much of the network we would need to transfer in order to have ultrasound in Tanzania work the same way as in Europe or the United States is an interesting empirical question. An ethical question would be whether it would be desirable to have scans ‘for foetuses’, ‘for fun’ and for formations that cannot be cured. Lack of knowing about the world of invisible agents in northwest Tanzania that afflict women’s wombs and worlds does not prevent ultrasound from working. The question whether ultrasound is a ‘good’ technology in this context may be tricky: ultrasound may inform women to go home rather than to stay in the hospital and be delivered by c-section or be evacuated after birth as the biomedical logic has it. Ultrasound may thus discriminate against a present child but may still serve women’s ability to fall pregnant. In order to be a ‘good’ technology, ultrasound thus has to do more than display bodily reality. It has to provide relevant information. Evaluation of technology then has to ask which information is relevant. Whom and what does it include or exclude? And whom does it serve? 257

Evidence, Ethos and Experiment

Which Better Is Best? A ‘good’ technology hence not only answers but asks the right question, and displays the reality that at a particular moment is of concern to the patient and to a further treatment trajectory. In recent years, a lot of attention has been devoted to asking whether medical technology really makes patients better. It has been asked whether practice is evidence-based. This question has fed numerous clinical trials that are increasingly also carried out in Africa (see Kelly, this volume). In the very early phase of the trial those conducting it set the objectives of diagnosis or treatment and decide which parameters they will follow in order to measure whether these objectives have been met. Most often a list is thereby made of potential undesirable side effects. ‘Good’ technology in this context is a technology that proves to be effective in the sense that it produces a high number of desired effects with minimal undesirable outcomes. Common medical research on technology builds on this differentiation between knowledge first and application later. The better and the more detailed the knowledge the better decisions can be made (by medical practitioners or policy makers). By examining ultrasound practices in Tanzania, I have shown that things are not so simple. Rather than asking whether or not technology is effective, the more interesting question to pose is ‘which effects does technology have?’ Not ‘do technologies have side effects?’ but ‘which different advantages and disadvantages do technologies carry?’ Not only ‘what do technologies cost?’ but ‘which kind of costs are attached to a certain technology and who is to pay?’ (cf. Willems 1995). Already for someone measuring the effects of technology on the body, it is difficult to decide which parameters should represent the ‘better’. For an attempt to measure the effects of technology on a network of things, this is even more the case. It makes the question of what a ‘good’ technology is, a broad question that encompasses many things.

How Broad Are Technologies? It should be clear by now that a technology cannot simply be called ‘good’ when it is efficient and an effective means for a pre-determined goal. Not only do goals change as the technology travels but also when achieving this goal, technologies may have diverging effects on the context they become part of. Answering the question of whether a technology is ‘good’ requires us to always pose the question anew, ‘Which reality is the most relevant in what case?’ Technologies, hence, have to be evaluated broadly. But how broad are technologies themselves? While common technology 258

Foetuses, Facts and Frictions

assessment makes a difference between technology and humans, expecting effectiveness from the former and ethical actions from the latter, I have attempted to break through that dichotomy in calling for a technology that has to be scrutinized alongside these normative dimensions too. Technology itself is not a thing, but always a technology (with)in context. That is unfortunate for clinical trials that isolate technology or technological elements from context and that ignore the ‘noisy’ variables although it may be exactly those that count. Instead, technography allows considering technology and society together in the temporary form of an ‘international’, a ‘state’, a ‘hospital’, or a ‘woman’s apparatus’.

A Good Technology A good technology – that sounds nice and legitimate. Yet ‘good’ is a broad term, and as my research on ultrasound in Tanzania suggests, we should perhaps rather ask ‘what are the specific kinds of good that are at stake with a particular technology?’ Between the terms of ‘good’ with regard to the technology itself and those used when discussing the usage of technology, there can be a big difference: technology itself comes in terms of utility, efficiency and efficacy that are linked with a research tradition based on epidemiology wherein clinical trials have become the golden standard. Technology is often used in terms that pertain to particular – Western – ethical traditions. Where ‘good’ in the epidemiological tradition is studied as a fact to be discovered – ‘good’ being the precise display of a malformed foetus, for example – dominant ethical traditions need to be evaluated when there is a clash of values over facts. Fixed ethical principles are thereby mobilized for the evaluation of the practices of humans: ‘autonomy’ is one of these principles, ‘doing good’ and ‘doing no harm’ another, ‘justice’, or ‘respect’ yet others. In order to formulate these principles, (local) contexts do not play a role as if too much ‘daily noise’ would threaten the sight of these big lines. As mentioned, though, it is precisely these details that do the work, and that have to be known in order to be able to (e)value medical technologies as local and situated configurations with their intended and unintended consequences. Rather than evaluating whether a technology works according to pre-determined principles, we should perhaps ask how it is enabling or disabling, beautiful, careful, heavy, helpful or painful, and so on; not whether a technology can make people better but which ideals it brings to those who use it as well as those who are treated with it. In sum there is no simple solution to the request of policy makers or ethical committees to be provided with a final, fixed set of variables that 259

Evidence, Ethos and Experiment

determine whether ultrasound is a good technology for a district or a regional hospital. Instead, ‘good’ technology is a choice between different worlds that emerge with its introduction.

Sound Research: Studying ‘Ultrasound in Tanzania’ Medical research on ultrasound in Africa is still scarce or altogether non-existent. Writing an ethnography of medical research, hence, could only simply state this lack – in the light of numerous studies on technical resolution, innovations (such as 3D scanners), abnormality detection and resulting surgical procedures or foetal mortality rates in ‘the North’. In presenting my research object to various audiences in the field, in academic and especially in developmental environments, my choice has repeatedly been commented upon as ‘unusual’ or rather ‘far fetched’ – with a hint that it is not part of the ‘real problems’ faced by Africa today. The advantage of having a surprising object of study was that my topic in the field elicited a lot of stories as many people – professionals, patients and planners alike – were eager to share their experiences since nobody had ever asked about them before. At the same time, the disadvantage of studying healthcare technology in Africa is writing at the margin of Science and Technology Studies and Medical Anthropology but also of healthcare development or public health sciences – disciplines that are still more concerned with the logistics of supply or the study of disease(s) and its outcome on populations. The aim of this chapter, in contrast to biomedical studies, has thus been different – namely to claim the necessity to study technology in (its) interaction with new contexts. Rather than assuming that a technology is the same regardless of context, and hence works or does not work (in this way ending up as another development failure), the above insights into ultrasound in Tanzania have demonstrated appropriation and different workings as technology inevitably is embedded into existing practices in order to deal with the health and disease of real bodies. As globalization is ongoing, travel of sophisticated healthcare technology increases in order to disseminate biomedicine across the world. In this context I hope that the ‘sound insights’ that my study has elicited – however different from those expected by the national ethics committee, I suppose – may be picked up to study other human technology interactions. Given the persistence of a high maternal and perinatal/infant mortality rate in Tanzania, despite all 260

Foetuses, Facts and Frictions

the ‘goods’ reaching the country, a technographic study of technologies meant to make facts, establish foetuses and save mothers, yet leading to innovative and deadly frictions with(in) context and between contexts would be a first step to bridge the technology divide and study rather than assume the emerging societies.4

Notes 1. Max Planck Institute for Social Anthropology, Halle/Germany. This chapter is based on my Ph.D. research on ultrasound in Ghana and Tanzania with extensive fieldwork between 2002 and 2005. This research was financed by WOTRO/NWO in the Netherlands. Many thanks go to the women and professionals who have shared their ultrasound worlds with me, and to the participants of the Kilifi Conference in Kenya in 2005 with which I discussed an earlier version of this paper. 2. All names used in this text are pseudonyms. 3. Clinical diagnosis of pregnancy time and physiology measures time in two ways: one is based on the woman’s report of her last menstrual period, thus establishing gestation age. The other is what I have called fundal height-time, which is established by professional palpation. Comparing these two times with each other allows the diagnosis of particular ‘risks’, such as twins or too much amniotic fluid (cf. Müller-Rockstroh 2007). 4. here, for example, about the WHO-promoted ‘saviour’ of women’s lives – the partograph – the fetoscope or dopton, or antiretroviral drugiv I am thinking s in the context of PMTCT (prevention of mother-to-child transmission) Programmes.

References Akrich, M. 1992. ‘The Description of Technical Objects’, in W. Bijker and J. Law (eds), Shaping Technology/Building Society: Studies in Sociotechnical Change. Cambridge, MA: MIT Press, pp. 205–24. ———. 1993. ‘Essays of Technosociology: A Gazogene in Costa Rica’, in P. Lemmonier (ed.), Technological Choices: Transformations in Material Cultures since the Neolithic. London: Routledge, pp. 289–337. De Laet, M. and A. Mol. 2000. ‘The Zimbabwe Bush Pump. Mechanics of a Fluid Technology’, Social Studies of Science 30: 225–63. Escobar, A. 1995. Encountering Development: The Making and Unmaking of the Third World. Princeton: Princeton University Press. Gammeltoft, T. 2007. ‘Sonography and Sociality’, Medical Anthropology Quarterly 21(2): 133–53. Geissler, P.W. 1998. ‘“Worms Are Our Life”, Part 1: Understanding of Worms and the Body among the Luo of Western Kenya‘, Anthropology and Medicine 5: 63–79.

261

Evidence, Ethos and Experiment

———. 2000. ‘The Significance of Earth-eating: Social and Cultural Aspects of the Geophagy among Luo Children’, Africa 70: 653–82. Geissler, P.W., R.J. Prince, M. Levene, C. Poda, S.E. Beckerleg, W. Mutemi and C.E. Shulman. 1999. ‘Perceptions of Soil-eating and Anaemia among Pregnant Women on the Kenyan Coast’, Social Science and Medicine 48: 1069–79. Goodman, N. 1978. Ways of World Making. Indianapolis: Hackett Publishing Company. Latour, B. 1988. The Pasteurization of France. Cambridge, MA: Harvard University Press. Mitchell, L.M. 2001. Baby’s First Picture: Ultrasound and the Politics of Fetal Subjects. Toronto: University of Toronto Press. Mol, A. 2002. The Body Multiple: Ontology in Medical Practice. Durham: Duke University Press. Mueller-Rockstroh, B. 2007. Ultrasound Travels: The Politics of a Medical Technology in Ghana and Tanzania. Ph.D. dissertation. Maastricht: Maastricht University, The Netherlands. van der Geest, S. and S.R. Whyte (eds). 1988. The Context of Medicines in Developing Countries: Studies in Pharmaceutical Anthropology. Amsterdam: Het Spinhuis. Whyte, S.R., S. van der Geest and A. Hardon. 2002. Social Lives of Medicine. Cambridge: Cambridge University Press. WHO (World Health Organization). 1985. ‘Future Use of New Imaging Technologies in Developing Countries’, Report of a WHO Scientific Group. Geneva: World Health Organization. Willems, D. 1995. ‘Tools of Care: Explorations into the Semiotics of Medical Technology’. Ph.D. dissertation. Maastricht: Maastricht University, The Netherlands.

262

Chapter 11

Healers and Scientists: The Epistemological Politics of Research about Medicinal Plants in Tanzania or ‘Moving Away from Traditional Medicine’1 Stacey A. Langwick

One June morning in 2006, Edmund Kayombo regaled me with the claims of people who travel at inhuman speeds without vehicles, of medicinal oils that prevent a person from being photographed, and of people rising up from the dead. After each anecdote he would say: ‘It can’t be proven, yet it happens’. The only social scientist among a team of medical doctors, chemists, botanists and ethnopharmacolgists at the Institute of Traditional Medicine (Taasisi ya Dawa za Asilia) in the Muhimbili University College of Health Sciences in Dar es Salaam, Tanzania, Dr Kayombo’s career rests on his ability to contribute to the process of subjecting therapeutic happenings to scientific proofs. Yet despite all his collaborative work, and his confidence in the benefits of coordinating social and natural scientific research, he maintains a strong interest in the ‘rituals’ of healers that cannot be ‘proven’. His individual work has examined the therapeutic importance of sacrifice,2 the role of ancestors,3 and the power of dreams.4 Over the past thirteen years that I have researched healing in Tanzania, I have had several opportunities to sit with Dr Kayombo, swapping stories about the medicines of healers with whom we have each worked. His refrain – ‘it can’t be proven, yet it happens’ – captures the friction that is generated by the Institute’s strong focus on laboratory research. This chapter takes the link between scientific proof and therapeutic happenings as an object of inquiry 263

Evidence, Ethos and Experiment

rather than the result of ‘successful’ research into traditional therapies. How and why do some therapies come to be proven and others do not? The answer to this question ties the content and epistemology of scientific research to the historical, political, ethical and technological regimes in which they are produced. As many scholars have suggested, the search for solutions to problems of knowledge about medicines, bodies, disabilities and misfortunes is also a search for solutions to problems of nation, economy and society.5 In light of such arguments, the rapidly intensifying investment in knowledge about traditional medicine deserves attention. Traditional medicine in Africa is currently enjoying a revival. The World Health Organization and the African Union are coordinating new forms of global and regional organization and stimulating greater national efforts to investigate plant, animal and mineral products. In line with these efforts, the Institute of Traditional Medicine for which Dr Kayombo works, as well as the Office of Traditional Medicine in the Ministry of Health, and the Traditional Medicine Research Unit in the National Institute of Medical Research have strengthened the bureaucratic management and scientific development of traditional medicine in Tanzania. These institutional organs have also expanded linkages between Tanzania and international partners devoted to the study and exploitation of medicinal plants. African interest in the development of medicinal plants dates back to the 1960s when many newly independent countries struggled to allocate a sufficient proportion of their hard currency reserves to the importation of pharmaceuticals (Akerele 1990). Yet, how to develop this interest has never been self-evident. The colonial persecution of traditional healers and the repression of many forms of traditional medicine pushed healing practices underground in some parts of sub-Saharan Africa (Feierman 1985). At independence, Tanzania, like many other newly independent African states, faced the enormous challenge of building the first national public health infrastructures (Ombongi, this volume). To respond to these desires under the constraints of limited resources, postcolonial governments began to imagine drawing on and elaborating popular therapeutic networks. In 1968, seven years after independence, the Tanzanian government explicitly recognized the right of traditional healers to exist through changes in the Medical Practitioners and Dentists Ordinance. Traditional healers and their herbal medicines were re-cast as human and natural resources. In 1969, the Chief Medical Officer in the Ministry of Health issued a directive ordering all Regional Medical Officers to collect information from healers on traditional therapies, including the medicines they used. While this directive does not appear to have been 264

Healers and Scientists

carried out, five years later, the Tanzanian government established a unit to investigate traditional medicines within the Muhimbili University College of Health Sciences. The Institute of Traditional Medicine is mandated to document healing practices, identify the medicinal qualities of plant, animal and mineral substances, and to develop traditional medicine in the service of national interests (Kilewo et al. 1987). Interest in traditional medicines through the 1980s waned. Then, in the 1990s, under the pressure of neoliberal reforms, the promise of traditional medicines captured attention once again. In 1989, the Tanzanian Ministry of Health expanded to include an Office of Traditional Medicine. This office helped to catalyse the formation of an association of traditional health practitioners, which was formally recognized in 1996. In 2002 the legislature passed the Traditional and Alternative Medicine Act into law.6 This act called into being the Traditional Medicine Research Unit in the National Institutes of Medical Research (opened in 2003) and the National Council on Traditional and Alternative Medicine (launched in 2006). These national efforts have been intertwined with regional and global initiatives from the beginning. In the 1970s, Tanzania’s strong commitment to non-aligned socialism led to multiple exchanges and connections with China. China’s more elaborate institutionalization of Traditional Chinese Medicine not only modelled one way that traditional medicine might be integrated with rural healthcare schemes emphasizing ‘barefoot’ doctors, but also some of the key leaders in efforts to develop traditional medicine in Tanzania in the 1990s were trained in China in the 1970s.7 More recently, the World Health Organization (WHO), through its own Traditional Medicine Unit, has compiled recommendations and guidelines regarding policy formation, research protocols and legal and ethical codes related to traditional medicine. The influence of these guidelines can be seen in the 2002 Tanzanian law as well as in the structure of government-supported efforts to develop traditional medicine. In addition, Tanzania joined other member states in the African Union (AU) to declare 2001–2010 the Decade of African Traditional Medicine. The AU adopted a plan of action for the implementation of the decade’s goals in 2003.8 To meet some of these challenges, Tanzanian experts together with colleagues from Uganda and Kenya formed the East African Network on Medicinal Plants and Traditional Medicine the same year (IDRC/ESARO Nairobi 2003). Research agendas, priorities and methods as well as legal frameworks are being developed, debated, shared and coordinated on national, regional and global levels. As each initiative formulates new policy, attempts to craft new legal frameworks, establishes and expands research programmes, and builds 265

Evidence, Ethos and Experiment

fruitful institutional collaborations, it also raises the question of which traditional medicine will be supported and elaborated (Last 1987). Evoked in an effort to meet national and international health development goals, these official articulations of traditional medicine aim to facilitate the integration of ‘traditional’ and ‘modern’ medical systems. Integration requires bringing scientists, healers and their medicines into the same frame. To date, this frame has been centred on the plant, animal and mineral substances used to prepare medicines. Research and development initiatives imagine the complementarity of traditional and modern medicine by articulating the power of healers, the meaning of medicines, the efficacy of therapies, and the ways in which various forms of healing did and could engage each other through medicinal plants. This privileging of medicinal plants is, at least partly, due to the ways that they can be made to travel and the possibilities that this travel evokes. Therefore, one aim of this chapter is to describe how plants travel in Tanzania and the way that these movements shape traditional medicine. I argue that plant substances travel as their attachments to healers and scientists are reconfigured. Scholars of science and technology studies have repeatedly shown that the circulation of scientific entities establishes new connections and evokes new relationships (Latour 1988, 1999; Callon 1999). Movement is, in essence, an effect of these new relationships. The effort involved in enabling medicinal substances to move around within and outside of Tanzania can be seen in the ongoing struggle over the place of healers in national, regional and global efforts to coordinate and elaborate traditional medicine. Among those who organize, fund and conduct research on traditional medicine, the role of healers is unsettled and often controversial. How are healers and their ‘rituals’ to be included or excluded from the national and international development of traditional medicine? Are, as Kayombo provocatively suggested, the official efforts to exploit the medicinal plants products in Tanzania and elsewhere in Africa actually ‘moving away from traditional medicine’? Is the distinctive ‘side’ (upande) of traditional medicine in Tanzania lost in the organizational structure and epistemological commitments of such centres of research as the Institute of Traditional Medicine? Answers to these questions are hotly contested because they suggest answers to other questions about the power, truth and utility of traditional medicine. The ways in which healers are responsible for their medicines varies as the practices considered traditional medicine vary. As a result, which traditional medicine is researched and institutionalized shapes debates over how healers should be held accountable by the state 266

Healers and Scientists

(and by consumers), as well as how the state, the scientific community, and industry should be held accountable to healers. To explore these questions I consider two broad sources: meetings about traditional medicine and research on traditional medicine. In the first half of this chapter, I examine how participants in three influential workshops concerning traditional medicine in Tanzania debate the position of healers. The first is the International Conference on Traditional Medicinal Plants in Arusha, Tanzania, from 18 to 23 February 1990. The second and third are more recent efforts to gather together ‘stakeholders,’ first regionally and then nationally. Tanzania hosted the initial Stakeholders Workshop on the Sustainable, Safe and Effective use of Medicine Plants in East Africa from 24 to 27 November 2003. To follow up on the commitments made at this 2003 regional meeting, a National Stakeholders’ Workshop on Traditional Medicine and Medicinal Plants was held from 4 to 6 August 2004 in Bagamoyo, Tanzania. These meetings sketch the changing relationship between healers and scientists as traditional medicine becomes further institutionalized in Tanzania. These changes have profound implications for what medical knowledge is as well as what medicines themselves are. The second half of this paper considers how these new kinds of knowledge and medicines are produced through scientific work. Epistemological commitments in research projects generate particular relationships between healers and scientists. I examine the structure of the Institute of Traditional Medicine through some of the key publications of the researchers there. These papers were selected from the bibliography produced on their website and in conversation with colleagues at the Institute of Traditional Medicine. They include ethnobotanical compilations,9 pharmacological studies,10 phytochemistry11 articles, and dietary analyses.12 Each of these studies explicitly articulates relationships between scientists and healers. In so doing, they also make medicines move. The lens of medicines and their travels, however, does not predetermine what these relationships are. The final section looks at the possibilities of a relatively recent project at the Institute of Traditional Medicine, entitled ‘Observation and Evaluation for Safety and Efficacy of Traditional Herbal Remedies Used in the Management of HIV/AIDS in Tanzania’. I consider the ways in which this research into therapies addressing the conditions of HIV/AIDS patients might (or might not) enable healers to negotiate new positions within the scientific community, as collaborators in research rather than merely as targets of research.

267

Evidence, Ethos and Experiment

Jockeying for Position In 1989, responsibility for the integration of Traditional Medicine in Tanzania into the national healthcare system moved from the Ministry of Culture to the Ministry of Health. One of the first tasks of the newly established Office of Traditional Medicine in the Ministry of Health was to coordinate Tanzania’s role as host of the 1990 International Conference on Traditional Medicinal Plants. Funded and organized by the South-South Commission, the WHO and the United Nations Development Programme, this conference attracted participants from Africa, Asia and Latin America. The five-day programme was opened by Dr A.D. Chiduo, the Minister of Health for Tanzania; Ali Hassan Mwingi, the President of the United Republic of Tanzania; Julius Nyerere, the Chairman of the South Commission and Dr G.L. Monekosso, the Director of the World Health Organization (WHO) Regional Office. This workshop marked an intensification of interest in and investigations of traditional medicine. In fact, for Tanzania, it served to set the stage for the range of debates and approaches that would come to shape the national development of traditional medicine. Later workshops have striven to establish strategies, techniques and institutional arrangements that would promote the forms of development first discussed in 1990. It is significant, then, that this conference focused on traditional medicinal plants. The conference’s focus on medicines and the positioning of plant materials as active protagonists in the development of traditional medicine left open the question of healers’ roles in this research. The majority of papers presented at the conference elaborated attempts to build inventories of medicinal plants or to analyse the activity of such plants during laboratory tests. In such studies, healers were typically represented as generic figures (x ‘is the key plant among six plants which are used in the preparation of an antiasthmatic drug regime by traditional doctors’ [Hermansson et al. 1991: 38]). At times they even disappeared behind ‘traditional uses’ (‘In Kenya the two species Warburgia ugandesis and W. stuhlmannii … are used traditionally as medicines against many ailments’ [Kioy 1991: 64]). Only once does a healer appear in a scientific paper as an individual interlocutor, and this is only very briefly before being displaced by the plant product itself (‘A traditional medicine was identified through interaction with a traditional healer, Mr Charles Obuya of Rangwe, South Nyanza’ [Aluoch et al. 1991: 9]). Traditional healers, however, did participate in the conference beside the policy makers and scientists (including botanists, pharmacologists, chemists and 268

Healers and Scientists

physicians).13 A few even made presentations. Only one authored a paper that was printed in the proceedings, however; and he is not listed as a ‘healer’ but rather as the Director of Operations and Lab Science Systems at African Medical and Technological Labs and Stores Services Centre. An uncertainty about the position that healers should have in national and regional efforts to promote Traditional Medicine ran through the conference proceedings – an undercurrent in all exchanges and emerging explicitly (often unexpectedly) in some. The juxtaposition of the following arguments best represents the poles of the debate as it occurred during the 1990 conference. The first was asserted by a group of scientists from the Institute of Traditional Medicine and the Department of Pharmacology at the Muhimbili Medical College. We are hereby calling scientists of all professions to come to the rescue of our culture which is useful to none, except ourselves … We believe that the duty of experts involved in traditional medicine research is to provide [a] scientific basis of the practice of our herbalists aimed at upgrading, improving or authenticating their practices, pointing out, without fear, the bogus medicinemen and the fake traditional healers and assure medical practitioners that the drugs made from indigenous plants are as good as modern ones. Thus by doing this the scientific community shall be making a very useful contribution to therapeutic innovation in primary health care in Africa. (Mutayabarwa et al. 1991: 140) The director of the Antiquities Unit in the Tanzanian Minister of Labour, Culture and Social Welfare made the following contrasting argument. [W]hen undertaking our various researches in traditional medicine, we should be aware of the fact that by getting this pool of knowledge from our villagemen, a process which to them appears to be unidirectional, we are in fact stripping them of their special place in the society and hence their status. It is largely because of this fact that the author advocates for cooperation and collaboration in both research and using traditional medicine by both parties. (Waane 1991: 214)14 Are scientists to be the mediators of truth and the authenticators of culture? Or are they to be collaborators, cooperating with healers in the name of different goals? 269

Evidence, Ethos and Experiment

In 1990, most participants seemed to be grappling with what these positions might actually mean in practice. For instance, if there are to be forms of integration between traditional and modern medicine, which rely on cooperation and collaboration with healers, then does the quality of collaboration not depend – at least in part – on the quality of the healer? How might ‘bogus medicinemen’ and ‘fake traditional healers’ be avoided? When the Director of the Traditional Medicine and Drugs Research Centre at the Kenya Medical Research Institute raised the question of how ‘real’ traditional medicine practitioners could be identified, Prof. Koumare, a WHO officer from the Africa Regional Office, suggested that ‘communities’ know those who are the ‘real’ traditional healers among them. A selection process could be devised and identity cards could be issued (Ministry of Health, Tanzania 1991: 292). The Programme Manager of the WHO Traditional Medicine Unit in Geneva suggested that traditional healers needed to ‘group themselves’ in order to ‘interface with the formal system’ (Akerele 1991: 7). The Tanzanian healers at the conference held a special session to discuss the possibility of building a healers’ association. Mr Shauri, the healer/lab director mentioned above, served as the chairperson for this meeting. Should the validity of a healer’s practice be judged by scientific laboratory tests, community nomination or an association’s membership criteria? The answer to this question was not solved in this 1990 conference. In raising it, however, the participants imagined the possibility of very different relations between scientists and healers. In this discussion, each of the participants, despite their differences deflects the obligation to discern who counts as a healer (no less a ‘good’ healer). They alternately put this evaluation in the hands of the(ir) ‘community’ or a professional organization or a certification process. In so doing, each suggestion imagines the establishment of a new infrastructure through which healers may be made or may make themselves. As these various processes are implemented, they deserve further ethnographic analysis (see, for example, Pigg 1995, 1997a, 1997b). I would like to suggest another point here, however. The deflections of the participants illustrate the ways these broader debates strategically avoid the more complicated issue of what ‘traditional’ therapeutic knowledge consists of. Many people in sub-Saharan Africa know medicinal plants and use them to care for themselves, their children, other kin and neighbours. In fact in one area of Kenya women were found to know as many plants used for medicinal purposes as healers (Geissler et al. 2002). Although my own work in Tanzania does not involve such careful or systematic ethnobotanical studies, my experience in Tanzania resonates with this claim. As I was trying to learn to 270

Healers and Scientists

identify and name plants used by healers with whom I studied, I often turned to neighbours or friends for repetition, clarification and assistance. At one point, when I lined the walls of my house with large sheets of paper to which I tapped the leaves and various parts of plant used for medicine, the help of literate friends who corrected my spelling in Kiswahili, Kimakonde and other languages spoken in the area was invaluable.15 Such knowledge was not held evenly, but some familiarity with these plants was held widely. Yet, in what sense does the ability to name or spell the names of plants used for medicines capture therapeutic knowledge? Geissler and Prince (2009) have shown that no medical knowledge generated in therapeutic practice can be reduced to the innate properties of plants. As women in Luoland (Kenya) debated treatments and medicines for children ‘more than a choice of tool was at stake’. Rather, healing was about ‘trying’ and the care, responsibility and affection that such action entailed. They argue that medicinal knowledge and therapeutic transformations ‘are not so much located within entities as between them – they emerge from encounters and their traces, rather than reside in things’ (2009: 1). For healers in Tanzania, the capacities of medicines similarly exist in relations of practice. Legitimating only the aspects of healers’ knowledge of medicines related to their ability to identify plants reduces their work to that of a dispenser of pharmacologically efficacious substances. Healers in Tanzania, like women in Luoland, catalyse the transformative capacities of medicines by putting them in relation – with particular bodies, needs, symptoms, intimacies and so on. Yet, healers’ efforts coordinate transformations between far more entities than non-specialists’ efforts do. Many healers in Tanzania learn about medicine through dreams and visions, which facilitate their engagement with ancestral shades and a variety of non-human agents. They often collect medicines at the impetus of these entities, which are also held responsible for ‘bringing’ patients to them. Healers consolidate, even embody these relations. As a result, their hands are said to have power. While lay people in Tanzania might know what plants they need to make a particular remedy, they will still (at times) choose to go to a healer to receive treatment rather than collecting the plants themselves. Medicines conceived, prepared and administered by a healer are different, more than the sum of the plants collected and the touching and attention given (Langwick 2007) – a fact that has implications in current debates about intellectual property. Perhaps for this reason Mr Shauri’s own presentation did not focus on how to weed out unsavoury healers or elaborate the favoured structures 271

Evidence, Ethos and Experiment

through which true healers might be discerned. He took for granted that healers range across a spectrum of abilities. There are ‘bad’, ‘good’ and ‘better’ healers. After all, healers themselves participate avidly in these evaluations. Without denying the complicated issue of credibility, Mr Shauri turned the discussion to traditional practitioners’ relationship with biomedical practitioners. He suggested that more equal collaboration was needed between credible biomedical and traditional practitioners. Assuming that both credible traditional and biomedicine practitioners exist, he challenged the bias of medical personnel who advocate that traditional healers refer complicated cases to the hospital and yet do not advocate that medical practitioners refer complicated cases to traditional healers (Ministry of Health, Tanzania 1991: 291). He then continued his talk by elaborating a number of herbal plants used as medicines. In response, he was asked to talk about the ways in which he standardized his methods and dosage. Mr Shauri discussed the difficulty of efforts to standardize treatments, even so he admitted to collaborating with the Institute of Traditional Medicine at Muhimbili in an effort to begin to establish standard dosages of his medicines for skin ailments (Ministry of Health, Tanzania 1991: 292). For Mr Shauri, while the extent to which medicines could be submitted to the discipline of scientific practice was uncertain, such translations were provocative and a valid topic for experimentation. Tied to the preparation and administration of medicinal substances, questions about dosage strike at the heart of the relationship between healers and their medicines. They are also at the heart of scientists’ interventions into the relationship between healers and their medicines. The regimes of measurement that determine pharmaceutical dosages establish medicines and persons as separate entities. The properties of medicines are seen as affecting the properties of bodies. The former can hold both danger and promise for the latter. Dosing then consolidates the forms central to scientific efficacy. Interestingly, some healers express interest in dosages, and even more in the aesthetics of dosing. Measurements – three times a day, one teaspoon, or one cup – are powerful even when, or perhaps especially when, disassociated from the structures that produce them. Therefore, several times during the 1990 conference, the issue of how healers control their dosages emerged both as a point of anxiety and of potential collaboration (Ministry of Health, Tanzania 1991: 300). Nkunya and his colleagues, for example, argued that: Unfortunately, the use of crude plant extracts, without any scientific evaluation, could lead to serious complications. Ineffective drugs 272

Healers and Scientists

could be used just as a matter of belief or tradition; under/over-dose could be taken; highly toxic drugs with short term, long term, or cumulative effects could be prescribed etc. The last two effects, however, are much more difficult to recognise than the others, and hence potentially more serious. In addition to these, the preparation, handling and storage of the drugs could lead to decomposition or transformation of the hitherto active constituents to ineffective and/or harmful products. Thus there is a need to evaluate and establish a scientific rationale for the use of the traditional medicinal plants, through chemical, pharmacological, toxicological and microbiological studies. (Nkunya et al. 1991: 101) Such critical evaluations of healers’ unscientific practices not only situated healers in relation to scientists, but also revealed the ways that scientists were trying to justify their own role in the promotion of traditional medicine. Science was proposed as a way to mediate danger, ensure efficacy and rationalize distribution. In the brightening international spotlight, scientists strove to assert the value of their own work and to make that work central to national and regional projects. Yet, scientists are themselves disciplinarily varied and methodologically diverse. Different research questions seem to propose, even require, different relations between healers and scientists. Not everyone at the conference wanted to focus collaborations between healers and scientists on issues of toxicity, chemical activity and appropriate dosage. Dr Bhat critiqued the ‘specialized nature’ of many of the papers at the conference and advocated for more attention to the interaction between drugs and patients (Ministry of Health, Tanzania 1991: 296). Dr Mapunda echoed this concern as he wondered at the majority of his colleagues who insisted on the isolation of active ingredients, rather than the examination of the ways in which traditional medicine is improving or protecting the health of citizens (Ministry of Health, Tanzania 1991: 296). Professor Elmi (Head of the Department of Morphology and Pathology at Somalia National University) argued that systematic evaluation of medical plants has not been particularly fruitful and suggested rather a more ‘applied’ approach in which healers could be supported to treat patients with certain medicines after toxicity studies (299; also Mshiu et al. 1991: 180). By questioning the centrality of pharmacological and phytochemical studies of plant medicines to all national and regional projects developing traditional medicine, these participants opened up a debate over the ways in which traditional 273

Evidence, Ethos and Experiment

medicine might be made compatible with science – its instruments and its practitioners. The shape of traditional medicine as a national and regional project has been forged through these debates over which scientists will define the agenda and which research will be considered essential. A Tanzanian healer turned the discussion from the importance of standardizing the dosages of medicine to the desirability of manufacturing traditional medicines. She hoped that collaborating with scientists would enable the large-scale production of herbal medicines in Tanzania. To underline the benefits of such a focus, she claimed to have successfully treated AIDS, as well as a number of other diseases (Ministry of Health, Tanzania 1991: 297). Dr Antonio and George Washington (pharmacists in Angola and Brazil, respectively) promptly challenged her to reveal the names of the plants she used in her treatments. The Sierra Leonean minister of health, however, insisted on a ‘respect’ between healers and doctors that shielded healers from having to reveal their formulations. He went on to argue that traditional medicine is also a ‘science’ (Ministry of Health, Tanzania 1991: 298). Professor Koumare from the Division Médecine Traditionnelle in Mali supported the notion that traditional healing was its own ‘system’. He suggested that ‘true exchange’ required that traditional healers use ‘local terms’ rather than ‘medical terms’ when describing their work (Ministry of Health, Tanzania 1991: 298). In 1990, participants still debated which kinds of healing knowledge could be translated into scientific knowledge and who would be competent to perform these translations. At a different moment in the conference, two researchers from the Institute of Traditional Medicine in Tanzania suggested that rather than Traditional Medicine being a different science, the traditional healer might be best thought of as ‘an accomplished traditional botanist’ (Mahunnah and Mshiu 1991). Are healers to be positioned as scientists, proto-scientists or inventors that can use science to support the mass manufacturing of their products? During the International Conference on Traditional Medicinal Plants in Arusha, there was no consensus on an answer to this question. Less explicit arguments about healers’ contributions to the process of revitalizing and rationalizing traditional medicine hinted at the need for consensus, however. In particular, claims about who had the rights to any profits that might accrue through the development of herbal medicines. Prof. Abondo Antoine, Director of the Institut de Recherches Medicals et des Etudes des Plants Medicinales in Cameroon, raised the question of how a traditional healer should be protected if ‘his plant’ proves to be commercially viable (Ministry of Health, Tanzania 1991: 294). In these debates, the 274

Healers and Scientists

position of healers in relation to science was not as important as the position of healers in relation to the nation. Mshui (1991: 136) argued that ‘developing countries should be compensated for their traditional knowledge and biological storehouse’. Mshui, Sayi and Sarungi (1991: 178) advocated for mechanisms that would ‘safeguard sovereignty over their genetic resources or for conservation of tropical products and traditional knowledge of the indigenous people’. What is to be the connection between healers’ claims to their medicines and the state’s claims to sovereignty and compensation? The issue of healers’ ownership of medicines (or their intellectual property rights) has become even more central to discussions in workshops and conferences over the past five years.

Revitalizing and Rationalizing Heritage16 By 2003, as the first regional Stakeholders’ Workshop in Arusha was being held and the Eastern Africa Network on Medical Plants and Traditional Medicine was being launched, the relative positions of healers and scientists in Tanzania had been consolidated. On 5 December 1995, healers in Tanzania registered the first major national organization of traditional healers (CHAWATIATA) with the government’s support. Other associations of healers also emerged in the subsequent years. These groups provided institutional partners with which policy makers and researchers could engage. In 2002, the Tanzanian legislature passed the Traditional and Alternative Medicine Act calling into being a more elaborate institutional structure concerning traditional medicine. These policy initiatives, and the workshops that were designed to implement them, strove to articulate traditional medicine as complementary to biomedicine. In this context, healers’ roles came to be increasingly defined by their ability to help perform a traditional medicine that is complementary to biomedicine. Individual healers negotiated their position through their participation in the bureaucratic management and scientific development of a national traditional medicine. Their success and power evolved from creative and strategic translations of traditional medicine into a healthcare system dominated by biomedicine. As the title of this 2003 workshop indicates, when traditional medicine is scaled up to regional and national levels, healers represent only one set of ‘stakeholders’. The structure of delegates at the workshop established healers’ claims on traditional medicine as equivalent to, even at times in competition with, the claims of researchers, scientists, policy makers, private entrepreneurs and journalists. 275

Evidence, Ethos and Experiment

Under these conditions, associations of traditional healers were increasingly being asked to serve as units of bureaucratic control while each research or clinical project was being negotiated with individual healers.17 Furthermore, research – rather than the cooperation of traditional and modern healthcare practitioners – came to be heralded as central to the project of integration. In his opening speech, the Deputy Minister for Health ‘emphasize[d] the importance of research as a necessary tool for evidencebased promotion of traditional medical practice. Research on traditional healing practices, traditional medicinal products and conservation of vital plants’, he argued, ‘is the foundation for scientific development of traditional medical practice’ (Mwinyi 2003: 17).18 In his welcoming remarks, Professor Mahunnah (2003: 10) representing Muhimbili University College of Health Sciences called on participants in the workshop to explore ‘how best we can exploit the rich folklore and medicinal plant resources that Africa is endowed with [in order] to alleviate the suffering of our people. This bio-resource should also be used as a vehicle to promote sustainable livelihoods of the communities through strategic and sustainable utilization of biodiversity’. As if in response, the Institute of Traditional Medicine described its mission as investigating the ways in which traditional medicine could fill ‘existing gaps’ in the national health service. A paper by four long-standing scientists at the Institute argues that the ‘potential of traditional medicine’ is to be articulated through ‘direct scientific research findings’ (Uiso et al. 2003). A consensus emerged among those considered national and regional stakeholders that scientific research would provide the language through which traditional medicine’s complementarity would be forged. During the first discussion session of the 2003 Stakeholders Workshop, Professor Mukiama, from the University of Nairobi Kenya, argued that the development of a [pharmaceutical] product is a process and that the source of information is traditional healers … The issue is that traditional healers are over ambitious on products. They should know that there is a process involving studies on toxicology, efficacy, and safety of the products being developed. It is also important to know that the benefits accruing from a drug are for the community and not for the individual, since the information is community knowledge. (178) While healers are ‘the source of information’, Mukiama asserted that the information itself is ‘community knowledge’. He implies that traditional therapeutic knowledge is less about the skill or innovation of a healer, and 276

Healers and Scientists

more about the materialization of social history and communal practice. This argument formulates healers’ roles in national development as custodians of common, historical knowledge about plants. By asserting a qualitative difference between healers (individuals) and their medicinal knowledge (communal), Mukiama’s argument frees ‘knowledge’ and ‘plants’ to travel, while it restricts the travel of healers. This double move is what makes such arguments so powerful and salient. With no inherent tie to healers, medicines may engage other actors – most importantly for Mukiama, scientists. He asserts his privileged position for scientists by linking their technological practice with ethical practice. ‘Over-ambitious’ healers concerned with individual benefits require stricter bureaucratic management so that those conducting ‘studies of toxicology, efficacy, and safety’ can develop medicines for the common good. As community knowledge is scaled up to serve as a national resource, the very body of the healer becomes the ground for extraction and science the image for unrestricted exchange.

The Secrets of Complementarity To make traditional medicine complementary to biomedicine, regional and national initiatives are re-casting not only the role of healers but also the power of healers. Both healers’ value and their efficacy are being evaluated in terms of their contributions to the exploitation of medicinal plants for the gain of the state. In this frame, healers may be good or bad sources, but they are no longer nodes of relation between the present and the past, the living and the dead, the human and the non human, the material and the immaterial. This re-casting has direct implications for healers’ claims to their medicines and any profits their development might garner. In this context, participants hotly debated the ‘secrecy’ of healers. For some, ‘secrecy’ is an act of hoarding community resources for personal gain. For others, ‘secrecy’ is an effect of demanding that specialized knowledge evoked through intimate relationships among a healer, his or her ancestral shades and a range of spirits be treated as communal knowledge. To examine this debate, I quote, at some length, from one of the discussion sessions as they were recorded in the workshop proceedings. Dr Orwa (Kenya Medical Research Institute) reported that the Kenya Medical Research Institute takes herbalists as collaborators. They are taken on board in all processes including co-authorship in publications emanating from joint research work. 277

Evidence, Ethos and Experiment

Mr Njoroge [Institute of Herbal Medicine, Kenya] asked Dr Orwa to give clarification on secrecy of research results and commercialization and their immediate use by communities. Dr Orwa responded that so far there is no secrecy. Dr Sekagya [PROMETRA, Uganda] emphasized that knowledge is very important to traditional medicine practice. Hence, there should be a mechanism to safeguard and reinforce secrecy. Dr Mnaliwa (Ministry of Health, Tanzania) commented that the problem of secrecy is due to lack of clear policy and intellectual property rights (IPR) on traditional medicine and medicinal plants. Therefore, a legal framework is required that will address these pertinent issues. Prof. Mahunnah (Institute of Traditional Medicine, Tanzania) stressed that it is important to realize that global bodies handle IPR issues, and that it is vital to make progress on this issue. There is urgent need to develop regional IPRs (based on Sui generis) so that there is a common position during discussions on the global front. This is the strategy adopted by the African Union in the AU (OAU) Model Law. (179) The comment that initiated this topic in the first discussion session was the description of a novel solution to the uncertain terrain of healers’ intellectual property rights. Dr Orwa said that in KEMRI healers co-author papers. They are not merely ‘informants’ but rather collaborators. As such, healers have claims to the authorship of particular plant extracts and compounds. Although the international criteria for ownership of plant materials remains unsettled, if it comes to be based on current legal understandings that restrict patentability to ‘novel’ (as opposed to existing) knowledge then authorship of scientific papers might establish evidence of a healer’s role in the innovation of particular medicines or herbal products. In response, a colleague asked Dr Orwa about the mechanisms through which his and his research team’s results are kept confidential or secret in order to hold out the possibility of exclusive rights to future commercialization. Even as Dr Orwa claimed that at this stage there is no effort to keep their research results secret, a representative of PROMETRA – an NGO ‘dedicated to the restoration of African traditional medicine and indigenous science’ – jumped in to advocate for secrecy. A new tension was introduced. The discussion slipped quickly from the co-authorship of scientists and healers to the secrecies necessary to safeguard commercial 278

Healers and Scientists

interests, to the secrecies necessary to safeguards healers’ value. Dr Sekagya’s comment suggested that he is savvy to the world of drug development where knowledge is a commodity and therefore wealth comes from controlling the flow of knowledge. Co-authorship may not adequately express or manifest healers’ control of their own medicines and medical knowledge. For one thing, claims made through the publication of scientific papers sustain precedents privileging ‘scientific’ transformations as ‘novel’ in contrast to a range of transformations made in therapeutic practice. The significance of authorship of any kind was debatable to some participants, however. The head of the Tanzanian Ministry of Health’s Office of Traditional Medicine spoke as a representative of the state when she argued that legal rights can safeguard commercial interests. In suggesting that legal rights will protect healers, she assumed that their interests are market-driven. Such legal rights, Muhunnah reminded his colleagues, are not being written only in the interests of the healer or of the nation but rather there are international and multinational agents who have a stake in the way that intellectual property rights are conceived in Africa. Alliances are important. He left aside the issue of how healers might negotiate their position within these regional and global alliances. A year later, interest in the kind of secrets that generate wealth drove the 2004 National Stakeholders’ Workshop on Traditional Medicine and Medicinal Plants in Bagamoyo, Tanzania. The establishment of councils and working groups and the discussion of Memorandums of Understanding (MoU) between collaborating parties dominated the agenda of this national meeting. In other words, as national-level officials considered the regional objectives of the Eastern Africa Network on Medical Plants and Traditional Medicine, they prioritized deciding who would be in the know and formalizing the hierarchies of knowing. Dr Kayombo presented the Institute’s plan to use MoUs to clarify and formalize ‘the intentions, obligations, and mutual responsibilities’ of various parties collaborating on research of traditional medicines. Three model agreements were discussed: (1) between the University of Dar es Salaam and universities in other countries; (2) between the Institute of Traditional Medicine and the Forestry Research Institutions; and (3) between the Institute of Traditional Medicine and Traditional Health Practitioners. Healers are again only one of the many actors with a stake in the development of traditional medicine. While the details of the first MoUs involving Traditional Health Practitioners are still under negotiation, they promise to further articulate the relationship between sci279

Evidence, Ethos and Experiment

entists and healers needed to move plant, animal and mineral products from therapeutic knowledge to scientific study. The discussion and debate among participants in the three primary workshops about traditional medicine in Tanzania concerned, most broadly, the organization of efforts to produce new medicinal knowledge. The therapies available to scientific study (and therefore open to elaboration through scientific proofs) are limited by: (1) national, regional, and international goals to integrate traditional and modern medicine; (2) the strength of scientists in articulating themselves as central to the project of integrating traditional and modern medicine; and (3) the emergence of medicinal plants as the ‘obligatory passage point’ (Callon 1999) through which integration is imagined. By defining ‘stakeholders’, explicating the role of scientific research and establishing the centrality of research on plant matter, the three workshops discussed here have demarcated a space for the problematization of traditional medicine. By dominating this process of problematization, scientists have consolidated their position in the development and integration of traditional medicine over the past fifteen years, and simultaneously narrowed the role of healers by casting them as mere manifestations of Tanzanian heritage. Framing the development of traditional medicine as the revitalization and rationalization of African, or in this case Tanzanian, heritage transforms healers’ relationships with their medicines, displacing them in favour of scientists’ relationships with plant, animal and mineral substances. Debates about secrecy and openness are also debates about the threatening consequences and possible benefits of scientists separating healers from their medicines. The agency of scientists, healers and medicines as well as their relationships with each other are, however, configured and re-configured in research practice. The organization of multidisciplinary research into traditional medicine and the methodological choices of the researchers constitute these actors and their relations. The next section turns to research conducted by scholars at the Institute of Traditional Medicine. Because the Institute has established itself as a key coordinating body for regional initiatives related to traditional medicine, these scholars are the same ones who played critical roles in the workshops above.

280

Healers and Scientists

Relocating Knowledge, Instituting Traditional Medicine Located at the Muhimbili College of Health Sciences, the Institute of Traditional Medicine articulates its centrality to national healthcare goals by imagining the integration of traditional and modern medicines as the transformation of the historical knowledge of healers into a practical tool for hospitals. The Institute strives to move traditional medicine from the purview of healers to the administration of doctors. This section explores how healers are conceptualized and engaged in the course of research into medicinal plants at the Institute of Traditional Medicine. In many healing practices in Tanzania, a therapy emerges at the intersection of the healer, various unseen agents including majini and ancestral shades, the person seeking treatment and the substances available to treat the malady. Medicines, maladies and healers come to meet, know and define each other in this relationship (Langwick, 2011). The embodied nature of this therapeutic knowledge confounds attempts to duplicate medicines and practices across time, place and patients. Researchers at the Institute of Traditional Medicine, therefore, intervene in these relationships in the hopes of translating some of the substances used by healers into objects of scientific investigation. Sociologists such as Kayombo may identify the initial stages of therapeutic work as conversations with ancestors, dream-like visions of medicines or spirit possession. In contrast, ethnobotanists offer plants as entities with (potential) inherent medicinal value. The travels of plants from sociologist to ethnobotanist to chemist to pharmacologist structure work in the Institute and also order the making of new ‘traditional’ medicines. Ethnobotanical work – both through field studies and in its place in the hierarchy of research on traditional medicines – establishes plants as an origin point. They form a baseline, a source, a moment from which (scientific) elaboration may begin. Therapies are reduced to plants with therapeutic value. Scientific investigation de-contextualizes therapeutic knowledge, so that plants can be re-contextualized within international biomedical agendas, within the laboratory and within national healthcare services. This process of de- and re-contextualization happens gradually as the medicines move through various stages of investigation, as they are disassociated from healers and associated with scientists. New knowledge about traditional medicine comes to rest on the new social relationships of plants with scientists. Both the Institute of Traditional Medicine’s focus on plant materials and the kinds of science that are deemed central to their research are evident in 281

Evidence, Ethos and Experiment

their organizational structure. The Institute is currently comprised of three departments: (1) Biology and Agronomy; (2) Natural Products Chemistry and (3) Pharmacology and Toxicology.19 Biologists (including botanists), chemists and pharmacologists direct the most important research conducted at the Institute. Participation of, or reliance on, healers for aspects of this research reveals the position of healers in the development of traditional medicine and the quality of their relationships with scientists. Healers typically figure in two moments of research conducted at the Institute: (the beginning) when plants to study are being identified and (the end) when research results are being disseminated. Specific plants come to the attention of scientists in two ways, through field investigations and literature reviews. One goal of the Institute is to contribute to the existing literature cataloguing ‘indigenous’ medicinal plants. Efforts to document and store ‘information and materials pertaining to all aspects of traditional medicines’ in Tanzania involve employing healers as informants (‘sources of information’). Ethnobotanists working in the Institute have published five main articles cataloguing field collection data from traditional healers and traditional birth attendants. Although ethnobotanists often record when and how healers harvest their plant medicines, these studies generally focus on the identification and systematic cataloguing of plants. Efforts to document therapeutic practices more broadly are separated through disciplinary specializations from efforts to document various plant materials that are used therapeutically. Kayombo, the sociologist at the Institute, records rituals and meanings while the ethnobotanists sort through the practices involved in learning about a medicine, collecting it and preparing it. They note only those ‘cultural features’ that are relevant to the future collection by staff and cultivation in one of the Institute’s five gardens. Other botanists may assist in the identification and preparation of samples for the herbarium located on the grounds of Muhimbili, as well as any future cultivation and collections from forests, roadsides and gardens. In conversations with colleagues at the Institute, I have heard healers referred to as the first ‘screening process’. Indeed, ethnobotanists interested in traditional medicine suggest that it is methodologically advantageous to identify plants for possible study through healers because their use has presumably correlated with symptomatic changes (Johns et al. 1995). Ethnobotanists recast plants in families of flora, translating local names to Latin ones and connecting them with relatives in many parts of the world. While the new kinships that they highlight establish the ground for 282

Healers and Scientists

the plants’ new social relations, this discipline is in itself insufficient to meet the Institute’s goal of developing traditional remedies to be administered by biomedical doctors. As the study of a plant intensifies, ethnobotanical field studies lead to chemical investigations, which lead to pharmacological research. As one of the first translations of healers’ medicines into plants for scientific study, the value of ethnobotanical work is in the ways that it is picked up and used by scientists with other disciplinary specializations. In phytochemical and pharmocolgical studies, healers garner only the passing statements referred to earlier in this chapter. Broad generalizations such as ‘traditional healers use x plant for various complaints’ are common lines in the introductions of these studies. Used only to justify the authors’ choice of a particular plant for study, healers appear as generic figures, living outside of place, space and time. One healer’s or some number of healers’ use of a plant comes to be a citation in a rationale for study. Botanists or their trained assistants, not healers, collect the actual plant materials on which these experiments are conducted. The WHO has invested in designing an international protocol for the collection of medicinal plants that will produce useful plant specimens for further study. The plant materials on which these phytochemical and pharmacological studies are conducted are linked with the activities of healers through specific forms of botanical identification and classification. This link allows the ‘same’ plants used in healing practices in one part of Tanzania to be collected in various parts of the world. It also allows studies on similarly identified plants to be compared, such as those on the much-used ‘neem tree’ common in the repertoire of herbal treatments in Tanzania, India and many other tropical countries. In the possession of the chemists and the pharmacologists these plant materials are pounded, spun, squeezed, filtered, freeze-dried and tested on mice. In Latourian fashion, scientists and their lab technicians make selected material act with, in and on a variety of entities. In this process, ‘whole’ plants are displaced by chemical compounds. These compounds emerge through relationships with acids, gels and ethers, expressing compatibilities or incompatibilities and showing ‘pharmacological activity’. Indeed, the chemists and the pharmacologists specializing in pharmacognosy in Tanzania articulate their work as ‘the search for new biologically active drugs … aimed at the enrichment of our present therapeutic arsenals and the control of newly found pathogens’ (Mbwambo et al. 1996: 335). Drugs are always drugs for something. They emerge in the context of questions about treatment. Plant extracts, for example, may be tested for 283

Evidence, Ethos and Experiment

antibacterial, antifungal or antiparasitic activity. Therefore, it is through these studies that plants develop relations with disease entities as well as with scientists, lab technicians and the numerous technologies with which they work. By focusing on plants (rather than on healers or on treatments), the Institute locates scientists’ interaction with healers as background preparation for scientific investigation. Healer’s critical transformations of plant materials as they are collecting medicines and preparing remedies and administering treatments occur outside the time of scientific investigation. Therefore, individual scientists, or the Institute, or the national government can assert that any ‘innovation’ of the plant material occurred in the laboratory. This also highlights both the radical and the conservative aspects of KEMRI’s move to have healers co-author scientific papers. Co-authorship acknowledges healers as innovators, but also reinforces the location of innovation as the laboratory. This perspective best situates the claim of the PROMETRA representative above who suggests that secrecy is the only way that healers can maintain their work as specialized. Secrecy, he suggests, holds a space outside of the time of scientific studies for innovation and transformation. While the best way for various actors to navigate the asymmetries of power is debatable, it is clear that these asymmetries come to be articulated through the methodology of research projects and sedimented in research institutions through the multidisciplinary organization of their research projects. Transforming healers’ knowledge of medicines into scientists’ knowledge about medicines raises questions of power, utility and truth.20 Because healers disappear from the activity of the Institute during the time that plants are transformed, they can be brought back at the end as an audience for the findings of the research. Plants that came to the attention of scientists because they were being used by healers are now introduced to healers. The reintroduced plants or medicines come back to the healers with their new social ties – their botanical kinships, their extraction procedures and their active ingredients. At times these new relations might suggest ways that the healer can improve or modernize his or her own practice. At other times the Institute starts manufacturing and bottling medications or herbal products that healers or patients themselves might access. While the research strategies I have described dominate the Institute’s work and the process of modernizing traditional medicine in Tanzania, they are not the only ways of imagining the ‘development’ of traditional medicine and coordinating it with biomedicine. Other relationships be284

Healers and Scientists

tween healers and scientists (as well as the state) can be, and are being, imagined. Healers in Tanzania, for example, are organizing to establish their own botanical gardens in the name of conservation;21 they are calling for the establishment of colleges of traditional healing;22 and they are setting up their own laboratories to produce medicines distributed in their own network of clinics.23 In each of these other options healers maintain an intimate and elaborate relationship with their medicines. Furthermore, changes in epidemics give rise to novel research strategies within the Institute which offer moments of possibility where old distinctions and hierarchies may be maintained or reconfigured. As Schumaker (this volume) illustrates, ‘research cultures’ potentially draw ‘different groups of people … together in a common project, despite their often radically different understandings of the project’s meaning’. A new research project at the Institute of Traditional Medicine allows us to begin asking: what possibilities might be opened up if healers move with their plants?

Changing Relations in the Era of AIDS During my recent conversation with Dr Kayombo, between his refrains of ‘it can’t be proven, but it happens’, a woman walked into his office. As a way of greeting, she asked him if he recognized her. After a moment of looking hard into her face, he smiled and inquired if a particular person was her father. She had gained weight. She beamed as she returned his smile. As a person diagnosed with AIDS, this women was participating in a study under way at the Institute. On this morning in June 2006, she came to show Dr Kayombo how well she was doing and how much healthier she looked, before she travelled back to her home in the south-western part of the country. This on traditional medicines that healers use to treat HIV/AIDS – Mradi wa Utafiti wa Dawa Asilia yainayotumiwa na Waganga wa Jadi kutibu maradhi ya Ukimwi Tanzania – involves the ongoing participation of healers who are actively treating people with HIV/AIDS. This study requires extended engagement between healers and scientists and significantly greater cooperation than any other project of this size at the Institute. Furthermore, healers are moving with their medicines; they – not just their plants – are collaborators in the research project. All aspects of their expertise – the ways that they know their medicines, collect them, prepare them and administer them as well as the ways that they engage and evaluate patients – will shape the results of this study. 285

Evidence, Ethos and Experiment

The extent of the impact of AIDS in Tanzania has provoked this methodological experiment. The inability of the state to provide the growing number of afflicted people with expensive pharmaceutical cocktails has led the government to explore the effects of traditional medicines. In the 2002 parliamentary session, the hopeful search for AIDS drugs was invoked as a rationale for supporting the Traditional and Alternative Medicines Act that was then under consideration. In this environment, where biomedically approved drugs are not available for large numbers of people who need them, ethical concerns shifted. The testing of traditional medicines used by a significant percentage of the population came to be framed as an imperative. The fact that people living with AIDS in Tanzania seek out traditional healers of their own accord and that they use the medicines administered by these healers is offered as evidence in justifications for conducting a form of ‘clinical trial’ before (and potentially during) work isolate to pharmacologically active compounds. Those participating in the Institute’s HIV/AIDS research project are tested for HIV as well as for a number of other maladies and health index measures. In addition, healers assess their patients’ conditions. The medicines they receive are collected and prepared by healers. Biomedical and traditional practitioners monitor participants receiving these treatments from healers. Any changes in biomedical measures including viral load will be noted in this ongoing treatment and exchange. Case histories will require multiple tellings and translations. While those treatments that show significant reduction of symptoms or viral load will be subjected to chemical and pharmacological investigations, the medicines are not likely to be fully displaced by their chemical compounds. Medicines are simultaneously accountable to patients, healers and scientists. This study offers the possibility of different relations between healers and scientists, which in turn enables the possible emergence of different, hybrid forms of proof. In August 2003, the Institute, with the support of the government, ran a workshop for all the healers who would initially be participating in the research project. The Principle Investigator of the project articulated the procedures. A doctor participating in the research project talked about the medical definition of HIV/AIDS as well as the symptoms and need of people living with AIDS. Other presentations addressed the counselling of people with AIDS and the legal standing of traditional medicines. Tellingly, however, the first presentation of the workshop addressed the issue of healers’ ‘ownership rights’ (hati miliki). Professor Mahunnah, from the Institute of Traditional Medicine, outlined three main points concerning intellec286

Healers and Scientists

tual property rights, each of which open a space for negotiation between healers and scientists. First, he asserted that: ‘Cooperation among the participants of research of traditional medicine must be guided by documents of agreement of different kinds, each to be appropriate for the kind of research’ being undertaken. These agreements he called Memorandums of Understanding. Second, he introduced the idea that property rights are linked to the ‘demonstrated results of the creativity of human intellect’. Finally, Professor Mahunnah told the healers present that the structure and authority of the Traditional and Alternative Medicine Council (the creation of which is called for in the 2002 Traditional and Alternative Medicine Act) would be critical, because this Council will have jurisdiction over such matters (Mahunnah 2003: 19). As in the other workshops discussed above, the unsettled question of ‘ownership’ led to statements about healers’ secrets. In his opening address, the Senior Deputy of the Muhimbili College of Health Sciences explained that: ‘The government through its Institute [of Traditional Medicine] intends to improve and to develop traditional medicine so that it comes out of the secret mysteries of individual healers and it becomes safer for more people’ (Lembariti 2003: 5). The Principle Investigator argued that: ‘The success of this research will depend a great deal on cooperation between researchers and healers and it is important that there are not be any secrets’. To demonstrate his commitment to this policy of disclosure he asserted that the researchers would not keep secrets. He promised that healers participating in the project would received the results of any investigation involving their plants and that they would be made aware of all scientific evidence of the benefits of medicines used in the project (katika mfuko wa mradi huu). In contrast, Kayombo explained to me that one of the advantages of this methodological approach is that it allows healers to ‘keep their secrets’. To imagine what this disclosure and secret-keeping might entail, I turn briefly to a controversy in a project in South Africa with some similar goals. In Durban, South Africa, President Bush’s initiative on AIDS is funding a project to train 375 healers to administer HIV tests, keep medical case histories, and refer people with AIDS to the clinic. As part of this project the medicine school at the University of KwaZulu-Natal has provided a list of guidelines to help healers recognize AIDS. One healer objected to these guidelines, which required healers to rationalize their treatment response based on a list of symptoms they elicit from each patient. She argued that more inclusive guidelines would recognize that many healers do not ask a 287

Evidence, Ethos and Experiment

patient to list symptomatic complaints but rather healers ask their ancestors what to prescribe to each patient. In reporting this incongruence and the healer’s resistance, the Wall Street Journal commented that: ‘In the end, Mr Hartzell [who is running the US-funded project] and his colleagues decided to leave the ancestors out of the clinical guidelines, reasoning that all spiritual mediums know about their role already’ (Phillips 2006). If, as Megan Vaughan has argued, people are more concerned with establishing a place from which to negotiate than they are with defending claims to authenticity, the medium’s appearance in the US press is noteworthy. This is certainly more of a place from which to negotiate than the marginalized position of a ‘custodian’ of a community’s historical knowledge. Yet, the South African project failed to find a way to recognize the agency of ancestors, entities central to the medium’s treatment. My point in relation to the Tanzanian project is: if this project on traditional medicines used to treat HIV/AIDS is to take advantage of its methodological uniqueness, then it must not write ancestors and other entities critical to healers’ transformation of plants into medicines out of the project’s guidelines. Rather the challenge is to figure out how to write them in. To write into collaborative projects the temporalities that allow the assessment of diverse innovations and transformations. To write in the possibilities that healers have contributed significant ‘creativity’ and ‘human intellect’ to the development of a therapy. As this chapter suggests, however, this is a radical undertaking, for it requires opening legal and bureaucratic arrangements as well as medical research protocols to the possibilities of other ways of knowing. It requires not only tracing out the political and economic investments in reified notions of knowledge, of subjects with knowledge in their heads, of knowledge as property, but also imagining and enabling different sorts of commitments, reconfiguring political and economic relations. Ancestors challenge the very relationship between science and capital that has called into being this revival of traditional medicine in Africa.

Epistemological Politics ‘It can’t be proven, yet it happens.’ This chapter has examined my colleague’s refrain by investigating the epistemological conditions that enable ‘proof ’ of some things that ‘happen’ and not others. I have argued that the nature of scientific proof is not self-evident, but rather emerges in the details of research methodologies and the structure of research institutions. As a result, 288

Healers and Scientists

the conditions that allow some things to be proven and not others change over time. As the Institute of Traditional Medicine’s project on AIDS treatments illustrates, research into traditional medicine could be dynamic ground for the elaboration of new hybrid forms of proof. Yet, the values and practices that bolster particular forms of proof are sedimented in institutional structures. Hierarchies of knowledge, not only between healers and scientists, but also between different branches of ‘science’, have been concretized in the organizational structure of the Institute of Traditional Medicine. Most dramatically, Dr Kayombo’s social studies serve to index all that is deemed ‘unnecessary’ and ‘uncaptured’ in his colleagues’ research at the Institute. He holds the place of the indigenous, of the Other, of the traditional. From one perspective, his presence among his colleagues re-inscribes the boundary created as ‘useful elements’ of ‘traditional’ medicines are identified and elaborated through scientific practice. However, Dr Kayombo’s particular interest in recent attempts to establish Memorandums of Understanding (MoU) with traditional healers challenges this position. Negotiations over these documents hold out the possibility of renegotiating what the integration of traditional medicine and biomedicine means. While integration will still require bringing scientists, healers and their medicines into the same frame, this frame could be expanded to include more than the plant, animal and mineral substances used to prepare medicines. Whether these first MoUs do or not, healers are demanding that entities critical to the efficacy of their work (ancestral shades, various unseen entities and religious spirits) be recognized within the framework of these collaborative projects, as compatible with scientists, laboratories and commercial forms of manufacturing. Which traditional medicine is elaborated through scientific study, then, emerges from a complex array of concerns. In the dominant epistemological form, medicines move as they develop new social relations, as their ties with healers are broken and their ties with scientists are strengthened. Translating therapies into scientific objects has traditionally relied on a complicated procedure of de- and re-contextualization. Central to the ideas of hybrid proofs is the possibility that healers might move with their medicine, that they might have a role in the re-contextualization of their medicines. As investigators and participants in projects examining traditional medicines negotiate their methodologies, research protocols and collaborative agreements, they also re-enforce or re-configure relationships between scientists and healers. Inherent in these relationships are the ways in which 289

Evidence, Ethos and Experiment

these actors are accountable to each other. For this reason, each time workshop participants attempted to articulate healers and scientists in the same frame the conversation swung to a discussion of intellectual property and healers’ secrets. Therefore, the question of how healers and their ‘rituals’ are included or excluded from the national and international development of traditional medicine is a methodological, a political and an ethical question. When Kayombo worries that initiatives to develop traditional medicine might actually be ‘moving away from traditional medicine’, he is not nostalgic for an authentic traditional practice but rather he is suggesting that healers and other entities critical to therapeutic practice have a place from which to negotiate.

Notes 1. Personal Communication with Dr Edmund Kayombo, Institute of Traditional Medicine, Muhimbili Medical College at the University of Dar es Salaam, 7 June 2006. 2. Kayombo (1996). 3. Kayombo (2003). 4. Kayombo (2000). 5. For a classic and powerful argument on how solutions to problems of scientific knowledge are also solutions to problems of social order see Shapin and Shaffer (1985). For the ethnographic examples in relation to medicine and bodies, see Farquhar (2002), Lock (1998), Martin (1994), and Weiss (1992). 6. The Traditional and Alternative Medicine Act No. 23 of 2002, United Republic of Tanzania. Dar es Salaam: Government Printer, 2002. 7. For a more elaborate discussion see Langwick (2010). 8. The African Union’s Plan of Action for the Decade of African Traditional Medicine, viewed on the AU website www.africa-union.org. 9. Chhabra et al. (1987, 1989, 1990a, 1990b, 1991). 10. Johns et al. (1995); Mbwambo et al. (1996); Malele et al. (1998); Kagashe and Moshi (2003). 11. Kapingu et al. (2000); Mbwambo et al. (2004, 2006). 12. Uiso and Mahunnah (2000). 13. Eight of the nine participants listed as traditional healers were from Tanzania, as was the one participant listed as an herbalist. 14. This quote is preceded by a direct comparison between the position of doctors and ‘medicinemen’. Mr Waane (1991: 214) stated that ‘Recently, the Government of Tanzania restructured its salary structure. In the new structure, medical doctors and pharmacists have been categorized as being in the rare professions. This is, in a way, a recognition of their special role and position in the Tanzanian Society. It is an acceptance and an ascription of special status. This status is based on the special knowledge that they have, the rigorous program that they have to go through 290

Healers and Scientists

15. 16.

17.

18.

19. 20. 21.

22. 23.

in acquiring this knowledge, and their special relation to the process of life and living. However, in assessing the place of the traditional healers in this and previous societies, we tend to forget the fact that in their case too, our societies gave them similar or even more consideration’. Many of the healers with whom I worked were not literate. This phrase is borrowed from a statement by Akerele made in 1990 at the International Conference on Traditional Medical Plants at which time he was the Programme Manager for the WHO Programme on Traditional Medicine. See, for example, the strategy of the Tanga AIDS Working Group (Chaze 2003) and the Institute of Traditional Medicines recent research into medicinal healers’ use to treat AIDS (Uiso et al. 2003: 53). In this quote, the Deputy Minister of Health flags the explicit link between the development of traditional medicine in Tanzania in 2003 and broader commitments to evidence-based medicine (EBM). The forms of scientific research that produce evidence, which supports the establishment of ‘best practices’ and the standardization of medical work, have implications that political leaders such as this deputy minister are mobilizing in sophisticated ways. As Timmermans and Berg (2003) have illustrated in their co-authored book The Gold Standard, the institutional transformations demanded by evidence-based medicine not only reconfigure medical treatment, but also the forms of autonomy and kinds of relationships implicit in that treatment. For instance, one of the effects of evidence-based medicine is the production of ‘a liberal subject endowed with a biomedical body requesting professional care’ (2003: 54). The regimes of evidence EBM promotes collude in making the subject critical for neoliberal reforms. Drawing traditional medicine into state discourse in this way mobilizes it in conjunction with very particular kinds of ‘development’ efforts, as well as an emerging ‘culture of neoliberalism’ (Comaroff and Comaroff 2000). Scientists and technicians at the ITM are working to expand three other areas of investigation: clinical evaluation, medical anthropology and pharmaceutical technology. This distinction resonates with Susan Renyolds Whyte’s distinction between knowledge about and knowing (see this volume). Personal conversation (June 2006) with Rene Gerrets, who has recently assisted a group of healers in the Infakara district write a grant which would support the building and maintaining of a botanical garden of the plants that they use in their therapeutic practices. F. wa Simbeye, ‘Call for Traditional Healers College Amid Suspicions of Witchcraft’, Daily News, 5 July 1999, p. 3. ‘Herbalists to Start a Lab?’, Daily News, 26 April/July 2003, p. 4.

References Akerele, O. 1990. ‘Registration and Utilization of Herbal Remedies in Some Countries of East, Central, and Southern Africa’, in Ministry of Health, Tanzania (ed.), Proceedings of the International Conference on Traditional Medicinal Plants, 18–23 February. Arusha, Tanzania: Dar es Salaam University Press, pp. 3–8. 291

Evidence, Ethos and Experiment

———. 1991. Registration and utilization of herbal remedies in some countries of Eastearn, Central and Southern Africa. In Ministry of Health, United Republic of Tanzania Traditional Medicinal Plants. Dar Es Salaam University Press. Comaroff, J. and J.L. Comaroff. 2000. ‘Millennial Capitalism: First Thought on a Second Coming’, Public Culture 12(2): 291–344. Callon, M. 1999. ‘Some Elements of a Sociology of Translation: Domestication of the Scallops and the Fisherman of St. Brieuc Bay’, in M. Biagioli (ed.), The Science Studies Reader. New York: Routledge, pp. 67–84. Chhabra , S.C. , Mahunnah, R.L.A. and Mshiu, E.N. (1987) Plants used in Traditional Medicine in Eastern Tanzania . I. Pteridophytes and Angiosperms (Acanthaceae – Canellaceae). Journal of Ethnopharmacology 21 : 253–77. Chhabra , S.C. , Mahunnah, R.L.A. and Mshiu, E.N. (1989) Plants used in Traditional Medicine in Eastern Tanzania . II Angiosperms (Capparidaceae - Ebenaceae). Journal of Ethnopharmacology 25: 339–59. ———. 1990a. Plants used in Traditional Medicine in Eastern Tanzania . III. Angiosperms (Euphorbiaceae – Menispermaceae), Journal of Ethnopharmacology 28: 255–83. ———. 1990b. Plants used in Traditional Medicine in Eastern Tanzania. IV. Angiosperms (Mimosaceae – Papilionaceae). Journal of Ethnopharmacology 29: 295–23. ———. 1991. Plants used in Traditional Medicine in Eastern Tanzania. V. Angiosperms (Passifloraceae -Sapindaceae). Journal of Ethnopharmacology 33: 143–57. Farquhar, J. 2002. ‘Sexual Science: The Representation of Behavior’, Appetites: Food and Sex in China. Durham: Duke University Press, pp. 211–42. Feierman, S. 1985. ‘Struggles for Control: The Social Roots of Health and Healing in Modern Africa’, African Studies Review 2/3: 73–147. Geissler, P.W., S.A. Harris, R.J. Prince, A. Olsen, R.A. Odhiambo, H. Oketch-Rabah, P.A. Madiega, A. Andersen and P. Molgaard. 2002. ‘Medicinal Plants Used by Luo Mothers and Children in Bondo District, Kenya’, Journal of Ethnopharmacology 83(1–2): 39–54. Geissler, P.W. and R. Prince. 2009. ‘Active Compounds and Atoms of Society: Plants, Bodies, Minds, and Cultures in the Work of Kenyan Ethnobotanical Knowledge’, Social Studies of Science 39(4): 599–634. Hermansson, K., L. Kenne, G. M. Rukunga, G. Samuelsson, and W.M. Kofi-Tsekpo. 1991. Chemical characterization of pharmacologically active compounds from Synadenium pereskiifolium. In Ministry of Health, United Republic of Tanzania Traditional Medicinal Plants. Dar Es Salaam University Press. Institute of Traditional Medicine, Tanzania. 2004. Proceedings of the National Stakeholders Workshop on Traditional Medicine and Medicinal Plants, Bagamoyo Oceanic Bay Hotel, Tanzania, 4–6 August. International Development Research Center (IDRC). 2003. Proceedings of the Stakeholder Workshop on the Sustainable, Safe, and Effective Use of Medicinal Plants in Eastern Africa, 24–27 November. Arusha, Tanzania. Organized by the IDRC Eastern and Southern Africa Regional Office in collaboration with the National Museums of Kenya, the Institute of Traditional Medicine in Tanzania, and the National Chemotherapeutics Research Laboratory in Uganda. 292

Healers and Scientists

Johns, T., M. Gaetan, M. Faubert, J. Kakwaro, R.L.A. Mahunnah and E.K. Kimanani. 1995. ‘Anti-giardial Activity of Gastrointestinal Remedies of the Luo in East Africa’, Journal of Ethnopharmacology 46: 17–23. Kagashe G, Moshi MJ. Screening of extracts of Bidens pilosa for anticonvulsant activity. Tanzanian Medical Journal 2003(18):28-30 Kapingu MC, Guillaume D, Mbwambo ZH, Moshi MJ, Uiso FC, Mahunnah RLA. Diterpenoids from the roots of Croton macrostachys. Phytochem 2000; 54(8): 767-70. Kayombo, E. 1996. ‘“Tambiko” as a Healing Therapy in Tanzania’, in S.N. Madu, P. Kakubeire and A. Pritz (eds), Psychotherapy in Africa. Vienna: World Council for Psychotherapy, pp. 71–81. ———. 2000. ‘Can Dreams be a Source of Knowledge in Traditional Healing Therapies? A Case of Six Traditional Healers in Tanzania’, in S.N. Madu, P. Kakubeire and A. Pritz (eds), Psychotherapy and African Realities. Pietersberg: World Council for Psychotherapy, pp. 95–119. ———. 2003. ‘Mzimu in Traditional Healing in Africa: A Case of Tanzania’, in S.N. Madu (ed.), Contributions to Psychotherapy in Africa. Pietersberg: World Council for Psychotherapy, pp. 144–63. Kayombo, E.J. and F.C. Uiso. 2003. Warsha Kuhusu Utafiti wa Dawa Asilia zinazotumiwa na Waganga wa Jadi kutibu maradhi ya Ukimwi Tanzania [Workshop on Research of Traditional Medicine That Is Used by Traditional Healers to Treat HIV/AIDS in Tanzania], 16 August. Institute for Traditional Medicine, Muhimbili University College of Health Sciences. Kilewo, J.Z.J., I.A.J. Semali, R. Msuya, E.N. Mshiu, D.M. Do Amsi, C.K. Makwaya and E.P.Y. Muhondwa. 1987. ‘Traditional Medicine Practice in Tanzania: History and Developments’, Tanzania Medical Journal 25: 7–10. Langwick, S. 2007. ‘Devils, Parasites, and Fierce Needles: The Politics of Translation in Healing in Tanzania’, Science, Technology and Human Values 31(1): 1–30. ———. 2010. ‘From Non-aligned Medicines to Market-based Herbals: China’s Relationship to the Shifting Politics of Traditional Medicine in Tanzania’, Medical Anthropology 29(1): 15–43. ———. 2011, Bodies, Politics, and African Healing: The Matter of Maladies in Tanzania. Bloomington: Indiana University Press. Last, M. 1987. ‘Traditional Medicine To-morrow’, Tanzania Medical Journal 25: 10–13. Latour, B. 1988. The Pasteurization of France, trans. A. Sheridan and J. Law. Cambridge, MA: Harvard University Press. ———. 1999. Pandora’s Hope: Essays on the Reality of Science Studies. Cambridge, MA: Harvard University Press. Lembariti, B.S. 2003. ‘Hotuba ya Mkuu wa Chuo ya Ufunguzi wa Mkutano wa Watafiti na Waganga wa Jadi Chuo Kikuu Kishiriki Muhimbili’ [Opening Speech of the Director of the College of Health Sciences for the Meeting of Researchers and Traditional Healers at Muhimbili University], in E.J. Kayombo and F.C. Uiso (eds), Warsha Kuhusu Utafiti wa Dawa Asilia zinazotumiwa na Waganga wa Jadi kutibu maradhi ya Ukimwi Tanzania [Workshop on Research of Traditional Medicine That Is Used by Traditional Healers to Treat HIV/AIDS in Tanzania], 16 August. Institute for Traditional Medicine. Muhimbili University College of Health Sciences, pp. 4–5. 293

Evidence, Ethos and Experiment

Lock, M. 1998. ‘Anomalous Ageing: Managing the Postmenopausal Body’, Body and Society 4(1): 35–61. Mahunnah, R.L.A. 2003. Mfumo wa Tiba Asilia na Hati Miliki [The System of Traditional Medicine and the Formalities of Ownership Documents], in E.J. Kayombo and F.C. Uiso (eds), Warsha Kuhusu Utafiti wa Dawa Asilia zinazotumiwa na Waganga wa Jadi kutibu maradhi ya Ukimwi Tanzania [Workshop on Research of Traditional Medicine That Is Used by Traditional Healers to Treat HIV/AIDS in Tanzania], 16 August. Institute for Traditional Medicine, Muhimbili University College of Health Sciences, pp. 16–19. Mahunnah, R.L.A. and E.N. Mshiu. 1991. ‘Ethnobotany and Conservation of Medicinal Plants’, in Ministry of Health, Tanzania (ed.), Proceedings of the International Conference on Traditional Medicinal Plants, 18–23 February. Arusha, Tanzania: Dar es Salaam University Press, pp. 83–86. Malele R.S., M.J. Moshi, D.A.S. Mloka, S. Melekzedeck and Z.H. Mbwambo. 1998. Invitro antimicrobial activity of some Tanzanian plants used by traditional healers. Tanzanian Medical Journal 3: 31–33. Martin, E. 1994. Flexible Bodies: The Role of Immunity in American Culture from the Days of Polio to the Age of AIDS. Boston: Beacon Press Books. Mbwambo ZH, S. Apers, M.J. Moshi, M.C. Kapingu, S. Van Miert, M. Claeys, R. Brun, P. Cos, L. Pieters and A. Vlietinck. 2004. Anthranoid compounds with antiprotozoal activity from Vismia orientalis . Planta Medica 70: 706–710. ———. 2006. Antiparasitic Activity of Some Xanthones and Biflavonoids from the Root Bark of Garcinia livingstonei. J. Nat. Prod., 69: 369–72. Mbwambo, Z.H., L. Luyengi and A. Douglas Kinghorn. 1996. ‘Phytochemicals: A Glimpse into Their Structural and Biological Variation’, International Journal of Pharmacognosy 34: 5. Ministry of Health, Tanzania. 1991. Proceedings of the International Conference on Traditional Medicinal Plants, 18–23 February 1990. Arusha, Tanzania: Dar es Salaam University Press. Mshui, E.N. 1991. ‘Biotechnology and Medicinal Plants’, in Ministry of Health, Tanzania (ed.), Proceedings of the International Conference on Traditional Medicinal Plants, 18–23 February 1990. Arusha, Tanzania: Dar es Salaam University Press, pp. 133–36. Mshiu, E.N., J.G. Sayi and P.M. Sarungi. 1991. ‘A Note on the Utilization and Commercialisation of Traditional Medicine’, in Ministry of Health, Tanzania (ed.), Proceedings of the International Conference on Traditional Medicinal Plants. 18–23 February 1990. Arusha, Tanzania: Dar es Salaam University Press, pp. 178–80. Mutayabarwa, C.K., S.C. Chhabra, G.M.P. Mwaluko, J. Fulgence and W. Msangi. 1991. ‘The Chemistry and Pharmacology of the Essential Oil from the Leaves of Hyptis Suaveolens (L) Poit’, in Ministry of Health, Tanzania (ed.), Proceedings of the International Conference on Traditional Medicinal Plants. 18–23 February 1990. Arusha, Tanzania: Dar es Salaam University Press, pp. 140–45. Mwinyi, H.A. Opening Remarks. Proceedings of a Stakeholders Workshop on the Sustainable, Safe and Effective use of Medicinal Plants in Eastern Africa, Arusha. 24–27 November. Nkunya, M.H.H., H. Weenen and D.H. Bray. 1991. ‘Chemical Evaluation of Tanzanian 294

Healers and Scientists

Medicinal Plants for the Active Constituents as a Basis for the Medicinal Usefulness of the Plants’, in Ministry of Health, Tanzania (ed.), Proceedings of the International Conference on Traditional Medicinal Plants. 18–23 February 1990. Arusha, Tanzania: Dar es Salaam University Press, pp. 101–111. Phillips, M.M. 2006. ‘Persuading Africans to Take Their Herbs with Some Antivirals: US-backed Program Pushes Doctors, Healers to Treat AIDS Patients Together’, Wall Street Journal 5 May, p. A1. Pigg, S.L. 1995. ‘Acronyms and Effacement: Traditional Medical Practitioners (TMP) in International Health Development’, Social Science and Medicine 41(1): 47–68. ———. 1997a. ‘“Found in Most Traditional Societies”: Traditional Medical Practitioners between Culture and Development’, in F. Cooper and R. Packard (eds), International Development and the Social Sciences. Berkeley: University of California Press, pp. 259–90. ———. 1997b. ‘Authority in Translation: Finding Knowing, Naming and Training “Traditional Birth Attendants” in Nepal’, in R. Davis-Floyd and C. Sargent (eds), Childbirth and Authoritative Knowledge: Cross-cultural Perspectives. Berkeley: University of California Press, pp. 233–62. Shapin, S. and S. Schaffer. 1985. Leviathan and the Air-pump: Hobbes, Boyle, and the Experimental Life. Princeton: Princeton University Press. Timmermans, S. and M. Berg. 2003. The Gold Standard: The Challenge of Evidencebased Medicine and Standardization in Health Care. Philadelphia: Temple University Press. Uiso FC and R.L.A.Mahunnah. 2000. Wild food plants in health and dietary sufficiency LINKIS Report 2: 65–67 Uiso, F.C., R.L.A. Mahunnah, Z.H. Mbwambo and M.J. Moshi. 2003. ‘Contribution of Traditional Medicine to Health Care and Nutrition in Tanzania’, Proceedings of a Stakeholders Workshop on the Sustainable, Safe and Effective use of Medicinal Plants in Eastern Africa, Arusha. 24–27 November, pp. 36–46. Waane, S.A.C. 1991. ‘The Use of Traditional Medicinal Plants: The Cultural Context’, in Ministry of Health, Tanzania (ed.), Proceedings of the International Conference on Traditional Medicinal Plants. 18–23 February 1990. Arusha, Tanzania: Dar es Salaam University Press, pp. 209–21. Weiss, B. 1992. ‘Plastic Teeth Extraction: The Iconography of Hay Gastro-sexual Affliction’, American Ethnologist 19(3): 538–52.

295

Chapter 12

Parasite Lost: Remembering Modern Times with Kenyan Government Medical Scientists P. Wenzel Geissler

Introduction Patience Visiting Kisumu in western Kenya, I pass by the station of the Ministry of Health’s Division of Vector Borne Diseases (DVBD) where 15 years ago I conducted research on worm infections with a group of Kenyan and European scientists.1 The offices and laboratories, situated adjacent to the District hospital in the centre of Kenya’s third largest city, were built between the 1930s and 1970s. In the early 1990s, we were still able to conduct simple scientific research on body parasites here, and the laboratory staff were busy. In the early years of the twenty-first century, the place is quiet. The compound, shaded by mature trees, some planted to commemorate the departure of the last British scientist employed by the Division up to the 1970s, others to mark the launch of overseas collaborative projects in the 1990s, is poorly maintained since the last ‘groundsman’ – who had continued cleaning the place well past his retirement – succumbed to old age. Among patches of high grass and bushes stand the dusty wrecks of vintage Land Rovers recalling past movements and achievements. The DVBD is the section of the Ministry of Health in charge of the control of diseases such as malaria, sleeping sickness, bilharzia, worms and viral infections –the most common diseases affecting Kenya’s citizenry. With the 297

Evidence, Ethos and Experiment

headquarters in Nairobi and about fifty stations across Kenya, DVBD employed about 300 staff in 2004 (down from about 800 in the early 1980s). Most of the employees are approaching retirement, and resources are scarce. Thus, DVBD Kisumu has no transport, no reagents or basic laboratory equipment and no functioning means of communication as telephone lines were closed down for unpaid bills some years ago. The gate, guarded by the security officer of an overseas research institution using one of the outbuildings for cool storage of specimens, is nowadays rarely opened for cars. A signboard beside it lists, underneath ‘DVBD/MoH’, the US Army’s ‘Walter Reed Project’, which used to collaborate with the DVBD, and the ‘Kenyan-Danish Health Research Project’, DVBD Kisumu’s last larger collaborative project. Some staff members sit beside the gate, observing the traffic outside and greeting passing acquaintances. The water tap outside the main building is running continuously – if there is water – to fill jerry cans for the staff members living in areas with erratic water supply. The DVBD, formerly DIBD (Division of Insect Borne Diseases), originated in 1938 as Yellow Fever Control programme, slightly ahead of the other ‘East African Research Organisations’ founded in the wake of the 1940 Colonial Welfare and Development Act (see Clarke 2007). It is the oldest medical research institution in Kenya and internationally leading in the field until 1978, after which the Kenyan Medical Research Institute (KEMRI), one of a new generation of ‘parastatal’ research institutes established at the onset of neoliberal restructuring, gradually took the helm of scientific research from government and academic institutions, through collaborations with transnational ‘partners’. DVBD’s main mandate is parasitic disease surveillance and control of insect and snail vectors of disease; as part of these activities, initially as a pastime of its then British officers, it conducted research, discovering, for example, the liver-stages of the malaria parasite, or the ecology of the blackfly that carries river blindness. After Kenya’s independence in 1963, the Division maintained its mandate, led by an emerging new generation of Kenyan scientists. When, due to economic crisis and austerity policies, funds for regular disease control vanished, the DVBD succeeded, because of its skilled staff, to attract collaborative research projects with European and North American research groups, and while the wider health sector rapidly decayed, DVBD staff – at least the staff of stations involved in such collaborations – experienced good, if very unsteady, incomes, and conducted recognised scientific research. 298

Parasite Lost

Inside the main building, the heavy prison workshop furniture, fans and half-high oil-paint coating leave no doubt that one enters a postcolonial government institution. The mid-1990s office equipment, left by the last collaborative project, awaits repair. The fridge, no longer needed for specimens, is used for soft drinks. The station head has gone to an NGO-funded workshop, hopefully providing him with new knowledge, contacts and some allowances. In his office, an organogram, drawn at another workshop, places the DVBD within ‘District Health Management’, one of the buzzwords in the 1990s ‘decentralisation’ of the national health system. In the laboratory, two men work at a microscope, reading slides for a malaria research project by an overseas university collaborating with KEMRI, to which they were seconded; other men sit and chat with visiting friends. One takes the result of a malaria test, scribbled onto the corner of an exercise book, to a relative waiting at the gate. At 10 am, as every day, the secretary prepares and serves tea. The scent of milky, sweet tea with a little ‘Nice’ biscuit on the saucer, evokes lost times – remembrance of my own past with the DVBD and historical memories far beyond that. During the 1990s boom of collaborative research projects, staff members attained higher academic degrees in European universities and even subordinate staff earned many times their salary through daily allowances paid by ‘projects’, initiated by overseas agencies (as opposed to national government). This age of bounty ended around the turn of the century due to a somewhat unstable political climate, to new mechanisms of collaboration and funding, and because of changes in medical science: in 1994, trained laboratory technicians with a microscope and cheap reagents could make a decent contribution to parasitology and epidemiology, and – through integration into government – to public health; post-millennial science, by contrast, has become focused on clinical trials that usually are designed transnationally, under the leadership of European and North American institutions, charities and corporations; such trials rely upon big sites with large populations under demographic surveillance, they need sophisticated, expensive and rapidly changing immunological and genetic laboratory equipment, and they are regulated by internationally agreed standards such as ISO or ‘Good Clinical Practice’; they require highly controlled environments – both to produce internationally recognised science and to manage large funding flows and staff bodies; and their results are directed towards ‘global health’ policies rather than immediate utilisation by 299

Evidence, Ethos and Experiment

local public health authorities. In this age of ‘big science’, DVBD with its ties to government and its public health mandate is a less suitable partner than the pure research institutes – ‘parastatals’ – that were founded from 1978 onwards. Deprived of government funds and overseas support, the DVBD became unable to produce science. For the people left in the station, a former colleague’s visit rekindles memories of fieldwork and after-work social life, of regular incomes and educational opportunity; and they are curious to hear whether I may be starting new research: ‘Bring a project, we are ready, we can work!’. My employment with the London School of Hygiene, where so many earlier DVBD scientists had come from or gone to, raises expectations that I cannot fulfil; unable to bring a new scientific project, we decide at least to have the building painted, which has become a source of embarrassment to the staff. Especially in contrast to the flash premises of the many HIV research projects and NGOs that have emerged in Kisumu during the past decade, the simple and so obviously governmental buildings of the DVBD are more than a little out of date. Buying the paint and seeing the work progress, we remember the days when we had commenced our last project in 1994. This was a different time, we concur, as we wait for the tea. It is delayed because the council has cut off the water supplies. In the meantime, the secretary plays ‘Patience’ on her ageing project computer. The present of this public health institution – as for many Kenyan government health institutions – is one of stasis. While the grounds and buildings are quietly transformed by rains, dust and plant growth, the people inhabiting them are plagued by boredom and by their desire to use their skills. ‘We are stuck’, they remark, and their frustration is exacerbated by the thriving cityscape – benefiting from the plethora of NGOs and projects that HIV has brought to the city – that expands outside their gate. Behind this present are memories of a better past, when the station had a staff of fifty people and four vehicles, when the men began the workday at dusk and gathered for drinks in the evening, when international scientists visited and Kenyan staff presented their work overseas, and when the DVBD was part of projects on various level of scale, comprising local endeavours as well as nation building, modernisation and scientific progress. These are memories of past futures, of aspirations and hopes. Within and beyond the two prevailing senses of time that surround DVBD as I know it – stasis and nostalgia 300

Parasite Lost

– there is a third temporality,: a sense of future, of directed movement. At this point, this is found mainly in the men’s recollections, although it survives in moments of hope as well as in the rethorics of ‘community development’ that facilitate engagements with potential funders and workshop audiences. In this chapter I search for this now almost forgotten sense of ‘project’ – in terms of science, of government and as lived by the people of DVBD. I explore this project through the stories of the ‘DVBD men’.3 Some were told while we were in the field or unfolded as we worked alongside each other. Others I heard in the homes, scattered across rural western Kenya, to which the old men had retired – some as long as 30 years ago. Understanding of our shared history and present predicament, as men of public science, might help us reconsider our sense of direction.4

People of the State This chapter deals with evolutions in African science and government, seen from the vantage point of a troubled present. Scholarly reflections on the ‘crisis of the African state’ often ask whether the state – represented as an institutional edifice – exists or not, or is weak or strong, successful or failing (e.g. Zartman 1995; Beissinger and Young 2002).5 This approach foregrounds static structures of statehood and underrates its openness and potentiality – the promise of the modern state. Spatial imaginaries of statehood – hierarchy, territory, structure – here overshadow temporal ones – process, plan, utopia.6 By contrast, I will not ask whether the state is there or not, or works or not (nor whether it has been there or not, and when, or if it was effective) – there is no doubt that it still is there, for example, in the form of millions of African civil servants, teachers, nurses, agricultural officers, and that it remains important as a tool of political control, but also as a source of institutions and standards, a potential source of order. Instead of searching and vetting the modern state as a thing, I want to explore it as a project – an anticipation of things to come – as it was lived by the men of DVBD and lives in their stories. I want to show, through an ethnography of people of the state – rather than, as often is the case, of the people versus the state – that the state need not be seen as a closed edifice, an external entity that stands in some relationship to (civil) society and to citizens, but as a representation of society to itself, an opening towards possible societal futures and a frame for social action. Focusing on the people of the state allows us to locate the workings of government in the everyday work of people and to trace, in their inter301

Evidence, Ethos and Experiment

twined narratives of professional and private life, connected themes and kindred intentionalities or temporalities across different domains and levels of scale. What, in these modern lives, links scientific discovery and public health, educational and economic ambitions and child rearing, is an overall notion of forwardness or ‘generation’. Generation has here, as will become clear in the life stories below, different, interconnected meanings: it is about the essence of enlightenment, the pursuit of truth and engendering of transformation; it is about growth and renewal – of oneself and one’s descendants, and of one’s group and one’s nation; and it is about relations, between young and old. This modern, generative project has withdrawn into our past; to retrieve it, we listen to old men’s tales.

Old Men of Modern Science In this chapter I draw upon the life stories of a group of old men from western Kenya – born in the 1920s, recruited by DVBD in the 1940s and 1950s and retired from government service in the 1980s – whom we found

Figure 12.1: The Staff of DVBD Kisumu, Around 1962 (photographer unknown) (personal archive, Alfred Lwoba). 302

Parasite Lost

through archival documents and younger DVBD staff ’s advice. Most of these men live in their inherited rural homes or, if they invested some money, in rural settlement schemes, tending their land and cattle. Some have retained a medical role, operate drug shops or diagnostic laboratories, and they tend to be well known and respected in their villages, where most are active in Church and other community matters. Their rural homesteads, often built from concrete or brick, with painted tin roofs, screened verandas and half-high oil painted walls, citing modernist Kenyan architecture, are easily recognised among the more common mud-huts; they embody their owners’ achievements in making modern Kenya, intertwining personal lives and scientific work, domestic growth and the development of the nation. Inside their houses, sofas and armchairs are covered with snow-white embroideries; glass shelves display books and decorative objects; calendars and clocks ornate the living room along with Christian mottos and family pictures, coloured ones of young people with academic gowns, cars and urban backgrounds, and some faded black and white ones of the DVBD’s scientific work (Figures 12.1 and 12.2).

Figure 12.2: Technologists Bench Training in a DVBD Laboratory, 1960s (photographer unknown) (personal archive, Ambrose Masime). 303

Evidence, Ethos and Experiment

Guests are generously welcomed here. Visits, commencing with prayers, begin with tea and sponge cakes and proceed with a substantial meal, as customary hospitality would demand. It was during visits like this that our conversations about the Kenyan nation, government and science, historical change and a shared sense of purpose arose. In the 1930–40s, when the men had left school and started working ‘with government’, modern, scientific colonialism had just been conceived (e.g. Cooper 2002); when they retired in the 1980s, national scientific and health structures gave in to the onslaught of economic crisis and austerity policies (e.g. Mbatia and Bradshaw 2003). These memories of better days – when government healthcare worked better than now, when salaries reached almost the end of the month, and when they were young – triggered nostalgia: today, little scientific work is produced by government institutions, and the government rarely legitimises its actions with science (drawing instead on ‘economics’ as justification, or not acting at all); state science workers have become living anachronisms inhabiting ruined institutions. But what exactly do these elderly Kenyan science-workers long for? Is it the stability of a past material, political and epistemological order, a time with less scarcity and confusion, as some interpretations of postcolonial African nostalgia suggest. If modernity was, as I suggested above, not only a disciplinary order but a transformative project, which was shared across the nation by people with different interests and aims, then maybe these old modern men long less for closures than for openings: not (only) for the past order’s stability and finality but for a past sense of direction and progress, which meaningfully intertwined professional and everyday life, government and science, state and citizens (see, for example Ferguson 1999, Tsing 2004). Below I will explore this proposition; first I will examine the men’s ‘fieldwork’ in scientific ‘campaigns’, and the social relations of ‘the field’; then I will attend to the men’s self-understanding as men of science and government, and to the relationships between citizenship and political interests, professional and educational curriculum and the personal quest for a modern life. Before I commence, a note on historical periodisation is on order. The men presented here worked with the DVBD during the half century of the ‘developmental state’ (late 1930s to early 1980s) – embracing both sides of formal independence in 1963: ‘imperialism of knowledge’ (Cooper 2002) and postcolonial scientific national government (see, for example, Ombongi, this volume). Their biographical accounts did not usually emphasise decolonisation, but the Second World War and the crisis of the 1980s (roughly from the oldest men’s entry into DVBD to their retirement). 304

Parasite Lost

Below, I will follow their periodisation and treat the era of developmental modernity as one whole, rather than contrasting the colonial and postcolonial period. The men had certainly welcomed independence, but they experienced it not only as a political change – accompanied by the abolishment of racist policies, not least at work – but as an expansion of the promises of scientific work, accompanied by deepened commitment to a new government and the collective of the young nation, and new responsibilities and opportunities for learning and advancement.

Fieldwork Campaigns: ‘Doing’ Diseases, Making the Nation The men’s lives were structured by ‘campaigns’ to research and control pathogens or vectors, such as malaria or tsetse flies. These were delimited geographically and in time (although the timeline often proved optimistic), and one campaign followed the other, accompanied by movements across the colony, and later the nation: ‘when we had finished doing malaria in Nandi, we did Bilharzia here in Nyanza ...’ (Sawe,7 24 January 2005).8 Some men specialised later, for example in ‘mosquito work’ or microscopy, but all acquired broad knowledge of tropical diseases: ‘I liked all diseases … There was none that I liked better than the other. I liked malaria because it is a very bad illness. Sleeping sickness also finishes people. I liked it. Coming back to do snail surveys, I liked it … Because I liked research work, I had to like any that came up’ (Otiende, 26 February 2001). ‘Came up’ could imply newly discovered diseases or epidemic outbreaks,9 but also the changing directions of international science and new treatment and control technologies which gave rise to campaigns.10 Not rarely, British scientists and medical administrators decided upon investigations or interventions based on informal consultations and local personal contacts. Later, wider networks including the World Health Organisation and linkages between Kenyan government and overseas development agencies became important to initiate interventions. In spite of many scientific and political changes, this basic pattern of ‘campaigns’ persisted from colonial times to the present.11 ‘Doing’ an illness or insect combined making knowledge and action, aiming towards an endpoint, ideally ‘eradication’ or ‘control’. Campaigns implied, in analogy to military interventions, an enemy and a trajectory from reconnaissance, via engagement, to victorious conquest. In the age of heroic entomology, when DVBD began, the enemies were often 305

Evidence, Ethos and Experiment

flies and mosquitoes, but the general narrative pattern remained, although vectors and pathogens changed along with politics and economics. The bellicose nature of the task was explicit; thus, when one of the men, then a lab assistant, wanted to become a soldier in the Second World War (getting higher salaries and chances of attaining a medical qualification as a hospital assistant, otherwise prevented by the ‘colour bar’), his officer, Garnham, admonished him: ‘this place where you are is a war, you can’t leave this place and go to another war’ (Okonji, 14 December 2004).12 To ‘go out’ to where the problems are, to work in and on ‘the field’, explore another place and improve it, was at the core of the old men’s stories. ‘Field’ designated an area of investigation and action as well as a territory inhabited by people and pathogens. It was usually a rural area, in agreement with the term’s pastoral connotations; an object of study, clearly separated from its scientific observers who resided in town.13 Thus, although the field in fact often resembled or even included the men’s rural homes, it was – probably partly embracing the British officers’ viewpoint – constituted as a site of adversity, marked by ‘hard physical work’ (Okonji, 14 December 2004) and exposure to the elements and diseases during camped life. Despite ‘hardship allowances’, it carried heroic connotations.14 The term ‘safari’ (Kiswahili ‘journey’) featured in all men’s accounts, and it seemed that the

Figure 12.3: Officer’s Tent with Kisumuead Head of DVBD, 1960s (photographer unknown) (personal archive, Ambrose Masime). 306

Parasite Lost

English implications of ‘safari’: excitement and maleness, expansive spaces and discovery, was cultivated as a source of identity and pride across the colonial colour bar, and shaped DVBD beyond colonial occupation (see Figure 12.3). Travel also meant freedom from local social ties and kinship, and to some extent from supervision by their (until late 1960s white) officers, who often operated from the urban station. ‘Safari’ captured both the actual travel to and work in remote areas, and a wider sense of movement, of progress and change – of the world and of the men themselves: Field research [original English]! I liked it better than office work. ... I stayed in the office in the afternoon, but in the morning I went to the field. I really liked the field because the field doesn’t make one lazy; it makes the body strong. Someone who walks becomes stronger than someone who just sits in the office doing nothing. (Otiende, 26 February 2001) Fieldwork implied moving from ‘station’ to ‘the field’, where the men spent weeks, returning only to receive allowances and take them to their wives, go to church and organise their farms. With the men, material resources moved out from the urban centres of government and science to the periphery of the rural ‘reserve’. The men’s fieldwork stories often began with the distribution of equipment at the DVBD station’s stores – an iron sheetprotected shed, which at the turn of our century was derelict and presumably empty. These stores, they recalled with marvel amplified by present scarcity, had contained ‘everything, flour, oil, lamps, tents, beds, blankets, even guns … all you could need for the field – we went out with everything’ (Luoba, 21 September 2005). This wealth of instruments and supplies demonstrated the government’s resources and capacity. But the stores fascinated the old men not just as a lost cornucopia; they were attracted by the equipment’s mobility: loaded on trucks, everything needed for scientific work, accommodation and sustenance could be moved out to where sickness had struck and scientific questions needed to be addressed. This was a portable modernity, a mobile capacity to make known and to act: ‘we did not need to go back [to the station]; we even had a mobile laboratory, on a truck, containing everything you need in the field’ (Luoba, 21 September 2005) (Figure 12.4). This fascination with the completeness of provisions and self-contained, mobile capacity showed also in enthusiasm for camp life, which seemed to have been shared across ranks, and which remained a feature of DVBD until the 1980s (Figure 12.3). Camp307

Evidence, Ethos and Experiment

Figure 12.4: Mobile Laboratory, 1950s (from ‘DDT vs. Malaria’) (photographer unknown).

ing moves one right out into the field, yet the independent, mobile and nonlocal nature of tents, camp-beds and kerosene stoves marks a clear separation from the stable local environment. Camping creates a community of ‘field men’ who lead, in principle, a self-contained existence without relying on or relating to local communities. This positioning, close to reality but detached from it, is the idealised position of the scientific observer; in field life, the men’s entire everyday embodied this stance.15

Wananchi – ‘Our Fellow Kenyans’ Scientific detachment notwithstanding, campaigns got entangled with the trajectories of people living in the field, where DVBD men met their fellow citizens. The DVBD men entered these encounters as ‘government people’: ‘We were the eyes of the government .... Anywhere there was a disease our government sent DVBD people to go and check out that disease, and they would bring a report and then the government would begin to treat people’ (Okonji, 14 December 2004). And ‘anywhere our government thought of doing something [such as agricultural schemes], our DVBD staff would first go and check that place in terms of the diseases that were there’ (Ogalo, 24 June 2004). DVBD men 308

Parasite Lost

were the eyes and arms of government, and they dispensed of the government’s knowledge, and not rarely its medicines: ‘we always walked with Chloroquine and Aspirin tablets for the villagers’ (Okonji, 14 December 2004). And yet the men were also citizens, dependent upon government decisions and resources. They described themselves thus variously as subjects and government agents, and used the term ‘wananchi’ (Kiswahili ‘citizen’) both for themselves and others: ‘we told wananchi to do …’ or ‘what could we small wananchi do about …?’ Characteristic ‘middle figures’ (Hunt 1999), DVBD men personified the government-citizen bond, and their work drew other people, too, into the middle, thus continuously expanding the scope of government and citizenship. In the field, the men relied – despite camping equipment and mobile laboratories – upon ‘local’ people, both for food and casual labour, and for legitimation: First we went to the chief [to] tell him that we want to look for insects that cause diseases … in [this] area: ‘Go and teach people that the mosquito people are coming to work here!’ … He has to announce it to people, and people must agree. So when we go in, they don’t disturb us. (Okonji, 24 June 2004) While there indeed was a chain-of-command from the ministry, via the chief, to ‘the people’ (wananchi), statements like ‘people must agree’ or, in other conversations, ‘there must be unity’ should not be misread as mere calls for obedience; they also imply a vision of a whole society and its common good achieved through the work of government. The use of the term ‘to teach’ is important here as it conveys a sense of learning, which joins DVBD men and villagers in a forward-moving community of modern Kenyans. Talking about a particular ethnic group, which was not represented among the DVBD staff, one man illustrated the link between government and knowledge: ‘in those days they hadn’t yet entered government. They had not begun learning. They were still backward’ (Otiende, 29 March 2001). The men considered the transfer of knowledge to the wananchi a corollary of successful scientific health work. Rather than denouncing this off hand as an insidious form of discipline, we should recognise the hopes entailed by this project, which both expanded knowledge and spread it. Somewhat surprisingly – in view of scholarly fascination with rumour and resistance to colonial medicine (White 2000) – western Kenyan 309

Evidence, Ethos and Experiment

wananchi generally seem to have welcomed the DVBD men and their endeavours. Memories of control campaigns might seem particularly positive from the vantage point of the desolate present, but the elderly villagers’ praise of government disease control and their chagrin over its present poor state, and the bewildered silence that often met our prompting for conflicts – ‘but there was lots of sickness … would people who are sick object to medicine?’ – suggests that wananchi indeed appreciated the work’s positive intentions. Blood-stealing rumours did occur and were remembered by the men when we asked, but they seem not to have led to resistance (see White 2000; Geissler 2005). Despite occasional disagreement with villagers about specific methods, labour recruitment, quarantine measures or drug side-effects, mid-century rural western Kenyans, maybe especially after gaining national independence, seemed to have accepted their role of citizens of public health, as members of a national collective guided by science. In this way, fieldwork campaigns extended the scientific project into a political one.

Becoming Men of Scientific Government Along with fieldwork campaigns and the emergent project of a healthy nation, the DVBD men pursued a personal transformative project: making themselves as modern men of science in a modern nation. This was as much a professional as a private trajectory. Education and learning were central to this, but also mundane aspects of everyday life, and the making of a family.

Becoming Scientific Subjects All DVBD men had received above-average schooling – the old men Primary, the young ones Secondary or Tertiary education. The oldest men had been recruited from school into government during the late 1930s: My father was just a human being. He had nothing. When I left school I was like naked because I had only the school uniform. On the first day a brother of mine took me down to stroll in Kisumu and look for employment. We walked around. … I first went to the police, but I was too short. From there we went to … the Provincial Medical Officer. We walked to the hospital and I was taken and he was left. I had luck. That is how I started the work of yellow fever inspector [later DVBD]. I was asked what 310

Parasite Lost

class I reached and I said ‘class six’; ‘Did you pass?’; ‘Yes, Sir, I passed’, ‘Then come on Monday and start work’. (Otiende, 26 February 2001) Some men had tried other government services or had failed to become soldiers before joining DVBD; for others, DVBD had been the only available opening into the medical field, which they had aspired to, often based upon early encounters with government doctors or with life-threatening illness. The men did not derive from the top layer of colonial rural life; unlike the progeny of the first Christian converts or of colonial chiefly families, many of their fathers had been pagans and none had been a chief. Indeed, they seemed to have pursued education and science in competition with wealthier Christian children, who from the 1940s onwards began entering clerical work and administration. Their childhood stories were full of hard-

Figure 12.5: Portrait with ‘The Mosquitoes of South Sudan’ (author’s photograph). 311

Evidence, Ethos and Experiment

ship, cattle herding and long walks to school; learning remained their lifelong aim; and books were among their valued possessions. One man, in his late eighties, thus asked to be photographed with what I first assumed to be a Bible, but which proved to be a 1939 edition of ‘The mosquitoes of South Sudan’, bought in instalments after the war (Figure 12.5). The most important book in the old men’s homes was nevertheless the Bible. Many had received baptism in Church schools, and described themselves as ‘strong Christians’. Several admitted that they only ‘found Christ’ late, and had lived life fully as young urban professionals, while others claimed they had led almost monastic lives, dedicated to acquiring knowledge, since their school days. Their commitment to Christianity, science and development – the struggle against insects and superstitions, and for hygiene and salvation, better souls and healthy bodies – was one, as it had been for medical missionaries (see, for example, Fabian 1991: 159); and several of the old men practiced after retirement as preachers, Church elders or medical experts in their villages. Education and Christianity were linked to self-discipline, and although the old men predictably lamented the youths’ lack of it, even younger men

Figure 12.6: Portrait of ‘Yellow Fever Inspector’, 1941 (photographer unknown). (Personal archive John Nguriethi). 312

Parasite Lost

Figure 12.7: Portrait with 1982 Uniform and 1970s Land Rover (author’s photograph).

claimed this DVBD virtue and praised ‘British’ punctuality, recalling the early morning parades, as well as dismissals due to minor delays during the early postcolonial period. And young as well as old cherished the uniforms that had been provided up to the early 1980s (Figures 12.6 and 12.7). The praise of self-discipline is common among older Kenyans, but fieldwork adds an edge to it. Exposed to the dangers of a hostile, remote and diseaseridden environment, ‘field men’, like soldiers, had to be enduring and punctual, fair, meticulous and self-controlled: We didn’t quarrel in the field. Because that was a bush and one could kill the other. The bush is bad. If there was a problem we sat down later. If we found a mistake, people accepted; then we could cook, eat together […] and that would be the end (Otiende, 26 February 2001).

Professional Family As this quote underscores, professional self-fashioning was not merely a process of individualisation, but produced new relations: ‘at DVBD, we were like a family’ (Luoba, 22 November 2005). Across ethnic and kinship 313

Evidence, Ethos and Experiment

differences people formed friendships within their age-cohort (Figure 12.8).16 These new forms of brotherhood were intertwined with conventional kinship relations: staff members recruited their kinsmen, others found wives among their colleagues’ families or during fieldwork. Kinship remained the men’s most important social concern, but professional relations and responsibilities were woven into this idiom. Crucial among these new relations were ties between older and younger men. Recruitment along kinship lines seems to have been so common that one old man lamented: ‘There used to be a rule that one could bring one son [to be employed by DVBD], but these days, they have cheated us on that, too’ (Sawe, 24 April 2005). More importantly, ‘bench training’ by older men (initially, British officers) was a mark of DVBD which even the youngest staff who had achieved academic certificates continued to praise as the source of DVBD men’s superior expertise: ‘In those days, people who were employed earlier and had been trained would sit down with us and show us [how things are done]. This made us know the work well from [down] there’ (Ogalo, 24 June 2004).17 This metaphor of growth was un-

Figure 12.8: Plate with Mixed Photographs of DVBD Colleagues, 1960s and 1970s (photographer unknown) (personal archive, Longinus Pamba). 314

Parasite Lost

derlined by a gesture of a tree growing that normally illustrates children’s growth; paternal bench training was seen as appropriate means of making younger men grow in a familiar, familial way, within a modern scientific project. ‘We guided them’ and ‘they showed us their ways’ were other ways of saying this. Some DVBD men had descended from older staff, but even without biological ties, the younger generation maintained lasting relations with their elders; decades after the old men’s retirement, younger staff knew about their wellbeing and whereabouts, as if they indeed were kinsmen. And whenever an old DVBD man passed away, his colleagues buried him, and until the last Land Rover broke down, the coffin would be taken on its last ‘safari’ in one of the DVBD vehicles. When they met, younger staff related to their professional genitors with respect and affection. They introduced them as ‘fathers’ and praised their achievements: ‘he could identify mosquitoes by ear’ (Staff Reunion, 12 December 2004). Due to the expanded educational opportunities after independence, generational respect and professional rank were sometimes at odds with each other. Thus, when we visited the home of one of the oldest staff, a former yellow fever inspector and lifelong DVBD man, accompanied by the retired Head of the DVBD – a generation younger, holding a UK doctorate and author on innumerable scientific papers – the

Figure 12.9: Urban Housing Estate in Shauri Moyo Kisumu Where DVBD Staff Lived With Their Families in the 1960s (author's photograph). 315

Evidence, Ethos and Experiment

two men, who were both equally enthusiastic about the encounter, kept exchanging polite and affectionate remarks of praise, such as ‘he is my officer’, ‘he taught me’. A feast was prepared on the occasion of the successful younger man’s visit, but when food was served the older man still washed his hands first. During dinner, the older man inquired about younger DVBD staff whom he had once trained, and was satisfied to learn that many of them had progressed to go abroad; he praised their Head, his ‘son’, for honouring his responsibilities and ‘taking well care of his people’ – much like a grandfather would praise his son for the grandchildren’s educational progress. The emphasis on parental bonds and intergenerational relations does not mean that all was respect and responsibility. Styles and attitudes of old and young differed (compare Figure 12.1 and 12.9). Among themselves, the old men bickered at times about their sons lacking discipline and ability, and they recalled, much like their erstwhile British officers, their irritation over bellbottom trousers, open shirts, sideburns and ‘Afro-look’ in the 1970s, when khaki uniforms went out of fashion. Meanwhile, the young men mocked their professional fathers’ ‘British’ accent and attire. Intergenerational conflicts were part of this family life. After independence, the old men who took over their colonial officers’ uniforms and privileges also inherited the attendant relationships and potential conflicts, as the 1960s fieldwork photograph of the DVBD’s first Kenyan Field Officer with ‘his men’ illustrates (see Figure 12.3). A particularly noteworthy conflict was about the practice of ‘man-baiting’, catching mosquitoes on the bodies of staff members, which the younger men resented and apparently got rid of after a strike in the 1970s.

Modern Kenyans The government scientists’ professional kinship created not only bonds across the discontinuities of scientific progress, epidemiological change and political ruptures; it also extended beyond the confines of locality and origin, to embrace the emergent Kenyan nation. Bench training was based on staff rotation between DVBD stations across Kenya, including stays in the laboratories of the Nairobi headquarters. In the 1960s, recruits spent, as part of their increasingly formalised training, periods of time in different stations, studying, under the supervision of local senior staff, the research and control of the diseases common in different parts of Kenya. These movements gave the men opportunities to enjoy life outside the immedi316

Parasite Lost

ate control of their kin, extended the social bonds between DVBD generations across the nation, and gave the men a shared knowledge of their country: its environments and diseases, its peoples and lifestyles, and its development projects and challenges. On their travels, they learned not only about microscopy and mosquito dissection but also about city life and about Kenya as a multiethnic nation (and not infrequently the men founded inter-ethnic families). They searched insects among pastoral tribesmen and in the backyards of Indian traders, and lived and worked with people from different parts of Kenya. During these peregrinations their primary attachment remained the capital city to which they regularly returned, before many of them, but not all, got posted in their area of origin. Much like their colleagues from the Kenyan Railways (Grillo 1973), DVBD men formed a truly Kenyan community, joined by their commitment to the lives of citizens and their modern nation.18 While staff rotation created a sense of community and of progressive movement, it also contributed to the production of an important separation, alluded to in the discussion of campaigns: that between the men and their field. While the men claimed to be at home anywhere in Kenya, in an important sense they were nowhere at home, because their lifestyle and professional detachment left them somewhere outside; as a then young trainee recalled from the 1960s, when he had learned about Leishmaniasis in the northeast: ‘Local people could never pass through the bush by the side of the village where our camp was. They feared we were in that business of blood’ (Luoba, 29 September 2005). DVBD men knew the Kenyan nation much in the same way they knew the villages in which they camped: they were right in the middle of it, and yet a little removed, as befits their status as men of science and men of modernity. This detachment from local ties in favour of a national network could be read as an expression of the modern state’s (and science’s) distanced gaze; alternatively, more in line with the men’s self-esteem, one could see it as nascent national citizenship.

The Generation of Science The large organogram in the DVBD Head’s office in Nairobi (Figure 12.11), displaying the entire institution panoptically – each man a cardboard slip in a brass holder, grouped by stations and regions – evokes stereotypes of modern government discipline: abstract, impersonal, hierarchical and static. Yet, if we add the narratives of professional kinship, above – about generational learning, mutual respect and responsibility, lifelong commitment – it begins looking more like a family tree. Not an instantiation of 317

Evidence, Ethos and Experiment

‘seeing like a state’, a static hierarchical simplification, but a moment in a larger movement through time. The Head himself, who began in one of the slots, moved around, learned and aged, and rose to the top: not the centre point of a bureaucratic order but an ancestor who watches over his continuously growing and developing family. The role of paternal bonds and respect for the past also qualifies the assumption that the progress of modern science relies on closures and on cutting its ties to the past – through discoveries and innovations – and that modern science thus is inherently presentist. DVBD men were keenly aware of the debts that they had towards older colleagues and situated their knowledge in long-term creative processes. This aspect of generation – making government and creating science – in and through lasting material relations, this role of time and relations in the making of state science, deserves further scrutiny in our reflections about science and government – colonial and otherwise. In order to look at generation from another angle, I will in the last section of this paper turn to the men’s private lives, and the way in which personal projects are intertwined with those of science and government.

Making Modern Lives The DVBD men usually resided in towns and cities. As they learned about science and contributed to the knowledge and health of the nation, they also transformed their personal lives, accumulating education and modest material wealth, participating in urban life and exploring their country (and in some cases beyond) and, last but not least, raising and educating their families. In the campaigns, with which I began this account, the men set out to transform the non-modern ‘field’ they had left behind; in their lives in town they pursued another facet of the same project: their mental and material self-transformation; in their rural retirement, which I will turn to in the end of the chapter, the two came together.

Coming to Town Most men described their early work years, moving between cities and stations, as a time of youthful license. One man, who had been active in the anti-colonial struggle and was still known within DVBD as ‘the freedom fighter’, remembered:

318

Parasite Lost

By then I had not known Jesus and I drank! At four after work we used to go to Kibera [in Nairobi] to drink, then we went to a dance hall; we were young and we enjoyed life. We did not think of [getting] a wife. We used to earn little money but things were not expensive, it was enough to buy food and even beer. And then it was still safe in Nairobi to move at night [i.e. before 1950s colonial repression], and it was much fun. (Nguriathi, 10 April 2005) Not all were so gregarious; some claimed never to have gone out at all, apart from to Church; but all remembered a time of relative affluence. Since the government provided the men with housing, allowances, uniforms and transport, the money they earned: ‘would remain. I would not spend it in a month, but bring money to mother. I would even buy a goat and take it home’ (Achieng, 17 December 2004). Only after they had ‘done’ several diseases, and often in conjunction with their promotion from laboratory attendant to technician and the attendant pay rise, the men married.19 With few exceptions, even the older men had one wife, unusual for their age group, but maybe more common among, usually Christian, ‘government people’. Most of the men belonged to the Anglican Church or Catholic Church; only a few had later joined an evangelical group, but very few had chosen one of the popular ‘African’ Independent Churches, underscoring their commitment to a modern, global community of faith – akin to universal science – rather than to an explicitly ‘African’ or ‘traditional’ identity. Talking about their marriages, the old men emphasised their wives’ rural lives in contrast to their own urban exploits: With us, once we are married your father gives you land. When you have been given land, you take your wife and you establish a home with her. When you are going somewhere even to work in Nairobi, you leave your wife there. So you would come to visit her after every end month if there is money and she would just stay there to till the land (Otiende, 26 February 2001). When we talked to their wives or widows, they claimed by contrast that they had only occasionally gone ‘home’ to till and look after rural kin, but had lived in the city most of their adult lives, and their styles in terms of dress, furnishing and cooking were urban. However, even though wives shared the men’s urban lives and mobility between stations, they never went to the 319

Evidence, Ethos and Experiment

Figure 12.10: Younger DVBD Staff Members, Mid-1970s (photographer unknown) (personal archive, Ambrose Masime).

Figure 12.11: Organigram of the Entire DVBD Displayed at the Director’s Office, DVBD HQ, Nairobi (author’s photograph). 320

Parasite Lost

field. Field was a male domain, outside, uncivilised and diametrically opposed to the modern home in the city.20 Like few other professional groups, the DVBD families lived the gendered distinctions implied by colonial forms of domesticity (e.g. Mutongi 2007); the men’s crusade against infection, wilderness and ‘backwardness’ was complemented by their wives’ efforts at producing a stable nuclear family within a modern, ‘permanent’ house. Until the 1980s decline, the men (and later their families) were housed, as part of their contract, in government housing, which initially was in the local hospital or the DVBD station, and later on in the housing projects that marked the development of Kenyan cities, especially after independence and up to the 1980s. These houses reflected both in their layout – kitchen, living room, bedroom(s), veranda, front and backyard – and in their organisation in ‘estates’, named ‘Lumumba’ or ‘Mboya’ after progressive African leaders, complete with schools, streetlights, shops and pubs, the modern ways of life that their inhabitants aspired to (Figure 12.8, 12.10). Here, the everyday of government men included forms of social life such as football, dancing and participation in congregational life, as well as drinks after work.21

Figure 12.12: Group picture of DVBD Staff Families on the Occasion of 1963 Christmas Party (photographer unknown)( personal archive, Alfred Lwoba). 321

Evidence, Ethos and Experiment

Urban living conditions, in conjunction with lifelong employment, facilitated close social relations among the men and their families (Figure 12.12). At the age of retirement they had known each other for thirty or forty years, had shared youthful city life as well as each others’ weddings, witnessed life crises and Christian revivals, had been godparents of each others’ children and supported their education and careers. They had seen times of plenty and times of scarcity, and shared the slow decline of government services, from post-independence heights into the present state of abjection. Like the abandoned laboratories in which their younger colleagues today wait for new ‘projects’, the decaying houses and overgrown streets of Kisumu’s estates, where rubbish and high grass harbour mosquitoes and disease, were to the old men evidence of ‘things falling apart’: during their time, ‘mosquito inspections’ and rubbish clearings would have prevented such dangers.

Propagating Modern Life One factor, apart from religion, modern self-image and housing conditions that may have contributed to the low rate of polygamy and to comparatively small families among these men was the emphasis on children’s education, another facet of the wider progressive project that the DVBD’s scientific work was part of. Most of the men’s children had been educated up to Secondary school, and many of the old men’s sons had gone on to university, often abroad. Their children’s achievements – even if these in many cases led them into other countries – were a source of great satisfaction, and their photographs, in academic attire, holding diplomas, in front of houses and cars, adorned the walls of the old men’s living rooms. Often the old men complained about their grandchildren’s unemployment and asked us to find them a place in government research. Despite their high educational levels, exceeding their grandfathers and fathers, today’s youths have little chance to proceed along the once clear-cut pathways that had led the older men to success (see Whyte and Whyte 1998). Education has little currency on today’s labour market, just as scientific knowledge has little impact on the workings of government – it is money not knowledge that rules, as the old men often lamented. In spite of this observation, education remained the men’s highest value, and science the standard of education; several of them remarked with chagrin that their grandchildren studied business or law rather than science. The progress of science and of modern government, of the nation and its citizens, and the growth of the own family in its most elemental sense 322

Parasite Lost

were intertwined tasks for the men, no matter whether they cleared bush and took skin snips, learned about mosquitoes and studied the Bible, invested in cash crops and sent their children to school, went to the Public Service Club for a beer on ‘members day’ or preached to their fellow Christians on the following Sunday. The totality of this modern project, and the appeal of its irreversible forward drive, rooted in close and lasting social ties and in a commitment to the past, became particularly clear to me when we stayed in the home of one of the oldest men, in the heart of Luoland. I end the account with this visit, a particularly pleasant moment in fieldwork, which provides a suitable counterpoint to the ruined field station with which I began.

Looking Down the Hill The appropriate starting point for approaching this homestead is the DVBD Kisumu station. In its back is an incongruous little round and grass thatched house built of bricks. It is the last remaining of a series of small experimental houses that in the 1930s had been built throughout the ‘African’ parts of the city to collect mosquitoes by night. In the 1940s, when the present station had been built around it, it had been given to one of the promising young ‘inspectors’ and his new wife, who had their first child here. A family from which many Kenyan science workers and doctors would spring was founded in an ‘African hut’ designed, according to colonial imaginaries, for scientific research, adjacent to a laboratory block. From there, the family moved to different urban estates and DVBD stations, while the household head, Mr O, built, like all DVBD men, a house in his rural ‘home’. Eventually, some years after his retirement, the family settled there. The home, a collection of large modern houses, several kilometres from the tarmac road, that stands out from the surrounding mud and grass homes, is set up as a customary Luo homestead: on top is the modest house of Mr O., built in the 1950s, characteristic for the generation of colonial civil servants, modelled upon the urban bungalow, with a small, mosquitoscreened veranda and surrounded by fruit trees. Across the hill slope below this house are the houses of the sons, substantial family homes of absentee owners living in town, growing in size as they descend to the mansion of the youngest son, a senior doctor. Following custom, the houses are positioned according to generation and seniority, ensuring that rainwater will run off ‘downwards’; yet they are also connected with electricity lines, water supplies and sewage pipes. 323

Evidence, Ethos and Experiment

These houses, uninhabited for most of the year, are the only houses with such amenities in the area, and each of them is larger than any other home on the surrounding hillside. Sitting on Mr O’s veranda, overlooking his son’s roofs, one appreciates the homestead head’s sense of achievement. It is a characteristically modern form of paternal pride; it goes beyond the mere satisfaction over one’s offspring. It is also an intentional actor’s project: this reproductive achievement is self-consciously creating new lives that are different from that of the father, let alone his ancestors’ lives; it moves beyond what has been and opens new trajectories, and it still remains part of a shared history. All sons, as well as the daughters – who for customary reasons do not build houses in the father’s homestead – have

Figure 12.13: Portrait of Mr O. 2004 and 1945 (2004)(author’s photograph). 324

Parasite Lost

attained the highest qualifications, funded from Mr O’s income, and have become urban professionals. Several sons are doctors, following their father in the health field, and fulfilling the father’s childhood dream, who when he joined DVBD had been inspired by the capacity of medical professionals. A granddaughter is a successful laboratory scientist – though not with government but with a transnational research collaboration – extending the family heritage. Their home embodies on the level of personal life the larger project discussed above in relation to scientific campaigns, and to the DVBD as an institution. The generation of change, through scientific work and ‘being with government’, is here intertwined with changing generations. Professional achievements and family success are intertwined sources of pride, which is reflected in the portrait of Mr O, with his photograph as a young man. Between the 1944 studio photograph of Mr O, aged 15, in his Yellow Fever Inspector uniform, shortly after leaving his father’s village, and seated above the rural family home sixty years later (Figure 12.13), lies a journey out into the world, a transformational ‘safari’, in which personal ambitions and the making of a successful family were integral parts of a larger professional modern project.

Generation Across professional and private aspects of the DVBD men’s life, links across time, bonds of generation, creative, transformative relations between the old and the new are critical. The men’s present was envisaged as being on the move, in the rapid transition between the past – as a source of pride, knowledge and achievements – and the future of better lives in a different society. From the vantage point of the men’s stories, the scientific government institution, DVBD, appears not as a static, hierarchical structure, made up of vertical connections that link the central state ministry to its farthest outpost across the territory. Instead of being a scientific segment of state bureaucracy, it is a current in a larger movement, a project composed of many, varied individual life courses.22 Whyte and Whyte (1998) observed, in relation to Ugandan villagers’ life with ‘development’, that the latter intertwines individual achievements, kinship values and universalist notions of progress. All of these are linked by the terminology of ‘growth’ (1998: 230–31). For the western Kenyan DVBD men, progress or development are dongruok, ‘growth’, applied equally to science and society, and children’s wellbeing and education (see 325

Evidence, Ethos and Experiment

Prince and Geissler 2010). This notion of growth takes form in the generational and generative ties in the men’s work, and in their investments – and their contributions to public health – into their children’s better future. Growth in this sense is keenly aware of its origins and of the open horizons that lie ahead. Contrary to the critique of ‘high modernism’ among some anthropologists, it is emphatically not a radical break with traditions and older relations, nor an imposition of a pre-determined future (see Scott 1998), but an open process, implying possibility and renewal – exemplified both in seeing one’s children outgrow oneself, and in making scientific observations beyond what was previously known.

Conclusion This paper examined the memories of Kenyan government scientists’ of my father’s generation; together with them I searched for their, or our, lost modern times. I showed how the DVBD men’s project – public health, implemented by the nation-state, guided by scientific research – was articulated throughout their professional and personal lives; similar narrative patterns, similar intentionalities or temporalities appeared across different domains and levels of scale. In fieldwork campaigns the DVBD men progressed, through investigations and discoveries, tied in with movements and actions; they transformed knowledge and made new spaces; the ‘field’ was simultaneously seen in different ways and made different; and this epistemological and governmental effort was seen as part of the making of an emergent collective – eventually, the Kenyan nation. Crucial for this work was the men’s own ambiguous identity as both citizens and being with government. This mediating role implied continuous personal transformation and self-improvement; these were men on the move, although their particular direction was never unequivocal. Learning and knowledge were central in this personal quest, which implied a lifelong process of individual education, and generational relations between more and less knowledgeable men. Finally, this personal project extended beyond work into the men’s family life, in which the itineraries of learning and improvement permeated the everyday and extended to future generations. This comprehensive project claims that it is possible to understand the world, and that it can be changed. The two kinds of transformation – of knowledge and of what has been made known – are intertwined: scientific discovery relies upon (field) work in the world, and governmental action upon the world is guided by ever-changing knowledge. Both forms of trans326

Parasite Lost

formations point forward, one towards truth the other towards the good life, without necessarily aiming for a definite end point. What links disease eradication and domestic life, scientific learning and discovery, public health and child rearing, is this overall notion of forwardness. It is this certainty of direction and movement, which once had engulfed their entire lives, which elderly Kenyans yearn for in their present state of, as they put it, ‘being stuck’. The material leads me to qualify some current anthropological and historiographic engagements with state and medicine that view the state as an entity vis-à-vis its people, and focuses on the controlling and disciplining nature of state science (as in the seminal works of Comaroff (e.g. 1983) or Packard (e.g. 1989)). Progressive narratives about welfare or equality, health or nation-building appear in such accounts as mere underpinnings of an apparatus of power, ‘grand narratives’ that legitimise domination; colonial medicine and public health have been used to underpin this argument (e.g. Hoppe 1997). Worse still, these narratives’ lack of fulfilment that often can be demonstrated retrospectively prove according to such analyses their deceptiveness (and, implicitly, the impossibility of intentional societal change). While such critique of governmental power and of science has been valuable to debunk older narratives of quasi-natural progress in terms of inevitable ‘discovery’ and linear ‘modernisation’, the DVBD men’s lives put into question the taken-for-granted duality of state and people, science and subjects; the men’s stories about negotiating the ‘field’ between these supposedly separate entities suggest that we should approach state/government and science/medicine not only as dispositives applied to people in a subject-object relation but always also as projects endorsed by them. These projects call up a collective whole, a sense of ‘society’ – in the terminology of their time, the nation, ‘Kenya’ – that comprises people and state, science and scientists and their subjects and gives them a direction in time. The imaginaries of people versus state, or science versus subjects, erect a spatial construct of sorts, an order of power, held together by the polarity of domination and resistance; this emphasis on structure – which is only reified by contrasting it with ‘resistance’ or ‘agency’ – tends to neglect or even belittle the distinct temporality that always also shapes modern government and science – as the DVBD men showed throughout their lives. This sense of progressive time, of human-made changes, within which government and science are efforts to know and transform society, is more central to the project of enlightenment than the rendering as ‘high modernism’ allows us to see; faced with the timeless, yet violent, presentism of the current world order, it might be time to get it back into focus. It is nei327

Evidence, Ethos and Experiment

ther mere ideology, deflecting from the brutal reality of power nor can it be measured, and rejected, against the yardstick of fulfilment, as the sober posthoc analyses of ‘failed’ state welfare projects claim. Rather than a disguise of domination, or a blueprint that continuously fails its realisation, this project is a vector of intentionality, a notion of forwardness; a longing, a desire. Talking to the old men it often seemed to me that they still held more of this desire – for truth and for the possibility of change, for unknowable horizons and forward movement – than I, two generations their junior, could muster. Their lives and itineraries are different from mine; they belong to the past; yet, their sense of direction inspires me, stuck in a present in which, indeed, structures of domination seem to obscure our sense of possibility.

Notes 1. I am grateful to the DVBD’s hospitality and support. Thanks to Philister Madiega who actively participated in the interview work, and transcribed many of them in an empty laboratory, which the DVBD had availed to us. I am grateful to Professor John Ouma and Dr Eric Muchiri, respectively previous and present Heads of the Division for their encouragement; and to Dr Alfred Luoba, late head of the Kisumu Branch, who participated in the fieldwork and continuously inspired the project. Drs Kenneth Ombongi of Nairobi University, and Maureen Malowany, McGill University, participated in this collaborative project and prepared publications on different aspects of it. We received valuable advice on all stages of the project from Professor Virginia Berridge. The idea for this research project was developed, and the earliest interviews conducted together with Dr Ruth Prince. The larger study of medical research around Kisumu, which this material is part of, are funded by Wellcome Trust, Leverhulme Tust and British Institute in East Africa. 2. While the DVBD Kisumu employed in its heyday about fifty people, KEMRI Kisumu employed in 2008 more than 1000 people to run overseas-funded clinical trials. 3. With the exception of two secretaries, the DVBD consisted of men. 4. This work is part of an ethnographic study of medical research in western Kenya, based in the research group Anthropologies and Histories of African Biosciences (http://aab.lshtm.ac.uk) at the London School of Hygiene. 5. Failed states are not ‘capable of carrying out functions that they themselves claim and that they are reasonably expected by their populations to carry out’ (Nodia 2002). 6. The relativist counterclaim that the state in Africa still ‘works’, just not as we expect it to, is equally unsatisfying as the normative ‘failing state’ argument (see Chabal and Daloz 1999). 7. Here and below I quote the informants with their own names, as all of them had requested when interviewed. 328

Parasite Lost

8. Kala Azar and sand-flies were found in arid areas, Bilharzia and snails, and malaria and mosquitoes in humid lowlands and so on. The men associated every DVBD station with specific diseases and campaigns. 9. The endemic diseases river blindness and bilharzia where thus described as ‘outbreaks’, when economic interests (labour recruitment and irrigation schemes, respectively) and new technologies (DDT and molluscicides, respectively) turned them into a focus of campaigns: ‘Bilharzia broke out. We took some people to go and do it’ (Okonji, 14 December 2004). Diseases that had been common for long appeared to staff (and villagers) as new diseases, because the research created them as named entities and targets of action; even diseases that did not actually exist in the colony but were international concerns – notably yellow fever – could in this way orientate DVBD’s work and the men’s professional identity (as Yellow Fever Inspectors, see below). 10. The epitome of such innovation was DDT, which after its first Kenyan trials was employed for spraying bush and houses and dousing rivers and ponds against insect-borne infections; the invention of other insecticides and molluscicides, and new applications, such as insect traps or aerial spraying, and of new chemotherapies generated continuously new disease foci and new campaigns. 11. Even new public health concerns, less determined by chemistry than by social science, such as ‘environmental management’, ‘gender sensitisation’, were inserted into the established pattern of campaigns: ‘doing x’. 12. Even the two DVBD activities that did not involve a distinct ‘field’ and had no clear beginning and end – urban malaria control (searching and destroying mosquito breeding sites) and yellow fever control (spraying ships, trains and aeroplanes that entered the colony) – relied on the same logic of territorial threat and conquest as ‘campaigns’; one maintained and expanded clean urban spaces in contrast to disease-ridden bush (designated as ‘reserve’ of traditions and pathogens); the other defended the borders, overlaying political demarcations with epidemiological ones. 13. DVBD fieldwork practices marked this separation. Thus, DVBD staff camped in their government tents within villages, and even within the popular areas of Kisumu they operated from specially constructed round tin huts, where they slept to night-catch mosquitoes. Staying in ordinary houses or living with local people would presumably have compromised the purity of their scientific and governmental position. 14. Mosquitoes, which the men presumably encountered every night, were in the context of fieldwork construed as threat that had to be countered by nets and insect repellent, underscoring the way the men re-made themselves and their environment as ‘fieldworkers’. 15. Tents also marked social differences within the camp, notably that between officers’ tents with floorboards, furniture and veranda, and the men’s communal tents, and continued these after independence. 16. Their lasting bonds of friendship became particularly evident at the funerals, where DVBD colleagues featured prominently and at times overshadowed the deceased’s brothers. 329

Evidence, Ethos and Experiment

17. The oldest DVBD men were initially employed as ‘yellow fever inspectors’ and subsequently became laboratory attendants or boys, then entomological laboratory assistants and, after further training, since the late 1940s laboratory technicians and senior technicians. Only after independence could they, through formal training, achieve the rank of laboratory technologist and (scientific) officer. 18. This trans-ethnic community was put on trial during the anti-Mau Mau counterinsurgency, and during postcolonial ethic clashes. Although the few Kikuyu-speaking DVBD men that had participated in the struggle said they had not been able to discuss politics at work, none of those we interviewed suggested that colleagues had reported on them (on the contrary, they told about support from non-Kikuyu men). And none of the men spoke about involvements in ethnic clashes, apart from stories in which DVBD staff had protected each other irrespective of tribe. In one case, a man and his family was housed for a long time in the DVBD premises because he, as a non-Luo, had been at risk during ethno-political conflicts. 19. For the oldest men, the general strike of 1947 brought welcome assistance, since: ‘after that strike we got forty shillings, and that was enough for me – I married … I bought cows from that money’ (Okonji, 24 June 2004). 20. The men shared this view with their officers who disapproved of the idea of staff wives in the field, although some of them did go to the field with their own wives. 21. Beer consumption illustrates the men’s struggles for their share of modernity: in colonial times bottled beer was not encouraged among Africans, who accordingly yearned for it; beer halls became the social joint after freedom was attained. This struggle about beer turned around in the 1980s, when President Moi abolished traditional beer brewing; now the men, due to their diminishing salaries and their old age, yearned for the neighbourhood beer parties they had had as young men. 22. The centrality of growth and generation, especially relations between men of different procreative generations, could be said to reflect something typically ‘African’ – just as some believe state patrimonialism and paternalism to be rooted in African kinship and sociality. A better approach would examine the links between the creativity and generationality of science, state and kinship not as African idiosyncrasies but as a universal, often neglected aspect of modern science and government.

References Beissinger, M.R. and C. Young. 2002. Beyond State Crisis? Postcolonial Africa and PostSoviet Eurasia in Comparative Perspective. Baltimore: Johns Hopkins University Press for Woodrow Wilson Centre Press. Chabal, P. and J.-P. Daloz. 1999. Africa Works: Disorder as Political Instrument. Oxford: James Currey. Clarke, S. 2007. ‘A Technocratic Imperial State? The Colonial Office and Scientific Research, 1940–1960’, Twentieth Century British History 184: 453–80. 330

Parasite Lost

Comaroff, J. 1993. The diseased heart of Africa. Medicine, colonialism, and the black body. Knowledge, Power, and Practice: The Anthropology of Medicine and Everyday Life. S. Lindenbaum and M. Lock. Berkeley, University of California Press: 305-327. Cooper, F. 2002. Africa Since 1940: The Past of the Present. Cambridge: Cambridge University Press. De Boeck, F. 1998. ‘Beyond the Grave: History, Memory and Death in Postcolonial Congo/Zaïre’, in R. Werbner (ed.), Memory and the Postcolony: African Anthropology and the Critique of Power. London: Zed Books, pp. 21–57. Fabian, J. 1991. Time and the Work of Anthropology. Critical Essays 1971–1991, Harwood Academic Publishers. Ferguson, J. 1999. Expectations of Modernity: Myths and Meanings of Urban Life on the Zambian Copperbelt. Berkeley, University of California Press. Geissler, P.W. 2005. ‘Kachinja Are Coming: Encounters around a Medical Research Project in a Kenyan Village’, Africa 752: 173–202. Grillo, R.D. 1973. African Railwaymen – Solidarity and Opposition in an East African Labourforce. Cambridge: Cambridge University Press. Hoppe, K.A. 1997. ‘Lords of the Fly: Colonial Visions and Revisions of African Sleeping Sickness Environments on Lake Victoria 1906–1961’, Africa 671: 86–106. Hunt, N.R. 1999. A Colonial Lexicon: Of Birth Ritual, Medicalisation, and Mobility in the Congo. Durham: Duke University Press. Mbatia, P.N. and Y.W. Bradshaw. 2003. ‘Responding to Crisis: Patterns of Health Care Utilization in Central Kenya Amid Economic Decline’, African Studies Review 461: 69–92. Mutongi, K. 2007. Worries of the Heart: Widows, Family, and Community in Kenya. Chicago: Chicago University Press. Nodia, G. 2002. ‘Putting the State Back Together on Post-Soviet Georgia: Beyond State Crisis?’, in M.R. Beissinger and C. Young (eds), Postcolonial Africa and Post-Soviet Eurasia in Comparative Perspective. Baltimore: Johns Hopkins University Press, pp. 413–43. Packard, R. 1989. White Plague, Black Labor: The Political Economy of Health and Diseases in South Africa. Berkeley: University of California Press. Prince R.J. and P.W.Geissler. 2010. The Land is Dying. Contingency, Creativity and Conflict in Western Kenya. Oxford: Berghahn. Scott, J.C. 1998. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press. Tsing, A.L. 2004. Friction: An Ethnography of Global Connection. Princeton: Princeton University Press. White, L. 2000. Speaking with Vampires: Rumor and History in Colonial Africa. Berkeley: University of California Press. Whyte, S.R. and M. Whyte. 1998. ‘The Values of Development: Conceiving Growth and Progress in Bunyole’, in H. B. Hansen and M. Twaddle (eds), Developing Uganda. Oxford: James Currey, pp. 227–44. Zartman, W.I. 1995. Collapsed States: The Disintegration and Restoration of Legitimate

331

Chapter 13

Is the Sharia of the Doctors Killing the People? A Local Debate on Ethics and the Control of HIV/AIDS in a Rural Area in Kenya Suzette Heald

The implementation of HIV/AIDS testing and therapy programmes is rapidly turning Africa into an arena not only for medical research but for medical experimentation. The urgency of providing ART on a populationwide scale in countries such as Kenya inevitably raises questions about infrastructure in rural areas with few facilities to monitor patient tolerance and treatment. Further, practice slides into research as medical staff routinely collect and report data on their patients and, with drugs and funding in many cases coming from trial programmes sited elsewhere, ‘interesting’ cases are referred onwards to research centres. It is not the intention in this chapter to comment on the discrepancy between the treatment offered in western and African settings and the ethical issues involved in global inequality. My focus in this paper is narrower. The emphasis on therapy is increasingly justified in humanitarian terms (Nguyen, this volume) and largely divorced from the need for prevention. Even though HIV is now taken as ‘endemic’ in Africa, provision for mass testing is not to the fore. Voluntary counselling and testing (VCT) remains the main, and for many, the sole option for those who wish to know their status. It has been strongly argued that the goals of public health have been jeopardised by the protocols surrounding VCT. De Cock and his colleagues (2002, 2003) have taken issue with the AIDS ‘exceptionalism’ position, as it has become known, where this disease alone is subject to special procedures that prevent it from being treated by the means that are used to contain other dangerous contagious diseases. Their arguments contrast the imperatives of public health with the particular interpretation of human 333

Evidence, Ethos and Experiment

rights which has dominated thinking in the AIDS area.1 The question posed is whether voluntarism and the ethics of normal doctor/patient interactions should remain dominant, with their mandatory confidentiality, in the face of the fact that such measures do little to prevent, and may indeed encourage, further transmission of the disease. If, then, there has been international scholarly debate, this paper must be seen to take up the issue at the other end. It reports on the opinions and criticisms of people in one small area of Kenya, Kuria District, in 2005 by the ethics of medical practice and their ideas of how the epidemic should be contained. Although healthcare staff reported a reluctance to test, this was not supported by interviews. Reluctance centred not on testing as such but on the procedures involved in VCT. A dominant view was that there should be widespread – ideally universal – testing with the results made public. At the forefront of people’s minds was the danger of transmission of this silent disease and they responded accordingly, with an eye to minimizing risk and enhancing their own safety. Some advocated quarantining. Their responses thus fall squarely on the side of public health and against the codes that govern the provision of services for HIV/AIDS. As one man, Mwita, put it, ‘the doctors must decide whether they will let their sharia (law) kill the people’.

AIDS in Kenya Kenya was extremely late to recognise the pandemic and mobilise resources with regard to HIV/AIDS. To the denial of the 1980s was added the economic stringency of the 1990s as structural adjustment policies led to a rundown of public services. It was not until 1999 that President Moi declared AIDS a national emergency,2 at a time when international donors, particularly the World Bank, were insisting on the establishment of a National AIDS Control Council (NACC). At that time, Kenya had an estimated national prevalence rate of 13.5 per cent3 but little in the way of dedicated medical services. What services there were came under NASCOP (National AIDS STI Control Programme) of the Ministry of Health. Capacity to treat the general run of sexually transmitted infections was limited and put as low as 43 per cent in Nyanza Province in 1999, the region where HIV infection was most advanced (Kenya, 2000). In the spring of 2001, NACC was promised funding by the World Bank Disaster Fund to the tune of USD50 million to establish the organization and to begin services but, by December of that year, none of this money had been disbursed because of doubts about reliable financial controls. Nor was this fear misplaced, in a country 334

Is the Sharia of the Doctors Killing the People?

notorious for corruption, in August 2003, the head of NACC was suspended (and subsequently charged with fraud) over the misuse of this money and NACC was also forced to cut off funding to fraudulent non-governmental organizations (NGOs).4 This created a further hiatus in actually gaining international assistance and the Global Fund did not approve its provision for AIDS until the Government showed willingness to tackle the corruption issue. The issue was resolved to an extent in August 2003 with the Global Fund disbursing funds through the Ministry of Health to provide free antiretroviral drugs at public hospitals for a five-year period. Things have moved much faster since then with a rapid scaling up of services, from virtually nothing in the public sector in 2000 to the provision of VCT and ART on a nationwide scale. For example, there were only three VCT sites in the whole country in 2000. By May 2005, this had increased to 555. Correspondingly, the number tested each year has risen from less than one thousand in 2000 to almost 400,000 in 2004 (Kenya, 2005a). Alongside this, there have been extensive programmes to raise awareness through public media, particularly the radio, supported at district levels by NGOs and, increasingly, through community-based organizations (CBOs), as World Bank and other donors stressed the value of community involvement and initiatives. Unfortunately, in the corruption stakes, these have proved equally guilty. Constituency-based AIDS Coordinating Committees (CACC) proved no more reliable than NACC in approving funding to organizations and there was little control over the proper use of money. New scandals involving these organizations dominated fieldwork in August 2005, with considerable local anger against ‘briefcase’ CBOs who had failed to deliver the promised services. In March 2005, in the light of yet more public scandals, the World Bank, IMF and Global Fund were all considering delaying further allocation of funds to Kenya and imposing new conditions.5 The future of services to Kenya was thus in some doubt. The late activation of services dedicated to the control of HIV means that Kenya moved rapidly from awareness and educational programmes to medical interventions in the form of VCT and ART. One of the key facets of this has been an almost immediate shift from a focus on prevention to a focus on treatment. Nor have the two been closely tied together. None of the health staff interviewed at district level emphasised the link between the two. Rather, VCT was seen as the first step to possible treatment rather than the means of persuading people to get tested in order to know their status. Kuria District, tucked away in a remote corner of Nyanza Province, on the border with Tanzania is something of a Cinderella district (Kenya, 2002). It 335

Evidence, Ethos and Experiment

was the last district in Kenya to receive antiretroviral (ARV) drugs, with the first consignment arriving only in June 2005, hard on the heels of VCT provision, which first became available in the district less than six months earlier, and was still in the process of development. The advent of testing and treatment raises the issue of infrastructure. Though Kuria suffered as other areas in Kenya in the early to mid-1990s as a consequence of structural adjustment policies and the introduction of cost sharing, this was probably not as noticeable as elsewhere because of the general paucity of medical facilities in the district. The only public hospital is sited near the district headquarters in Kehancha and was given District Hospital status in the mid-1990s. Little was done to develop its infrastructure until 1999 (Kenya, 1997). Even by August 2005, it was poorly equipped, with two in-patient wards unsupported by an operating theatre. For operations and many clinical tests patients had to go to Migori in the neighbouring district, where there is both a large government hospital and a private Mission hospital. For Kuria living in the western sections of the district, these hospitals are in fact nearer than the district hospital in Kehancha, but this does not hold for about two thirds of the 170,000 population.6 The introduction of VCT and ART, together with a large proportion of inpatients said to have AIDS, effectively turned the district hospital into an AIDS hospital. Yet, as has been said, the provision of ART in the district was in its infancy, with only ten patients receiving it by the end of August 2005.7 It only had first line treatments and does not have the facilities to do CD4 counts or to test renal and liver functioning – and patients must go to Migori for these tests. Nevertheless, there has been a noticeable scaling up of staffing and services. ART was introduced at the beginning of 2005, and VCT at the much the same time, with twelve nurses and clinical officers trained by the middle of the year. There were programmes for ART, for the prevention of mother-to-child-transmission (PMTCT), VCT and homebased care (though this still had to mobilise for AIDS). As other rural hospitals in Kenya, it is chronically understaffed, and Government policy of providing counselling by training existing personnel has, in the main, simply added a further task to already over-stretched staff. Further, those in charge of these programmes not only run those in the hospital itself but government clinics throughout the district, as well as supplying drugs and testing kits to NGO, CBO and Mission centres for VCT and other services. AIDS as an issue of general public concern is also relatively recent, with awareness campaigns a strong factor only in the previous two to three years, at a time, though, when the consequences of the disease in rising death rates were clearly identifiable. What is reported here thus has to be under336

Is the Sharia of the Doctors Killing the People?

stood as a slice in time, a picking up of people’s views at a particular stage of the epidemic and at a particular stage of awareness.

Research Methodology The research was designed to canvass knowledge and opinion among people in Kuria District at this particular stage of the pandemic in 2005. HIV positives were not targeted. Instead, I visited families already known though my previous research, in some cases dating back to 1984/85.8 In most cases, these were arranged visits and thus, in effect, social occasions. After initial greetings, I would ask about the problems the family was facing, just as I had done in the past, and this rapidly proceeded to discussions of AIDS, in many cases without any prompt from me. Discussions ranged over how and when they first recognised the disease, their response, in how they explained it, and, lastly, their reaction in the measures they advocated to contain it and protect themselves. There was no regular format to these conversations, however, as they took up issues as they were presented by the families. It should be said at this point that these were unusual social events, in that the topic is not one normally discussed in such family settings as it raises issues of impropriety. However, despite this overriding of normal kinship etiquette, there was no reluctance – indeed, rather the contrary – even though, in the great majority of cases, members of more than one generation were present and women as well as men. Kuria homesteads are still based around a modal two adult generational extended family. It was notable, however, that only women of the senior generation freely entered into debate and discussion, whether they were young and old. Wives of the junior generation, even when in earshot, usually drifted away, probably because they had more pressing duties to fulfil. In all, there were some ten family interviews of this kind. This material was supplemented by other discussions with friends and acquaintances, as well as with healthcare staff, in the course of the research.

Kurian Recognition and Response to the Epidemic It is only recently that anything like reliable figures on prevalence have been produced for Kenya. They still do not exist for Kuria, but one could say that awareness of the epidemic follows the same trajectory as elsewhere in rural areas of Kenya.9 If I go back to the situation in 1994/95, then, many people were aware of AIDS and talked freely about it. There was no specific coun337

Evidence, Ethos and Experiment

trywide educational programme at this time but, through the radio and newspapers, information had filtered through of this new disease. As significant, the first people were being brought home dying, and for burial. I came across three cases during my fieldwork which were attributed to AIDS then.10 These were all of people who had been ‘outside’, as it is put. One was a man who had served in the army, not only outside Kuria but outside Kenya (Namibia), another, a young doctor who had been working in another district, and the last a young woman in her mid-twenties who had left her husband, worked as a bar girl in nearby Migori, and now returned at the point of death. Others questioned in 2005 gave the same witness evidence, with the earliest case quoted dating back to 1993, again of a man who worked elsewhere in Kenya and whose wife died soon afterwards. This early experience has framed the knowledge of many people who associate the risk of contracting AIDS with ‘going out’ and with the ‘working class’, a term that has come to be used not of people located at the bottom of occupational and prestige structures but usually of those at the top, of the educated, particularly those in professional employment. They are the ones with ready money. However, in interviews in 2005, not all put the incidence of deaths so early. Many gave far later dates for the first deaths of which they were aware, the late 1990s or even as late as 2002 or 2003. This is probably no coincidence since it corresponds to the influx of AIDS money and the first real attempts at public awareness. AIDS was mentioned in chiefs’ barazas and, in 2003 and 2004, there were a number of NGO and CBO campaigns in the district. This was clearly crucial because it was notable during my fieldwork in 2001/2 that, although there was knowledge of AIDS, there was little talk of it at that time. It was not a topic on everyone’s tongue. This had all changed by 2005, and AIDS was linked to people’s concern over new patterns of death. People were dying in numbers never experienced before. Further, people had also seen what was happening among neighbouring Luo and Suba populations, where the death toll was higher and there were many ‘closed houses’, spelling the end of families. It is important to specify what people meant when they talked of an ‘AIDS death’. These were not deaths where the person was merely suspected of having the disease but those publicly acknowledged as such, sometimes prior to death and sometimes, but rarely, by an announcement at the funeral. For, in the interregnum, as we could put it, between the mid-1990s and early 2000s, another explanation of the increasing number of deaths had gained currency in the area. They were attributed to a particular form of witchcraft, ‘to be made to drink the chameleon’ (inyawambu), with the 338

Is the Sharia of the Doctors Killing the People?

verb being that used to take cattle to water. The poison is said to be made by drying a chameleon and then grinding the remains, which is then added to food or drink, often porridge. Poisoning is not a dominant form of Kurian witchcraft, and it is double-edged as its effects are believed capable of backfiring on the perpetrator. It is thus particularly significant that this came to be widely believed in the late 1990s, many people saying that they thought it was a new form of witchcraft. Others said it had been known earlier but they could not remember deaths being attributed to it. Despite widespread knowledge of HIV in 2005, many, and probably all, continued to believe that this was a separate complaint. Its symptoms were said to mirror the wasting and weakness of ukimwi, AIDS. The key diagnostic difference was that with the chameleon the victim changed colour, light-skinned people going darker, darker-skins lightening. Some said this became apparent during the course of illness, others that it was apparent in the corpse. Interestingly, no one questioned specifically about the symptoms of AIDS mentioned colour change; the most frequently reported symptoms were wasting, thinning and loss of hair, coughing, diarrhoea, loss of appetite (or the opposite, greediness), one person mentioned boils and another that the nails as well as hair dropped out and the skin became loose. Thus, even in 2005, the recognition was of two distinct fatal complaints. One, AIDS, was a disease of the ‘outside’, transmitted sexually, and dangerous to others. The second, that of chameleon poisoning, was an affliction of the ‘inside’, transmitted through the malignancy of others in the form of witchcraft but not transmissible sexually. It threatened only the individual. The distinction is perhaps not so clear-cut in practice now as it once was since it is recognised that the time has past when only those who ‘went outside’ were at risk of AIDS. Not only have many people now travelled to other areas of Kenya, but the disease, as most people stressed, is now in Kuria, and threatens to cut a swathe through the community. Further, where illness and death is attributed to the chameleon, people admitted that there was always a question at the back of their minds, as this is the more comfortable option for the sufferer and family. Aware that poison could be used as a euphemism, the question of what it was ‘really’ was always posed during sickness and after death because the implications were crucial with regard to transmission. And so, people said that they ‘tracked’ subsequent events to make sure. If the spouse or another known sexual partner died soon afterwards, one could be certain that it was AIDS. If not, the assumption was that it was the chameleon and the sexual partners of the deceased were ‘safe’ and, once sufficient time had 339

Evidence, Ethos and Experiment

elapsed, sexual relations could be entered into with them. Tracking also went backwards, to locate the source of the infection. This was relatively easy with those who had been ‘out’, so that a visit to Tanzania, for example, a few months before the illness could be cited as the source. Even those who knew about the latency period, and in other contexts knew the difference between HIV and AIDS, cited such examples and, when queried, responded that here in Kuria the progress of the disease was much more rapid than in the West. It was widely believed that someone could die within two years, even a few months. But tracking after infection was difficult, as all admitted, for one could only track one or two partners. People demonstrated a very clear awareness of the vast network of possible infection. One image used of the effects of the disease was that it was like a bus accident. It didn’t kill just one but many. The most immediate response to AIDS, then, was to use knowledge, based on observation, in order to minimise risk. In this discourse, it was not sexual morality as such that was the issue, but stupidity, carelessness and lack of a proper sense of responsibility that made one vulnerable to contracting the disease. Cases were quoted of men who had seen a friend die but who still formed a relationship with the widow after only waiting a month or two after the death. Witchcraft as a way of accounting for deaths gives notice of the way the advent of HIV leads to new explanations as, working through indigenous knowledges and concepts, people seek to comprehend and give meaning to new afflictions.11 Further, and significantly, witchcraft causation did not reflect on an individual’s moral probity and thus did not threaten Christian ideas of sexual morality. Church membership, particularly in Pentecostalist and charismatic sects, has soared in the past few years, with there being said to be 56 different churches in the single division of Kehancha today. Roadside preachers attract large crowds, and others, since 2002 when electricity reached the district town of Kehancha, blare out their message through amplifiers day and night. Almost everyone attends church on Sunday, or Saturday in the case of Seventh Day Adventists, and some form of Christian ritual has become the norm at funerals. Thus shame (inchinsoni) has become another key idiom for talking about HIV. Most churches, it seems, avoid the topic of AIDS directly, but preach a sexual morality based on abstinence and faithfulness. Some churches, however, particularly the Pentecostal, take a hard-line view, and people admitting to being HIV-positive risk being thrown out of the church for rejecting the teachings. A few have now begun to insist on premarital testing, with the 340

Is the Sharia of the Doctors Killing the People?

pastor often accompanying the couple to the VCT centre and asking them for the results afterwards. If one or other of the partners proves sero-positive, the marriage is cancelled and this is announced to the congregation. This moral discourse links to the identification of other ‘sources’ of infection. As elsewhere, bar girls, who flood into the district during the tobacco markets, are an obvious one.12 But, given Kurian marriage practices, there are others, such as those women who, because of their position, have to take lovers to bear children. These are the widows and those married to a ‘house without’. There have always been many widows in Kuria, young as well as old, because of the intensity of raiding in the district. Unlike surrounding areas, they do not remarry as, apart from the fact that they are deemed ‘unlucky’, the bride-wealth paid for them gives rights to all children born to the husband, irrespective of actual paternity. They thus, after the husband’s death, take effective charge of his property and continue bearing children for their ‘house’. To do this – and for economic help in the form of ploughing and gifts – they take lovers. A similar situation exists for those married as nyumba mboke, to a ‘house without’, referring in this instance to a house without a son.13 Such women become daughtersin-law to a woman, or sometimes a couple, who have no son to continue their line. Again, they must take lovers. The role of lover in such instances was well-recognised and incurred not shame but some respect in the past, as a man is seen as taking on responsibility not only for his own family, but offering help to another. Over the last ten years attitudes have become more ambivalent. Church morality based on shame and sin tends to interpret all relationships out of wedlock, including the two above, as promiscuous and thus against the teachings of Christ. This is reinforced by AIDS educators who tend to see such practices in ‘risk terms’, as a source of infection, and as one of the ways in which the virus could rapidly spread throughout Kuria. In the space of a few years, thus, a once honourable pattern of behaviour has become dangerous and even dishonourable. Fear of early death and the inability to care for oneself or one’s family; shame, associated with sexual misconduct; witchcraft; and the need to protect oneself from the infected are the main ways in which people talked about the dangers of AIDS. Together they form a concatenation of ideas, which mingle together in various ways. Faithfulness, for example, might be enjoined both as indicating proper responsibility and care for one’s family and as in accordance with Christian teachings. In the end it all has to do with character (eteemo). People were condemned as ‘careless’ for sleeping around, regardless of the danger that they then brought back to their fam341

Evidence, Ethos and Experiment

ilies. Christian morality did not necessarily or normally enter into such discussions. Further, it was subject to some comment that, despite almost universal church membership, there was also a deal of hypocrisy. Several spoke of the practice of church vigils, seminars and camps, where the evidence left around afterwards hardly spoke of chastity. How do people react to risk? At one level, one could say that there are as many responses as individuals, but there are common threads. Fatalism is one. Many spoke of how people getting a diagnosis of HIV could die of ‘shock’ or ‘stress’. They went back to their homes and gave up, dying within a week or two. Others spoke of it as an accident, something which would happen if it happened and there was little to do to prevent it. One Pentecostal pastor argued that it might even be ‘God’s will’, a punishment, Armageddon, and thus any attempts to counter it would be doomed to failure. However, a more active approach to managing the crisis was dominant. The problem for most was to be able to assess the risk as it affected them. Knowledge was the key as has been seen in the ‘tracking’ response. And this extended to HIV testing and openness with regard to the diagnosis.

Medical Options Hospitals and clinics, as we know, are only one possible avenue for healthcare in Africa and not the first and often not even the last port of call. Partly, medicines for common ailments are widely known – those for malaria, for example, can be obtained from pharmacies and others from herbalists. Indeed, with the erratic nature of supply to government facilities, even those accessing them often have to go and buy medicines elsewhere. Corruption does not stop at the clinic; medical staff are widely believed to siphon off drugs and sell them privately. But, there are further disincentives: the distance and difficulties of travel, the ‘harshness’ of nursing staff, of waiting all day for treatment or going through the ‘back door’ by bribing one of the staff and, with confusion over what has to be paid for and what not, uncertainty of cost.14 Where illnesses can be easily treated, people usually prefer to go directly to buy medicine or to a private clinic, where they have hope of being treated with respect, as a patron and not just as workload. When an ailment is deemed more serious, most people go to a diviner or healer. This has been widely reported for Kenya and elsewhere in Africa (Janzen 1992; Birungi 1998; Mwanzo 2000). With HIV, the problems have been even greater, partly because of the prevalence of ideas of witchcraft causation, with which only diviners and 342

Is the Sharia of the Doctors Killing the People?

healers can deal. This goes some way towards explaining something which had puzzled me. In 2002, hospitals in the area which had been well-patronised in 1995 were deserted. Since I had not then heard of the chameleon poison, I did not make a connection and it remained a mystery. However, a sudden growth in witchcraft fears might well explain the eschewing of biomedical provision, adding to the frustrations caused by low levels of staffing and cost at the time. Diviners claim that they can treat the chameleon and, undoubtedly, some claim too that they can cure AIDS. One diviner said that she could tell which of her clients were curable and which were not. In the latter case, she advised the relatives (never the patient) not to waste more money on searching for a cure. In many cases, this was and is leading to the serious impoverishment of families, as they have recourse to one healer after another in a desperate search for health. Only at the point of death would patients be taken to the district hospital, a view supported by hospital staff. With the late introduction of ART, it is too early to assess whether attitudes will now change and how quickly. Though most people had heard that there was such medicine ‘in the west’, and some that it was available elsewhere in Kenya, very few were aware of availability in the district. With the late introduction of VCT, the same consideration applies, though it could be accessed earlier through facilities in the Migori District hospitals, or a dedicated site attached to the district headquarters there. Within Kuria district itself, there were in August, apart from those at the district hospital, few up and running facilities, some of which were run by missions or NGOs. More, however, were planned and several ‘mobile’ government centres were under construction. These centres are mobile not in the sense of a travelling clinic but in that the counsellors are mobile, visiting such centres on a rotational basis. Even so, the provision that was there was hardly overwhelmed and, taking this evidence alone, one would not gain the impression of any great demand for testing. The explicit recognition of the need for testing apparent in the interviews and discussions thus seems to be conflated by the poor uptake of, admittedly, the little on offer. This might be taken as an index of the distinction between what people say and what they actually do, if it were not for subsequent discussion. These reflected the general mistrust of modern sector medical provision. More specifically, there were doubts about the protocols surrounding testing, with its mandatory counselling and confidentiality.15 Pre-test counselling was deemed by most to be unnecessary, or something which could be done on a group basis, not individually. It was 343

Evidence, Ethos and Experiment

frightening in itself and dissuaded many from taking the test. Testing alone did not appear to elicit this reaction. A CBO the previous year had called a mass meeting at one site. After counselling the whole meeting, volunteers were asked to come forward for testing and many did, indicating that there is indeed demand and that in public settings people are willing to test. This goes against the conventional wisdom based on the stigma of AIDS but, as one interviewee who took such a test explained, he saw it as demonstrating his ‘responsibility’. Thus, in the context, it indicated not shame but responsible conduct, the person taking the test showing that he cared for his own and his family’s wellbeing. More critical than this was the secrecy associated with diagnosis. Kuria are aware of medical ethics and confidentiality and, in general, condemn it, as very often even in ordinary consultations neither patient nor carers are told the diagnosis, or they are told it in language that they do not understand.16 This is one of the major criticisms of modern sector healthcare. With HIV, this takes on further force because it defeats the goal of knowledge which, as indicated earlier, people consider as their main hope. More than this, secret testing was held to promote a further danger. A psychological response might not be to die of ‘shock’ on being given an HIV-positive result but rather to be determined to take as many people as possible with you. People would no longer care about their life, or that of their family, and deliberately set out to infect others.17 Everyone believed this to be the case. Subsequent discussions often centred on those who were ‘responsible’ and those who were not, and who would carelessly disregard the interests of others in this way. Thus secret testing carried danger in itself to the community. Almost everyone wanted the results of tests to be made public, both so that people had the knowledge to protect themselves from the infected and to prevent the infected wittingly spreading the disease. Several thought that the onus should be on medical staff to inform the location chief of any person found to be HIV-positive so that he could announce it at a baraza. It was widely believed that health staff tested all blood, for example, that given for other diagnostic purposes or for transfusion, and thus they knew people’s status. And, in fact this was probably so – though the results were not given to the patients. The lack of transparency in medical procedures, the lack of ‘straightforwardness’ on the part of healthcare staff, was clearly a factor that fed into the mistrust that many expressed. Given that it is currently illegal for doctors to disclose HIV status, some people were thinking up their own alternatives. One man had the idea of 344

Is the Sharia of the Doctors Killing the People?

forming a cooperative group under the chief of his area. The location chief would furnish them with the names of those who had died (burial certificates are issued by the chief) so that they could attend the funeral to find out the cause of death that way. If they thought it could be AIDS, they would then go to the home and persuade the surviving spouse to go for a test. If it proved positive that would be announced. He did not see any difficulty with that: the spouse could be persuaded to agree and then others would avoid sexual relations with him or her. On being questioned on how that family would survive, the response was that the group would then also give support, for example, by ploughing for a widow. Most argued for a more radical change in medical procedures and were in favour of some form of mandatory testing. Testing of those about to go up to University had been discussed in the press and on radio and some thought that secondary school children should also be made to test. Parents felt that they had a right to know the child’s status, since they would be making considerable sacrifices in order to pay the increasingly high secondary school fees. Given that HIV is still thought of as a death sentence – and in fact even with the advent of ART is likely to remain so for a considerable time for the poor – this represented for them a squandering of resources, a lost investment. It was also pointed out that it would provide a strong disincentive to early sexual debut. Others were in favour of compulsory testing with all results made public. What I want to stress here is that it was clearly a matter to which people had given considerable thought. Several advocated quarantining. The government, I was told, should arrange special places for those found positive. Even within families, this often gave rise to heated debate, with some members arguing that it would not be sensible to send away people who were still well and could still look after their families. What would the family do? The usual counter was that relatives and neighbours would help out. The majority favoured a more limited type of quarantine, a sexual quarantining of those found positive. They would be avoided sexually but not socially. These people would still be relatives, friends and neighbours and would continue to be members of the community and cared for. The problem, admittedly, with both approaches was how to make sure everyone tested, and that the knowledge of their status was known. The logistics of how to do this were discussed at length. It would need official sanction, preferably a law which mandated universal testing. Then, it could be organised through the chiefs or through the iritongo, the assemblies of the people who, since the introduction of sungusungu in 1998, have tried cases 345

Evidence, Ethos and Experiment

of theft and successfully controlled the raids and banditry that once dominated in the district (Heald 2002, 2006). The problem – as with the thieves – is that some people would still ‘escape’ and there might be no way of knowing about strangers coming into the community. But, at least to this extent, they would be safer and it would constitute a major step in arresting the epidemic. All those questioned said that they would go for such a test.

The Doctors are Killing the People What has been reported in this paper is something which has been neglected in previous social research on AIDS in Africa. Far from stigma being the dominant idiom and concern of people, the need for knowledge in order to protect themselves and considerations of public health came to the fore. It could be that Kuria represent an exceptional population but the research methodology also possibly allowed this issue to rise to the surface. The research was not concerned with the difficulties of those suffering with HIV/AIDS or facing discrimination, or with other categories of people held particularly to be at risk. In the NGO rhetorics of community involvement, such involvement is usually interpreted in ‘activist’ terms, that is, with the voices of those dedicated to the needs of those living with HIV/AIDS.18 This represents a very small and biased sample. By contrast, this research canvassed opinion as expressed by families and people who, with a single known exception, did not know their HIV status. Clearly, they, and the communities they represented, wanted the knowledge that would give them power to take more control over their own health. The introduction of VCT is justified in terms of the need for people to test and to know their status. Since it is justified in terms of utility, it is important to question whether the particular ethical constraints which it imposes are conducive to the task. The advent of rapid testing kits has technically transformed the possibilities of testing, but the orthodoxy around VCT continues, slowing down the potential benefits that mass testing could bring both to individuals and to public health. To undertake VCT, counsellors are required to be trained at specialised centres. The government has responded not by employing additional staff but by sending staff already employed for training, thus increasing their duties and putting added pressure on facilities already inadequately staffed. This, inevitably, slows down both the development of programmes and the ability of hospitals to offer the service. The same applies to the voluntary sector which, whether Mission, NGO or CBO, must raise a considerable 346

Is the Sharia of the Doctors Killing the People?

sum of money to pay for the training. Options are also restricted in other ways, as setting up even a mobile centre is surrounded by protocols that require facilities to be built according to a particular pattern, with sufficient space to allow for privacy during the counselling and testing process, and so on. What seems not to have been fully considered in such policy is whether it really creates the appropriate context for testing on the scale required. AIDS educators tell people that they must now ‘face up to the disease’. Dedicated special sites surrounded by secrecy might well continue to be read as a tacit marker of individual shame. Public testing, by contrast, might well provide a more conducive environment, indicating civic virtue, an index, as it seems to be in Kuria, of responsible rather than irresponsible conduct. Counselling too might be read as a process that reinforces the dilemmas of testing rather than resolving them. And, it slows down the provision of testing on a mass scale. The test now takes only twenty minutes but the whole process, with its mandatory pre- and post-test counselling is much, much longer. It takes time and this too is costly for both patients and providers. While the test kits are provided free, the counselling is not, and charges are made to cover this, often of only fifty shillings (less than USD1), but that can be more than a poor person can afford. Most facilities said that they waived the charge for the very poor but mere knowledge of a charge may in fact deter many. Further, testing is not a one-off process, but must be repeated after three months to take account of the window period and, in current pandemic circumstances, regularly thereafter. Then, there is the question of disclosure. The procedure current in 2005 is that those found to be HIV-positive were encouraged to disclose their status to at least one other person. This was explicitly for the support of the patient. Usually, in Kuria, this was someone of the same gender, a close relative in the case of a woman, a friend of the same age in the case of a man. Disclosure to the sexual partner or spouse was and remains unusual (Family Health International, HAPAC Report 2002). While this is understandable enough on a personal level, given the undoubted distress that the news will give the partner and the threat of recrimination or even abuse, it is clearly a procedure again helping to spread rather than to curtail the epidemic. The 2003 Demographic and Health Survey gave evidence of what seems like a high proportion of discordant cohabiting couples; 7.4 per cent nationwide, but 17.1 per cent in Nyanza, the Province in which Kuria lies. As de Cock et al. (2003) have pointed out, those most at risk could well be faithful HIV-negative partners in relationships where the other is infected. 347

Evidence, Ethos and Experiment

It also runs against existing ST1 practice where patients are refused treatment until they bring in their sexual partner so that both can be treated together. A nurse in Kehancha hospital claimed there was little difficulty of persuasion in such cases. An underlying assumption is that a human-rights approach privileges the rights of the autonomous individual and this creates a tension between private and public health. It is argued that this is not necessarily the case, and the Siracusa principles allow for legally authorised measures (ECOSOC 1985). Indeed, there is little reluctance to impose restrictions when global influenza pandemics threaten. The issue of why such measures have not been applied with regard to HIV takes us back to the origins of policy in the USA in the 1980s and the increasing coalescence of human rights and humanitarianism in interventions (Nguyen, this volume; see also Epstein 1996; Heald 2003). These have from the first privileged the rights of the infected, yet in the legal sphere, human-rights courts operate in terms of a balance of rights, with those of the individual adjudicated against the rights of others and the values or good of the community. The human rights of one party do not take automatic precedence. Given the failure of current AIDS policy as it has been applied in Africa to curb the pandemic, it is now time to seriously consider whether the rights of the (as yet) uninfected should not be given greater weighting. The exceptional status of HIV has involved a whole new stress in medical ethics. While, on the one hand, it is clearly based on ethical principles embedded in the therapeutic encounter, many normal procedures seen as in the patient’s interests have been set aside. It has become more like the protocols of medical experimentation where ‘informed consent’ is taken as a prerequisite. (But then, perhaps as the forms filled in by counsellors indicate, it is as much about medical data collection as the specific problems of those being counselled.) With de Cock et al. (2002, 2003), it is hard to avoid the conclusion that this is detrimental to the cause of public health and the urgent need to find a way of curbing the epidemic. Indeed, with HIV now deemed ‘endemic’ in Africa, the need for adequate prevention measures has to an extent been overwhelmed by the roll-out of ART. Testing has become the means for recruiting people into therapy programmes rather than the provision of knowledge which would allow people to act responsibly for their own and others’ safety. And, above all, the question is raised of whose ethics should dominate in this area – those of the donor community or those of recipient communities throughout Africa? In the early months of 2006, by general agreement, two locations in Kuria lob348

Is the Sharia of the Doctors Killing the People?

bied the District Hospital for the provision of mass public testing in their areas. They were refused on the grounds that it would be unethical. I leave this paper as I began it, with Mwita’s view that the codes of the medical profession threaten to kill them all.

Notes 1. See Heald (2002, 2003, 2006) and Allen and Heald (2004). 2. It was not until 2003 that incoming President Kibaki make what appears as the now mandatory ‘War against AIDS’ speech. 3. Surveillance during the 1990s was poor in Kenya with Sentinel survey data subject to the vagaries of inadequate testing kits and staff strikes. In 2000, prevalence was put at 13.5 per cent nationwide (Kenya, 2005 a). This was scaled down dramatically after the Kenya Demographic and Health Survey of 2003 (KDHS) (Kenya 2005b), with prevalence nationwide lowered to 6.7 per cent. However, there is some debate over the figures, as only 70 per cent of respondents in this survey agreed to be tested (Barnett 2004). Prevalence rates vary considerably, however, with the KDHS estimating it as 14 per cent in Nyanza Province and rates soaring to almost 40 per cent in Suba District, near to Kuria. 4. Kaiser Daily HIV/AIDS report, 18 August 2003 (accessed 8 October 2005). Most donor money during this time was delivered via KANCO, the Kenyan AIDS NGO Consortium. 5. The Standard, 13 March 2006; allAfrica.com, 17 March 2006 (accessed 18 March 2006). Further allocation to Kuria District had already been suspended until the funds already dispersed in 2005 had been properly accounted for. 6. In the early 1990s, as doctors and other healthcare workers moved out of government service because of the low pay, a number of private hospitals and clinics were opened up in the area, but many of these have since disappeared. 7. Because of late delivery, the promised funding of the programme for five years was effectively cut to three years and the reliability of supply was still an issue. But, as the medical officer in charge of the programme said, three years was better than nothing, and one cannot but sympathise with the difficulties of his position as he must still tell patients that they must take the medicines daily for life. 8. Most of these families were those selected on a random basis in 1984/5 in an investigation of the social and economic effects of tobacco contract farming in the area, funded by the ESRC. The survey was repeated in 1995, and over time and a succession of visits, many of the families have become friends (see Heald 1991b, 1999). 9. The Kenya Demographic and Health Survey of 2003 did not survey sufficient numbers of people in Kuria District to produce a valid statistic of HIV prevalence. However, health staff at the District Hospital gave estimates in line with the 14 per cent for Nyanza Province as a whole, well above the Kenyan average. 349

Evidence, Ethos and Experiment

10. There were other possible cases recorded in my field notes where the diagnosis was said to be typhoid or unknown. 11. The way this is done cannot be predicted in advance. The advent of a new form of witchcraft in Kuria contrasts with Botswana, where its symptomatic forms were recognised as known complaints and incorporated into an aetiology based on ideas of pollution (see Heald, 2002, 2003, 2006; Allen and Heald 2004). In Kuria, by contrast, it opened up a nosological divide between diseases known via biomedicine and those attributed to witchcraft. 12. BAT set up small-holder tobacco contract farming in the area in 1975 and this, together with maize, is now the major cash crop in the region (Heald 1991b, 1999). 13. This is the usual English translation given of the situation, though a more poetic rendering of the ‘reawakened house’ is given by Malcolm Ruel (private communication). 14. Uncertainty of cost has in fact increased with international funding since some provision is free – for some STIs, for malaria, TB and HIV. But, patients may still be charged (illegally) and the free provision does not cover all treatments. Thus while ARV drugs are free, patients put on these still have to pay for tests (for CD4 count, liver and renal functioning and so on) and must pay for treatments for opportunistic infections. Susan Whyte et al. (2004) reports that nurses in Uganda talked of their ‘blanket test’ by which they assessed the wealth of a patient and thus whether it was kinder not to inform a patient about the availability of drugs. Observations in Kuria suggest that only the wealthier and better educated have and are currently accessing ARV. 15. Suspicion and lack of trust also came out clearly in discussion surrounding condoms. Condoms have been distributed in their thousands in the last two years. In Kuria, one found the same suspicion regarding them as in other parts of Africa and, behind this but rarely directly stated, the underlying belief that HIV had been introduced by the west. People voiced fears that they were poorly manufactured or sabotaged by having tiny holes. More were concerned about the lubricant, and its potentially harmful effects. Meningitis was mentioned by one person. However, for others, the problem was that they posed a different kind of health risk. In an area where latrines are far from universal, used condoms were often disposed of in the bush (as well as being in evidence around church centres after vigils), posing a risk to children who might pick them up unawares. 16. See Heald (1991a) on the significance of the exoteric discourses of medicine as compared with the exocentric ones of diviners when it comes to both power and trust. 17. HAPAC report, 2002, records the same reactions from focus-group discussions and interviews with Kuria attending Ombo Mission Hospital in Migori. ‘Infected people have been noted and heard to say that they are not dying alone’ (Appendix 4). 18. The NGO HIV/AIDS Code of Practice Project, 2004, has further endorsed this practice, arguing that people living with HIV/AIDS must be incorporated into policy and project development at all levels. 350

Is the Sharia of the Doctors Killing the People?

References Allen, T. and S. Heald. 2004. ‘HIV/AIDS Policy in Africa: What Has Worked in Uganda and What Has Failed in Botswana?’, Journal of International Development 16(8): 1141–54. Barnett, T. 2004. ‘HIV/AIDS: ‘How Bad Does Bad Have to Be Before We Believe It Is Bad? How Can We Translated Words into Deeds?’, Journal of International Development 16(8): 1181–84. Birungi, H. 1998. ‘Injections and Self-help: Risk and Trust in Ugandan Health Care’, Social Science and Medicine 47: 1455–62. De Cock, K., E. Marum and D. Mboori-Ngacha. 2003. ‘A Serostatus-based Approach to HIV/AIDS Prevention and Care in Africa’, Lancet 362: 1847–49. ———. 2002. ‘Shadow on the Continent; Public Health and HIV/AIDS in Africa in the 21st Century’, Lancet 360: 67–72. ECOSOC. 1985. ‘The Siracusa Principles on the Limitation and Derogation Provisions in the International Covenant on Civil and Political Rights’, Geneva: ECOSOC. Epstein, S. 1996. Impure Science: AIDS, Activism, and the Politics of Knowledge. Berkeley: University of California Press. Family Health International. 2002. HAPAC End of Project Report on Syndromic Management Therapy in Church Health Facilities in Nyanza Province, Kenya. Heald, S. 2007. ‘Controlling crime and corruption from below; Sungusungu in Kenya’. International Relations 21(2): 183–99. ———. 2006. ‘Abstain or Die: The Development of HIV/AIDS Policy in Botswana’. Journal of Biosocial Science 38: 29–41. ———. 2005. ‘State, Law and Vigilantism in Northern Tanzania. African Affairs,105: 265–83. ———. 2003. ‘An Absence of Anthropology: Critical Reflections on anthropology and AIDS policy and practice in Africa.’ In G. Ellison, M. Parker and C. Campbell (eds.) Learning from HIV/AIDS: a Biosocial approachCambridge: Cambridge University Press, pp. 210–37. ———. 2002. ‘It’s Never as Easy as ABC: Understandings of AIDS in Botswana’ African Journal of AIDS Research 1(1): 1–11, 2002 ———. 2000 ‘Tolerating the Intolerable: Cattle raiding among the Kuria’. In G.Aijmer and J.Abbink (eds.) Meanings of Violence. Oxford: Berg, pp. 101–21. ———. 1999 ‘Agricultural Intensification and the Decline of Pastoralism: a Kenyan case study Africa 69(2): 213–37. ———. 1991a. ‘Divinatory Failure: An Examination of the Religious and Social Role of Gisu Diviners’, Africa 61(1): 299–317. ———. 1991b. ‘Tobacco, Time and the Household Economy in two Kenyan Societies: The Teso and the Kuria’. Comparative Studies in Society and History 33 (1): 130–57. Janzen, J. 1992. Ngoma: Discourses of Healing in Central and Southern Africa. Berkeley: University of California Press. Kenya, Republic of, 2005. Kenya Demographic and Health Survey 2003. Nairobi: Central Bureau of Statistics, Ministry of Planning and National Development Kenya, Republic of, 2005. AIDS in Kenya: Trends, Interventions and Impact. National AIDS and STI Control Programme (NASCOP) 7th edition, Nairobi. 351

Evidence, Ethos and Experiment

Kenya, Republic of. 1997. Kuria District Development Plan, 1997–2001. Nairobi: Government Printer. Kenya, Republic of, 2002. Kuria District Development Plan, 2002–2008, Ministry of Finance and Planning. Nairobi: Government Printer Kenya, Republic of. 2000. Kenya Service Provision Assessment Survey, 1999, Ministry of Health and National Council for Population and Development, Nairobi. Kenya, Republic of. 2002. Kuria District Development Plan, 2002–2008, Ministry of Finance and Planning. Nairobi: Government Printer. Kenya, Republic of. 2005a. ‘AIDS in Kenya: Trends, Interventions and Impact’, National AIDS and STI Control Programme (NASCOP), 7th ed. Nairobi. Kenya, Republic of, 2005b. Kenya Demographic and Health Survey 2003. Nairobi: Central Bureau of Statistics, Ministry of Planning and National Development. Kenya, Republic of, 1997. Kuria District Development Plan, 1997–2001. Nairobi: Government Printer Mwanzo, I. 2000. ‘Recasting Local Understandings and Management of Affliction: The Case of Malaria among Abalogoli of Rural Western Kenya’, Ph.D. dissertation. Brunel University. The NGO HIV/AIDS Code of Practice Project. 2004. Renewing Our Voice: Code of Good Practice for NGOs Responding to HIV/AIDS. Oxford: Oxfam. Whyte, S.R. et al. M.A. Whyte, L. Meinert, B. Kyaddondo. 2004. ‘Treating AIDS: Dilemmas of Unequal Access in Uganda’, Journal of Social Aspects of HIV/AIDS. 1(1): 14–26.

352

Chapter 14

The Historical Interface between the State and Medical Science in Africa: Kenya’s Case Kenneth S. Ombongi

Introduction By the 1980s the abundant optimism that had accompanied the birth of modern African nations in the 1960s, and belied imperial pessimism about the viability of the African state, had waned. What had seemed as the African ‘golden age’ transited into institutional decay, policy failures and civil strife in many postcolonial states. The subsequent state weakness undermined the hegemonic colonial legacy which had earned the post-independence state the lexical label of ‘postcolonial’, which referred to the colonial origins and characteristics of the African state (Young 2004). Scholars have explained the diminished dominance of the African state by its incapacity to sustain the high modernist agenda inherited from its imperial predecessor.1 The focus of this chapter is to analyse how the postcolonial state’s gradual decline influenced medical science (biomedical research and policy) in Kenya, bringing to an ‘end’ an interesting historical interface between the state and public health development. It provides a parenthesis of the changing role of the state in independent Kenya’s medical policy. The paper posits that while independent Kenya in the 1960s inherited a biomedical policy which hinged on state ascendancy, the progressive weakening of state apparatus in the 1980s and 1990s resulted in a confluence of non-state and international actors who currently play a vital role in Kenya’s public health system. The convergence produced a much more complex medical trajectories than those witnessed during the colonial and immediate postcolonial 353

Evidence, Ethos and Experiment

period, in which the state dominated. In a way, Kenya’s post-independence medical policy was a ‘mixed bag’ of sorts. It was, unlike during colonialism, characterized not only by the government’s interventionist programmes but also international collaborative networks, with diverse objectives, cutting across official and non-official organizations. The paper contains two main sections. The first section generally traces the colonial evolution of the state’s role in biomedicine in Kenya. It demonstrates that biomedicine developed as an integral part of the imperial state’s hegemonic system. The intricate relationship between biomedicine and the state would be a salient colonial legacy in independent Kenya. The second section examines the attributes of independent Kenya’s medical policy, outlining the modernist trajectories of state interventionism. Despite the diminished ascendancy of the Kenyan state, the section relates developments which prevented the local medical system from the anticipated total collapse.

The State and Evolution of Colonial Medical Policy James H. Sequeira, a medical officer in Kenya, delivered the Chadwick public lecture at the British Medical Association Hall, London, on 28 April 1932. His lecture focussed on the significance of African public health education in Kenya. In reference to the contemporary situation he said: When we talk about tropical diseases, we are, I think, inclined to consider afflictions as they occur in the European, whose duty, occupation or inclination, causes him to reside in those parts of the world which we call the tropics. (Sequeira 1932) Three decades later, in 1961, a newly Makerere-trained Kenyan doctor, Joseph Maina Mungai, was an intern at the King George VI Hospital, Nairobi.2 In a recent autobiography Mungai reminisced about his experiences as an intern: Although I had been brought up to accept, without any question, the segregated systems of education, health services and land ownership, the experience at the hospital left me curious as to how and why the system had arisen. The segregated health and education services made me feel professionally unhappy. (Mungai 2002: 87) 354

The Historical Interface between the State and Medical Science in Africa

While Sequeira indicated the reality that colonial tropical medicine was geographically but not socially tropical, Mungai expressed dejection towards state-centred modernism which was a vehicle of imperial and cultural chauvinism. Throughout the colonial period biomedicine hinged on, and was defined by, its imperial and social attributes. It alienated social aggregates, rationalized the existence of disparate groups, configured ethnic identities and buttressed oppressive routines. In independent Kenya, biomedicine strove to rid itself of what made the likes of Mungai, at the eve of independence, ‘feel professionally unhappy’. It helped in making sense out of the formal sovereignty which accompanied political independence and was a ‘channel’ through which hitherto latent anti-colonial medical struggles became manifested. Consequently, the centrality of state agency in shaping medical policy in both colonial and independent Kenya cannot be gainsaid. First, I will examine the situation in colonial Kenya. At the advent of the British rule, Kenya was an ecologically and epidemiologically strained environment. In the 1890s drought, epidemic smallpox, epizootic rinderpest and famine ravaged many areas in the Rift Valley and Central Highlands (Dawson 1979). Furthermore, the emerging colonial urban settlements, such as Nairobi, contended with frequent rodent-transmitted bubonic plague while malaria was a real danger to the early European residents in Kisumu and Mombasa (Achola 2002). With such a grim backdrop, scientific medicine held the promise of transforming the disease-ravaged colony into a citadel of ‘progress’. The epidemiological dangers Kenya posed and the urgent need to protect the clique of ‘men on the spot’, the colonial officials tasked with the responsibility of governing the colony, provided the twin basis for the nascent colonial medical policy. The colonial medical system was small and limited in focus and scale. By the turn of the twentieth century it had evolved from the days of the proverbial ‘medicine chest’ in the 1880s to mundane hospital institutions.iii Socially, it was tailored for the interests of the white minority while epidemiologically, it focused on diseases, such as malaria, perceived to be of maximum danger to Europeans. Geographically, colonial medical policy started as a purely urban phenomenon but gradually, particularly from the 1920s onwards, radiated to the African reserves following the Second World War epidemiological revelations that the African disease burden was more than hitherto thought and appreciated (Iliffe 1998: 38). By the 1930s, however, little was known about the African disease situation in the reserves. In 1932 there were only thirteen medical units in the reserves, over 355

Evidence, Ethos and Experiment

100 subsidiary dispensaries, with seventeen European medical officers, nine European sisters, six European sanitary inspectors and about 1,000 African staff (Adalja 1962: 108). Despite being thinly spread among the Africans, biomedicine was an inherent ‘tool of empire’. First, it was a tool of empire either in Headrick’s sense in which scientific medicine heralded European tropical expansion or in Beck’s understanding that scientific medicine was a necessary condition for colonial modernism.4 Second, it was a ‘tool of empire’ for playing more to the subjective whims of the imperial state than to the objective rules of scientific medicine.5 As a tool of colonial power consolidation, medical policy was an effective mechanism of social control in the hands of the state. It divided social aggregates into hierarchies of hygienically ‘clean’ and ‘unclean’ and called for racial segregation ‘both from a moral and hygienic point of view’.6 Ultimately, ‘biomedical’ arguments rationalized residential racial segregation in Kenya’s urban centres through ‘The Public Health Ordinance, 1913’.7 Consequently, public health became a crucial component in defining the relationship between the ruler and the ruled, the dominant and the dominated. As elsewhere in the British Empire, biomedicine removed focus from the real cause of disease, the socioeconomic conditions of the ruled in a colonial setting, and placed it on hereditary and racial differences.8 In a number of ways, colonial public health measures caused suspicion and largely came to represent the oppressive colonial command and social control in the eyes of the non-Europeans. Disease control routines, such as quarantine, acquired social rather than the sanitary meaning which they were meant to convey. Highly contagious disease conditions such plague were shrouded in secrecy among non-Europeans for fear of exclusion, loss of property and mistreatment from the authorities.9 For example, in 1902 ‘a case or two of plague’ in the mushrooming Indian Bazaar in Nairobi attracted the wrath of J.H. Patterson, the chief railway engineer, who ‘promptly burned [down] the whole place’ after an hour’s notice to the residents (Patterson 1907: 291–92). The ‘rage’ for sanitary improvement, literally the triumph of environmental determinism, particularly in urban public health in colonial Kenya, mirrored the monopolization of power and unbridled hegemony of the state. Consequently, environmental sanitation could not operate only on the basis of the ultimate and impersonal authority of biomedicine. It had to be ‘built’ around the colonial cultural and racial differentiation, a crucial component in the relationship between the ruler and the ruled. Sanitation, 356

The Historical Interface between the State and Medical Science in Africa

just like any colonial social-control mechanism, operated at the level of the divisive barrier of colour that had failed to put humanity above race. The colonial medical research agenda, the processes which produced medical knowledge, like attempts to control disease, was defined by what Megan Vaughan called ‘particular historical circumstance and particular social forces’ (1991: 6). In most of its aspects, medical research was tributary to the exceedingly hegemonic ascendancy of the state, and racial-cultural dominance. Indeed, in the moral matrix of imperial legitimacy, ‘tropical’ medicine operated a less contradictory way to its historical and social context. In that way, it was ‘tropical’ in so far as geography was concerned but socially it was ‘temperate’ and imperial. There are numerous historical examples to illustrate the above state of affairs, but for purposes of brevity I will confine myself to only two in this paper. The first example was the colonial doctors’ attempts to use eugenics to justify the African subordinate position in the colonial order and the second one was the chequered history of malaria control. To many colonial biomedical experts Africans, as subjects of study, were a very fascinating race. In a sense, they fitted well within their long-held Eurocentric racial prejudices. Colonial biomedicine ‘provided a way’ to decipher the actions of this ‘eerie’ lot – Africans. For example, in the early 1930s F.W. Vint, an assistant bacteriologist at the Nairobi Medical Research Laboratory, studied ‘the stage of cerebral development reached by the East African native’. Despite the obvious differences among diverse ethnic groups occasioned by ecological settings, psychological orientation and processes of socialization, Vint ‘treated the natives of Kenya as homogenous’ (1932:48). Vint concluded: Thus from both the average weight of the native brain and from measurements of its prefrontal cortex I have arrived, in this preliminary investigation, at the conclusion that the stage of cerebral development reached by the average native is that of the average European boy of between 7 and 8 years of age. (Vint 1932: 48) One wonders whether it would ever be possible to ascertain in much greater details all of the forces that motivated Vint and his ideas other than the larger historical and social context in which he was socialized, lived and worked. Vint’s analysis was later viewed, rather contemptuously and cynically, by African medical scholars as an expression of prejudiced, preconceived and racialist sentiments that needed to be countered.10 Largely, 357

Evidence, Ethos and Experiment

it is important to note that colonial medicine’s incongruity of playing to the popular cultural gallery, as opposed to objective science, was as a result of being a ‘tool’ in the hands of the colonial state to which racial chauvinism was a legitimizing agency. In the promotion of biomedicine, the state, through its agents, tailored it to the advancement of its hegemony and political needs – hence the desirability of studies such as Vint’s. The second example which demonstrates the role of biomedicine in the consolidation of colonial rule and definition of power relations in Kenya was that of malaria control policy. By the turn of the twentieth century, malaria was generally an endemic problem confined to the lacustrine regions of Lake Victoria, and the Mombasa coastal strip. By the 1920s malaria, aggravated since its epidemiology had dramatically changed because of the colonial ecological and social transformation. In 1926 the first malaria epidemic affected Nairobi and Central Kenya, the heart of the agrarian colonial economy, and sent shockwaves throughout the colonial establishment. Malaria further proved its intractable ascendancy during the epidemic outbreak of 1928, which affected the north Rift Valley grain belt – Eldoret, Kapsabet, Kitale and Trans Nzoia. Subsequently, high altitude malaria epidemics became common in the 1930s and 1940s.11 Malaria would remain the single most important public health problem in the colony until the hey-days of DDT insecticide and synthetic antimalarials – especially chloroquine – during and after the Second World War. The developments in the science of bacteriology informed patterns of antimalarial measures in such a way that it created effective social control mechanisms that defined colonial power relations in Kenya. During the late nineteenth century the development of microbiology had led to the discovery of the etiologic and transmission agents of malaria, and of course those of other communicable diseases. Generally, this resulted in a fundamental shift of malaria control interests from detailed epidemiologic studies to control based on measures against malaria’s etiologic and transmission agents, an attack on plasmodia parasites and anopheles vectors. In early urban Kenya antimalarial policy started out with puritanical town sanitation. In due course, it assumed an approach that blended sanitation with quinine prophylaxis. From 1910 onwards the colonial administration encouraged, albeit with expedient versatility, the administration of quinine, a primordial antimalarial specific, to civil servants, regardless of race. Quinine was distributed either in known malarious towns or non-malarious ones that had occasional malaria upsurges. The targeting of only specific sections of the African public for quinine pro358

The Historical Interface between the State and Medical Science in Africa

phylaxis at that early period was a clear departure from the official policy that targeted the dominant group, Europeans, among urban residents. It was underpinned by an underlying desire to protect the health of Europeans by treating African malaria ‘health carriers’, real or perceived. This was clear from the official conviction that ‘if the [African] servants are kept free from malaria or, at least, as free as possible, there is less risk of infection to the Whites concerned, because mosquitoes in the vicinity of the bungalows are less likely to be infected’.12 As another measure against malaria, the state effected social distance through racial segregation by the promulgation of ‘The Public Health Ordinance, 1913’. This was the first official institution of racial segregation on health grounds in Kenyan towns.13 The racially selective patterns of settlement set the foundation of the standard layout of Kenya’s colonial towns. It included a high ground, often called Milimani (hills), a European enclave that contained an administrative boma (office) and official residence, a crowded Indian bazaar and African slums in the ‘east-ends’ (Zeleza 1989). The white residents existed in favoured circumstances, followed by Indian traders and both were served by the underprivileged Africans in slum dwellings. The emphasis on segregation in urban Kenya represented a discontinuity with the holistic traditional miasma control practices of the pre-germ era. It narrowed the scope and focus of public health in dealing with malaria in urban settings. ‘Vector (mosquito) sanitation’ and residential segregation transformed public health from targeting ‘whole environments’ and ‘whole populations’ to specific environments and social/racial groups. With this, scientific medicine provided legitimacy to practices compatible to the imperial policy that subordinated economically weak urban racial groups to powerful ones (Achola 2002). However, rapid urbanization and colonial capital gradually tore through racial boundaries and intensified interaction between the ruler and the ruled while weakening social distance as a malaria prevention strategy. However, the aggravation of malaria in Kenya’s countryside, African Reserves and Settled Areas, since the epidemics of the 1920s, resulted in the colonial authorities increasingly looking at it as a social disease – ‘caused’ and spread by the social and economic conditions, dearth and dirt. On one hand, the projection of malaria as a social condition reinforced the socialcultural dichotomy between Africans and Europeans, between the so-called uncivilized and civilized, which, in turn, led to the delimitation of areas, like in urban centres, where Africans could live and where they remained forbidden. On the other hand, it gave way to the administration’s expedient 359

Evidence, Ethos and Experiment

medical attention directed towards African reserves. In November 1928 the editor of the Kenya and East Africa Medical Journal (KEAMJ), owned by the local branch of BMA, captured the medical opinion of the time: It must not be forgotten that conditions in the native reserves have a definite bearing on the position in the settled areas. The reservoirs of malaria, from which the disease is carried by labourers seeking work on farms, must receive their quota of attention. To confine effort to any single area of the country will not permanently remove the danger of further outbreaks in it.14 However, attempts to incorporate as opposed to exclude African reserves, in relation to settled areas, in the control of malaria that developed within this wide policy framework provided the post-epidemic period not only a fundamental, and but unique, malaria policy posture. Predicated on the stimulus that the Kenya highlands were no longer safe in the wake of epidemic malaria, the new policy derived from the African despicable social conditions. According to the contemporary medical opinion, Africans were a race at the ‘verge of extinction’ because of three interrelated conditions: their ‘insanitary’ environment, deprivation, and malaria. In 1926, for example, the DMS in his annual report indicated: The sanitary problem of Africa is how to improve the standard of living among a population of some millions of persons; in many cases poorly developed physically; at a low stage of civilisation and comparatively uneducated; high birth-rate, a high death-rate and a high infantile mortality; suffering from preventable disease; in occupation of fertile land without the ability to use that land to the best advantage.15 With such racial rhetoric, colonial officials were convinced that it was possible to redeem Africans from their state of dirt, dearth and disease through public health education instead of addressing the social and economic deprivation that was the real cause of African malaise.16 With education, the authorities argued, could come prosperity, with prosperity, sanitary progress, and with sanitary progress, alleviation of disease.17 Thus, the state sought to improve sanitary conditions as a critical project of imperial statecraft. In the late colonial era, the increased faith in technology and progress as transforming agents in official circles reinforced its attempts to ‘remould’ African social habits conducive to disease. 360

The Historical Interface between the State and Medical Science in Africa

The State and Postcolonial Medical Policy At independence, Kenya was engulfed in the contemporary global ideological divide. Jomo Kenyatta, its first president, took a western inclination. He enhanced the political and administrative hegemonic colonial structures and practices while adopting state capitalism underpinned by populist welfare, and cultural nationalism. The state’s attempts in infrastructure and service provision combined with the unbridled nationalist aspirations to shape independent Kenya’s medical trajectories. The hitherto blocked nationalist sentiments geared towards repudiating colonial paternalism and racial discrimination found expression in biomedical developments. If biomedicine was a vehicle of British cultural chauvinism in colonial Kenya, it was a mechanism of reasserting African cultural nationalism in independent Kenya. The consequent feeling that ‘we are finally in charge’ enhanced the centrality of the state in medical personnel training, direction of medical research and the provision of services and facilities. Broadly, the ascendancy of the postcolonial state as a significant agency in medical policy took the form of a predilection towards Africanization at the professional, social and epidemiological levels. Professionally, the emerging cadre of African doctors, often graduates from Makerere and later Nairobi, aspired to Africanize the medical profession. Throughout the colonial era racial discrimination kept Africans at the lower ranks of medical practice. Iliffe documented how a few Africans who qualified as doctors were often posted to remote areas, where professional advancement was either minimal or took too long to come if and when it happened (Iliffe 1998: 79). The bid to Africanize the medical profession found expression in the creation of the Nairobi Medical School (NMS). While Makerere was the first medical school in East Africa, NMS became the first in producing a large number of doctors and beacon of light for the regional research agenda. The mission of NMS according to Maina Mungai, its first dean, was to train ‘doctors to work among the masses’ (Mungai 2002: 85). NMS’s rise was characterized by the eagerness to dispense with the rigid conventional curriculum of Makerere which curtailed African advancement in the medical profession (Iliffe 1998: 174–76). The pioneers in Kenya’s medical training, such as Mungai, sought to liberate the training of doctors from curriculum rigidity, as well as continue the colonial phenomenon in which academic achievement rather than ability to pay through the medical school determined entry to the medical profession or a career in medical research. Today, at the dawn of 361

Evidence, Ethos and Experiment

the twenty-first century, this has apparently changed following pressures on the NMS, and the University of Nairobi which hosts it, to raise funds from outside the exchequer to meet its extra expenses. It seems that many academically average students, from financially stable families, are pursuing a career in medicine, something that was unthinkable in Kenya as late as the early 1990s. In addition to that, a number of medical schools, probably with very doubtful infrastructural facilities and personnel support, have sprung up in recent years.18 In the 1960s and 1970s medical and research training was informed by some level of quality, the need for African achievement and mass service delivery. Whether today’s training considers imparting lifelong doctors’ values, knowledge and skills, treasured by the immediate post-independence generation, is a matter open to debate and a fairly controversial subject in Kenya. However, it seems that while the current system has opened doors to many and certainly encouraged diversity, it has allowed for flexibility and career progression without adapting to fast changing demands in infrastructure and personnel development. At the social level, the Africanization of medical policy was manifested in the populist ‘mainstreaming’ of medical services to Africans. The medical administration came under the strong influence of the ultra-nationalists, usually doctors and researchers at the NMS and later the Kenya Medical Research Institute (KEMRI) who supported the state declaration of ‘service to our people’. The immediate postcolonial era witnessed mass expansion of government medical services in the countryside (Iliffe 1998: 172–73). This was further stimulated by the fast waning of the pre-existing local ambivalence towards biomedicine. Many people turned to hospitals, health centres and dispensaries for medical attention compared to the previous years (ibid.: 169–70). Despite the undertaking to correct colonial errors of discrimination and exclusion, the independent state inherited the imperial public health structures, the modernizing mentality of the late colonial period and the dominance of curative, rather than preventive, medicine which would be salient features of Kenya’s medical ‘postcoloniality’. The medical modernization in the form of extending services and facilities to the previously ‘forgotten’ African reserves became a crucial symbolism of the Kenyan state’s hegemonic ascendancy. Conversely, the colonial government extended medical services to the African areas only when and to where it was economically expedient (Gilks 1933). In his independence-day speech, Jomo Kenyatta declared disease as the number one national enemy, an enemy of progress. In a sense, the expansion of 362

The Historical Interface between the State and Medical Science in Africa

medical services became a way of ‘speaking’ directly to the independence euphoria of the 1960s. Government presence at the grassroots had not only to be represented by the colonial legacy of uniformed local administration – chiefs, District Officers (DOs), and District Commissioners (DCs) – but also through the local dispensary, health centre and general hospital. In the period between 1962 and 1972 government hospital beds increased from 10,617 to 18,186. The fact that by 1969/1970 the Ministry of Health accounted for a staggering 48 per cent of all state expenditure demonstrated the significance government attached to this sector (Iliffe 1998: 169–72). By the late 1960s, however, the demand for medical services outstripped government effort and financial capacity. Things worsened in 1969 when government took up to finance rural dispensaries and health centres, a responsibility hitherto undertaken by local authorities (ibid.). Gradually, this started a process in which the very symbols of government presence, institutions of public health services, decayed. The final level of the Africanization of biomedical was epidemiological, in which Kenyan researchers sought to Africanize disease and research. This was done by seeking to disapprove the long-held colonial belief that the so-called ‘diseases of civilization’ were uncommon among Africans (ibid.: 169–70). For the African postcolonial medical researchers, like Maina Mungai, it was ‘payback’ time to their colonial counterparts. They refuted colonial social constructions of the African disease ecology and epidemiology. Indeed, they fought vigorously against the racially lopsided colonial medical knowledge in which they studied. For example, the pioneer Kenyan cancer specialist, Edward George Kasili, spent his professional life proving that leukaemia, a ‘disease of civilization’, was as common in Kenya as elsewhere in the world (Kasili and Taylor 1970). As if fate destined him to prove by his own life what he stood for professionally, he died of cancer in 1997 (Iliffe 1998: 178). In 1978 a team led by Professor Nelson Awori undertook the first renal transplant in sub-Saharan Africa (ibid.). If Awori spent his professional life proving that renal surgery was not an exclusive preserve of western scientists, Hilary Ojiambo, a long-serving professor of medicine at the NMS, did so through cardiac research and surgery. The African researchers’ nationalist inclinations blended well with the ambitious modernist postcolonial state. In a sense, while the nationalist stirrings of the 1950s culminated in the Mau Mau war that was fought in forests, the nationalist stirrings of the 1960s and 1970s resulted in a ‘war’ which was fought in medical and surgical wards, NMS departments, frontier health centres and clinics. 363

Evidence, Ethos and Experiment

Perhaps Kihumbu Thairu, Chairman of Kenya Medical Research Institute (KEMRI) founder committee and professor at NMS, pushed the malleable boundaries of the nationalist euphoria in biomedicine a bit too far.19 Thairu authored a book, The African civilizations, in which he analysed the African culture and its relevance. He used the word ‘pink’ to refer to Europeans, arguing that it was ‘more scientifically correct’ than the usual one, white. He dismissed western education as a form of cultural ‘conditioning’ and argued that modern medical procedures of surgery long existed in African pre-colonial societies (Thairu 1975: 79–83). In Thairu’s view: We should be proud of our medicine men. It is they who bore the torch of medical knowledge in the pre-microscope days in our society. Indeed it is truly amusing to find that Europeans use an ancient African symbol, the snake, for European medicine. Our symbol for African medicine is not snakes! The snake was the emblem of only ONE African doctor called IMHOTEP in ancient Egypt. (ibid.: 83) Thairu’s enduring legacy would for long determine KEMRI’s key research agenda – a concentration on common people’s diseases: malaria, bilharzia, tuberculosis, HIV/AIDS and so on. Even indigenous medicine, a ‘no-go area’ for modern researchers hitherto, got a place in KEMRI through the establishment of the Traditional Medicines and Drugs Research Centre in 1984 (Iliffe 1998: 192, 197–200). Therefore, while Tom Mboya and Emilio Mwai Kibaki, Kenya’s 1960s ministers for planning and finance respectively, implemented economic Africanization, affirmative action for Africans in trade, industry and land settlement, Joseph Maina Mungai, Edward George Kasili, Nelson Awori and Hilary Ojiambo, the pioneering Kenyan medical researchers, implemented a parallel medical Africanization. Indeed, biomedicine, trade, industry and African resettlement provided a ‘meeting point’ for political and medical elites of independent Kenya. Just like the economy, biomedicine became a vital tool through which the new-found freedom and self-determination were expressed. Medical research was of value not only because it was an instrument of state dominance but also because it provided important ways through which African scientists reclaimed their stake in the wider world dominated by the west. The congruency between medical research and personal aspirations, on the one hand, and medical research and state hegemony, on the other, was aimed at interrogating the pre-existing intercultural and interracial barriers nurtured by colonialism. 364

The Historical Interface between the State and Medical Science in Africa

That way, biomedicine in independent Kenya, like during colonialism, became a symbol for both state power and individual aspirations. Although the Kenyan state essentially, and successfully, enhanced the core hegemonic attributes of the colonial legacy through biomedicine, the situation changed for the worse from the 1980s onwards. In biomedicine, like in economics and politics, external and internal demands eroded the state capacity and resources to sustain the 1960s and 1970s developments. Buoyed by an unprecedented rise in the number of people seeking health services and the financial expenditure involved, the government health institutions simply buckled. It did not help matters that there was rampant corruption in the ranks of the health ministry, particularly in the supplies of medical stores. The 1996 scandal in which the Ministry of Health Permanent Secretary was dismissed for signing a provisional tender of antimalarial chemicals costing 79 per cent of the ministry’s annual budget was a classic example of entrenched corrupt practices. Infrastructural decay was evident in government health institutions while lack of transport for emergencey, referral and mobile services crippled the once popular medical system. Overcrowding and acute shortage of drugs become the norm not the exception in government health facilities. It is estimated that annual per capita public expenditure on health was USD 9.5 during 1980–81. By 1996/7 it had fallen to a paltry USD 3.5. Poor pay for public service doctors and state ruthlessness in handling doctors’ demands for collective bargaining, which culminated in the 1994 doctors’ strike, pushed many doctors to private practice. All these problems exposed the Kenyan medical system, just like the state, to international scrutiny, ridicule and distrust (Iliffe 1998: 72, 182–84, 193–94). The public health crisis was aggravated by the local political fluidity of the 1980s, which opened political space to many actors, and the shift in international paradigm, following the end of the Cold War, which thoroughly undermined state command (Young 2004: 37). With all these, the colonial legacy of state hegemony in the country’s medical system was surely coming to an end. The cross-currents which made the Kenyan state less dominant as an agenda setter and ‘ultimate’ command in medical policy, created a chasm that was filled in by strengthened non-governmental and international medical organizations, especially in research and therapy on malaria and HIV/AIDS.20 The efforts of non-state actors or international collaboration arose from the combination of technical and political reasons. Compared to her sister states, Uganda and Tanzania, Kenya had sufficient medical infrastructure to support international collaboration.21 Besides the Division of 365

Evidence, Ethos and Experiment

Vector Borne Diseases (DVBD), Ministry of Health, which had a long history of research on communicable diseases (see Geissler, this volume), there was the Nairobi Medical School and KEMRI which emerged, as alluded to earlier, in 1967 and 1979 respectively. Also, the confluence of international interests earmarked Kenya as a key western ally where British and American medical research efforts converged. Apparently, there was a perceived need to protect Kenya during the Cold War, which the west considered under its influence. Among the prominent examples, an international research initiative was started seventeen years ago with funds from the Wellcome Trust in the KEMRI Malaria Research Unit in Kilifi, on the coast of Kenya.22 Together, local scientists and those from the Liverpool University, Department of Tropical Medicine, and the Kilifi unit conducted clinical trials of a new antimalarial drug, lapdap, following the widespread resistance to the existing drugs. The unit also embraced the newfound community-researcher collaboration, a new trend which increasingly took researchers away from the laboratory to work for and with local communities. The unit doctors did not only relate to the locals as the source of their study material but also undertook projects of social responsibility. In the adjacent Kilifi District Hospital they operated a paediatric ward to treat children infected with malaria. To forestall the threat of multidrug-resistance, they trained local shopkeepers and kiosk owners on how to administer antimalarial drugs.23 If the Wellcome Trust concentrated in funding biomedical research, other international non-state organizations provided preventive and curative services. The African Medical and Research Foundation (AMREF) was one such organization. AMREF, with its headquarters in Nairobi, was founded in 1957 with the aim of providing plastic surgery and medical care to remote areas. It pioneered outreach clinical medicine (mobile medical units) among the nomadic populations in Kenya. During the 1980s its focus shifted more to community-based health schemes which emphasized community participation, training and health education.24 Furthermore, AMREF, besides its well-known airlift rescue services, has over the years collaborated with the Ministry of Health to promote both primary healthcare and preventive medicine particularly through a programme of providing health learning materials on malaria for schools and high-risk groups such as expectant mothers. In 1983 AMREF went into a research collaboration with KEMRI, whose findings confirmed malaria resistance to chloroquine.25 Recently, between 1998 and 2002, they undertook an employer-based malaria control programme in collaboration with the Malaria 366

The Historical Interface between the State and Medical Science in Africa

Control Unit, Ministry of Health. The project focussed on the improvement of health and socioeconomic status of workers, farmers, members of women groups and their families in the malaria endemic to the coast and the lake region. AMREF, like the Wellcome Trust, sought to move out and strengthen industries, local institutions and the capacity of local communities in malaria control and prevention.26

Concluding Remarks Over the years, biomedicine has had a fundamental impact on the African state and its apparatus. The state was, during the colonial era, and even more so in the post-independent period, the single most important purveyor through which people reaped the benefits of medical science in Africa. However, the colonial medical system permitted little equal access racially, socially and geographically. It was a significant ‘tool’ of imperial domination and exclusion. In Sierra Leone, for example, the vector theory of malaria and concept of Africans as ‘health disease carriers’ reinforced the use of biomedicine to ratify the practice of racial segregation as a measure against malaria. In 1899 Ronald Ross, the man who discovered that it was mosquitoes that were responsible for malaria transmission, lead a team from the Liverpool School of Hygiene and Tropical Medicine to Freetown, Sierra Leone. Following his Indian experience, where separate quarters for Europeans and ‘natives’ kept the former’s infection low, Ross recommended racial segregation in Freetown, a measure which the local colonial administration pursued with profound alacrity (Spitzer 1968). Consequently, in the rest of Africa, just like the Kenyan example has shown, colonial medicine was closely intertwined with imperial political interests of power and domination. While the postcolonial state perpetuated medical modernization of the 1950s, the late colonial period in Africa, it failed to ‘liberate’ biomedicine in Africa from the historical state hegemony. In this chapter I have tried to show that colonial biomedicine in Kenya was a significant aspect which afforded imperial leverage to the state. Incorporating most of the colonial features, especially institutions of hegemony, the immediate post-independence state acquired the ‘postcolonial’ characterization. But this was only for some time. From the 1980s onwards, the role of the state in biomedical development dwindled. To fill up the emerging ‘void’ non-state and international organizations strength367

Evidence, Ethos and Experiment

ened their role in Kenya’s disease and research policy. If, traditionally, the country’s medical system and research largely evolved in the historical and social context in which the state dominated, what does the more than two decades of state decline portend for tropical medical research? Perhaps the key highlight of this narrative was the role the postcolonial state played in altering the posture of tropical medicine from being only geographically ‘tropical’. For the first time, in the massive proliferation of medical facilities and the emergence of government-funded research institutions, we saw traces of tropical medicine for the South by the South to some extent. Recent developments seem to suggest the increasing dominance of non-state and international actors in biomedical research in Kenya. These organizations usually were, and continue to be, supported financially from the North; they are made up of key researchers from the North, and a few collaborators from the South, and conducted in ‘fortified enclaves’ which tap research material from communities nearby. Does this phenomenon represent a shift from tropical medicine with less social connotations to one with more geographical ones? Also, the immediate postcolonial Kenyan scholars showed a predilection towards social and cultural imperatives in the understanding of their mission and research output. This was partly because of the ‘protection’ they enjoyed from the hegemonic state. Their dominant concern and fascination was what the ‘African’ was or should be regarded to be. They emphasized, if not exaggerated, biomedical progress in explaining how the ‘African’ was either at the ‘same’ level or was ‘like’ the hitherto dominant races. One wonders whether, in an increasingly collaborative research environment and given the dismal state support, the leverage and ‘freedom’ Kenyan medical researchers enjoyed in the 1960s has any future.

Notes 1. The literature on this subject started to appear from the mid-1980s but mushroomed in the late 1990s. Since 2000 the theme has acquired unprecedented prominence in historical and political literature. See Young and Turner (1985); Mamdani (1996); Mbembe (2001); Cooper (2002) and Young (2004). 2. King George VI Hospital, Nairobi, was renamed Kenyatta National Hospital at independence. 3. Due to the fact that porterage was the main means of transport in the tropics before colonial railways, Burroughs, Wellcome & Company in London produced a selection of combined drugs in the tabloid form packed in neatly made bottles and 368

The Historical Interface between the State and Medical Science in Africa

4.

5.

6.

7. 8. 9.

10. 11.

12. 13.

14.

15. 16.

fitted inside a strong steel medicine chest specially made to be carried on the porters’ heads. The medicine chest was a key characteristic feature of the many adventurers and explorers to Africa in the 1880s and 1890s. See Boedeker (1935). The phrase ‘tools of empire’ was used by Headrick in 1981. Macleod (in Macleod and Lewis 1988: 2) took it up and used it in reference to the relationship between medicine and imperialism. See also Beck (1970). See Achola (2002). For a detailed discussion on how biomedical functioning in Africa was a product of subjective socially constructed whims unique to colonialism see Vaughan (1991). This argument was given by Dr Burkitt, a private practitioner in Nairobi, in his recommendation to the Sanitation Commission set up to look into the sanitary conditions in the emerging colonial towns. See The Leader of British East Africa, 15 March 1913, p. 11. The Official Gazette of the East Africa Protectorate, 15 October 1913, pp. 875–78. For a detailed study on how this happened during the great Bombay plague epidemics of the 1890s, see Chandavarkar (1998). For example, K.V. Adalja, one of the earliest Indian private medical practitioners in Nairobi reported cases of his patients who strongly resisted the notification of plague: ‘Transfer of [plague] patients to the infectious Disease Hospital was considered to mean certain death and every possible pressure was therefore brought to bear to prevent notification of cases of contagious diseases, more particularly of cases of plague. Six families of my patients once left me simply because I refused to accede to a request of this kind. To examine a blood-slide for malaria was regarded with deep suspicion lest, in fact, one were seeking to confirm a diagnosis of plague’ (Adalja 1959: 444). Maina Mungai, a leading pioneer Kenyan medical researcher, termed Vint’s views as ‘erroneous conclusions’; see Mungai (1998). For example, Europeans in Nairobi suffered epidemic malaria at an interval of five years since 1926. There was an epidemic in 1930, 1935 and 1940; see De Mello (1947). KNA, PC/NZA 1/7, Nyanza Province Annual Report, 1911–1912, p. 26. The promulgation of this ordinance followed the visit to Kenya of W.J. Simpson, a Professor of Hygiene at Kings College, London, detailed by the CO to advise on the sanitary policy in the colony. On the basis of his findings Simpson recommended the formalization of racial residential zoning on health grounds (The Official Gazette, 15 October 1913: 875–8; Medical Department 1913: 18, 34) ‘Editorial’, in November 1928 in KEAMJ 5(8): 249. This editorial commentary was as a result of the conviction among medical officials that the epidemic of 1928 was caused by increased intercommunication between African reserves and settled areas through the medium of migrant labour. Medical Department Annual Report 1926, p. 47. Ibid., pp. 47–48. 369

Evidence, Ethos and Experiment

17. Ibid., p. 48. 18. Moi University in Eldoret town started a medical school a few years ago while Kenyatta University in the outskirts of Nairobi is in the process of establishing one. 19. Much of the funding that supported the establishment of KEMRI and its various research programmes came from JICA, a Japanese agency for international development. 20. The argument that the role of non-state and international actors increased with the dwindling of state dominance in medical policy is a normative notion as of now. It will need more research of on the role of these actors to come up with an empirical picture of their contribution. 21. The 1970s and 1980s state crisis and subsequent collapse in Uganda made Kenya overtake it as a regional centre for medical research. The Makerere Medical School quickly lost its former reputation and was replaced by Nairobi Medical School. The west was reluctant to deal with Nyerere, Tanzania’s president, because of his socialist inclination. 22. The Wellcome Trust started its operations in Kenya in 1957 in funding what they regarded as inadequately supported subjects. Its activities picked up since its collaboration with KEMRI in the 1980s onwards. See Wellcome Trust first report covering the period 1937–1956 (London, 1956), pp. 35–37; The Wellcome Trust 19th report, 1989–1990 (London, 1991), p. 92. Copies of these reports are held at the Wellcome Research Library at the DVBD building, Nairobi. 23. The Wellcome Trust 19th report, 1989–1990 (London, 1991), p. 92. Also, the activities of the KEMRI malaria unit at Kilifi were reported in the Daily Nation, 15 July 1999, p. 3. 24. AMREF, General Background. This document is held at the Wellcome Trust Research Laboratories, Nairobi. Also see Daily Nation, 6 October 1998. 25. AMREF, General Background. 26. AMREF’s Malaria Programme: The First Year, October 1992 – September 1993, p. 4–5.

References Achola, M.A. 2002. ‘Colonial Policy and Urban Health: The Case of Nairobi’, in A. Burton (ed.), The Urban Experience in Eastern African, 1750–2000. Nairobi: BIEA. Adalja, K.V. 1959. ‘Thirty-two Years in General Practice in Nairobi’, East Africa Medical Journal 36(8): 444. ———. 1962. ‘The Development of the Medical Services in Kenya’, East African Medical Journal 39(42): 108. Beck, A. 1970. A History of the Medical Administration of East Africa. Cambridge, MA. Boedeker, H.A. 1935. ‘Some Sidelights on Early Medical History in East Africa’, East African Medical Journal 12(4): 100–7. Chandavarkar, R. 1988. Imperial Power and Popular Politics. Cambridge University Press, Cambridge.

370

The Historical Interface between the State and Medical Science in Africa

Cooper, F. 2002. African since 1940. Cambridge: Cambridge University Press. Dawson, M. 1979. ‘Disease and Social Change: Smallpox in Kenya, 1880–1920’, Social Science Medicine 13B(4): 245–48. De Mello, J.P. 1947. ‘Some Aspects of Malaria in Kenya’, East African Medical Journal 24(1): 115. Gilks, J.L. 1933. ‘The Medical Department and the Health Organization in Kenya, 1909– 1933’, East African Medical Journal 9(12): 340–52. Headrick, D.R. 1981. The Tools of Empire: Technology and European Imperialism in the Nineteenth Century. Oxford University Press, Oxford. Iliffe, J. 1998. East African Doctors: A History of the Modern Profession. Cambridge: Cambridge: University Press. Kasili, E.G. and J.R. Taylor. 1970. ‘Leukaemia in Kenya’, East African Medical Journal 47(9): 461–68. Macleod, R. and M. Lewis. 1988. Disease, Medicine and Empire. Routledge: London. Mamdani, M. 1996, Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton: Princeton University Press. Mbembe, A. 2001. On the Postcolony. Berkeley: University of California Press. Mungai, J.M. 1998. ‘Research Publications are Important Performance Indicators of University Medical Education in African Countries’, East African Medical Journal 75(6): s31. ———. 2002. From Simple to Complex: The Journey of a Herdsboy. Nairobi: Kenway Publications, p. 85. Patterson, J.H. 1907. The Man-eaters of Tsavo and Other East African Adventures. Macmillan, London. Sequeira, J.H. 1932. ‘The Educational Aspect of Public Health Work in the Tropics with Special Reference to Kenya’, East African Medical Journal 9(3). Spitzer, L. 1968. ‘The Mosquito and Segregation in Sierra Leone’, Canadian Journal of African Studies 2(1): 54–56. Thairu, K. 1975. African civilization: Utamaduni Wa Kiafrika. Kampala: East African Literature Bureau. Vaughan, M. 1991. Curing Their Ills: Colonial Power and African Illness. Cambridge: Polity Press. Vint, W.F. 1932. ‘A Preliminary Note on the Cell Content of the Prefrontal Cortex of the East African Native’, East African Medical Journal 9(3): 48. Young, C. 2004. ‘The End of the Post-colonial State in Africa? Reflections on Changing African Political Dynamics’, Africa Affairs 103: 23–49. Young, C. and T. Turner. 1985. The Rise and Decline of the Zairian State. Madison: University of Wisconsin Press. Zeleza, T. 1989. ‘The Establishment of Colonial Rule, 1905–1920’, in W.R. Ochieng (ed.), A Modern History of Kenya 1895–1980. Nairobi: Evans Brothers, pp. 63–65.

371

Chapter 15

The Intimate Rules of the French Coopération: Morality, Race and the Postcolonial Division of Scientific Work at the Pasteur Institute of Cameroon Guillaume Lachenal

This article aims to observe the ethics of medical research through the lens of historical ethnography. It focuses on the most important institution of biomedical research in Cameroon, the Pasteur Institute of Cameroon, created in Yaoundé in the late 1950s. The Pasteur Institutes of Dakar, Bangui, Abidjan, Brazzaville and Yaoundé have been the emblems of the French scientific presence in Africa. Between 1960 and 1990, French expatriates ran these prestigious laboratories and maintained close relations with the Pasteur Institute of Paris. The African Pasteur Institutes embodied and symbolised a certain idea of French grandeur, in line with the ambitions of ‘La Coopération’, the official name of the French foreign policy in postcolonial Africa. At the same time, they played a pivotal role in the emerging scientific system of African nations. Their vitality manifested how successfully the Pasteurian international venture, which had been closely articulated to French imperial expansion,1 adapted to French postcolonial redeployment. The term ‘Coopération’2 was coined at the moment of independences, and aimed at redefining the French African relations as equalitarian and friendly. The change coincided with the advent of ‘Development’, when French ‘civilizing mission’ gave place to the so-called ‘technical assistance’. The policy consolidated a network of political, diplomatic, military, economic, administrative and techno-scientific links. Incarnated by the massive presence of French expatriates – the ‘coopérants’ – in African cap373

Evidence, Ethos and Experiment

itals and bushes, the Coopération paradoxically tightened the FrenchAfrican bond. At the top of African and French states, it took shape as a set of informal and secret personal connections. Mighty figures stood behind the scenes of official diplomacy, and violently secured French strategic influence and economic interests in Africa. Between 1960 and 1990, ex-French Colonies not only ‘remained’ in the orbit of France but also formed a peculiar extension of its domestic space, known as the African ‘pré carré’. The Coopération thus catalysed ambivalent and long-lasting relations of ‘brotherhood and subjection’.3 It was only in the late 1990s that the Franco-African tie gradually loosened; that the relation of France with Africa was said to be ‘no longer a family affair’ (Chafer 2005). A series of recent studies have analysed the political history of the ‘FrancoAfrican family’, its role in French politics and its successive adjustments to a changing international context.4 Less has been done to understand the centrality of science and medicine in the system of the Coopération, or to explore, at the level of development institutions, the social meaning of Franco-African relations. The African Pasteur Institutes are particularly appropriate case-studies to analyse such issues: the highs and lows of their histories illustrate the ambitions and ambiguities of the Coopération. They enable us to investigate how this original arrangement has organised the practices and ethics of medical research for several decades. The history of Pasteur Institutes during the time of the Coopération offers several analytical entries into the politics and polemics of medical ethics. First, this paper is concerned with a major element of current ethical guidelines for medical research in Africa, which is often termed as ‘capacity building’ by international funding agencies. In the case of overseas medical research, the notion embraces the policies that are looking for the training and involvement of local scientific communities (from research auxiliaries to academics) through cooperative research. ‘Capacity building’ is often presented as modern ethics’ antidote to North-South inequities as regards scientific institutions, staff and resources. It is essentially a new expression for an old issue: the moral and political questions raised by the division of scientific work in colonial and postcolonial contexts. Our analysis of that issue will confront one of the most obvious paradoxes of French development policy in Africa: though the training of Africans as technical experts was the initial raison d’être of the French Coopération, the everdelayed implementation of this programme, evidenced by the sustained presence of French ‘advisors’ in leadership positions, became the main reason of its contestation in Africa and in France. How the professional 374

The Intimate Rules of the French Coopération

hierarchies of the Coopération were defined and disputed inside the walls of a biomedical institution will be the main focus of this paper. Secondly, the very fact that the Coopération promoted an ethos for postcolonial relations will be examined. Being a good coopérant was above all a matter of style: French expatriates not only had to be expert and generous, they also had to be friendly with their patients and their subordinates. In other words, familiarity, in the sense of informality and intimacy, defined the morale and the aesthetic of the postcolonial encounter. It also inspired nationalist imageries. The Coopération, on this point, was in continuity with French colonialism. The French colonial ethos had been imagined in opposition with the British ‘distance’ towards colonised people:5 policies of inclusive citisenship and assimilation, though limited in practice, exemplified French universalism and idealism; administrators as well as doctors insisted that the intimate knowledge of natives was an essential part of colonial government; ambiguously enough, ‘melting into the African crowd’ distinguished great French colonials. This ethics of intimacy, which then poeticised French postcolonial redeployment, was eventually reproached to the Coopération: according to activists, the friendly ‘françafrique’ turned into a criminal ‘mafiafrique’ (Verschave 2004). Our study will draw on this paradox, and analyse the Franco-African exception as a deeply ambivalent ethical experience. Thirdly, our study assumes that a critical commitment to the ethics of present biomedicine in Africa requires, in addition to the formal discussion of ethical codification, analytical efforts dealing with the asperities and uncertainties of medical research. Our study does not attempt to discuss specific infractions to existing guidelines, nor does it give a complete monograph of the history of Pasteur Institute of Cameroon. It rather tries to focus on selected moments of controversy, assuming that scandals are not only occasions when ethical norms can be revealed and analysed, but also moments of ethical institution and regulation, as scholars from the fields of science studies, micro-history and pragmatic sociology have shown (Revel 1996; De Blic and Lemieux 2005: 9–38; Boltanski 2007). Our approach to archives reflects this assumption. This work has faced archival constraints, which most historians of postcolonial Africa are familiar with: erratic organization and informal – but drastic – regulations of accessibility. In the case of the Pasteur Institute of Cameroon, successive institutional ruptures, as well as accidental and voluntary destruction led to fragmented, if not absent, local archives.6 On the other hand, the trans-national connections of the institute imprinted their multi-polar organization on documents, 375

Evidence, Ethos and Experiment

and distributed relevant sources in Paris, Marseille and Dakar.7 Constituting serial documentation has turned out to be impossible,8 which reinforced our choice to focus on the ‘hot spots’ of the institute’s history. The use of oral sources was crucial: not only did it complement archival lacunae but it also helped identify themes of particular intensity, which then oriented a subjective, rather than extractive, use of existing archival sources. The following analysis, focused on a set of exceptional circumstances or figures, then includes an interpretation of the very structure of archives, assuming that their form mirrors the succession of contractual liaisons and scandalous ruptures that shaped the institution’s history.9 The moments when documents were destroyed, the international pathways of their exportation, as well as the strategies of their ‘confidentialization’, become significant parts of the story. This study thus deliberately combines micro-historical analysis and a reflexive engagement with dispersed and controversial sources.10 Most of the sources of this work are informed – and inflamed – by a current statement: funded in 1959, the Pasteur Institute of Cameroon (which has been subsequently renamed 9781444332865) is still, in 2005,

Figure 15.1: Cameroonisation as Technical Emancipation. A Laboratory at the Centre de Recherches Médicales du Cameroun (Future Pasteur Institute) in 1957. © Centre Pasteur du Cameroun (Infocam, Série Santé, 151615-A) 376

The Intimate Rules of the French Coopération

under the direction of French expatriates. Nevertheless, their presence has been harshly questioned, almost since the creation of the Institute. French Pasteurians11 themselves have regularly claimed their will to integrate Cameroonians as their equals, and eventually as head of the Institute – which would result in the so-called ‘Cameroonization’ of the Institute. Year after year, Cameroonization gradually became the dramatic thread of the postcolonial period: everyone wished to achieve Cameroonization. It was demanded by Cameroonian doctors, it was repeatedly promised by Pasteurian and Cameroonian authorities, but it was systematically postponed. An apparent compromise has thus been negotiated and re-negotiated, not without severe crises. We will analyse a crucial element of this compromise, which lies at the core of the ‘Cameroonization controversy’: the hierarchies which ruled the division of work in the Institute, and which maintained locals in subaltern positions – regardless of their effective academic curriculum. As Gabrielle Hecht has shown in a similar case (the ‘gabonization’ of uranium mines), the hierarchies which separated expatriates and local staff were not only defined by technical skills and diplomas. Their content was socio-technical, as it also involved, for example, discipline at work or domestic behaviour (Hecht 2003). At the Pasteur Institute of Cameroon, morality was crucial to define postcolonial boundaries: professional hierarchies were displaced from the domain of technique to that of private and corporal conducts. For example, it was stressed that Cameroonians were ‘incontinent’ as regards sex, alcohol, money or food, and this was contrasted with the supposed French ‘self-control’, which justified the constant exclusion of Cameroonians from managerial positions. Such intimate rules thus maintained socio-technical divisions which frequently overlapped with racial ones. This article analyses how, in the confined scene of research laboratories, these moral frontiers were simultaneously erected and crossed by locals as well as expatriates.

The ‘Cameroonization of Pasteur’: A Narrative On 1 January 1959, the Institut Pasteur du Cameroun (IPC) was officially inaugurated. The vast and modernist building, in the heart of Yaoundé’s administrative district, opened its doors to French colonial doctors and to their local auxiliaries – nurses, laboratory technicians, and numerous unskilled staff. As was the case in Pasteurian offshoots throughout the Empire, the Institute’s activities combined prestigious missions (research in tropical medicine, public health and hygiene) with alimentary tasks (medical 377

Evidence, Ethos and Experiment

analysis and vaccine delivery). As it was also the rule in imperial Pasteurian outposts, a colonial military doctor, who had been trained in the Pasteur Institute of Paris, was appointed director.12 Exactly one year before the programmed independence of Cameroon (1 January 1960), the allegiance to a French scientific empire was not anachronistic – although it may appear so, at the precise moment when the French army was crushing the radical nationalists throughout the country (Joseph 1977; Mbembe 1996). The commercial and scientific links with the Parisian mother house were organised by a ‘convention’ between the Pasteur Institute and the new government of Cameroon.13 One of the first administrative acts of the Prime Minister Ahmadou Ahidjo,14 the agreement left room for Cameroonian national pride: the new state was proprietary of its Pasteur Institute, which was not the case elsewhere in French Africa where Pasteur Institutes were ‘subsidiaries’ of the Parisian mother house. The director was chosen by the Cameroonian government ‘on the proposal of the Pasteur Institute of Paris’. As a monument of late-colonialism and as the scientific pillar of the emerging nation, the new laboratory captured the ambiguity of the moment: the independence, far from abolishing the colonial relation, redefined it as one of friendship and assistance. What was inaugurated, thus, was also a new period where French grandeur and Cameroonian national development were playing a win-win game, called the Coopération.

Cameroonization as a Development Policy, 1944–1967 The Coopération apparently transformed the meaning of French doctors’ presence in Cameroon. In contrast with the ‘Pasteurization of Cameroon’ (Latour 1988)15 undertaken by the hygienist medical programmes of the colonial period – especially ambitious and well funded in this territory, the postcolonial policy was rather the ‘Cameroonization of Pasteur’. Although the IPC cemented the elective link between French Pasteurians and Cameroon,16 the Coopération defined French presence as temporary. The first director of the IPC, André Gamet, was proud to announce in his first report the future arrival of the first ‘Cameroonian Pasteurian’: Day after day, the new nation is showing a growing interest in the young institute; one Cameroonian doctor, who had been trained during one year in our disciplines, has gone to the mother house to 378

The Intimate Rules of the French Coopération

qualify for entrance into the great Pasteurian family. Two of his compatriots have already been enrolled for the next years to come. (Gamet 1964: 7) Cameroonization, that is to say the progressive replacement of French experts by locals, was on the initial agenda. This meant no revolution: Cameroonization had been designed as a colonial policy, as elsewhere in French and British Africa.17 In the name of equality, but also for economic reasons, the policy of ‘l’africanisation des cadres’ (the Africanization of managerial staff) had been progressively implemented since 1945.18 In the case of medicine, hundreds of ‘African doctors’ were trained in Dakar, the imperial capital. During the 1950s, they were in charge of most of the practical work in medical services throughout the French Empire. In Cameroon in particular, ‘African doctors’ were among the most prominent elites, although they were maintained in subaltern (and underpaid) positions; they embodied a technical version of the colonial ideal of assimilation – turning Africans into Frenchmen. The new deal of the Coopération made promises of equality come true. From 1958, top-ranking French military doctors, such as the director of the IPC, became the ‘technical advisors’ of their former auxiliaries: during the first years of self-government, two prime ministers and most ministers of Health were former auxiliaries of the colonial medical service. Inside the Pasteur Institute, Cameroonians were offered the possibility to become Pasteurians: laboratory researchers and scientific ascetics. Prosper Etoundi Essomba Ntsama (‘Etoundi’) was the first Cameroonian doctor to attend the Cours Pasteur in the Parisian mother house, the course which academically defined the ‘Pasteurian’ pedigree since 1898. Born to one of the most prestigious families of Yaoundé, Etoundi was trained as a doctor in the Faculty of Medicine of Paris, from which he graduated in 1961. He started the Cours Pasteur in October 1962, and eventually qualified as a genuine ‘Pasteurian’. When he returned in 1963 to the IPC, he was appointed head of the medical analysis department. In 1965, he even began to teach in the technicians’ school. In 1966–67 he went back to the Pasteur Institute in Paris for a series of internships, and attended the course of mycology: his professors in France stated that he was ‘among the best elements of his class’.19 Etoundi, at that time, was a figure of the commitment of Pasteurians to the development of Cameroon. French doctors initially presented Etoundi’s ascension as yet another success. At the IPC, staff and incomings grew, international research pro379

Evidence, Ethos and Experiment

Figure 15.2: Professor Etoundi, Head of the Centre Pasteur/IMPM, Interviewed for the National Daily Cameroon Tribune (28 December 1978). Article collected in the Archives de l’Institut Pasteur de Dakar, Carton n°2, Institut d’Hygiène et de Microbiologie du Cameroun.

grammes were launched, and dozens of Cameroonian nurses were yearly trained as laboratory technicians. In an apparent continuity, four French military doctors were appointed as directors between 1959 and 1974, while a dozen others headed respective laboratories. According to annual reports, the Coopération meant a familial atmosphere within the walls, and a promising future for the new Cameroonian nation. Other sources tell a very different story.20

Cameroonization as a Failed Experiment, 1967–1979 Prosper Etoundi, ‘trained as the dauphin’, actually became a pathological problem for French Pasteurians – it was referred to as the Etoundi Case. Tensions emerged when in July 1967, French director Henri Brottes had to be repatriated for health reasons, leaving the director’s office vacant. Etoundi, who was just coming back from France, was more qualified than 380

The Intimate Rules of the French Coopération

the other Frenchmen in the Institute. He was logically ready to become director for an undefined interim. But Brottes, just before his evacuation from Yaoundé, summoned up its remaining energy to solemnly warn French and Cameroonian authorities: Etoundi must not become director. The same day, he called Louis Chambon, the director of the Pasteur Institute in Brazzaville (Congo), to come ‘in emergency’ in Yaoundé and restore order by assuming the Institute’s direction. Everyone in the Pasteurian world feared that Etoundi might attempt to hold the direction – what they understood as a coup. During the following months, anxious letters were exchanged – and carefully archived in a file named ‘Affaire Etoundi’ – between the Parisian bureau for the ‘Pasteur Institutes Overseas’,21 the French Ministry of Coopération, the French military health service and the African Pasteurian subsidiaries. The situation in Yaoundé was said to be ‘very difficult’, ‘exceptional and very serious’. A French Pasteurian substitute for Brottes, allying ‘competence and authority’, had to be urgently recruited. ‘For imperious reasons’ and as a provisional solution, the director of the Pasteur Institute of Bangui (Central African Republic) was asked to replace his fellow from Brazzaville as director in Yaoundé.22 Pasteurians eventually chose as director André Poirier, a military doctor with solid African experience, who arrived in Yaoundé in October 1967. Very quickly, his relationships with Etoundi went ‘very bad’. The Institute was going from one clash to another. Facing an epidemic of robberies and rumours during the year 1968, Poirier had the impression that ‘there [was] a plan aiming to destroy the Institute, … certain persons being ready to use all means … to become the director of what would replace the I.P.C’. He desperately begged for ‘the complete eradication of Etoundi’. Some Frenchmen in the Institute ‘could not take it anymore’ (‘à bout de nerfs’). Etoundi ‘felt he was rejected by the Pastorian family’.23 Ambiguously enough, Cameroonian authorities vetoed any eviction of Etoundi, while supporting a French presence at the head of the Institute. According to Pasteurians, Etoundi was a ‘threat’ for the Institute’s reputation and morality: sending him abroad for supplementary training appeared as the only possible compromise. In 1969, Etoundi, with the support of eminent French doctors, was chosen for a two-year training in Europe sponsored by the WHO. In 1972 Ravisse, the new director of the IPC (who took up the position after an ‘exhausted’ Poirier) confidently saw Etoundi appointed as a Professor of Epidemiology and Hygiene at the newly created Faculty of Medicine of Yaoundé. Ravisse thought that Pas381

Evidence, Ethos and Experiment

teurians, with this prestigious appointment, ‘had found the solution to the Etoundi’s case’.24 This was only the first act of the melodrama. In 1975, Professor Etoundi made a staggering comeback: his appointment as director was officially announced. He benefited from the large-scale reorganization of public administrations, banks and enterprises undertaken by President Ahmadou Ahidjo through a series of decrees issued in 1974.25 The reform, still remembered as the ‘Cameroonization’,26 attempted to incorporate an emerging generation of qualified elites into the state bureaucracy, and to accelerate the advancement of previous generations. This was enabled by the creation ex nihilo of numerous directorial positions, as new structures and new organizational charts were designed. In particular, the IPC was merged into a bigger institution of biomedical research, the Institut de Recherche Médicale et des Plantes Médicinales (IMPM), itself included in a central research agency (the Office National de la Recherche Scientifique et Technique (ONAREST), under the Prime Minister’s authority). The policy had clear nationalist implications. Cameroonization targeted the most visible signs of French presence at the head of local institutions, so that French expatriates were asked to step back. Such national affirmation echoed more general changes: the Cooperation Agreements (Accords de Coopération) signed with France at the moment of the independence were being contested and renegotiated by many African states including Cameroon.27 For the first time – but not the last – the system of the Coopération was diagnosed ‘in crisis’ and ‘profound changes’ were announced as a new president was elected in France.28 In order to keep up with the times, the appointment of Etoundi as head of the IMPM was a clear message to Pasteurian expatriates: time had come for equality. While the Cameroonization policy was smoothly implemented in most research institutions – French expatriates adapting to the equalitarian requirements of the ONAREST – Pasteurians could not think of collaborating with Etoundi in a ‘nationalised’ institute. They understood Cameroonization as a ‘coup de thatre’, one which the cordiality of their relations with Cameroonian officials could not have foreshadowed – the last IPC director bitterly remembers ‘ministers wishing long life to the Pasteur Institute’ and the president ‘Ahidjo laughing … while [he was] injecting him in the bottom’ just a few days before the Cameroonization reform.29 Rather than accepting confrontation with their ‘disappointing’ pupil, Pasteurians decided to leave Cameroon with the agreement of the Parisian direction. The head of the overseas bureau in Paris, Louis Chambon (who had struggled with Etoundi during the summer of 1967), came to Yaoundé to make clear 382

The Intimate Rules of the French Coopération

that Paris would not allow the Cameroonian institute to use the brand name ‘Institut Pasteur’ anymore. On the 21 June 1975, the IPC director left his office and removed the inscription ‘Institut Pasteur’ from the entrance. The building was renamed, under Etoundi’s direction, the ‘Centre Pasteur’. A new future was being imagined: Etoundi announced forthcoming scientific breakthrough. The local production of ‘all vaccines’ was promised;30 to make room for the future production unit, the archives of the old Institute were destroyed. But the Centre Pasteur quickly came across difficulties: all relations with the Pasteur Institute in Paris, whether commercial or scientific, were halted, while the usual subsidy from the Cameroonian government was not paid. Inside the laboratories, the sudden departure of French managers had painful repercussions on work quality: the ‘great house’ was crumbling. An anecdote, omnipresent in oral narratives of the institute’s history, is said to capture what happened to the ‘Cameroonised’ institute: The President Ahidjo was ill and sent his urine [for analysis] at the [ex-] Pasteur Institute. The day after, there was a mistake, a serious mistake. He got his results […] and what was written was: ‘Pregnancy test: positive … or negative’. He was furious, took his phone, and called back Pasteur [in Paris].31 In Paris, military doctors at the Pasteur Institute and at the Ministry of Coopération quickly reacted. In December 1978, a Pasteurian, Jean-Pierre Digoutte, was sent for an ‘evaluation mission’ in Yaoundé.32 Digoutte had just succeeded Louis Chambon as head of the Delegation for the overseas Pasteur Institutes; like his comrade, Digoutte personally knew Etoundi since 1967, when he too had rushed to Yaoundé as a provisional director, to keep Etoundi away from the Director’s office. He spent one week at the Centre Pasteur, and met the most important local authorities – Cameroonian ministers as well as the French ambassador. His report unsurprisingly described a catastrophe: the centre had ‘fallen in a progressive lethargy’ while the projects launched by Etoundi were said to be ‘megalomaniac’. According to Digoutte, Cameroonian authorities themselves pointed out the cause: ‘the character of the director and his inaptitude to administer and organise such a centre’.33 They also prescribed a solution: French staff had to be brought back, and Etoundi evinced. Early in 1980, a French military Pasteurian was appointed by President Ahidjo as head of the Centre Pasteur, with support from the Pasteur Insti383

Evidence, Ethos and Experiment

tute and the French Ministry of Coopération, which promised funds and staff. Etoundi was forced to leave his office and later, his sumptuous official villa – his French successor repeatedly demanded it as it was ‘practically the only advantage … of [the director’s] bleak task’.34 The fall of the nationalised Institute, assessed by a grotesque anecdote and detailed reports, was not only wrecking Etoundi’s personal career. Since Etoundi had incarnated the experiment of Cameroonization, the assessment of his own failure carried with him a more general result. He was also taking on his shoulders a test on aggregated categories (‘Cameroonians’ and ‘Pasteurians’), their technical content and the ability to move between them. Though the experiment was subtly biased by the sudden suspension of the French aid, the fall demonstrated that French Pasteurians were indispensable for decent governance. As an implicit corollary, it also proved that Cameroonians were, as ‘Cameroonians’, inapt for the direction of such a research institute. Ahidjo’s Cameroonization thus worked as a test (épreuve):35 as the reports carefully evidenced the cause of the IPC’s material disorders, with pictures, tables and testimonies, they also produced generalizations and re-ordered categories. The return of French expatriates ratified these new ontological definitions.

The Compromise of La Coopération, 1980–2000 A new institutional compromise was set up in 1980. The institution, with a new status, was renamed ‘Centre Pasteur du Cameroun’ (CPC). The new director undertook the reorganization of its operations with authority. ‘Some measures were hard and led to the dismissal of the employees who did not understand … the real goals of our action’, he stated in his first report. Those who were not ‘attached to the renaissance of the house’, including Etoundi, were sent to other medical institutions, to be replaced by younger recruits.36 A new balance was in the making: the involvement of the Pasteur Institute in Paris was parsimonious, and the organization, budget and evolution of the institution was placed under the control of a Board (‘C.A.’, Administration Council) of twelve Cameroonian members, representing different ministries and authorities.37 The compromise proved stable throughout the 1980s. While incomes grew, new buildings were constructed and research programmes were launched. Several Cameroonian doctors, pharmacists and highly skilled technicians were integrated in the staff and put in charge of medical analysis services; proper scientific work, 384

The Intimate Rules of the French Coopération

by contrast, remained exclusively the task of French managers.38 At the same time, Cameroonian medical elites were able to control the Centre’s operations, and had no difficulties finding positions in other institutions or in the emerging field of international co-operation. Inner hierarchies were still clear, but the local staff were appeased by a rather privileged status: the Cameroonization, during the 1980s, was not an issue anymore. Starting from 1988/9, a dramatic economic crisis shook Cameroon, amplified by the devaluation of the CFA Franc in 1993. It hit the CPC and destabilised that arrangement. With its budget depending on state subsidies and on local customers, both heavily affected by the crisis, the CPC went bust by 1994. With the guidance of its Board and of the French Coopération agency in Yaoundé, it tried to rethink generally its financing strategies and its inner functioning: costly audits and ‘reorganization projects’ followed one another. In accordance with the guidelines of structural adjustment plans, the main adaptation was to make half of the staff redundant in 1996, and to cut the salaries of remaining employees by 50 per cent. In a tragic and tense ambiance, books and archives were burned.39 The Cameroonization problem also resurfaced. In 1996, French Ministry of Coopération officials requested the ‘Cameroonization of the director’ as a prerequisite for financial support, officially in the name of Cameroon’s ‘national sovereignty’, more pragmatically because of the cost of military doctors’ salaries. While French policy-makers repeated ‘no more substitution’, several Cameroonian scientists could claim their own capacity to manage high-profile transnational research, as their skilful involvement in the AIDS field had been demonstrating for a couple of years (Eboko 1999: 123–40). The traditional lines of labour division inside the CPC were also fading: local Ph.D. students began to be trained in the laboratories, and in 1998 a full-time Cameroonian researcher was recruited after his training in France. Throughout the late 1990s, the implementation of Cameroonization was said to be imminent. But successively in 1996, 1998 and 2002, as a new director had to be appointed, Cameroonization was postponed. Despite obvious changes, the most symbolic – and controversial – rule of French Coopération at the CPC seemed too essential to be overcome. The French expatriate in charge of medical Coopération for Central Africa conceded in 2002: ‘Lab managers are all French. We have only been cameroonizing “bench managers”. While the scientific level [of Cameroonians] is phenomenal … at Pasteur they say: it’s gonna happen’.40

385

Evidence, Ethos and Experiment

The ‘Cameroonization of Pasteur’: Theories and Tensions The second part of this study further dissects the Cameroonization controversy. It examines the successive theoretical arrangements which regulated the division of work at the laboratory, their tensions and their articulation to the postcolonial situation.

The Impossibility of Technical Assimilation As forty years of failed attempts have demonstrated, Cameroonization is not a technical issue. Africans’ socio-technical mobility was one of the most significant projects of late-colonial and postcolonial development. Technical assistance, academic patronages and student grants were thought to guarantee a quick relieve of French coopérants. From the 1960s numerous Cameroonians actually produced impressive academic curricula; secure positions for themselves in local ministries, universities, hospitals and enterprises often made their return to their homeland attractive. However, in the bastions of French science, Cameroonization has proved very difficult.41 While French policy makers, as well as Cameroonians, acknowledged that French presence was abnormal, it became hard to justify it by diplomas’ insufficiencies. Etoundi’s career has indeed revealed the inner contradictions of the postcolonial policy of ‘technical assimilation’. The rupture of 1975 manifested its failure publicly, when Parisian authorities did not allow Etoundi, though a theoretically perfect ‘Pasteurian’, to use the name ‘Institut Pasteur’. Precisely because ‘enough is never enough’ to be a Pasteurian, as a former French director implicitly suggests – although he knew well that Etoundi was a Pasteurian: – [former French director]: When it was Cameroonised, the Cameroonians didn’t not want to keep the label ‘Institut Pasteur’, and on the other hand, the Pasteur Institute in Paris didn’t want to give the name ‘Institut Pasteur’ to an institute that was not headed by a French person … – [author]: It really had to be headed by a French person? – [former French director]: By a French person or by someone trained in the Pasteur Institute of Paris…But Etoundi had absolutely no qualification for … [silence].42 386

The Intimate Rules of the French Coopération

During the years of heated conflicts with Etoundi, even Pasteurians acknowledged that ‘as of right, his complaints were justified’. Yet they added that ‘as of facts, it was useless to rely on him for the direction of a service, or even to work on a steady basis’. The equation was thus simplified: what ultimately determined the ability to manage was not the curriculum – Etoundi’s was impressive. The reciprocal relation was true too: some devoted Pasteurians were not academically Pasteurian. Indeed, when looking for the replacement of their ill comrade in 1967, Pasteurians in Paris and Africa were alarmed: their ‘backup director’ Poirier, although a ‘man of great value … had unfortunately no diplomas from the Pasteur Institute’. Which, they added, ‘may create problems due to the presence of M. Etoundi who turns out to have graduated from the Pasteur Institute.’ But unable to find a candidate with adequate titles, Parisian authorities had to resign themselves to choose Poirier, and to hide his imperfect pedigree to the Cameroonian government. So that by 1968, the Pasteurian version of ‘republican elitism’ – equality and exams as the condition of social mobility – was already undermined by two symmetric counter-examples. Another coherent theory of French Coopération had to be elaborated, which could justify the prolongation of coopérants’ mandate.43

The Moralization of Professional Categories Technical and academic qualities being irrelevant, debates about the Institute’s management were led with blurred notions of what were the indispensable attributes of Pasteurians. Efforts to define the ‘Pasteurian essence’, recurrent from the ‘affaire Etoundi’ in the 1960–70s to the 1990s, have been ‘precisely ambiguous’, so that it has always been possible to circumvent definitions based on academic paths: what mattered above all was the loosely defined ‘Pasteurian spirit’. On the other hand, the definition of the undesirable ‘Cameroonality’ was a lot more clearly cut: it was one of greed and appetite – an incapacity to face temptation. As he is remembered, Etoundi is the very embodiment of these Cameroonians’ attributes: that is, not only being culturally at risk for mismanagement but also, as Achille Mbembe (2000: 152) put it, expressing a corporal habitus ‘hostile to continence, frugality, sobriety’: [Etoundi] was a very ambitious man, who had not a few faults … He was keen on women, he was keen on booze, he did not really fit in, 387

Evidence, Ethos and Experiment

one can say. […] He didn’t know how to manage or administrate the Pasteur Institute for its wellbeing. Of course, for his own wellbeing … yes, he did have a sense of management.44 Similarly assessed by his contemporaries, Etoundi’s lacunae were presented as intimate failures. Brottes made this explicit in 1967: He was not reproached with any bossy acts, which would be incompatible with [the other doctors’] functions. But his general behaviour and attitude make the everyday life much harder. Besides, his technical efficiency is almost inexistent. Is this failure the consequence of a currently troubled family life? It is hard to tell.45 The inadequacy of Etoundi was thus connected to moral and domestic incapacities. In other writings, Pasteurians felt necessary to multiply allusions to Etoundi’s familial life: married to a French woman, Etoundi was in divorce court during 1967–68. On this point too, Etoundi certainly ‘did not fit’ and contrasted with his fellows – military doctors with respectable families and mundane spouses. In other words, the boundaries which confined Etoundi’s ambitions were informal and intimate frontiers. ‘L’affaire Etoundi’ explicitly transformed Cameroonization into a matter of culture. In the following years, the narratives of Etoundi’s rise and fall served as edifying examples to illustrate and legitimise stereotypes about Cameroonians’ inability to govern without ‘tribalism’. Through narration and commemoration, the individual case of Etoundi took on an essential meaning: Cameroonian’s culture was equated with immorality, which justified the division of work at the laboratory. This was not specific to Pasteurian institutions. Many commercial firms and development organizations working in postcolonial Africa developed similar rationalizations of the postponed ‘Africanization’ of their staff.46 Culturalism became a crucial element of postcolonial management, precisely because it guaranteed the coherence of its theoretical system: what explains the failure of technical assimilation is not the inconsistency of the project itself but an incompatibility between the ‘African culture’ and the sense of direction and duty necessary to medical and scientific ‘good practices’. Ironically enough, the explanation was contained in the rationales of late-colonial development plans. The psychiatrist J.-C. Carothers, who categorised the ‘African mind’ as essentially immoral, had considerable influence among medical planners in the fifties (Carothers 1953). Memo388

The Intimate Rules of the French Coopération

randa about medical education would insist that Africans, though satisfying in their technical work, had to be confined to subordinate tasks. Quoting Carothers, the head of the colonial Health Service in Cameroon explained in 1954 that the ‘sense of responsibilities’ needed to replace European doctors in managerial functions ‘will be for a long time inaccessible to African agents …. The ameliorations will depend on the evolution of [their] mentality, a phenomenon which does not (or in a very limited fashion) devolve upon [European] technicians’.47 A statement which seemed to predict and rationalise Cameroonization’s failure even before the policy was effectively implemented.

The Needs of Reproduction Yet the cultural hierarchies defined after the return of French expatriates in 1980 did not simply manifest some resurgence of ‘colonialist thought’. Their importance was not merely theoretical. What was also at stake was the reproduction of a professional space. A few points can be recalled. Pasteur Institutes in Africa formed a closed circuit, where dozens of military Pasteurians would spend their entire career.48 The vast majority of scientists working at Yaoundé’s institute were part of this corps. Rotations would be organised every four years, among directors as well as other scientists. Moves would carry them round between Bangui, Yaoundé, Dakar, Brazzaville, Abidjan, Tananarivo – in whatever order. Those postcolonial careers – ’les carrières africaines’ – were constructed along inter-African connections, and France only served for summer vacations. So that Pasteurians coopérants not only shared a common scientific ethos, but had close relations with each other and with members of their Parisian hierarchy, who had followed the same tracks before them. Their professional positions enabled particularly comfortable lifestyles, as the already attractive salary was completed by ‘entertaining costs’ (frais de representation) and other perks (such as enviable houses, household staff and an official car). Therefore, it is unsurprising that French doctors were frequently accused of ‘hanging on to their jobs’. For them, preventing Cameroonization guaranteed a professional status quo. Dr Waddy, a former colonial doctor in British colonies and a notorious admirer of French Coopération, explained it frankly in 1965: The only object [of French doctors] is to do the best possible work not only through dedication but also in order to gain promotion in 389

Evidence, Ethos and Experiment

the corps, at the head of which – all in the same service – they can see the post of generals, professors, directors of Pasteur institutes and specialists of all disciplines. … I have heard … of the difficulty arising from the fact that in the African states local doctors are glad to take the urban specialists’ posts …. This is not a matter of jealousy or envy by the French: … if such a service is to be kept in being, there must be smooth avenues for promotion through all grades. (Waddy et al. 1965: 252–53) It can be said, without exaggeration, that the rigidity of Pasteurian hierarchies buttressed an entire scientific enterprise, as well as the alluring life overseas of dozens of French families. This system shaped medical research practices. Between 1960 and 1980, research in the Pasteur Institutes in Africa benefited from the inter-African job swap of Pasteurians. Scientific practices used the pyramidal inner organization of Pasteur Institutes, with few French scientists topping an armada of local unskilled auxiliaries. In Yaoundé as in other places, the most important scientific activity was research on arboviruses. Arboviruses being by definition present in insects,49 the project, launched in 1962, consisted of massive collections of mosquitoes, in collaboration with entomologists50 and numerous ‘mosquito boys’. Viral isolation required inoculation of mosquito extract to newborn mice, therefore huge animal facilities had to be kept. Here again, local unskilled staff was crucial to their ‘taylorised’ organization. In laboratories, various techniques then aimed to identify and isolate viruses. Each isolated viral strain was sent to the Pasteur Institute of Dakar, which received samples from all African Pasteur Institutes and worked in close links with the WHO and American organizations. Yaoundé was thus part of an inter-African network, where technical knowledge and managerial ethos circulated with moving Pasteurians. Etoundi insisted on being integrated to the arbovirus lab. But he was kept away from the benches, explicitly because he ‘did not fit’ into the inter-African planning of the research.51 Being Cameroonian and responsible for the future Cameroonization of the IPC, he obviously could not hope for a mobile carrière africaine. The inter-African routes of Pasteurian science thus relied on – and reinforced – local hierarchies perpetuating the colonial order.

390

The Intimate Rules of the French Coopération

Race and the Control of Professional Mobility The elaboration of a cultural theory of Cameroonization filled the needs of scientific (re)production, and indeed reproduced colonial hierarchies. The importance given to domestic behaviour in the debates about Etoundi – one French Pasteurian officially reproached Etoundi with not coming to the expatriates’ daily aperitif – suggests another continuity with the colonial situation: the fact that private conducts were the ‘social and cultural space where racial classifications were defined and defied, where relations between coloniser and colonised could powerfully confound or confirm … the categories of rule’ (Stoler 2001: 830). In that sense, defining an essential ‘Cameroonality’ in moral, domestic and prosaic terms participated in a racialization of professional hierarchies. This racialization, certainly euphemised through the notion of ‘culture’,52 responded to the dissolution of colonial hierarchies by technical emancipation and postcolonial familiarity. Racialization of professional hierarchies sometimes consisted in explicit procedures. A famous case is still remembered today as the ‘affair of the coloured coats’. In his attempt to ‘sanitise’ the Institute after the disorders of the Cameroonization, the French director of the early 1980s – a military doctor remembered by some as ‘an old school colon’ – decided to use coats

Figure 15.3: ‘We Were Wearing the Coat of a Mechanic’. A Laboratory Technician in Dark-Green Coloured Coat, c1986. © Centre Pasteur du Cameroun. 391

Evidence, Ethos and Experiment

Figure 15.4: Unskilled Staff Wearing Dark Blue Uniforms at the Centre Pasteur du Cameroun, c1986. © Centre Pasteur du Cameroun.

of bright colours to distinguish the different classes of staff. White was destined to medical doctors and pharmacists, marine blue was for laboratory technicians, dark green for other subordinate laboratory staff, and orange for administration. What may appear to the reader as a conventional policy of staff classification – similar to those applied today throughout European hospitals – was actually interpreted by the staff as a demonstration of racialist management. Highly skilled Cameroonian lab technicians – trained in France, well-paid, belonging to the state ‘bourgeoisie’ – had to wear the dark coat of ‘a mechanic or of someone who was sweeping the backyard’.53 On the other hand, a French lab technician, incidentally the nephew of the director, was given a white coat, which was hardly justified by his diplomas and by its local status of ‘petit blanc’. The loss of the white coat was not incidental in Yaoundé’s postcolony, where ostentation mediated social stratification (Mbembe 2000: 181–86). Many felt humiliated, and one Cameroonian technician even chose to resign, irritated by such blatant injustice. Quite originally, the policy entailed an inversion of the social order: a ‘black bourgeois’ had to wear the uniform of the ‘bas peuple’. The coloured categorization did not perform class hierarchies (as is the case in Europe) but seemed to reverse them by the superimposition of racial distinction. The second originality of the policy was its initial motive: pre392

The Intimate Rules of the French Coopération

venting usurpation. The general use of the white coat by Cameroonian staff had led to ambiguities. ‘Docta’54 had become everyone’s function, at least in public eyes, from the lab cleaners to the French director. Competent lab technicians were often able to diagnose and prescribe, and get paid for it. A ‘clarification’ was needed, not only to rationalise the commercial and scientific operations of the laboratory but also because the French director’s own prestige was challenged, tells a member of the staff: It is because of [Mr Xxx, a Cameroonian nurse and lab technician] that this story of multicoloured coat happened. [Mr Xxx] did his medical studies, he was a graduate nurse. He was the head of a medical district [during the colonial period]. He was able to cure, to prescribe, to consult. He did all that very well. One day one minister came [to the Centre Pasteur]. Someone whispered to [the French Director]: ‘Here’s a minister’. The Director came to the minister: ‘Hello, I am Director of the Centre Pasteur, let’s go to my office’. The minister said: ‘No, wait, I came here to see my doctor’. [The Director asked]: ‘Which doctor?’ [The Minister]: ‘Doctor [Xxx]’. [The Director]: ‘But he is not a doctor, he is a technician’. ‘ [The Minister]: ‘He has been my doctor for 10 years. He gave birth to my daughter’. This is why the Director dressed [Mr Xxx] in blue.55 As an attempt to define professional categories, the ‘story of coloured coats’ replicated the mechanism of the Affaire Etoundi. In both cases, the racialization of management was articulated to the control of professional mobility; in both cases, Cameroonians did not stand as the passive objects of the disciplinary projects of French expatriates, but they actively challenged the internal order of the Pasteurian house; in both cases, movements through boundaries were deemed moral infractions and immorality was pointed out as a cultural characteristic of Cameroonians. We should add here that (im)morality, which organised postcolonial hierarchies in the Pasteurian house, has been a major matter of concern for French managers. Not because of Cameroonians’ licence – which is expected and even reassuring – but because Pasteurians themselves have been worrying about their own moral rectitude. Indeed, in Yaoundé, the limits of ‘respectable’ behaviour have been regularly crossed by expatriates: audits and 393

Evidence, Ethos and Experiment

evaluations recorded some of the domestic ‘problems’ of French doctors,56 while the expatriate community had always been fond of gossips featuring Pasteurians’ infractions to usual decency. So that the depiction of Cameroonians’ moral failings takes an internal vocation: it draws, by rebound, distinctions inside the expatriates’ moral space, just as Etoundi, as he was presented as the figure of Cameroonian greed, also served as a counter-definition of a Pasteurian ethos of disinterest and self-control. When gossiping about Etoundi, French doctors were also, and maybe first and foremost, talking about themselves. Cameroonization then takes another meaning, inherited from the French colonial concept of ‘bougnoulisation’57 – the racialised equivalent for the British ‘going native’. For example, speaking of the latest escapade of the French director, one retired – and perfidious – expatriate could insinuate: ‘I’ve heard that [he] is completely Cameroonised’.58 Eventually, if we are to account for the many stories about the appetite – for money or mistresses – of genuine ‘French Pasteurians’59 we might suggest here that the Pasteur Institute has actually been Cameroonised all the way through.

Conclusion By the 1990s the fall of the Coopération, as a period and as a system, was established and sometimes celebrated. A conjunction of geopolitical changes, domestic orientations and African crises prescribed and accelerated a relative ‘withdrawal’ of France from Africa. The ‘normalization of relations with Africa’ – a significant expression – was officially demanded. The politics of ‘Franco-African intimacy’ were considered outdated. The microcosm of Pasteur Institutes reflected and even pre-empted the demand. It was reorganised into the ‘Réseau International des Instituts Pasteur’, a new structure which promoted equalitarian scientific relations. It was aiming for a more global reach, beyond the African ‘pré carré’ of the Coopération, and new Institutes were created or revitalised in southeast Asia and China. The most spectacular internal change was the ‘civilization’ of the expatriate staff. In 2002, a new manager was appointed as head of the International Network: for the first time, he was not a military doctor. In the previous decade, the retirement of an entire generation of military doctors had already left space in Pasteur Institutes for civil scientists, who had not spent their entire career overseas and a fortiori in Africa. African scientists were also more visibly integrated to laboratories. In many ways, their recruitment followed a wider movement, provoked by local scientists 394

The Intimate Rules of the French Coopération

themselves, who proved more skilful than French institutions to adapt to the new rules of internationally funded ‘research by project’. Cameroon is a paradigmatic case: the explosion of the AIDS epidemic, and the following rise of AIDS research, saw a very discrete involvement of the Centre Pasteur on the local scene, which was dominated by local big men based in universities and hospitals. In that case, the loosening of the Franco-African tie coincided with the affirmation of local scientists with global connections (Lachenal 2005: 123–49, 2006: 309–30). Does the end of the Coopération also mean the ‘normalization’ of medical research practices? Such a change, in the light of this article, can appear desirable: most of the inconsistencies of the Coopération were a matter of style, as familiarity often led to obscenity. Formal ethics, with their codes and guidelines, as well as the new modes of regulations of North-South collaborations which emerged in the 1990s, appear both as a promise and a menace. The promise of a reform, that of the unmaking of the FrancoAfrican exception, which had been resistant to citisens’ control for decades. The threat of a loss, that of a dimension of ethics which was embodied – with many imperfections – by the ‘intimate rules’ of the Coopération, but which formal guidelines hardly captures: the fact that ethics emerge from uncertain face à face (encounters) and painful scandals.60

Acknowledgements I would like to thank Sophie Vasset, Guillaume Clément and Linda Amarfio for their help with English. I would also like to thank the organisers and participants of the Kilifi Conference for their precious remarks, as well as participants of various seminars where this work has been presented (Medical Anthropology Seminar, London School of Tropical Medicine and Hygiene; Wellcome Unit for the History of Medicine, Oxford).

Abbreviations AIP: Archives of the Pasteur Institute, Paris, France AIMTSSA: Archives of the Institut de Médecine Tropicale du Service de Santé des Armées, Marseille, France. ACPC: Archives of the Centre Pasteur du Cameroun, Yaoundé, Cameroon. AIPD: Archives of the Pasteur Institute of Dakar, Sénégal. ANY: Archives Nationales, Yaoundé, Cameroon. 395

Evidence, Ethos and Experiment

Notes 1. On the link between Pasteur Institutes and French imperialism see Moulin (1992: 307–22); Dozon (1991: 269–78) and Pelis (1997: 583–622). 2. I agree here with Tony Chafer (2005: 10) that coopération is ‘untranslatable into English, [as] the term denotes something specifically French that is both far more wide-ranging in its compass and more ambiguous in its meaning than its most commonly used English equivalents: development aid or assistance’. 3. This expression, coined by Hannah Arendt, has been used as the title of Jean-Pierre Dozon’s (2003) essay on French African relations, Frères et sujets. La France et l’Afrique en perspective. 4. Dozon (2002, 2003); Mouralis (1997); and the special issue of Modern and Contemporary France, ‘France and Africa in the Global Era’, February 2005, 13(1). 5. I do not refer to essential differences between French and British colonial style, but to the fact that such differences had been actively imagined and constructed by experts in colonial administration (Dimier 2004). 6. The archives available at the CPC mainly concern the 1990s. All other archives were destroyed, by 1978 and by 1995. Colonial archives concerning the construction of the Institute in the 1950s were consulted at the National Archives of Yaoundé (ANY). 7. Research was conducted at the archives of the Pasteur Institute in Paris (France), at the Société de Pathologie Exotique (Paris, France), at the archives of the Pasteur Institute of Dakar (Dakar, Sénégal), at the archives of the Institute de Médecine Tropicale du Service de Santé des Armées (Marseille, France) and the Centre des Archives d’Outre Mer (Aix en Provence, France). 8. No institution, including the CPC, has the complete collection of annual reports. 9. On the ‘content of the form’ of archives, see Stoler (2002: 87–109). 10. More than 60 interviews have been conducted in France, Cameroon and Sénégal with retired expatriates, current member of the staff of the Centre Pasteur of Cameroon, and with various French and Cameroonian medical authorities. Ethnographic observation and archival work was conducted in Yaoundé (Cameroon) on April–May 2002, August–October 2003 and January–March 2005, with the support of the Institut de Recherches pour le Développement (IRD) and of the Agence Nationale de Recherche sur le Sida (ANRS). 11. The term ‘Pasteurian’ (in French Pasteurien, Pastorien) has never been clearly defined. In the more restrictive sense, a Pasteurian is someone who has succeeded in the Cours Pasteur, a course at the Institut Pasteur in Paris. In an extended meaning, Pasteurians are the employees of Pasteur Institutes throughout the world. ‘Overseas Pasteurians’ more particularly refers to the structured body of colonial doctors working in Pasteur Institutes of (former) French colonies. 12. On the link between the Pasteur Institute and the colonial Health Service, see Dozon (1991). 396

The Intimate Rules of the French Coopération

13. The convention was signed early in 1959, after exchanges between Paris and the government of Cameroon at the end of 1958. The convention is partially reproduced in the first annual report of the IPC. AIP, IPO RAP 5–6, A (Gamet 1960). 14. Ahmadou Ahidjo became the first president of independent Cameroon. He left power, officially for health reasons, in 1982. 15. On the history of colonial medicine in Cameroon see Lachenal (2010). 16. For an example of Pasteurians’ writings on Cameroon see Martin (1921). 17. On Kenya see Kenneth Ogombi, this volume. 18. The term ‘Camerounisation des cadres’ appeared in 1957 (Anonymous 1958: 240). 19. AIP, Box IPO-B-10, Fonds Marneffe, 1966–1970, Letter of Dr Segretain to Dr Marneffe, 28 April 1967. 20. The controversy concerning Etoundi resulted in numerous correspondences (most of them informal and hand-written) and reports among the Pasteurian network – representing the main part of existing archives concerning the history of the Pasteur Institute of Cameroon. Significantly, the only available files concerning Cameroon at the AIP were constituted on the occasion of the ‘Affaire Etoundi’ in 1968–1970 (AIP, Box IPO-B-10, DG IPOM Fonds Marneffe, 1966–1970.) 21. The coordination of overseas Pasteur Institutes had been centralised in the Pasteur Institute of Paris, from 1939, by a ‘Délégation Générale aux Instituts Pasteur d’Outre Mers’, exclusively headed by military doctors. 22. AIP, Box IPO-B-10, DG IPOM Fonds Marneffe, 1966–1970. Letter of Dr Chambon to Dr Marneffe, 21 July 1967. Letter of Dr Marneffe to Dr Oddou, 24 July 1967.; ibid. Letter of Dr Marneffe to Dr Poirier, 21 July 1967; ibid. Letter of Dr Marneffe to the Minister of Armies, 28 July 1967. 23. All quotes this paragraph: AIP, Box IPO-B-10, DG IPOM Fonds Marneffe, 1966– 1970. Letter of Dr Poirier to Dr Chambon, 21 April 1968; Letter of Dr Poirier to Dr Chambon, 11 March 1968; Letter of Dr Poirier to Dr Chambon, 1 April 1968; Letter of Poirier to Chambon, 11 March 1968; Letter of Dr Brottes to Dr Etoundi Essomba, (Draft), 19 November 1967. 24. Interview with former French director (dir2), Paris, 11 March 2002. 25. The decree concerning the reorganization of research institutions was the Presidential Decree 74/538 issued on 6 June 1974. 26. The policy had equivalents elsewhere in Africa, such as the radical Zairianization (1973–75) or the more subtle Ivoirianization (1978). See Mundt (1995: 112–13); Bobb (1999: 230–31). 27. The contestation of the ‘Accords de Coopération’ began in 1972 (Anonymous 1973: 95–115). 28. Such announcements have been ritualistic in France. See Cumming and Langford (2005); Bayart (1984). 29. Interview with former French director (dir2), Paris, 11 March 2002. 30. ‘Tous les vaccins bientôt produits au Cameroun’ was the headline of the national daily newspaper Cameroon Tribune on 29 December 1978. AIPD, Box n°2. 397

Evidence, Ethos and Experiment

31. Interview with former French director (dir4), Paris, 11 January 2002. 32. The file concerning this mission was found in the archives of the Pasteur Institute of Dakar (AIPD, Box n°2). It presumably followed Digoutte, who collected it in 1979 just before he was appointed director of the Dakar Institute. The presence of these documents in Dakar reveals the importance of inter-African circulations in the ‘Coopération’ system. 33. All quotations of this paragraph: AIPD, Box n°2, ‘Institut Microbiologie Cameroun/Centre Pasteur/Yaoundé’, ‘Rapport de mission auprès de l’Institut de Recherche Médicale et des plantes médecinales de Yaoundé, J.P. Digoutte, 9 au 15 Décembre 78’, p. 2. 34. AIPD, Box n° 2, Copy of the letter of Jean-Paul Albert to the Minister of Public Health of Cameroon, Yaoundé, 19 March 1980. 35. On the notion of épreuve, see Boltanski and Chiapello (1999: 73–80). 36. AIP, IPO RAP 5–6, J.P. Albert. 1982. Rapport d’activité. Centre Pasteur du Cameroun (1980–1982), Introduction. 37. The new status was given by the Presidential Decree n°80/240 of 30 June 1980 on the reorganization of the Centre Pasteur du Cameroun. 38. Medical analysis has been constantly considered as an unrewarding task by French Pasteurians. For example: AIP, IPO RAP 5–6, H. Brottes. 1967. ‘Rapport sur le fonctionnement technique de l’Institut Pasteur de la République Fédérale du Cameroun’, Année 1966, p. 27. 39. The library of the IMPM, where some archives of the Pasteur Institute were stored, is said to have burned by furious staff on the occasion of one lay-off. Interview with Cameroonian member of the scientific staff of the CPC, CmStaff1, 3 October 2003. 40. Interview with French expatriate (Coop4), Yaoundé, Cameroon, 17 April 2002. 41. Similar statements could be made in Senegal, Ivory Coast or Gabon. 42. Interview with former French Director (Dir1), Paris (France), 19 December 2002. 43. All quotes this paragraph: AIP, Box IPO-B-10, DG IPOM Fonds Marneffe, 1966– 1970, Letter of Dr Brottes to Dr Marneffe, 20 October 1967; Letter of Dr Marneffe to J.R Thybert, 30 August 1967; Letter of Dr Chambon to Dr Marneffe, 1 September 1967. 44. Interview with former French expatriate at the IPC (Coop1), Marseille (France), 12 December 2003. 45. AIP, Box IPO-B-10, DG IPOM Fonds Marneffe, 1966–1970. Letter from Brottes to Chambon, 21 July 1967. 46. On the Gabonization of Uranium mining, see Hecht (2003). On the Africanization of the French trade company CFAO, see Bonin (1987: 356–60, 444–52). 47. AIMTSSA, Boite 101, Raymond Beaudiment, ‘Réponses au questionnaire émanant de la mission d’inspection de la France d’Outre-Mer’ (1954). 48. The rotations of directors between Pasteur Institutes can be retraced in Dedet (2000). 398

The Intimate Rules of the French Coopération

49. Arbovirus for ARthropode-BOrn-VIRUSES. This group of virus includes yellowfever virus. 50. Entomologists were appointed by the Office for Scientific and Technical Research Overseas (Orstom). The Orstom was created after the Second World War as the office for colonial science, and became the main scientific actor of French ‘Coopération’. On the early history of Orstom see Bonneuil and Petitjean (1996: 113–61). 51. Being part of the arbovirus research required mobility through the continent, beginning with a training in Dakar. 52. For a discussion on culture as the new name of race, see Balibar and Wallerstein (1988). 53. Interview with member of the Cameroonian staff of the CPC (Cm 2), Yaoundé (Cameroon), 26 March 2005. 54. ‘Docta’ is a common expression used in salutations of medical workers in Cameroon – derived from broken English. ‘Boss’, ‘Chef ’, ‘Big man’ are other similar salutations, which are both respectful and ludicrous. 55. Interview with member of the Cameroonian staff of the CPC (Cm 2), Yaoundé (Cameroon), 26 March 2005. 56. For example, evaluation of militaries by their hierarchies could include careful notes on one’s ‘problems with his wife’. ACPC, File Mission Militaire Médicale, manuscript notes, 1995–1996. 57. ‘Niggerization’ could be an English translation. ‘Bougnoule’, from the wolof word for ‘black’, became a widely used anti-Arab insult in Algeria and France. 58. Informal discussion with a former French expatriate at the CPC (Coop 2), Paris, 6 November 2003. 59. French directors have been regularly accused – but never frontally – of funds misappropriation, particularly during the crisis (1988–96), by French expatriates as well as Cameroonian officials. 60. See in Geissler and Kelly, this volume, for a discussion on the ethics of intimacy as opposed to formal ethicism.

References Anonymous. 1927. Rapport Annuel du Gouvernement Français sur l’Administration sous mandat des Territoires du Cameroun pour l’année 1926. Paris: Imprimerie Générale Lahire. Anonymous. 1945. La Conférence Africaine Francaise. Brazzaville. 30 janvier-8 février 1944. Paris: Ministère des Colonies. Anonymous. 1953. ‘La crise de la coopération franco-africaine’, Revue française d’études politiques africaines 90: 95–115. Anonymous. 1958. Rapport annuel du gouvernement français à l’assemblée générale des Nations Unies sur l’administration du Cameroun placé sous la tutelle de la France. Année 1957. Paris: Lavauzelle. 399

Evidence, Ethos and Experiment

Balibar, E. and I. Wallerstein. 1988. Race, nation, classe: les identités ambiguës. Paris: La Découverte. Bayart, J.-F. 1984. La politique africaine de François Mitterrand. Paris: Karthala. Bobb, F.S. 1999. Historical Dictionary of Democratic Republic of the Congo (Zaire). Lanham, MD: Scarecrow. Boltanski, L. and E. Chiapello. 1999. Le nouvel esprit du capitalisme. Paris: Gallimard. Boltanski, L., E. Claverie, N. Offenstadt and S. van Damme (eds). 2007. Affaires, scandales et grandes causes: de Socrate à Pinochet. Paris: Stock. Bonin, H. 1987. C.F.A.O., cent ans de compétition. Paris: Economica. Bonneuil, C. and P. Petitjean. 1996. ‘Les chemins de la création de l’Orstom, du Front Populaire à la Libération en passant par Vichy, 1936–1945: recherche scientifique et politique coloniale’, in P. Petitjean (ed.), Les sciences coloniales, figures et institutions 2: 113–61. Carothers, J.C. 1953. The African Mind in Health and Disease. Genève: WHO. Chafer, T. 2005. ‘Chirac and “La Françafrique”: No Longer a Family Affair’, Journal of Modern and Contemporary France 13(1): 7–23. Cumming, G. and R. Langford. 2005. ‘France and Africa in the Age of Globalization: A Fading Affair?’, Journal of Modern and Contemporary France 13(1). De Blic, D. and C. Lemieux. 2005. ‘Le scandale comme épreuve. Éléments de sociologie pragmatique’, Politix 18(71): 9–38. Dedet, J.P. 2000. Les Instituts Pasteur d’Outre-Mer. Cent-vingt ans de microbiologie française dans le monde. Paris: L’Harmattan. Dimier, V. 2004. Le gouvernement des colonies, regards croisés franco-britanniques. Bruxelles: Editions de l’Université de Bruxelles. Domergue Cloarec, D. 1988. ‘Les problèmes de santé à la Conférence de Brazzaville’, in Institut Charles de Gaulle (ed.), Brazzaville, janvier-février 1944: aux sources de la décolonisation. Paris: Plon, pp. 157–69. Dozon, J.-P. 1991. ‘Pasteurisme, médecine militaire et colonisation’, in M. Morange (ed.), L’Institut Pasteur. Contributions à son histoire. Paris: La Découverte, pp. 269–78. ———. 2002. ‘L’Etat franco-africain’, Les Temps Modernes 620–21: 261–88. ———. 2003. Frères et sujets. La France et l’Afrique en perspective. Paris: Flammarion. Eboko, F. 1999. ‘Logiques et contradictions internationales dans le champ du sida au Cameroun’, Autrepart 12: 123–40. Gamet, A. 1960. ‘Rapport sur le fonctionnement technique de l’Institut Pasteur du Cameroun, Année 1959’. ———. 1964. ‘L’Institut Pasteur du Cameroun’, Bulletin de liaison des anciens élèves de l’Institut Pasteur 21: 11–16. Hecht, G. 2003. ‘Rupture Talk in the Nuclear Age: Conjugating Colonial Power in Africa’, Social Studies of Science 32(5–6): 691–727. Joseph, R.A. 1978. Gaullist Africa: Cameroon under Ahmadu Ahidjo. Enugu: Fourth Dimension Publishers. ———. 1977. Radical Nationalism in Cameroun: Social Origins of the U.P.C. Rebellion. Oxford: Clarendon Press. Lachenal, G. 2005. ‘Les réseaux post-coloniaux de l’iniquité: pratiques et mises en scène de la recherche médicale au Cameroun’, Outremers 93: 123–49. ———. 2006. ‘Scramble for Cameroon: Virus atypiques et convoitises scientifiques au 400

The Intimate Rules of the French Coopération

Cameroun, 1985–2002’, in C. Becker and P. Denis (eds), L’épidémie de sida en Afrique subsaharienne. Regards historiens. Paris: Karthala, pp. 309–30. ———. 2010. ‘Le médecin qui voulut être roi. Médecine coloniale et utopie au Cameroun’, Annales HSS 65(1): in press. Latour, B.1988. The Pasteurization of France. Cambridge, MA: Harvard University Press. Martin, G. 1921. L’Existence au Cameroun. Études sociales, Études médicales, Études d’hygiène et de prophylaxie. Paris: Larose. Mbembe, A. 1996. La naissance du maquis dans le Sud Cameroun, 1920–1960. Paris: Karthala. ———. 2000. De la postcolonie. Essai sur l’imagination politique dans l’Afrique contemporaine. Paris: Karthala. Moulin, A.M. 1995. ‘Les instituts pasteur de la méditerranée arabe: une religion scientifique en pays d’Islam’, in E. Longuennesse (ed.), Santé, médecine et société dans le monde arabe. Paris: L’Harmattan, pp. 129–64. ———. 1992. ‘Patriarchal Science: The Network of the Overseas Pasteur Institutes’, in P. Petitjean (ed.), Science and Empires. London, pp. 307–22. Mouralis, B. 1997. République et colonies: entre histoire et mémoire. La république française et l’Afrique. Paris: Présence Africaine. Mundt, R.J. 1995. Historical Dictionary of Côte d’Ivoire (the Ivory Coast). Lanham, MD: Scarecrow. Pelis, K. 1997. ‘Prophet for Profit in French North Africa: Charles Nicolle and the Pasteur Institute of Tunis, 1903–1936’, Bulletin of the History of Medicine 71(4): 583–622. Revel, J. (ed.). 1996. Jeux d’échelles: la micro-analyse à l’expérience. Paris: Seuil-Gallimard. Stoler, A.L. 2002. ‘Colonial Archives and the Arts of Governance’, Archival Science 2(1– 2): 87–109. ———. 2001. ‘Tense and Tender Ties: The Politics of Comparison in North American History and (Post)Colonial Studies’, Journal of American History 88: 829–65. Verschave, F.-X. 2004. De la Françafrique à la Mafiafrique. Bruxelles: Tribord. Waddy, B.B., C.C. Chesterman and J.L. Mcletchie. 1965. ‘[Discussion]’, Transactions of the Royal Society of Tropical Medicine and Hygiene 59: 252–54.

401

Chapter 16

The Mosquito Taken at the Beerhall: Malaria Research and Control on Zambia’s Copperbelt* Lyn Schumaker

Introduction This article describes a historical case of medical research and its application – the study of mosquitoes that made malaria control possible on colonial Zambia’s Copperbelt. Scholars have pointed to the ways that colonial doctors and sanitarians labelled Africans as ‘wild’ vectors or reservoirs of infection dangerous to ‘civilised’ European settlers. While not banishing racial divisions, the Copperbelt malaria control programme made an important step away from such medically justified segregationist visions, replacing them with new metaphors of health and risk, and new geographies of human/mosquito movement, consumption and reproduction. In addition to these European medical and industrial views, I will analyse African miners’ and malaria control workers’ responses, including their resistance to some aspects of malaria control, but also their frequent identification with its modernising, scientific mission. Malaria control on the Copperbelt succeeded because its interventions to counter mosquito breeding shared many elements with new industrial and city planning ideas about worker health and housing. Research, experiment and intervention characterised these endeavours, too, as well as the larger interwar African context. In practical terms, however, the success of malaria control depended upon the ability of medical experts to translate their research on mosquito behaviour and habitats into routine practices of surveillance and environmental intervention – integrating research and application. Sharing of practices, if not always of interpretations, 403

Evidence, Ethos and Experiment

enabled diverse actors to participate in malaria control – including mining engineers familiar with problems of drainage and African antimalaria teams who saw their work as hygienic, improving the environment where they lived. The research culture that emerged during antimalaria work also shared elements found in the emergent African urban industrial culture of the Copperbelt. This is revealed in the interactions, positive and negative, between African malaria control workers and Copperbelt residents. I will use evidence gathered during archival work, participant observation and interviews on the Copperbelt, as well as during visits to rural sites with a high population of retired miners (the Lake Bangweulu area in Zambia’s Luapula Province). Interviews done by a research assistant, Timothy Mgala, also took place in May and October 2006 in and around Luanshya. These interviews with antimalaria workers mainly focused on the late 1950s to 1990s, while a few remembered early mine history, 1928 to the 1940s. The following sections describe the background to the Copperbelt programme, the malaria experts’ approach and their research practices. I then consider the parallels and interactions of malaria control with the mines’ organisation and control of workers. The final section will discuss Zambian miners’ views and the antimalaria team members’ perceptions of their work. Ultimately, the antimalaria programme played a role not only in the mine’s approach to controlling workers and moulding them into a modern (and politically moderate) working class. It also appealed to African miners’ views of themselves as a modern urban people, though they sustained continuing grievances about the ways that malaria control practices underlined their subordinate position in the Copperbelt’s racial and class hierarchy.

The Ross Expedition to Roan Antelope Mine, 1929–1930 Interviewer: Some people think that when [the mining company] cut the reeds and used dynamite to straighten the river, the loud noises caused the snake to go away. BaKacembele: Maybe the snake was just annoyed!1 In 1929, the Roan Antelope Mine in Northern Rhodesia (colonial Zambia) embarked on a campaign against malaria based on advice from the Ross Institute for Tropical Diseases in London. The mine, founded in 1926 by A. 404

The Mosquito Taken at the Beerhall

Chester Beatty’s Selection Trust Company, had been rapidly developed after the discovery of deposits of copper sulphide ore the previous year. Beatty’s personal history of ill-health caused by his early work as a miner in the American West may have predisposed him to consider worker health crucial to industrial success. It was not until a major labour recruitment crisis overtook the mine, however, that he called in the Ross Institute. Roan Mine had suffered a high death rate among European and African workers due to malaria, dysentery, typhoid, pneumonia and blackwater fever (the latter caused by inadequate treatment of malaria infection with quinine). Both African and European miners, the former largely recruited from other parts of the region and the latter largely recruited from South Africa, began to avoid Roan Mine. Africans in particular deserted the mine after the death of a surveyor’s assistant, Joseph Zgambo, in 1928 (Watson 1953: 14). Zgambo drowned in the Luanshya River, which flowed through the mining area, and African stories circulated about a snake spirit that caused people to die at the mine (ibid.: 14–15; Berger 1974: 13). The mine’s African personnel manager C. F. Spearpoint and the Ross Institute’s malaria expert Malcolm Watson believed the snake was an African metaphor for malaria. During the labour crisis Spearpoint visited an elderly African headman who said that people had always avoided the river because of the many illnesses that struck people down ‘like a snake’. ‘In spite of the many interesting stories I have heard from natives about the Snake’, Spearpoint wrote in a chapter for Watson’s later book on the malaria control programme, ‘I feel that the one about sicknesses should be accepted as the true origin of the reptile’ (Watson 1953: 15). Watson was respectful of Spearpoint’s knowledge of ‘the African’ and echoed this metaphorical interpretation by comparing the snake to the ‘Scylla’ of Greek mythology. Watson also added another metaphor – the witches’ cauldron from Macbeth – for the stew of diseases Africans supposedly brought to the mine from their villages (ibid.: 8–11).2 The Ross Institute’s methods of malaria control focussed on the anopheles mosquito – the vector of malaria – using drainage, management of vegetation and channelling to increase the speed of flowing water to discourage breeding. The programme at Roan was the Ross Institute’s first application of these methods in Africa, but it was carried out in the shadow of a failed attempt at Freetown in Sierra Leone (1899–1902) that had been implemented by Ronald Ross, discoverer of the role of mosquitoes in the transmission of malaria (Harrison 1978: 126–27).3 Roan Mine was ideal for mosquitoes, situated in a region crossed by sluggish, marshy rivers. Prior to the mine’s construction the area had been 405

Evidence, Ethos and Experiment

thinly settled and possibly not exceptionally malarious except at the missions and slave-trade centres that existed before colonial rule. Although Watson, going on information from mine managers, believed the area to have been depopulated by malaria and other diseases prior to the mine’s arrival, Africans recall numerous villages along the Luanshya River that were displaced by the mine; they also believe malaria was brought to the region by Europeans.4 Regardless, the mine’s concentration of workers in a small area, its poor housing, its badly designed water system and its construction programme – that ‘pockmarked’ the landscape with holes that filled with water in the rainy season – all contributed to a series of devastating malaria epidemics in the late 1920s. Although the Ross Expedition later recognised the construction work as a contributing factor, they still laid the greatest blame on the African environment, as did the mine’s owner, Chester Beatty (Watson 1953: 3, 26). The mine’s expenditure on malaria control made sense given the scarcity of workers on the Copperbelt and the competition for labour with nearby Belgian Congo mines, known for their good housing and medical provision (Utzinger et al. 2001). Most important to the mine were the highly paid skilled white miners, increasingly reluctant to work in Northern Rhodesia because of its reputation for malaria. Thus even after suffering the worldwide economic depression of the early 1930s, Roan Mine sustained the malaria control programme because of its immediately obvious benefits. The Ross Institute had been founded in 1926 to honour Ronald Ross. It aimed to promote his method of malaria control, as well as provide sanitation measures for other diseases in the tropics. Advising colonial commercial enterprises (and some mission and government health services) became the chief aim of the Institute, especially after Malcolm Watson arrived in 1928, fresh from his successful application of Ross’s method to Klang District, Port Swettenham and the rubber plantations in the Malay States (Watson 1953: 5). He also developed his skills for dealing with colonial business interests through his work for rubber planters as an independent consultant from 1907 to 1928 (Litsios 1996: 47). Already a noted philanthropist, Beatty had been invited to the opening of the Ross Institute’s headquarters on Putney Heath in 1926.5 Thus, he already knew of its work when the London-based managing director for his Copperbelt mines, Arthur Storke, sought advice in July 1929 (Watson 1953: 88; Storke 1933: 88, cited in Buchanan 2005: 65).6 With funding from Beatty’s company, the Ross Institute Expedition left for Northern Rhodesia in September 1929. The famous colonial sanitarian, W.J.R. Simpson, led 406

The Mosquito Taken at the Beerhall

the expedition, accompanied by a doctor trained at King’s College Hospital in London, A.C. Dalzell, who specialised in bacteriology and pathology, and by C.R. Harrison, a former rubber planter and expert on drainage who had met Watson in Malaya.7 After Simpson returned to London, Harrison stayed on at Roan until late February 1930, putting into place the antimalarial drainage scheme (Watson 1953: 26). Dalzell and a mining engineer from Roan continued the drainage and other anti-mosquito measures after Harrison left (ibid.: 35). Watson visited the mine periodically until 1950 to assess the malaria control measures and other sanitation efforts. During some of his visits in the 1930s, he carried out experiments with the new antimalarial drug, atebrin, using African miners’ children as research subjects (ibid.: 29; see below; also see Manton in this volume).8 Drainage works to control the most important vector, Anopheles gambiae, began in late 1929 and had an immediate effect. Deaths due to malaria fell dramatically in the early months of 1930 and continued to be low throughout that year’s rainy season. Subsequently Watson made much of the fact that the mine took his advice not to build a ‘300-bed native hospital’ in 1930, because he believed it would not be needed once the malaria control programme was established (Watson 1953: 30–32). His triumphant observation that the smaller hospital they eventually built had only 39 patients during his 1932 visit, however, does not take into account the drop in the numbers of Africans employed at the mine due to the Depression. Nevertheless, the fall was real and proved the effectiveness of Watson’s use of the Ross method: the incidence of malaria fell from an average of 61.1 per thousand per month during the 1929/30 rainy season to 31.6 in 1930/31 and 17.5 in 1931/32 (Rodgers 1962). Malaria incidence among European miners and their families fell from 105 per thousand in 1929 to 47 per thousand in 1932 (Fisher 1969: 5). The programme’s success at Roan led to its adoption by other Copperbelt mines. Indeed, Roan and the contiguous government township of Luanshya (which had subsidised the work in its area and used the mine’s malaria control team) began to promote the town as a ‘garden city’, to tempt skilled European miners into long-term employment.9 The mining company (and Watson during his later visits) emphasised the high quality of housing for both African and European miners, and the well-watered gardens surrounding the white miners’ family houses featured prominently in Watson’s subsequent book on malaria control (Watson 1953: 68–69). Although malaria control surely contributed to Roan’s later reputation among African miners as a good place to work, those who worked there in 407

Evidence, Ethos and Experiment

the 1930s recall stories of the mine’s early days that blamed the numerous deaths on the Luanshya River snake, not malaria. Most interviewees today believe that the deaths came to a halt when the mine owners agreed to fund a ritual to appease this nature spirit and to show respect to the true African ‘owners of the land’ – the original Lamba-speaking inhabitants. As one interviewee put it, ‘This is Lamba land. Now the chief of this land…got upset saying, “Why don’t you give me something since this is my land where you dig the copper?” It’s like the same chief caused these amafya – black magic’.10 European observers, meanwhile, pointed out that the cutting of reeds that destroyed mosquito habitat may have also destroyed the snake’s supposed habitat, thus calming African ‘superstitions’ (Watson 1953: 15; Berger 1974: 13; Kalusa 1993: 29). When Zgambo disappeared, white supervisors had also tried ‘blasting in the river in case the body had become lodged beneath a rock or trees’ (Watson 1953: 14). Some claimed that this convinced Africans that the snake had been frightened off. Nevertheless, as the opening quotation in this section illustrates, neither the snake nor the Africans were impressed.

Mosquito Control at Roan The Ross method demanded meticulous research and surveillance into mosquito habitats to reduce mosquito numbers to the point where malaria transmission could no longer be sustained. The method was called ‘species sanitation’ by the Dutch malariologist, Nicholas Swellengrebel, who was influenced by Watson’s focus on anophelines when they met in Sumatra in 1913 (Bradley 1994; Litsios 1996: 48). For species sanitation, Watson advised the use of a variety of methods underpinned by good drainage. This factor may have explained his success with planters and mining industrialists, for tropical agriculture relied on drainage for land reclamation while mine safety also required effective drainage, a mine being in many respects like a ‘very large well’ (Buchanan 2005: 67). Given the many different anophelines, each with different habits and habitat, species sanitation was complex. Depending on the species, drainage had to be directed to drying marshes, or filling depressions in the ground, or increasing the speed of a river’s flow to stop larval development, or introducing either shade or sunlight to a river or marsh environment. Cutting reeds and removing vegetation, or ‘scuffling’ grass to ground level, were also required before ponds could be oiled to kill mosquito larvae (Watson 1953: plate 30 opposite p. 66). 408

The Mosquito Taken at the Beerhall

In late 1929 Harrison dealt with the natural and artificial depressions in open sunlight that provided A. gambiae breeding sites, and by April 1930 Watson found none of this species within the control area (Watson 1953: 23). A. funestus took longer, for it bred in marshy areas in light shade – characteristic of much of the Copperbelt and especially of the vicinity of the sluggish Luanshya River (ibid.: 24). For the attack on A. funestus, Harrison focussed on dambos – low, treeless, marshy areas, with shallow pools and light shade provided by reeds and grass. As Watson observed, ‘On the Roan Antelope Mine, the damboes mark the course of the Luanshya River and its several tributaries, sub-dividing the land so that hardly a house does not come within half a mile of a dambo’, while blackwater fever cases were associated with houses built near their edges (ibid.: 24). Watson had selected a half mile as a rule-of-thumb for siting houses beyond the reach of mosquitoes flying out from their breeding grounds, though the reliability of this distance was subject to other conditions, providing opportunities for local research (see below). Watson initially drew a half-mile perimeter around the mine and its housing areas, and around the small town of Luanshya contiguous with the mine, the area that would be subject to drainage and other control measures (ibid.: 38). Then Harrison attacked the dambos with a series of drains – with Watson commenting on the fertility of the soil revealed by drying (ibid.: 25). Harrison and his team also deepened the channels of the Luanshya River and its tributary streams, creating a faster current that prevented breeding of both A. funestus and A. gambiae. Other control measures dealt with vegetation and the depressions created by the mine’s building programme. Some of the mine’s other features, such as its tailings dam, a drinking water reservoir and a ‘drift’ or shallow crossing that obstructed the river’s flow, were replaced by alternative arrangements that drained the stagnant water or created a more rapid flow (ibid.: 37–39). This campaign also included weekly spraying of oil on breeding sites that could not be drained. Meanwhile Dalzell instituted a worker health regime that included effective treatment of malaria infection in white and black miners, first with quinine but in the later 1930s with atebrin and then mepacrine (Fisher 1969: 5–6). White miners’ houses were also screened, though the houses of black miners were not, and the drop in malaria deaths among black miners was consequently less dramatic (Kalusa 1993: 30–31). New laboratory facilities also improved diagnosis of malaria, which was most significant for white miners’ health, while improving black miners’ health largely through combating pneumonia, the biggest killer of Africans on the mine.11 409

Evidence, Ethos and Experiment

Mosquito control work became routinised during the 1930s under the supervision of the Ross Institute, with follow-up studies by Watson, including experiments with antimalarial oils for spraying bodies of water (Watson 1953: 74ff). Prior to the Second World War, routine work focussed on drainage and the development of larvicides, with teams of African mosquito control workers deployed to cut reeds, oil ponds and maintain and extend the drainage works. These medical auxiliaries also participated in the experiments to discover the flying range of the local mosquitoes, as well as locating dambos to be drained or stream beds to be deepened as the mine housing areas expanded (ibid.: 151). The experiments led to the zone of antimalarial measures being increased in the later 1930s to three miles from the nearest habitation (Anonymous 1960: 10). African workers routinely surveyed mosquito species, numbers and locations either using ‘catching stations’ for the capture of adult mosquitoes or dipping for larvae in ponds.12 They also searched both African and European housing areas for unused tins, tyres and other containers that could fill with water and provide habitat for larvae. After the Second World War, this work changed with the arrival of DDT and other insecticides. Although the mine continued with drainage, vegetation control and oiling (using mixtures with DDT from 1946), much of the malaria control workers’ routine involved spraying inside African miners’ houses with residual insecticides at regular intervals, as well as painting insecticides onto the screens of European miners’ houses.13 Under this new regime malaria dropped to the ‘vanishing point’ on the Copperbelt (Fisher 1969: 5).

Welfare and Control at Roan Malaria control was part of a programme of welfare services developed by Roan Antelope Mine from 1930 onwards. Race remained a factor in the distribution of services because the mines saw segregation of housing as essential to attracting white miners and their families. Integration of housing and mine welfare services (including hospitals) only began during the African independence struggle of the late 1950s and early 1960s. Although Roan developed welfare services and housing in order to attract and keep a sufficient number of workers, the urgency with which it pursued these improvements varied with its labour needs. During the1920s development phase until 1932, Roan experienced severe labour shortages which stimulated its investment in malaria control. From the outset the 410

The Mosquito Taken at the Beerhall

mine encouraged African miners to bring their wives to live with them to increase stability and improve health through their domestic services (Parpart 1986: 36). After 1931, however, the Great Depression caused shedding of labour. Roan remained open but with a reduced labour force which suffered from the mine’s cost reduction measures, as well as sharing their accommodation and food rations with any unemployed who did not want to return to the impoverished countryside (Kalusa 1993: 14, 45). With economic recovery in 1935 the need for workers rapidly rose again. An African miners’ strike on the Copperbelt in that year, which led to the killing of six miners at Roan, increased management’s concern for worker control. Despite the ability of Africans to organise the strike across ethnic lines without a formal union, neither the mine nor government considered them ready for unionisation. Instead the mine strengthened an already existing ‘tribal representative system’ that employed so-called tribal elders to communicate the problems of their ethnic compatriots to the white compound manager. Government officials, too, felt that Africans could not (or should not) break their rural ties to become truly urban, industrial workers like Europeans, for whom union organisation was assumed to be appropriate (Cooper 1996: 43–50). The practical anthropology engaged in by colonial administrators who supported this migrant labour system saw Africans as vulnerable to social and individual breakdown under the stresses of a sudden permanent shift to urban life, a view that was bolstered by colonial psychologists and the culture contact school of professional anthropology in Britain (McCulloch 1995; Schumaker 2001; Tilley 2001). Most missionaries agreed about the dangers of the towns to African morals and family life, but in the interwar period some began to argue for the benefits of stabilisation in towns, particularly in a 1933 mission-funded study called Modern Industry and the African (Cooper 1996: 52–53). The anthropologists of the Rhodes-Livingstone Institute based in Northern Rhodesia also emphasised Africans’ ability to adapt and take advantage of the opportunities available in urban industrial life, beginning with Godfrey Wilson’s study of Northern Rhodesia’s Broken Hill lead mine in 1938. After the wave of strikes in the mid to late 1930s, concern for African welfare was bolstered by a shift in Colonial Office thinking towards viewing labour problems in terms of ‘development and welfare’ (see Cooper 1996: 62–67). Nevertheless, vast differences continued between African and European miners’ housing and services, while for grievances the tribal representative system persisted until 1953 when a new African union demanded it be abolished (Harries-Jones 1975: 159). Even when stabilisation 411

Evidence, Ethos and Experiment

of African labour became more widely accepted after the Second World War, white miners maintained their elite status using the threat of strikes. Segregation dragged on, with the dual wage structure lasting until 1969 (five years after independence) and segregation of housing continuing on a de facto basis well into the post-independence period. Medical reasons for segregation had already lost their force in the 1930s, however. The Ross Institute method of malaria control did not require keeping African workers’ wives and children at a distance from European settlements but brought them within the cordon of anti-malaria measures and provided them with curative treatment. Indeed, Watson’s experiments with atebrin on African miners’ children targeted them specifically to undermine the previous segregationist approach that had seen African women and children as a reservoir of malaria infection. Watson was also led by his belief that the mine was an ideal laboratory.14 It provided him with a population of Africans on which to experiment, using as a control the population of nearby Luanshya where Africans were still exposed to conditions that Watson considered similar to those of African villages. This allowed him to compare spleen rates for Luanshya children with those of children born on the mine to demonstrate the effective reduction in malaria infection on the mine (Watson 1953: 107–8; enlargement of the spleen is an indicator of underlying malaria infection). Other features at Roan contributed to its status as an ideal research site – its fine pathology laboratory set up by Dalzell and its trained staff. African miners’ wives soon disabused Watson of his belief that the mine provided willing subjects, however, for despite being bribed with food, ‘they hid when the [mine] police went to collect them’ instead of bringing their children back to complete their treatment with atebrin.15 Historically their response was similar to Zambian responses to other biomedical technologies – to pills, injections and blood-taking in particular – when these targeted children or an adult’s reproductive capacity. Roan Mine developed its programme of worker welfare services with the aim to oversee most aspects of workers’ production, reproduction and consumption. Its approach to company town paternalism derived from the same matrix of interwar environmental, social and medical concerns that stimulated the Ross Institute’s approach to tropical sanitation. Just as changes in tropical medicine led to a greater emphasis on the health of colonial workers, industry began to move away from its previously mechanistic, Taylorist-driven view of workers. In addition, both Roan Mine managers and Malcolm Watson responded to the influence of a new wave of urban health thinking – the Garden City movement. 412

The Mosquito Taken at the Beerhall

Beerhalls and Mosquitoes The British-based Garden City movement idealised the psychological and physical healthfulness of the English village. In the late nineteenth century this movement stimulated the design of new British suburbs and cities, while some Garden City thinkers, such as Patrick Geddes and Charles Compton Reade, attempted to solve the urban problems of the colonies. These included rigid imperial designs that excluded the Africans or Asians on health grounds, designs preferred by older sanitarians such as W.J.R. Simpson (in his work in India before his arrival at the Ross Institute). In contrast, Geddes and Reade advised flexible responses to the colonial city’s cultural, social and physical environment, showing a greater respect for indigenous design (King 1980; Home 1997). Colonial businesses often selectively employed Garden City ideas and rhetoric while rejecting the Garden City planners’ actual proposals, as was the experience of Reade in Northern Rhodesia and South Africa (Mutale 1999; Home 1997 161–65; Home 2000). Nevertheless, because of the later timing of urban development in Northern Rhodesia, Garden City ideas had a greater impact on mine housing at Roan and on the nearby town of Luanshya than they did in South Africa’s already well-developed industrial centres. (For South Africa see Parnell 1993; Parnell and Mabin 1995 and Crush 1993.) Thus, after 1930 a new concept of the use of space and its relationship to health guided the construction of both African and European miners’ separate living areas at Roan. Although the barriers between

Figure 16.1: Garden City – Roan Mine. (no date given) (Plate 31, in Malcolm Watson, African Highway; with permission from John R. Murray).

413

Evidence, Ethos and Experiment

racially defined populations remained, managers paid increasing attention to the careful positioning of medical, leisure, marketing, gardening and administrative activities within African and European housing areas. This reflected an emerging vision of the worker as a consumer of (healthful) activities and services and the worker’s body as a site for the demonstration of industry’s social concern (Sturdy 2000). This goal also demanded a focus on micro-environments within housing areas – and the movement of people between them. This resembled the attention to microenvironments and mosquito movements found in the method of selective anopheline control employed by the Ross Institute. The initial half-mile cordon between mosquito breeding habitats and human habitation was just one of the features of the mine’s new environmental vision of health: the placing of beerhalls, football pitches, sanitation facilities and small neighbourhood maternity and child welfare centres all reflected this attention to micro-environments and issues of access.16 The mine showed a similar concern for placement of facilities in the European area, employing a rigid class and status hierarchy of housing and amenities for whites, to control white miners’ behaviour (Holleman and Biesheuvel 1973). And for both African and European miners, as for mosquitoes, their movements (or flight range) became the object of medical concern. In the 1930s, Watson invoked the idea of the Garden City as an achievable goal made possible by water management and species sanitation. By the 1950s, Copperbelt towns had indeed become relatively healthful places to live for both Africans and Europeans, with luxuriant vegetation, lavish use of water and greatly reduced levels of malaria and water-borne disease. Nevertheless, stark differences persisted in the health of Africans and Europeans at Roan. Most of these had to do with differences in wages and housing, with the majority of African miners and their families suffering greater levels of disease associated with overcrowding and their children prone to malnutrition into the 1960s (Kalusa 1993: 117–19). Roan’s profits and its ability to pay high enough wages to attract skilled white workers depended on reducing outlay on the greater numbers of African miners. The mine’s popular policy of encouraging African miners’ families to live with them in mine housing also aided the mine to shift the costs of reproducing labour onto miners’ wives, who provided domestic services, food from gardening and money from beer brewing (ibid.: 16–17; see also Chauncey, Jr. 1981, Parpart 1986 and Hansen 1984). As a result, the ‘garden city’ at Roan looked very different when one moved from spaces occupied by Europeans to those occupied by Africans. 414

The Mosquito Taken at the Beerhall

European areas contained screened houses on relatively large plots, surrounded by fruit trees, flowers and small vegetable gardens near the servants’ quarters behind each house (Watson 1953: plates 31 and 32, opposite pp. 68– 69). In the African areas much smaller unscreened houses shared communal water, toilet and bathing facilities, while the spaces behind houses and in nearby undeveloped areas were used for maize and vegetable gardens and grazing animals. Higher paid skilled African workers enjoyed larger houses (still unscreened) and better sanitation. Malaria control also took on a different aspect in each area, with the greatest difference in incidence of malaria no doubt due to the lack of screening in African houses. Differences in European versus African garden style would also attract different kinds of attention from both mosquitoes and malaria control workers (see below). The mine used the provision of amenities to encourage miners to socialise within the mining community or nearby Luanshya, controlling their movements. The main beerhall was built by the government in Luanshya’s African township, later joined by a sub-beerhall built by the mine in Roan township nearer African miners’ houses (Epstein 1958: 66 and plate II opposite p.10). Movement was also controlled through work hierarchies and policing. At work, Africans fell under the scrutiny of African ‘boss boys’ and their white superiors, but in their housing areas, the white compound manager dealt with every aspect of African life, assisted by African mine police. Employed directly by the mine, they maintained discipline within the compounds, and in the early days provided the sole route of communication between miners and the compound manager, acting as translators and reporting (or failing to report) complaints or infractions (ibid.: 27). They could

Figuyre 16.2: European House, 1935, (Plate 32 in Malcolm Watson, African Highway, with permission from John R. Murray).

415

Evidence, Ethos and Experiment

also abuse their power, by beating miners, harassing miners’ wives, showing favouritism or taking bribes (ibid.: 27; also see van Onselen 1973 and Henderson 1975). Spearpoint recognised this problem when he instituted the tribal elders system in 1931 to provide a more direct channel of communication between himself and African miners (Epstein 1958: 17). Although a worker’s skills, class level and longevity determined his place in mine work gangs and the quality of housing he received, the tribal elders or tribal representative system ensured that, for purposes of organising around grievances, workers were isolated in controllable ethnic microenvironments, a system that discouraged cross-ethnic organisation such as unionisation. Although these ethnic microenvironments did not map spatially onto compound housing, they existed on the sociopolitical level as distinct cells of ethnic relations and activities. This system worked well for domestic concerns, but African miners did not feel their political and labour grievances were effectively handled and in the 1950s channelled these concerns into political parties and unions (Harries-Jones 1975). In addition to mine police and tribal elders, the mine also used beerhalls as a means of controlling African workers. A beerhall had first been built in Luanshya township as part of a government scheme for funding urban social services for Africans (Epstein 1958: 18). Beerhalls kept African miners within the environs of Luanshya for their leisure activities, congregating under the eyes of tribal elders who might be treated with beer by their ethnic compatriots. They exemplified the mine’s controlled and medicalized environment, serving beer of a standard strength in a setting regularly monitored for the presence of mosquitoes, and recirculating miners’ wages into social services. Various elements conspired to undermine this controlled environment, however. Marginal populations endangered the orderliness of mine life, contributing to ill health and violence. Mosquito habitats, and especially dambos, coincided with areas where dangerous populations could be found and where dangerous activities took place, both social and medical. The local Lamba people comprised one of these troublesome populations on the fringes of Roan. Notorious for refusing to work in the mine, Lambas near the mine had developed a thriving trade based on growing and selling food to the miners and providing beer and access to local women (Siegel 1989). Moreover, their lives on the fringes of the mining community demonstrated a viable alternative to the dangers of underground mining work and a means of survival when the mine laid off workers during economic slumps, allowing unemployed miners to avoid repatriation to their distant home villages. 416

The Mosquito Taken at the Beerhall

This troublesome population of the unemployed was referred to as ‘loafers’ and ‘squatters’, while the Lamba themselves were called ‘wild and lazy’ by colonial administrators and the mining companies (Siegel 1989). Worse, the Lambas’ successful agricultural activities were believed to contribute to the creation of mosquito habitat – due to the water that collected in the leaves of some crops, the small wells constructed to water gardens and the constant pockmarking of the land caused by grazing cattle, whose hoofprints filled with water in the rainy season.xvii Dambos, already pinpointed by Watson as ideal mosquito habitat, were also ideal for raising crops and grazing animals in an area subject to variability of rainfall, a purpose they had served in precolonial times in conjunction with local types of shifting cultivation on higher ground (Scoones 1997: 618; Wilson 1989: 371). Even more disturbing, illicit beer-brewing by Lambas and by miners’ wives (who used the sale of beer to supplement their husbands’ meagre wages) also tempted African miners away from the legitimate beerhalls, to the margins of the compound outside the notice of mine police, in grassy areas on the fringes of the forest at night – thus in places and at times when the drinkers could be exposed to malaria.18

‘I liked Science’: Mosquito Control as Medical Mission Malaria control on Zambia’s Copperbelt was also a supremely scientific mission, with Roan Mine acting as its field laboratory. Because of its fine pathology laboratory – the first one north of the Zambezi – the mine attracted a high class pathologist from Southern Rhodesia for its staff (Rodgers 1962). During the 1930s, other mines learned from the Roan model and rapidly applied it to their own unique settings. This process involved much local research into mosquitoes and mosquito habitats, an experimental activity eagerly taken up by the medical officers at each mine. For example, A.J.P. Coetzee, chief medical officer at the Nkana Mine, reported to Watson on the ‘flight tests’ carried out in 1937 by Botha de Meillon, noting with concern that ‘the mosquito (A. funestus) taken at the beer-hall must have flown at least 2.8 miles and that in one night’ (Watson 1953: 151–52). Coetzee used the results of this experiment to justify extending the controlled area by a considerable amount [to] include the swamps near the mine farm and those towards the Kafue river (ibid.: 152). As a perhaps not 417

Evidence, Ethos and Experiment

unwelcome by-product, the mines’ reclamation of land and extension of their farms also undermined the Lamba people’s local marketing of produce to the mines (their unregulated sale of food to African miners would later come in for criticism as the source of dysentery outbreaks, compared to the controlled supply of fruit and vegetables from the mine farms). The sense of a modernising scientific mission is also found in interviews with former African mosquito control workers. Asked about why he chose to do antimalarial work, Jackson Zgambo said ‘…at work it’s where I liked science a lot. That’s when I realised that science is a good job, it is a very good lesson even at my house. I said, ‘how will I live?’ The surroundings should always be clean. That’s what made me to join the health department’.19 Although Zgambo worked during the post-independence period (1973–1995), his comments resonate with those of older malaria control workers and older residents at Roan, who look back with nostalgia on the 1930s through the 1950s (for a similar sense of scientific mission, but among government workers, see Geissler’s chapter in this volume).20 Indeed, in 2006 some used the mine’s past malaria control regimen (which was maintained after the nationalisation of the mines in the 1970s) to critique the failures of the recently re-privatised mine. As former malaria control worker James Mulenga complained, ‘even me here, my house has not been sprayed by the Luanshya Copper Mines. Now how do you eradicate mosquitoes if other houses have not been sprayed, because spraying for mosquitoes should be every house, but they only sprayed houses for those who worked for LCM’.21 Where gardens were concerned, however, Africans had reason to object to malaria control regulations. Despite Roan’s policy of encouraging miners and their wives to supplement their rations through gardening, maize and cassava gardens came to be seen as dangerous malaria habitat in the late colonial and post-independence periods.22 One malaria control worker recalled, ‘Now when they gave us notices that they don’t want us to grow maize around our houses because they bring mosquitoes, it was me who first slashed my maize to give example to my fellow miners’.23Despite giving a good example, malaria control workers experienced considerable resistance: Here in Mpatamato people used to grow food in dambos, so stagnant water was always there…they used to threaten us that they are going to beat us… The work was so sensitive that sometimes we worked in the presence of [mine] policemen.34 418

The Mosquito Taken at the Beerhall

A kind of hygienic aesthetic also propelled this work – the mines insisted that maize and cassava gardens made the compound look ‘dirty’, according to one retired miner.25 Although Luanshya might be a garden city, African gardens did not conform either medically or aesthetically to the garden-city. With respect to house spraying, too, Africans experienced a more invasive style of malaria control. John Ntenje, an elderly Luanshya resident who worked as a gardener and servant in the 1950s, recalled that the interiors of white miners’ houses were not sprayed – screening made that unnecessary.26 Instead, the anti-malaria team would spray the screens from the outside of the house (Anonymous 1960: 4). Spraying the interiors of African miners’ houses with DDT became routine, however, and for the most part miners and their families welcomed it. As Benard Mulenga recalled of the 1940s and 1950s, many saw their work as ‘good because mosquitoes will reduce and diseases in townships will decrease’.27 Other miners contacted the health department when they noticed mosquitoes, cockroaches or other insects in their homes, requesting a visit by the antimalaria team.28 The general reduction of pests and the ability to sleep without being disturbed by mosquitoes was mentioned by all interviewees as one of the greatest benefits of house spraying. Some, however, resisted spraying, even into the post-independence period, but, as one mosquito control worker remarked: ‘we used to force them until that house is sprayed under what[ever] circumstance’.29 Resistance to house spraying stemmed from a number of concerns. People objected to spraying when a newborn baby or sick person was in the house . Some malaria control workers recalled that they would delay the spraying, though others said they would persuade the family to take the baby or patient outside. The latter, however, violated the indigenous medical practice of secluding newborns and patients to protect them from witchcraft. Others refused house spraying because it required them to take their meagre possessions outside, revealing their poverty to their neighbours.30 Still others, especially miners’ wives, would object because of the odour of the spray, complaining that it might cause spirit possession.31 Malaria control workers also speculated that people kept one room in the house locked because they were hiding charms.32 Retired miners said that during the independence struggle refusal to open a room for spraying could also mean someone was hiding stolen dynamite.33 The mine’s health department viewed resistance as a product of ignorance, as did most malaria control workers. Miners who refused house spraying were threatened with eviction to the squatter settlements or gov419

Evidence, Ethos and Experiment

ernment townships.34 Few would have chosen this medical exile, agreeing with Benard Mulenga: ‘No, even us, these things of spraying houses, we welcomed them. We did not stop them because we knew it was for our benefit’.35 Most miners saw themselves as a modern urban labour elite deserving of the attention to their health and comfort that house spraying and other malaria control measures demonstrated, part of a larger communal project such as that of nation-building described in Geissler’s paper in this volume. As one observed about the slashing of maize in the 1960s: ‘At that time people were not starving [so] that they needed to grow maize – the mines cared for them a lot, and the mines were mapped in such way that people did not grow food in restricted areas’.36 African miners’ general identification with the medically prescribed conditions of urban life and the goals of the malaria control programme ensured its success until the late 1970s, when economic stresses in Zambia, the failure of the global malaria eradication campaign and the advent of economic structural adjustment programmes began to erode the industrial, state and social commitment that had sustained it.

Conclusion The medical vision that informed the Ross Institute’s malaria control programme interacted in a number of crucial ways with the Roan Antelope Mine’s paternalistic vision of worker welfare. This interaction took place at the level of practices as well as visions, and African miners who experienced malaria control – sometimes enforced by mine police – saw similarities between malaria control and the mine’s control of its workers’ lives. Malaria control measures took a more invasive and coercive form when applied to African miners and their families than to European miners, particularly when dealing with the very different gardens that some Africans relied upon for subsistence even within the privileged environs of the ‘garden city’. Mosquito habitat mapped onto the wetland areas in which these gardens thrived, and it also mapped onto the troublesome population of squatters and the unemployed living on the margins of the mine, a population from which the mines also tried to insulate its workers. Thus, the medical segregationist view, which had labelled African women and children as ‘wild’ vectors or reservoirs of disease infecting vulnerable, civilised and ‘domesticated’ Europeans, gave way to a more complex view in which marginal habitats – were populated by marginal people as well as 420

The Mosquito Taken at the Beerhall

mosquitoes. These in turn endangered both African and European workers and their families by threatening the mining town’s ever-expanding but permeable boundaries of sanitation and control. Ostensibly, malaria control and tropical sanitation should have had similar benefits for both races and all classes of miner. In practice, the programme was more effective and less invasive for white miners because of their screened houses. Screens protected them prior to the introduction of DDT and later allowed them to avoid unpleasant house spraying. More importantly, both malaria control and the mine’s control of workers’ behaviour had negative consequences for the poorest and most vulnerable. The dangerous margins of the mining compound were essential to the wellbeing of the poorest mine families, who used them for maize and cassava gardens, for illicit beer selling and for leisure activities that were cheap and free from the surveillance associated with the mine compound. When unemployed, miners and their families also took refuge in the squatter settlements that grew around the mining town, to subsist on these illicit or disapproved activities. Malaria control measures that required destruction of maize and cassava gardens, thus, directly threatened their survival and inspired their resistance. The study of malaria control also has significance for the ethnography of medical research in Africa. Malaria control at Roan started with the Ross Institute’s initial research which later became routinised as part of the monitoring efforts behind mosquito control measures. When we study cases of medical research such as this, we are attempting to understand ‘research cultures’. A research culture is a particular kind of work culture – one associated with laboratories, field research, research institutes or museum collecting (Griesemer 1990; Schumaker 2001). Colonial research cultures often existed as ephemeral moments of medical or scientific exploration, ‘laboratories in the field’, those established during scientific surveys or major colonial and postcolonial interventions such as public health surveys or disease campaigns, urgent one day and gone the next. The research culture at Roan, however, endured well into the post-independence period. This research culture expressed itself in a number of ways, depending on the people engaged in its activities. C.R. Harrison, for example, did field research on mosquito habitats and the hydrological characteristics of the mine’s environment. Malcolm Watson saw the mine as a natural laboratory, with a controlled population and trained staff suitable for conducting experiments with new anti-malarial drugs. Mine medical officers and outside researchers from the region did experiments on mosquito flight ranges. 421

Evidence, Ethos and Experiment

And African mosquito control workers saw their work as a hygienic field science and called themselves ‘searchers’ when they captured mosquitoes or dipped for their larvae.37 (See Geissler, this volume, for the perspective of African research workers on the ‘research culture’ of the Division of Vector Borne Diseases in Kenya.) These activities also fit the larger context of social, scientific and labour experimentation that characterised the interwar period and the 1950s in Africa. Thus, the mining companies viewed Northern Rhodesia as a laboratory for African labour stabilisation (Cooper 1996: 336, 346), while members of the British imperial scientific community carried out the African Survey, which characterised Africa itself as a ‘living laboratory’ (Tilley 2001). My use of a ‘research culture’ approach for examining the work of the Ross Institute at Roan has focussed especially on the ways that different groups of people come to work together on a common project, despite their often radically different understandings of the project’s meaning. The Ross Institute, for example, engaged in malaria control research with a sense of mission – to prove that Ross’s method worked and could provide the springboard for tropical development. The mining industry, on the other hand, appreciated malaria control because of its implications for mining profits. In contrast, African mosquito control workers characterised their work as ‘science’, understanding it as a field activity akin to domestic hygiene (cleaning the surroundings of the home). When malaria control workers acted as ‘searchers’, the mining population also generally accepted this work as in their interest, reducing the nuisance of mosquitoes and other pests and protecting them from disease. Certain practices, such as the slashing of maize, might be resisted, but the mine’s malaria control workers did not attract the ‘vampire rumours’ that government workers of various kinds attracted in this period (White 2000). All of the mine residents interviewed for this study saw the antimalaria team as miners like themselves, while malaria control workers said that their mining company uniforms and the regular notices of spraying ensured that they were known by the people and never accused of being bamunyama, the blood-suckers or thieves of body parts associated at that time with the government hospital. Because of these links to the wider context, the Copperbelt malaria control programme helps us to understand the development of colonial society on the Copperbelt. Mining industrialists and malaria experts worked together towards a vision of African health that shaped the subsequent material, political and social life of the mining company towns. They were able to do this because malaria control and the mines’ organisation of 422

The Mosquito Taken at the Beerhall

labour derived from similar sources and mutually influenced each other during the crucial first phase of the establishment of Zambia’s industrial towns in the 1930s. Malaria experts and mining industrialists possessed an overlapping vision of health and progress through the application of scientific research. Equally important, they shared a range of similar practices for surveillance, monitoring, mapping, experimenting and intervening in the environment, practices that impinged not only on the local mosquito population, but on African miners. Most importantly, however, African miners took up and adapted these practices, shaping Zambia’s industrial urban identity.

Acknowledgements * The Wellcome Trust funded the research on which this article is based – University Award Grant No. 056339, ‘Mining, medicine and the body in Zambia’, 1999–2002. Thanks are due to Timothy Mgala of the University of Zambia Institute for Economic and Social Research for assistance with the interviews.

Notes 1. James Sankwe Kacembele and other retired miners who worked on the Copperbelt in the 1930s, focus-group discussion, 4 August 2004, Chishi Island, Lake Bangweulu. 2. To be fair, Watson used this comparison to show that African superstitions were similar to European ones and based on reasonable fears. The focus of his discussion of myths, moreover, was on those of the malaria experts who attacked Ross’s method. 3. The ‘Freetown tragedy’ was Malcolm Watson’s phrase for this failure, because Ross’s rivals, S.R. Christophers and John W.W. Stephens, used it (together with similar failures in Lagos, Nigeria, and Mian Mir in India, during the years 1902 to 1904) to discredit Ross’s method (Watson 1953: 26; Bynum 1994; Litsios 1996: 40–41). They subsequently reaffirmed the standard practice of racial segregation of urban housing on the grounds that it alone could protect Europeans from African women and children, whom they saw as the chief vectors of malaria (Litsios 1996: 42). Controversy over segregation to protect Europeans from malaria continued throughout the interwar period, coupled with debates about African immunity versus use of treatment against malaria, and the question of whether combating malaria among Africans had to wait for a general improvement in standards of living brought about by development. For a summary of these debates see Litsios (1996).

423

Evidence, Ethos and Experiment

4. Focus-group discussion with retired miners, 4 August 2004, Chishi Island; interview with Jackson Kabwe, 10 May 2006, Luanshya; for early prospectors’ views of the area as healthy see Gelfand (1961: 178). 5. ‘The Opening of the Ross Institute & Hospital for Tropical Diseases’, 15 July 1926: 17, Ross Institute Annual Report and Accounts 1926–1932, London School of Hygiene and Tropical Medicine (LSHTM) Archives, London. 6. Also see a different account in which a Roan Mine doctor initiated the request for the Ross Institute’s involvement (Rodgers 1962). 7. See Watson (1953: 6, 35), and the Ross Institute Annual Report for 1930, LSHTM. 8. Ross Institute Annual Report (1932: 38), LSHTM; Schulemann – Watson Correspondence, 1931–1932, BAL, 367–270, ‘England-Korrespondenz’, Bayer Archives, Leverkusen, Germany. Bayer supplied Watson with atebrin. He also visited the Bayer research laboratories and hosted visits by its researchers to the Ross Institute. 9. For government subsidies of malaria control work see Ross Institute Organising Secretary’s Report for March 1933, LSHTM. The reference to garden cities comes from the Garden City movement, an architectural and city-planning movement in Great Britain in the later nineteenth century which lasted well into the twentieth century. 10. Jackson Zgambo, Buntungwa, Luanshya, 9 May 2006. Amafya means ‘troubles’ or ‘problems’ in iciBemba, the language most widely used on the Copperbelt. 11. Pneumonia was the cause of fifty per cent of African deaths in 1930, and Charles Fisher, Roan’s chief medical officer from 1936, implied that for African workers pneumonia epidemics in the early days were more important to the bad reputation of the mine than malaria (1969: 5). 12. For the use of African houses and other sites as catching stations see the Second Annual Report of the Medical Department of Roan Antelope Mines for 1931, ‘Mosquito Catches’, ZCCM. 13. ‘Notes on Malaria and its Control for Planters and Miners’ by G. MacDonald, ‘Ross Institute – Temp’, Box 23–26, LSHTM; Horizon, February 1960: 10). 14. Ross Institute Annual Report (1931: 17), LSHTM. 15. Ross Institute Industrial Anti-Malarial Advisory Committee, 1 February 1933: 18, LSHTM. 16. Some of these facilities were subsidised by both the mine and government and shared between mine and government housing locations, including the welfare centres established in Luanshya in the 1930s (Gelfand 1960: 22). Subsidies for these welfare centres derived from beerhall profits to which African miners contributed (Kalusa 1993: 53). 17. Ronald Ngosa, 9 May 2006, Mpatamato, Luanshya; Jackson Zgambo, 9 May 2006, Buntungwa, Luanshya. 18. Illicit beer brewing and the shebeens (unregulated places for sale of beer), like the official beerhalls, also functioned as foci for the growth of new forms of sociality among urban Africans, as well as for the development of new types of African en424

The Mosquito Taken at the Beerhall

19. 20.

21. 22. 23. 24. 25.

26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.

trepreneurship and domestic relations (see Parpart 1986 and Chauncey, Jr. 1981 for women’s use of beer brewing to increase their independence on the Copperbelt). Interviewed at Buntungwa, Luanshya, 9 May 2006. Interviews with John Ntenje (b. 1936), Jackson Kabwe (b. 1941) and Benard Mulenga (b. 1942), Luanshya, 9 May 2006; and interviews with Harrison Mwelwa Kosamu (b. 1922), Kosamu Chakumana Village, Mpongwe; Norice Samuel Maila (b. 1924), Katanga Village, Luanshya Rural; and Goodwell Steven Nondo (b. 1934), Mikomfwa, Luanshya, 14 October 2006. Mpatamato, Luanshya, 9 May 2006. John Ntenje, Roan, 9 May 2006; see also Parpart 1986: 50, on miners relying on food from their wives’ gardens to maintain a lengthy strike. James Mulenga, Mpatamato, Luanshya, 9 May 2006; see also Ronald Ngosa, Mpatamato, Luanshya, 9 May 2006. James Mulenga, 9 May 2006, Mpatamato, Luanshya. John Ntenje, Roan, Luanshya, 9 May 2006. This view is similar to the reaction of Zimbabwean agricultural officers, that trees left standing in farmers’ fields look ‘messy’; see Wilson (1989: 374). Roan, Luanshya, 9 May 2006. Ibid. Jackson Zgambo, Buntungwa, Luanshya, 9 May 2006. Ibid. Ronald Ngosa, Mpatamato, Luanshya, 9 May 2006. Jackson Zgambo, 9 May 2006, Buntungwa, Luanshya; Ronald Ngosa, Mpatamato, Luanshya, 9 May 2006. James Mulenga, Mpatamato, Luanshya, 9 May 2006. Anonymous comments made by other miners during John Ntenje’s interview, Roan, Luanshya, 9 May 2006. Jackson Zgambo, Buntungwa, Luanshya, 9 May 2006. Roan, Luanshya, 9 May 2006. Goodwell Steven Nondo, Mikomfwa, Luanshya, 14 October 2006. Jackson Zgambo, Buntungwa, Luanshya, 9 May 2006.

References Anonymous. 1960. ‘Malaria Control on the Copperbelt’, Horizon: The Magazine of the Rhodesian Selection Trust Group of Companies 2 (2) (February): 4–10. Berger, E.L. 1974. Labour, Race, and Colonial Rule: The Copperbelt from 1924 to Independence. Oxford: Clarendon Press. Bradley, D.J. 1994. ‘Watson, Swellengrebel and Species Sanitation: Environmental and Ecological Aspects’, Parassitologia 36(1–2): 137–48. Buchanan, D. 2005. ‘Why Did the Investors in the Roan Antelope Mine in Northern Rhodesia Decide in 1929 to Make an Investment in Public Health?’, unpublished 425

Evidence, Ethos and Experiment

MSc. dissertation, University of Manchester. Bynum, W.F. 1994. ‘An Experiment that Failed: Malaria Control at Mian Mir’, Parassitologia 36(1–2): 107–20. Chauncey, Jr., G. 1981. ‘The Locus of Reproduction: Women’s Labour in the Zambian Copperbelt, 1927–1953’, Journal of Southern African Studies 7(2). Cooper, F. 1996. Decolonization and African Society: The Labour Question in French and British Africa. Cambridge: Cambridge University Press. Crush, J. 1993. ‘Scripting the Compound: Power and Space in the South African Mining Industry’, Society and Space 12: 301–24. Epstein, A.L. 1958. Politics in an Urban African Community. Manchester: Manchester University Press. Fisher, A.C. 1969. ‘40 Years of Medicine on the Copperbelt’, Horizon: The Magazine of the Rhodesian Selection Trust Group of Companies 11 (12) (December): 4–9. Gelfand, M. 1960. Proud Record: An Account of the Health Services Provided for Africans in the Federation of Rhodesia and Nyasaland. Salisbury, Southern Rhodesia: Government Printer. Gelfand, M. 1961. Northern Rhodesia in the Days of the Charter: A Medical and Social Study, 1878–1924. Oxford: Blackwell. Griesemer, J.R. 1990. ‘Modeling in the Museum: On the Role of Remnant Models in the Work of Joseph Grinnell’, Biology and Philosophy 5: 3–36. Hansen, K.T. 1984. ‘Negotiating Sex and Gender in Urban Zambia’, Journal of Southern African Studies 10(2): 219–38. Harries-Jones, P. 1975. Freedom and Labour: Mobilization and Political Control on the Zambian Copperbelt. Oxford: Basil Blackwell. Harrison, G. 1978. Mosquitoes, Malaria and Man: A History of the Hostilities since 1880. New York: E.P. Dutton. Henderson, I. 1975. ‘Early African Leadership: The Copperbelt Disturbances of 1935 and 1940’, Journal of Southern African Studies 2(1): 83–97. Holleman, J.F. and S. Biesheuvel. 1973. White Mine Workers in Northern Rhodesia, 1959– 60. Leiden: Afrika-Studiecentrum. Home, R. 1997. Of Planting and Planning: The Making of British Colonial Cities. London: E & FN Spon. ———. 2000. ‘From Barrack Compounds to the Single-family House: Planning Worker Housing in Colonial Natal and Northern Rhodesia’, Planning Perspectives 15: 327– 47. Kalusa, W.T. 1993. ‘Aspects of African Health in the Mining Industry in Colonial Zambia: A Case Study of Roan Antelope Mine, 1920–1964’, M.A. dissertation. University of Zambia. King, A.D. 1980. ‘Exporting Planning: The Colonial and Neo-colonial Experience’, in G.E. Cherry (ed.), Shaping an Urban World (Studies in History, Planning and the Environment). London: Mansell Publishing. Litsios, S. 1996. The Tomorrow of Malaria. Wellington, New Zealand Pacific Press. McCulloch, J. 1995. Colonial Psychiatry and ‘the African Mind’. Cambridge: Cambridge University Press. Mutale, E. 1999. ‘Urban Management in Kitwe, Zambia’, Ph.D. dissertation. University of East London. 426

The Mosquito Taken at the Beerhall

Parnell, S. 1993. ‘Creating Racial Privilege: The Origins of South African Public Health and Town Planning Legislation’, Journal of Southern African Studies 19(3): 471–88. Parnell, S. and A. Mabin. 1995. ‘Rethinking Urban South Africa’, Journal of Southern African Studies 21(1): 39–61. Parpart, J.L. 1986. ‘The Household and the Mine Shaft: Gender and Class Struggles on the Zambian Copperbelt, 1926–64’, Journal of Southern African Studies 13(1): 36– 56. Rodgers, L.M. 1962. ‘The Development of Medical and Health Services at the Roan Antelope’, NCCM/HO 11.8.5F, Part One, Zambia Consolidated Copper Mines (ZCCM) Archives, Ndola. Ross Institute Annual Report (1932: 38), held by the Archives of the London School of Hygiene and Tropical Medicine. Schumaker, L. 2001. Africanizing Anthropology: Fieldwork, Networks and the Making of Cultural Knowledge in Central Africa. Durham: Duke University Press. Scoones, I. 1997. ‘Landscapes, Fields and Soils: Understanding the History of Soil Fertility Management in Southern Zimbabwe’, Journal of Southern African Studies 23(4): 615–34. Siegel, B. 1989. ‘The “Wild” and “Lazy” Lamba: Ethnic Stereotypes on the Central African Copperbelt’, in L. Vail (ed.), The Creation of Tribalism in Southern Africa. London: James Currey. Sturdy, S. 2000. ‘The Industrial Body’, in R. Cooter and J. Pickstone (eds), Medicine in the Twentieth Century. Amsterdam: Harwood Press, 217–342. Tilley, H. 2001. ‘Africa as a “Living Laboratory”: The African Research Survey and the British Colonial Empire: Consolidating Environmental, Medical, and Anthropological Debates, 1920–1940’, D.Phil. dissertation. University of Oxford. Utzinger, J., Y. Tozan, F. Doumani and B.H. Singer. 2001. ‘The Economic Payoffs of Integrated Malaria Control in the Zambian Copperbelt between 1930 and 1950’, Tropical Medicine and International Health 7(8): 657–77. Van Onselen, C. 1973. ‘The Role of Collaborators in the Rhodesian Mining Industry 1900–1935’, African Affairs 72(289): 401–18. Watson, M. 1953. African Highway: The Battle for Health in Central Africa. London: John Murray. White, L. 2000. Speaking with Vampires: Rumor and History in Colonial Africa. Berkeley: University of California Press. Wilson, K.B. 1989. ‘Trees in Fields in Southern Zimbabwe’, Journal of Southern African Studies 15(2): 369–83.

427

Chapter 17

Trial Communities: HIV and Therapeutic Citizenship in West Africa Vinh-Kim Nguyen

Today, from many African vantage points it is possible to glean the contours of a truly postmodern future littered with failed states, humanitarian interzones and acephalous conflict. Nonetheless, among these ruins a troubled social order can be found camped around the gleaming engines of extraction that plug Africa’s oil and minerals into the global economy – and the growing humanitarian and NGO biomedical infrastructure that embraces Africa in a global moral economy (Ferguson 2006). Africa is still the global hub of modern medical humanitarianism, from Médecins Sans Frontières and the Biafran war to ‘Brand Aid’ for HIV Today (Richey and Stefane 2006). As this volume shows, it has also been a place where medical research has been conducted from colonial times onward. With the advent of the HIV epidemic this research has become increasingly humanitarian in character, preoccupied with alleviating the suffering caused by the epidemic and with achieving greater involvement of people living with HIV and HIV by subscribing to a broad human rights and empowerment agenda. In this chapter, I will examine how medical research, spurred by the HIV epidemic, catalysed the formation of community-based organizations and self-help groups focussed on gaining access to resources (including treatment) for people living with HIV. Over time, these groups have become veritable therapeutic communities, embedded in a moral economy which allows members to leverage access to medicines, food, social services – and new social ties. This too has precedents in colonial history – Eric Silla, for example, has shown how French leprosy research conducted in a leprosarium near Bamako led to a network of sociality that still binds to429

Evidence, Ethos and Experiment

gether a postcolonial community that lives in the grounds of the old leprosarium, now engulfed by the sprawling city (Silla 1998). I will then conclude by exploring how these therapeutic communities lay the groundwork for future medical research for HIV pharmaceuticals, vaccines and microbicides.

The Response to the Epidemic: A Brief History While the global history of HIV has yet to be written, it is already possible to distinguish three generations in the response to the epidemic. Each generation disseminated a different set of technologies that layered onto the previous set – to varying biological, social and institutional effect. The pre1994 classical public health strategy was sharpened by a focus on rights (1994–2000) which paved the way, after 2000, for an aggressive interventionist approach to testing and treatment, in some ways resurrecting a more orthodox public health approach – albeit with important differences. The ideal-type of the first generation of the response was incarnated in the World Health Organization’s model medium-term plans, used by national Ministries of Health as blueprints for designing programmes to respond to the epidemic. A mix of classical public health approaches to controlling infectious diseases were reflected in these plans, which stressed epidemiological surveillance (with a particular focus on high-risk groups and rapid ‘knowledges, attitudes, behaviours and practices’ or KABP studies) and achieving behaviour change. These programmes deployed a range of social technologies: voluntary counselling and testing (particularly for pregnant women and high-risk groups), social marketing of condoms and ‘IEC’ (Information – Education – Communication) campaigns. These aimed to raise awareness of HIV and how to prevent it, as well as to gather data that would paint a more finely grained epidemiological picture. Considerable efforts were made to broaden the scope of this response beyond narrowly defined ‘affected and vulnerable’ populations by reframing HIV as a problem of development, rather than as merely a public health or biomedical issue. The broadened response was to be multi-sectoral, concerned with addressing the social determinants of vulnerability by integrating HIV programmes within all branches of government and development assistance. This first generation of response was characterized by a managerial and epidemiological ethos concerned with mapping the epidemic and deploying standardized policy responses. Remarkably, in many parts of the 430

Trial Communities

world where the epidemic was already ravaging populations, this translated into little or no actual services to meet actual needs – HIV testing rates remained low and the epidemic coursed, largely invisible and silent, even in the hospital wards that were filled with patients suffering from opportunistic infections (Mann and Tarantola 1996). Even as this public health framework began to take hold across the globe, a coalition of HIV and social activists (some of whom had become powerful HIV bureaucrats) came together to contest what they feared was an ‘exclusion and control’ approach, arguing that instead ‘rights and empowerment’ should be stressed (Seidel 1993). This crystallized as a policy of ‘greater involvement of people with HIV and AIDS’ or GIPA in 1994. This policy aimed to ‘give a face’ to the epidemic through active empowerment of people with HIV and HIV by involving them at the front lines of the response as prevention workers and lay counsellors. GIPA helped shift the response to an emphasis on individual empowerment. People with HIV were trained to ‘come out’ and affirm their rights in workshops that used what have been called ‘confessional technologies’ by myself and others (see Hunt 1997) to build the skills to testify about being HIV in public (for a fuller account see Nguyen 2005a). These technologies had originated in small and large awareness-training groups pioneered in the civil rights, feminist and counter-cultural movements of the American 1960s (Lee 2002). They included open-ended questions, role-plays, trust-building exercises and many others. In the late 1970s they migrated to a host of other social change movements including gay liberation, playing a central part in the gay community’s response to the epidemic in America in the 1980s particularly through the ‘buddy’ system that matched HIV patients with caregivers who would support them through their illness. As I have argued elsewhere, they were powerful technologies for training, but also for producing new socialities and subjectivities (Nguyen 2005b). This second generation of the response was marked by an ideology of rights and empowerment and the practical deployment of a range of individualizing technologies that, in effect, addressed a trans-national public of people living with HIV. As a result, a small vanguard of HIV-positive activists was trained across the world, which helped to shift the paradigm of the response by 2000. Access to effective treatment for the infection, which had been available in the North since 1995 but had been considered too expensive, too complex and not cost-effective in the South, became a political issue (most notably in the context of the 2000 US Presidential campaign). Health was force431

Evidence, Ethos and Experiment

fully argued to be a human right, and this came to be viewed through the lens of access to treatment – and the complex welter of institutional, policy and even geopolitical issues that hindered that access. The stage was set for casting the epidemic as a humanitarian medical emergency. It was not long before a wave of ‘emergency’ programmes emerged to treat HIV – most notably, WHO’s ‘3x5’ initiative inspired by Paul Farmer’s group at Harvard and President Bush’s Emergency Plan for HIV Relief (PEPFAR). It is worth stressing that it was advances in biomedical research (that is, the discovery of life-saving highly active antiretroviral therapy in 1994) that were able to reframe a public health crisis deeply rooted in decades of neglect of global public health as a humanitarian emergency. Biomedical technologies – antiretrovirals, but also the technologies to diagnose HIV and monitor the medications – have become the driving force of the response. This is now increasingly concerned with scaling up access to treatment, and in the future will be called to attend to the complex logistical and policy challenges of managing millions of people in poor countries who depend on continual treatment for survival.

Catalysis Distinguishing three generations in the response provides a shorthand to historicizing the epidemic, although of course ‘in the field’ the reality was, and continues to be, more complex. I started working with community-based organizations in Burkina Faso and Côte-d’Ivoire in 1994. In those early years, HIV testing programmes were relatively few and far between – in the Ivoirian metropolis of Abidjan, for example, there was only one site that provided free and anonymous testing in 1995. In Burkina Faso, there was little or no access to HIV screening – including for screening blood transfusions. As a result, only a handful of people knew their HIV status. The few who were tested were often not told of their diagnosis. In the main hospital in Ouagadougou, Burkina Faso, for example, HIV testing was done sporadically for in-patients in 1994–95, but the majority were not told; the few who were informed of their diagnosis had an A scrawled in red on their charts. In the mid-1990s, two large clinical trials of AZT for the prevention of mother-to-child transmission of HIV (commonly abbreviated as ‘PMTCT’) were conducted in West Africa; one sponsored by the French national HIV research agency (ANRS) and the other by the rival American Centers for Disease Control (CDC). The French study tested 14,385 women in BoboDioulasso (Burkina Faso) and Abidjan while the American study tested 432

Trial Communities

12,668 women in Abidjan. All the women, of course, received state-of-theart pre-test counselling for HIV. Of the total of over 27,000 women tested, 3,424 were found to be HIV-positive. However, in the American study 618 HIV-positive women never returned for their results and the post-test counselling; in the French study the figure was 648 – almost a third. Researchers told me that the rate of drop-out was lower in HIV-negative women, suggesting that women suspected their diagnosis and decided not to return for results. Of the HIV-positive women who returned for their results, only 711 were included in the actual trials, the remaining 2,713 women having either not returned for follow-up as discussed, been excluded for various reasons or simply not consented (Dabis et al. 1999; Wiktor et al. 1999). The trials had an enormous impact in shaping the early response to the epidemic – simply through the sheer number tested. Based on my own fieldwork with HIV-prevention and care programmes in both countries in the mid-1990s, the majority of people who knew their HIV status had found out by having been recruited by the trials. Only a minority were eventually enrolled in the trials, which had to screen a very large number to have a significant pool from which to enrol a necessarily smaller number of selected, eligible study subjects. There were numerous reasons which disqualified women – not arriving early enough in pregnancy to receive the AZT, or suffering from anaemia (which could be dangerously worsened if they received AZT) were among the most common. Women who had tested positive but not been eligible to enrol in the trials complained bitterly that they had been ‘discarded’. They resented that they did not have access to what they perceived to be a panoply of services for the women who had been included in the trials and therefore had been randomized to receive either placebo or AZT. Indeed, women who were included in the trial did receive medical care and social services that were not available to others. Many women found their way into community groups after having been tested in search of material and social support; some even set up organizations for their fellow would-be trial subjects. They made up the bulk of the membership of early groups of people living with HIV in both countries. At the same time, international donors were stressing ‘empowerment’ of people living with HIV and HIV as a human-rights based response to the epidemic. The first step of empowerment was disclosure: being able to be ‘out’ and talk about being HIV positive. This was a difficult step for many of the women, for whom disclosure could lead to blame, violence and even being disowned by husband and family. Women – most of whom had found out about being HIV positive through the clinical trials – were swept 433

Evidence, Ethos and Experiment

up in the wave of empowerment-training workshops that coursed across Africa after 1995. The workshops trained Africans to give the testimonials about being HIV positive (Nguyen 2005b). They did so using experiential and active listening techniques first experimented by social psychologists after the Second World War that then disseminated into management schools and self-awareness movements in America. These confessional technologies were remarkable because they were so effective – participants reported to me putting them into practice immediately after the workshops finished: mirroring the posture of their interlocutors, asking open-ended questions, practising active listening. These social technologies were the foundation of what were to become therapeutic communities.

A Social Tissue of Disclosure When they returned to their community associations, workshop participants set up discussion groups to encourage other people with HIV to talk about their affliction. I attended these groups regularly from when they first started in 1998. In Côte-d’Ivoire and Burkina Faso, however, in the first few years, the groups did not seem particularly successful. Very few men attended. Attempts to get participants talking were at first met by silence and, eventually, halting attempts at self-expression that more often than not were a litany of complaints that seemed devoid of affective content. But over time, the dynamic in the groups began to change. After awkward, embarrassed beginnings a more convivial atmosphere began to prevail. Previously laconic participants became voluble and animated. Gradually, the charismatic side of some of the participants emerged. Their narratives were frequently couched in an evangelical idiom, describing the process of being diagnosed with HIV as the beginning of a conversion-like process, the first step on a road that led to greater enlightenment and the adoption of a more responsible, moral life (Figure 26). These declarations were inevitably followed by exhortations to the audience to get tested. These evangelical idioms disturbed many of the Western aid workers employed by the agencies that funded these efforts. These workers had come to international HIV work through HIV activism in the North, and many from the gay community. Needless to say, the moralizing tone of the message conflicted with aid workers’ personal values that stressed empowerment, sexual openness and tolerance. Initially, I suspected that these evangelical forms were either historical residues of the colonial period or a reflection of the growing popularity of 434

Trial Communities

Figure 17.1: Slogans at a ‘Community Meal’ of Positive Women in Ouagadougou (author’s photograph).

Pentecostal Churches. I spent considerable time interviewing volunteers and adepts in religious organizations. These organizations were responding directly to the epidemic in various ways: volunteering at the hospital, holding prayer services, providing emotional and material support; others were not so directly involved with the epidemic but concerned with affliction more broadly understood. Some spoke of volunteering to work with the ill and dying as a religious experience, while others expressed the desire to offer solace to those afflicted. Common to all was a powerful sense that what was at stake was the way in which HIV reframed moral dilemmas and required transformation in response. In other words, faced with the prognosis of certain disease and death – and little in the way of resources to alter that prognosis – what was at stake was how the good life was to be defined and how it might be attained. In this way, the moral dilemmas surrounding the HIV diagnosis (framed by popular understandings of HIV as a disease of sexual immorality) were transformed into ethical predicaments. Sufferers had little to work with – in fact, they had only themselves. Ethics was about attaining the goal of the good life (no matter how damaged the prospects might seem) through self435

Evidence, Ethos and Experiment

transformation; ethics was about the care of the self and the relationship to others. These circulating, global discourses about HIV were taken up in local ethical projects, in local economies of self-fashioning. What these ethical predicaments share, perhaps, with evangelical movements is the way in which they make available instruments and strategies that individuals may take up to equip themselves to better navigate the moral economies in which they are enmeshed. In the era of globalization and NGO discourses about AIDS, these moral economies are starkly juxtaposed, making this task particularly acute for people living with HIV and HIV in settings where the only source of solidarity is family. The market for testimonials, anchored in Western notions of self-help through confession, offered the best opportunity to gain resources that could help feed family and maintain one’s position in the kinship networks that, in the absence of a viable State, are the only available forms of social solidarity. Until recently, in the absence of any real political or economic engagement to address the structural issues driving the epidemic, it was possible to view these testimonials as only so much postcolonial theatre. But confessional technologies not only produced the desired testimonials, they were also instrumental in turning people living with HIV, many of whom first learned of their diagnosis through a clinical trial, into activists. More significantly, perhaps, they created a tissue of social relations organized around shared disclosure.

Triage As the supply of donated drugs increased from 1998, self-help groups of people living with HIV were increasingly faced with the gut-wrenching prospect of deciding who should get the drugs. No matter how many donations were received, demand always outstripped supply. Any organization offering even the most minimal services was quickly overrun. This was certainly the case of the HIV/AIDS groups where fear of stigma did not appear to be much of a barrier to a steadily increasing stream of would-be beneficiaries. Many were ill, or suspected themselves to be HIV positive because they had lost a spouse. The concept of triage was developed on the battlefield, as a way to most rationally use scarce treatment resources: those most likely to live are prioritized to receive care, while those whose prognosis is poor are left to die. Many organizations of people with HIV, were faced with the same kind of situation. They made the difficult decision of who should benefit from the 436

Trial Communities

limited source of drugs by adopting a form of triage. Those whose continued health was most likely to translate into increased resources for the group were the first beneficiaries. Those who were most charismatic, most able to deliver effective testimonials would be the best advocates for getting more drug donations. The discussion and self-help groups, which were ideally suited to cultivating and demonstrating testimonial skills, were sites where charisma and other social arts could emerge – and where they could be translated into access to medications. It was a subtle, implicit process, but it highlighted how the discussion groups were veritable social laboratories, where confessional technologies could be used to experiment new forms of disclosure and subtly constitute triage. Sometimes the decision as to who should get the drugs was more directly pragmatic. Prioritizing access to drugs for beneficiaries who could be counted upon to facilitate the group’s work in virtue of their professional position, for instance as a customs officer, was an example of how groups used drugs to increase access to medications. These strategic forms of social triage contrasted with the rhetoric that framed international donors’ aid, which was meant to target the most vulnerable members of society – not the most valuable. Over time, those who were gifted communicators also became those with the most direct experience with the drugs. Echoing the experience of HIV activism in the North, these patients were often the most knowledgeable people around when it came to their treatments. Now, as drug programmes expand, they are ideal candidates for assuming leadership roles in treatment literacy and expanded access programmes. (Some, however, had already left as part of a therapeutic migration that brought them to Europe – and antiretrovirals. In some countries, those living with HIV benefit from a ‘humanitarian exception’ from deportation by virtue of their biomedical condition. This therapeutic migration was not without consequence for domestic and foreign policy in Europe. See Fassin 2001, 2005; Ticktin 2006.) These stories are representative of many whose early experiences with the AIDS industry, its discourses of empowerment, and its confessional technologies trained them to become effective advocates. While South Africa’s Treatment Action Campaign (several media profiles of its founder, Zackie Achmat, have made him the best known African HIV activist in the North (see Power 2003) is an example of an organization that explicitly harnessed this process to identify and train future activists from the ranks of patients, throughout Africa the inchoate strategies of other groups and 437

Evidence, Ethos and Experiment

activists amounted to the same (Robins 2005, 2006). While some individuals, particularly well versed in the social arts or endowed with charisma, were naturally well suited to draw on the repertoire of confessional techniques to mobilize others, even those less skilled were able to benefit from the drills, exercises and training that proliferating workshops disseminated throughout the continent. But this ‘production of activists’ was not exclusively driven by social forces. The issue of access to ARV treatment in effect simultaneously provided both a therapeutic telos and the means to sustaining the therapeutic quest; it grafted seamlessly onto the ethical project that had been fashioned by sufferers through their involvement with HIV groups. Bluntly put, skill at telling the right stories got activists drugs, and kept them alive. For this vanguard of people living with HIV the confessional technologies drawn upon on in this therapeutic quest were conjugated with the growing availability of antiretrovirals to fashion, biologically and socially, therapeutic citizens. The activist slogan of ‘drugs into bodies’ tellingly illustrates how social forms are incorporated into individual biology.

From Triage to Community The early years of triage were a conflicted time. While the discourse of empowerment imagined a community of solidarity and self-help, differential access to medications belied the message. In Abidjan, for instance, not all of the women who had met through the trials and gone on to form self-help groups were able to subsequently get access to the treatments. As time wore on, the women in the groups on treatment grew healthy and plump, while those who did not became thinner. In one group, three women died in one year. Not surprisingly, much rumour and suspicion surrounded the question of access to the medicines, at times poisoning group life. In the context of generalized poverty, material issues also generated tensions. GIPA policies meant that donors funded organizations of people living with HIV to provide peer-counselling services for prevention and treatment programmes. Micro-credit programmes were also experimented by the agencies, while some resorted to outright hand-outs of food. Taken together, these faltering attempts at something between empowerment and charity generated competition between group members. Eventually, it led to a paradoxical phenomenon: rather than promoting solidarity between group members, these attempts to help created a dynamic of fission as 438

Trial Communities

members split off to form their own groups. They reasoned that having one’s own group would make it easier to compete for resources from the outside rather than from within the group. As resources became more plentiful, and donors adjusted their policies to encourage teamwork amongst organizations, competition receded, and the tissue of social relations amongst those who had survived the early years thickened. The shared experience of triage bound together the first generation of activists. Even between rivals, attendance at workshops and conferences fostered a hesitant and ambivalent comradeship that nonetheless, over time, smoothed over previous conflict. The divisive experience of triage was burnished into an ethic of care of the self – and a sense of obligation to others, who should no longer be allowed to die for lack of treatment. This bivalent ethic of transformation was reflected in the growth of treatment activism, as members of therapeutic communities, trained in the arts of public testimonies, became increasingly politicized and advocated for greater access to the medications. It is difficult not to see in this activism the reflection of the searing experience of the early years of the epidemic. Beyond overt activism, however, this ethic of care of the self and of others was further nurtured by the growth of ARV clubs and discussion groups aimed to support members’ being adherent to their medications. Peer support and peer counselling in effect worked to bind people living with HIV and their families into communities marked by shared therapeutic work and the increasingly dense moral economy into which this work embedded them. This moral economy, initially built on an ethic of disclosure and positive living, became increasingly material as the flow of resources (drugs, food aid, micro-credit, credits for school fees for children and orphans, and peer-counselling jobs) from donor agencies increased. With the drugs came better health and with better health the urge to plan for the future and, for many, start a family. Women, who make up the majority of those in the self-help groups and associations that make up this therapeutic community, were mainly tested in the early years in clinical trials or because their husbands had died. It was thus inevitable that, with improved health, people within the groups began to marry and have children. Ironically, many women had come to the groups through trials aimed at preventing mother-to-child transmission of HIV. They were now having more children, and safely – not so much because of PMTCT interventions but because they themselves were now on treatment. While the core of these groups was formed around a first generation of activists largely drawn from the ranks of those tested through the PMTCT trials, the subsequent 439

Evidence, Ethos and Experiment

expansion of PMTCT and HIV treatment programmes rapidly swelled the mass of the community and self-help groups, already densified by the earlier dynamic of fission. Today these groups offer a plethora of activities and services that in materially powerful ways transform the condition of being HIV positive into social identity and social relations. The glue that holds this all together, both synchronically and diachronically, is antiretroviral treatment – as a common goal, experience and the chance to have a future together. Remarkably, in these groups it is possible to witness a flattening of hierarchies (for example, between patients and doctors) as the experience of being HIV positive – or of being on ARVs – confers legitimacy and authority. These characteristics – a shared social identity and social relations, a common moral economy, and a flattening of hierarchies – have been described before, in therapeutic communities pioneered in post First World War psychiatry, and are not without reminding us of Goffman’s ‘total institutions’ (Goffman 1961; see also Mills and Harrison 2007). They certainly merit that we see HIV groups as therapeutic communities in the most robust sense. While confessional technologies and discourses of empowerment provide the tools for building these therapeutic communities, being diagnosed with HIV focuses attention on what is at stake with this activism. Growing availability of ARV drugs, which dramatically restore health in those who are ill and keep them healthy, in effect sustains activism not only by keeping activists alive and healthy but also by putting flesh on the biomedical discourse that preaches the very real biological efficacy of these drugs. Living through a diagnosis, with all the uncertainties and challenges it raises, is one thing; living through the ‘resurrection’ one experiences after one has been ill and has recovered because of the drugs is a whole other magnitude of experience. No wonder that the medicines have only confirmed the evangelical aura that surrounds HIV and its treatments. This ‘therapeutic citizenship’ contrasts with other forms of ‘biological citizenship’ that are also mediated by biomedical categories. Biomedical forms of citizenship have been described by medical anthropologists and sociologists in other settings. Adriana Petryna, for instance, introduced the term ‘biological citizenship’ in her ethnographic study of the aftermath of the Chernobyl nuclear disaster to highlight how its victims mobilized evidence of their damaged biology to make compensation claims (Petryna 2002). Rose and Novas also use the term in a more general sense to refer to how ‘citizenship projects organized in the name of “health”’ (Rose 2006: 24) are increasingly articulated in biomedical terms (Rose and Novas 2004). 440

Trial Communities

Heath, Rapp and Taussig (2004) have used the term ‘genetic citizenship’ in a similar manner. Biological citizenship indexes the way in which biomedical science and categories are used to categorize and manage individuals and adjudicate their claims for compensation; it also refers to how individuals act on their lives through biomedicine. Central to these notions of biological citizenship has been an understanding of the role of the State and other large, stable institutions, as guarantors of healthcare and social security. Therapeutic citizenship is also conditioned by biological knowledge and biomedical practice. It differs from the ‘biological citizenship’ described by Rose and Petryna in that it arises where large, stable institutions that can grant access to life-saving therapy are absent. Significantly, as I have shown, it is co-constituted with forms of community whose connective tissue is a charged, biomedical condition. Nonetheless it is a thin citizenship, solely focused on a particular disease; but since it is active in a setting where the disease is the only way to get any kind of citizenship, it takes on a particular poignancy.

The Future of Trial Communities This story started out with a clinical trial of AZT to prevent HIV transmission in pregnant women. While it may seem that the convergence of factors that leveraged the trial into the subsequent development of therapeutic communities was just a historical accident, the other studies in this volume suggest that this is not just an isolated case. The disintegration of the public health infrastructure in Africa means that today, trials are often the only way in which the ill can gain access to biomedical care – echoing colonial examples. Indeed, for many, clinical trials will be one of the few – or maybe the only – interaction they have with the kind of modern institutions those who live in industrialized countries take for granted. In effect, the growing numbers who take part in clinical trials – and the discourses and forms of discipline they embody – do so in an environment where they are otherwise largely disconnected from the cardinal features of global modernity. These features can be summed up as the ability to live in a world where the future has promise because large, stable institutions manage the domains of the social world necessary to individual and collective thriving: health, education, public services, economy and so on. This management in effect insures against the vicissitudes of chance events (an illness, a drought, a conflict) that would otherwise always threaten to compromise the future. 441

Evidence, Ethos and Experiment

Today, abysmal health conditions in Africa have finally caught international attention. Large international donors are now focussed on finding biomedical solutions to the epidemics that rage across the continent, even as the conditions that produced them are ignored. Mechanisms, of which the International HIV Vaccine Initiative is a model, have been developed to create markets for diseases of the poor, so that there is an incentive to develop drugs and vaccines for diseases that affect those too poor to pay. Identifying effective biomedical technologies (vaccines, drugs, microbicides and so on) requires clinical trials to be done in populations that are at risk for these diseases or are already affected by them. Paradoxically, those who were excluded from modernity find themselves reintegrated as experimental subjects. The biggest challenge for conducting clinical trials is access to a suitable population that meets study criteria and that will be available for followup. In the case of vaccines, microbicides, and other prevention technologies for instance, one would wish to have readily available ‘high-risk’ populations such as sex workers who are known to be HIV negative but whose chances of acquiring HIV is high and for whom, therefore, it is relatively easy to show a prevention effect in a clinical trial of a prevention technology. Similarly, trials of treatment interventions would need to have ready access to patients who are HIV positive and eligible for treatment. In his description of the patterns of resort of Indian women organ donors, medical anthropologist Lawrence Cohen shows how enrolment into family planning programmes – often the only contact these poor women had with the State – initiated a ‘therapeutic’ trajectory that continued on to selling organs. Cohen uses the term ‘bioavailability’ to refer to this condition. The term plays on the concept of bioavailability in clinical medicine, where it is used to refer to the portion of a biologically active substance that can actually be used by the body once it is ingested (some drugs, for instance, are more ‘bioavailable’ than others because they are more readily absorbed or are metabolized by the liver into more active forms). In the sense introduced by Cohen, bioavailability refers to how historical interconnections between biomedical knowledges and technologies (family planning and transplantation) drive programmes that seek out and target specific populations for intervention (Cohen 2005). Etymologically, Cohen links this concept with Foucault’s historical studies of the rise of the modern, governmental state. This, Foucault argued, was inextricably linked to the dissemination of practices designed to manage the population through the internalization of rules and procedures aimed at keeping bodies docile and productive. This ‘governmental’ form of state power focussed 442

Trial Communities

on producing a healthy, lively population rather than on asserting territorial sovereignty through force – hence Foucault’s term biopower, where ‘bio’ refers to the emphasis on ‘life’, in distinction to classical notions where state power resided in the ability to enforce territorial boundaries and the law through the use of violence. This study points to a similar kind of ‘bioavailability’, whereby early clinical trials produced populations through their attempts to screen and enrol patients. These populations did not disappear after the clinical trial was over; rather, they aggregated into therapeutic communities that in fact persisted because they ‘reproduced’. In the case I have discussed above, reproduction of these therapeutic communities was both social – through group fission – and biological, as group members marry and have children. These therapeutic communities hold the expertise and the local knowledge crucial to making experimental subjects available to clinical trials. They will probably therefore continue to ‘reproduce’ through the deployment of further trials. While it is too early to say, Silla’s history of the leper colony in Bamako evokes a future where today’s clinical trials will have persisted in the fabric of social life as therapeutic communities.

References Cohen, L. 2005, ‘Operability, Bioavailability and Exception’, in A. Ong and A. Collier (eds), Global Assemblages: Technology, Politics and Ethics. London: Blackwell, pp. 79–90. Dabis, F., P. Msellati, N. Meda, C. Wiffens-Ekra, B. You, O. Manigart, V. Leroy, A. Simonon, M. Cartoux, P. Combe, A. Ouangré, R. Ramon, O. Ky-Zerbo, C. Montcho, R. Salamon, C. Rouzioux, P. van de Perre, L. Mandelbrot. for the DITRAME Study Group. 1999. ‘6-month Efficacy, Tolerance, and Acceptability of a Short Regimen of Oral Zidovudine to Reduce Vertical Transmission of Hiv in Breastfed Children in Côte-D’ivoire and Burkina Faso: A Double-blind Placebo-controlled Multicentre Trial’, Lancet 353(6): 786–93. Fassin, D. 2001. ‘Quand le corps fait loi. La raison humanitaire dans les procédures de régularisation des étrangers en France’, Sciences Sociales et Santé 19(4): 5–34. ———. 2005. ‘Compassion and Repression: The Moral Economy of Immigration Policies in France’, Cutural Anthropology 20(3): 362–87. Ferguson, J. 2006. Global Shadows: African in the Neoliberal World Order. Durham: Duke University Press. Goffman, E. 1961. Asylums. Essays on the Social Situation of Mental Patients and Other Inmates. New York: Doubleday Anchor. Heath, Deborah, Rayna Rapp and Karen-Sue Taussig 2004 Genetic Citizenship In. A companion to the anthropology of politics David Nugent and Joan Vincent eds. London: Blackwell, pp. 152-67, 443

Evidence, Ethos and Experiment

Hunt, N.R. 1997. ‘Condoms, Confessors, Conferences: Among HIV Derivatives in Africa’, Journal of the International Institute 4(3). Available online from http://quod.lib.umich.edu/j/jii/. Lee, L.K. 2002. ‘Changing Selves, Changing Society: Human Relations Experts and the Invention of T Groups, Sensitivity Training and Encounter in the United States, 1938–1980’, Ph.D. dissertation, University of California, Los Angeles. Mann, J. and D. Tarantola. 1996. AIDS in the World II. Oxford: Oxford University Press. Mills, J.A. and T. Harrison. 2007. ‘John Rickman, Wilfred Ruprecht Bion, and the Origins of the Therapeutic Community’, History of Psychology 10(1): 22–43. Nguyen, V.-K. 2005a. ‘Uses and Pleasures: Sexual Modernity, HIV/AIDS and Confessional Technologies in a West African Metropolis’, in V. Adams and S.L. Pigg (eds), Sex in Development: Science, Sexuality, and Morality in Global Perspective. Durham: Duke University Press, pp. 245–68. ———. 2005b. ‘Antiretroviral Globalism, Biopolitics, and Therapeutic Citizenship’, in A. Ong and S.J. Collier (eds), Global Assemblages: Technology, Politics, and Ethics as Anthropological Problems. London: Blackwell, pp. 124–44. Petryna, A. 2002. Life Exposed: Biological Citizens after Chernobyl. Princeton: Princeton University Press. Power, S. 2003. ‘The AIDS Rebel’, The New Yorker, 19 May, http://www.pbs.org/ pov/pov2003/stateofdenial/special_rebel.html (accessed 11 June 2007). Richey, L.A. and P. Stefane. 2006. ‘Better (RED) than Dead: ‘Brand Aid’, Celebrities and the New Frontier of Development Assistance’, Copenhagen, Danish Institute for International Studies. http://www.diis.dk/sw27885.asp (accessed 11 June 2007). Robins S.L. 2005. ‘From ‘Miraculous Cures’ to ‘Normal(ized)’ Medicine: HIV Treatment, Activism and Citizenship in the UK and South Africa’, IDS Working Paper 252, October 2005, Institute of Development Studies, Brighton. ———. 2006. ‘From Rights to ‘Ritual’: HIV Activism and Treatment Testimonies in South Africa’, American Anthropologist 108(2): 312–23. Rose, N. 2006. The Politics of Life Itself: Biomedicine, Power and Subjectivity in the Twenty-first Century. Princeton: Princeton University Press. Rose, N. and C. Novas. 2004. ‘Biological Citizenship’, in A. Ong and S.J. Collier (eds), Global Assemblages: Technology, Politics, and Ethics as Anthropological Problems, 439–63. London: Blackwell. Seidel, G. 1993. ‘The Competing Discourses of HIV/AIDS in Sub-Saharan Africa: Discourses of Rights and Empowerment vs Discourses of Control and Exclusion’, Social Science and Medicine 36(3): 175–94. Silla, E. 1998. People Are Not the Same: Leprosy and Identity in Twentieth-century Mali. London: Heinemann. Ticktin, M. 2006. ‘Where Ethics and Politics Meet: The Violence of Humanitarianism in France’, American Ethnologist 33(1): 33–49. Wiktor, S.Z., I. Ikpini, J.M. Karon, J. Nkengason, C. Maurice, S.T. Severin, T.H. Roels, M.K. Kouassi, E.M. Lackritz, I.-M. Coulibaly, A.E. Greenberg. 1999. ‘Short-course Oral Zidovudine for Prevention of Mother-to-child Transmission of HIV-1 in Abidjan, Côte-d’Ivoire: A Randomised Trial’, Lancet 35(6): 781–85.

444

Chapter 18

Differences in Medicine, Differences in Ethics: Or, When is It Research and When is It Kidnapping or is That Even the Right Question? Luise White

I In 1899 a smallpox epidemic broke out in Kampala, Uganda. The founders of the medical mission Church Missionary Society vaccinated two young men with all that was left of the vaccine they had brought from England. When the youths did not come back as instructed, the missionaries proceeded to protect the local population with all the violence of modern science and a reinscription of local methods of cure, particularly variolation against smallpox. One missionary saw a young boy in a market ‘with an obvious vaccination mark .... So she collared him and brought him into the dispensary and, with somewhat trembling fingers, knowing the issues at stake, we extracted the precious drops of lymph, and with them vaccinated half a dozen others’. The other wrote that they managed to keep these six ‘under observation’ and ‘From them a week later, we inoculated twenty, and from them fifty, till one day we vaccinated over 800 by this arm to arm method, with but very few failures, as indeed might have been surmised in an unprotected population’ (Cook 1945: 52).1 Was this ethical? Part of the triumphalist narrative of western biomedicine is, who cares? If curing Africans requires that they be abducted, inoculated and observed, even as the well-intentioned doctors and nurses denied the evidence for local methods of disease prevention, even as those local methods enhanced their own vaccine therapy, is it really a problem? Africans, missionaries argued, fully understood the benefits of such prac-

445

Evidence, Ethos and Experiment

tices: they wanted vaccinations: ‘In their eagerness, they almost stormed the dispensary to get in’ (Cook 1945: 52). But were ethics, or even disease, the issue? A year later on the coast of Kenya, near Mombasa, CMS missionaries inoculated 1800 Africans against smallpox. They did not report such eagerness, however, but local competition: a year later, when a maneating lion prowled around the settlement, a healer grew wealthy ‘vaccinating’ Africans against lion bites. ‘When remonstrated with he showed the vaccination marks on himself, which had been done by us for smallpox, and could not see the difference between his work and ours’ (Mercy and Truth: The Magazine of the CMS Medical Mission 31 [1899], quoted in Vaughan 1991: 59). While I do not want to use these stories to produce a simple dialectic of colonial power and its subversion, I do want to argue that both stories allow me to talk about the ethics of medical practices in Africa in a somewhat different way than the other papers presented here, that western biomedicine was not always effective nor did Africans regard it as such. Indeed, there is only scanty evidence that Africans believed vaccinations to be efficacious. In Kampala a few years later, for example, it was noted that Africans avoided smallpox vaccination, claiming it poisoned them (White 1995: 1393). The lion bite story challenges colonial abilities to protect African bodies on the most conceptual level: the technology appropriate to prevent smallpox could be appropriated to prevent other misfortunes. The power of healers vaccinating for lion bite lay in its elision of the specificity and nature of the smallpox vaccine. The technique protected. The accoutrements of western medicine did not always protect, and they could become so muddled that they frequently harmed people. There is the widespread comment that people go to hospitals to die, while women in Nairobi in the 1920s claimed that ‘skilled’ men from hospitals entered their rooms at night with ‘rubber sucking tubes’ and drained all their blood. Twenty years later, on the East African coast, vans from the colonial medical department ‘patrolled’ the streets, and when they came upon an unwary African, they drew blood from his veins ‘with a rubber pump’. In many parts of East Africa, it was said that a generic fire brigade captured Africans and took their blood, but an old man in Uganda insisted that this was false: it was the Yellow Fever Department that took Africans’ blood, but they had no vehicles of their own, so they had to use fire engines (White 2000: 89–129). Medical missionaries and lion bites raise two linked questions, questions that are in fact joined by blood-thirsty ambulances. First, does the 446

Differences in Medicine, Differences in Ethics

concept of ‘disease’ cross cultural boundaries with any clarity at all? While few people would argue that smallpox or plague had the same meaning to Africans in the late Nineteenth Century as it did to European physicians, can historians begin to look at the broader range of concepts of misfortune that might problematise notions of disease even for those afflicted with them? Second, if smallpox is a misfortune caused by a deliberate act, and thus can be easily categorised with a lion bite, would a single curative practice ameliorate the affliction or elide the differences between the two? My point here is not to list the host of examples of colonial doctors and officials misunderstanding what Africans’ ideas about the cause and cure of various sicknesses were.2 Instead I want to suggest how they are linked. For example, Yoruba peoples of Nigeria and Fon peoples of Dahomey considered smallpox the greatest affliction in the gods’ arsenal of punishments (Ajose 1957; Morgan 1979; Buckley 1985). In the 1930s Dahomeans maintained that any attempt to prevent smallpox, such as vaccination, would simply bring the wrath of the god’s ‘little brother’, lightning, to the offender. Thirtyfive years later the CDC found Dahomey the West African nation most resistant to smallpox eradication (Herskovits 1938: 134–36; Ogden 1987: 67–69; Fenner et al. 1988: 887–92). Thus our own biomedical notions of ethical practices may not adequately address these terrains of differences and similarities. Willing subjects, volunteers, informed consent – these terms may not describe what happens when one notion of disease is cured, or even prevented. This paper argues that the very use of these terms obscures other ways to understand and contain illness. What follows is speaks to the concerns of both historians and policy makers: it is a critical history of ideas about smallpox in Africa and how the worldwide smallpox eradication campaign was conducted in Africa. I have chosen smallpox in part because it is a disease that was prevented both by African methods – variolation, scratching the surface of the skin to introduce diseased matter as a prophylactic against the disease – and western methods, vaccination, and in part because of the many accounts of the eradication of smallpox in Africa (Ogden 1987: 67–69; Fenner et al. 1988: 891–93).3

II The serious colonial conquest of Africa took place a century after Jennerian vaccines were developed. The association of the project of colonial control with health was already in place, with considerable discourse and 447

Evidence, Ethos and Experiment

perhaps less policy than had accompanied it in India. Precolonial African bodies seemed more ravaged by vector-borne diseases than anything that had occurred in Europe and the coincidence of tropical medicine and colonisation of Africa seems to have made bilharzia a more urgent topic of study than measles. Tropical diseases were sexier than contagious ones.4 In this context there was no immunisation: smallpox vaccinations only existed in response to epidemics, and response may be a generous term. Administrators, doctors, missionaries, or an official on tour would report an outbreak and orders would be sent to assemble the population for vaccination, finding district messengers to administer vaccines regardless of their efficacy (Headrick 1994: 218). Something in this particular history made variolation less visible than it was in many other colonial societies. Travellers and officials encountered it, were horrified by it, and made no effort to control it; they rarely noticed that the areas where variolation was most visible had the greatest resistance to smallpox vaccination.5 But variolation never captured the official gaze – medical or administrative – for very long, and it takes a careful and surprised reading of medical anthropologists and memoirs to learn that it continued well into the 1970s. Even in those texts variolation was presented in specific contexts for specific audiences. James Pascal Imperato, the American physician in charge of the AID’s smallpox campaign in Mali, wrote detailed studies of variolation in the Sahel. But he began the memoir he published about his experiences in Mali with an orientalist description of an isolated village ravaged by smallpox. They blamed the epidemic on a genii and went to Muslim holy men for help and eventually to nomads for variolation (Imperato 1975: 1–13). What did the persistence of variolation mean, both in cultural and epidemiological terms? Eugenia Herbert’s survey of secondary sources on variolation in Nineteenth and Twentieth Century Africa suggests that the risks of variolation were great for children because they often succumbed to the milder infection; for adults, particularly during epidemics, the survival rate for people variolated far outweighed any risks (Herbert 1975). In research carried out in 1967 Imperato found that in a village in rural Mali the death rate during a smallpox epidemic for people who had neither been vaccinated or variolated was about 17 per cent, whereas no one who had been variolated or vaccinated rarely succumbed to smallpox (Imperato 1968, 1974).6 Colonial officials and some of their historians had nevertheless routinely blamed major smallpox epidemics on variolation (Ajose 1957: 269–70; Herbert 1975: 546). In terms of cultural history – and I am writing about too broad a range to do more than generalise – African 448

Differences in Medicine, Differences in Ethics

variolation was very different from vaccination with live vaccines, and it was applied to many diseases – and misfortunes – that were not epidemic. Men and women were routinely variolated for everything from wealth and fertility to protection from lion bites; children were variolated for yaws.7 Much variolation was ritual. From the early nineteenth century to the 1920s European observers witnessed the variolation of hundreds of Ethiopians in religious ceremonies; the pustular fluid was diluted with butter and honey to purify it, and to make it go further. Smallpox victims’ huts were decorated with leaves and flowers. In Ethiopia in 1905 the emperor’s proclamations encouraged people to get vaccinations because it was a simpler procedure, not necessarily a more medically efficacious one. ‘The smallpox which is the result of vaccination ... has no bad effects; it is not necessary to diet; it does not stop work …’ (Pankhurst 1965; see also Fenner et al. 1988: 1007). Nevertheless, analyses of variolation should not homogenise the procedure nor African ideas about it. If nothing else, the purchase of variolation to prevent miscarriage or adultery was voluntary; and there is some evidence that Africans resisted variolation for smallpox because they believed it could cause the disease (Leakey 1977). Given the amount of literature that suggests that Africans chose healing therapies in sequence and that western biomedicine was never an alternative to local methods of healing but a complement to them,8 it would be difficult to label even mid-1960s African variolation as resistance to statesponsored vaccination. It could well be a supplement or another kind of treatment altogether: the concept of resistance implies that Africans saw vaccination the way medical professionals did, rather than something quite powerful and wonderful that had nothing to do with preventing disease. The evidence from Ethiopia suggests an layering of variolation and vaccination that cannot be explained in terms of resistance or compliance alone. During the 1960s smallpox eradication campaign many peoples in the central highlands, around the capital, avoided vaccination and participated in increasingly elaborate variolation rituals. In the southern parts of the country, however, Ethiopians actively sought vaccination, even when they continued variolation. Many people returned several times for vaccinations, and many villages falsely reported epidemics to encourage vaccination (Fenner et al. 1988: 107). Still, variolation during a smallpox epidemic in Africa in 1920 or 1930 is one thing; variolation in the 1960s in the midst of intensive state and international campaigns to eradicate smallpox is another – whether variolation was locally thought to be an alternative to state medicalisation or an essential part of local religion, in the national 449

Evidence, Ethos and Experiment

and international gaze it was seen as an assault on state and scientific abilities to prevent epidemic diseases.

III In what is normally considered the optimistic first years of African independence –decades before anyone would say postcolonial – inoculation campaigns should, in theory at least, have been free of the taint of colonial health endeavours and should have represented African nations’ ability to better the lives of their citisens. Epistemological concerns rarely entered into 1960s rhetorics of modernisation and development; although there was fleeting attention paid to traditional healers (see Langwick, this volume), healthcare for the citizens of African states was to take place within western notions of healing and curing just like courts were to function within western notions of crime and punishment. Yet the history of the smallpox eradication campaign reveals the fragile and defenceless statehood achieved by African nations in the 1960s and the triumph of global humanitarian concerns at the expense of local sovereignty and local humanitarian concerns. It is in fact a history that cannot be written from the vantage point of African states; the story must be told through its creators, the WHO, the CDC and to a lesser extent AID. And herein lies the irony of the question of ethics. In 1960 the ministers of health of four new West African nations visited the US on a State Department Leadership Grant. At the National Institute for Health they met with Dr Harry Meyer who was working on a new measles vaccine that was not yet licensed for use in the US. Measles rather than smallpox was the largest killer of children in West Africa. The minister of health for Upper Volta (now Burkina Faso) asked that this vaccine be tested in his country, and Meyer was keen to test the vaccine in Africa. In 1961 the NIH funded Meyer to test the vaccine in Upper Volta where it was found to be safe, and did not require gamma globulin as did its use in North America. For the next two years Meyer and his associates vaccinated 700,000 children in Upper Volta with dramatic results. Six other Francophone African nations immediately requested measles vaccine in US aid. Late in 1963 AID allocated the funds for vehicles, jet injectors, and measles vaccine for 25 million West African children. Only by bringing in the CDC could AID expand a national programme run by a drug-researcher to a regional programme. But the regional pro450

Differences in Medicine, Differences in Ethics

gramme floundered amidst administrative problems, non-existent local infrastructures, the impossibility of repairing US-made vehicles in formerly French Africa, and the frequency with which measles vaccine spoiled in the heat. At the same time, however, AID in its enthusiasm had offered the programme to four additional African nations. As the CDC’s personnel was stretched to their limits by bureaucratic challenges and equipment repairs, the new project evaluator, Ralph Henderson of the CDC’s Smallpox Unit, reported in 1965 that the programme was not worth the investment. The eradication of measles was not in sight, and any gains would be temporary, so that vaccination campaigns would have to be repeated every few years for the foreseeable future and as measles vaccine was expensive it was unlikely that AID would continue to provide it indefinitely. Henderson was of course aware of the developments in smallpox inoculation that made the possibility of world wide eradication feasible and relatively inexpensive. Earlier that year the CDC had completed its studies of the foot-powered injector that administered intradermal smallpox vaccine. Behind the scenes efforts by Henderson and many others helped persuade the American government to support the WHO plan to eradicate smallpox in the next decade. Armed with WHO support, the CDC then asked AID to combine the two projects – that the measles vaccination programme in West Africa be combined with a smallpox eradication campaign. Only by inclusion of smallpox eradication could the CDC justify the continued involvement of US public health resources, which by law, they noted, were intended to protect the health of the American people and who were in danger of smallpox only from infection from abroad or the routine inoculation itself. AID had a difficult decision, and no choices. They were committed to an eleven country measles inoculation programme, which they could not administer without the technical support of the CDC and US Public Health Service, which demanded that the measles campaign be coupled with smallpox eradication. This dilemma, such as it was, problematises the notion of sovereign states and bodies this volume addresses: policies of the US were negotiated and manipulated and bullied within the US by one agency acting against another, by medical possibilities and legal responsibilities being defined and defended. In this case, a state’s ability to mark and immunise bodies had to do with who within the state had the greatest claim to technical expertise. But the bodies marked in this campaign were African bodies, not American ones. Moreover, they were marked because of global health needs 451

Evidence, Ethos and Experiment

rather than local demands and concerns. Many West African officials, especially the directors of epidemic disease services in Francophone Africa, actively opposed smallpox eradication because they believed measles control was more urgently needed. The expansion of the measles programme to include smallpox – an expansion that was to reverse the conduct and priorities of the programme – meant that Ghana and Nigeria had to be included in West African vaccination programmes. Smallpox could only be eradicated if the entire region received vaccinations, but AID had long hoped that another donor would take on these two former British colonies. The WHO, established at the end of the Second World War to provide the health services European colonial powers were now too poor to provide, eventually funded the immunisation campaigns there, and as the global project continued, ran the programmes in the rest of Africa. Indeed, the involvement of WHO and US funding problematises the very use of terms ‘global’ and ‘local’: in the late 1960s smallpox eradication in Ethiopia was delayed as US advisors argued that a smallpox programme would jeopardise their malaria control programme (Ogden 1987: 21–29). Perhaps more important than discerning what was global and what was local is noting that a regional health programme had its conception, design and funding from outside the region. Given this, the ‘best’ sources for this period in West Africa are the official ones; after all, smallpox eradication is their history. And in this history the attributes of the nation-state are ascribed more frequently, and forcefully, to WHO (and occasionally the CDC) than they are to Upper Volta, Gambia, or even Nigeria. Indeed, the CDC described as ‘fiercely independent’ the WHO regional office in West Africa. The ways that WHO acted as a sovereign body, whose ability to prevent disease and whose allocation of technology and funds allowed them to act over and above the boundaries, bodies, and alliances of the nation-states whose populations were being vaccinated. The sense that this eradication campaign was a race against time strengthened the ways that WHO acted as a sovereign nation and Nigeria or Sudan did not. Clearly the humanitarian impulse was strong in these cases, but as we shall see, humanitarianism in and of itself was not the issue, vaccination was. It is not that the official literature of international medical aid bodies reveals any disdain for the workings of African nation-states. Although American sensibilities were tested again and again, agencies’ critique and disdain were reserved for other agencies.9 African bodies lay outside the sphere of bureaucratic rivalries. Without irony or apology, WHO’s official 452

Differences in Medicine, Differences in Ethics

history boasted that smallpox eradication was achieved most rapidly in Africa, the continent with the most limited health services. The WHO literature contains numerous examples of how African sovereignty was compromised so that WHO could go ahead with the business at hand: almost all are told in passing, as anecdotes. The casual quality with which the CDC and WHO describe WHO’s easy dealings with both state and rebels reveals how matter-of-fact these dealings were to the international organisation. In mid-1977, when the last smallpox cases in the world were found in Somalia, then very hostile to the US, WHO negotiated permission for CDC epidemiologists to work in the country (Ogden 1987: 64). The CDC’s own hagiography – as opposed to individual authors – presents themselves as innocents abroad: despite being outfoxed by West African realities, they boldly stumbled along. But the CDC was not as naive as they liked to pretend: a year after the Peoples’ Republic of Guinea had expelled the Peace Corps as agents of neo-colonialism, CDC epidemiologists negotiated access throughout the country – which had the world’s second highest death rate for smallpox – for their vaccination teams (ibid.: 109). The most compelling example of the subordination of African politics to WHO’s overrule is the Nigerian civil war. In 1967, when smallpox eradication was accomplished in much of Nigeria, the eastern region seceded and became the Republic of Biafra, on which Nigeria then declared war. WHO had just begun vaccinating there, and although the campaign was slowed by the fighting, it continued for almost a year as ceasefires were arranged specifically to transfer vaccine into Biafra. But no ceasefires were arranged to bring food into the new nation, and by mid-1968 severe famine was accompanied by a measles epidemic. Late in 1968 the government of Biafra appealed to the International Committee of the Red Cross and the World Council of Churches to begin smallpox and measles inoculation, which they did, the smallpox vaccines supplied by the WHO in Geneva. Throughout this period, suspected cases of smallpox were investigated as flights from Lagos, the capital, were allowed into Biafra to take out specimens (Fenner et al. 1988: 882).10 But even where the cause was common, where local medical services thought smallpox was as big a problem as WHO and the CDC thought it was, the endeavour was not. The picture that emerges from WHO’s own literature is of an agency impatient with national concerns and national problems. Zaire (now DRC), which had along with Ethiopia the highest smallpox mortality on the continent had approached WHO about smallpox eradication during periods of stability in the early 1960s. By the 1970s, 453

Evidence, Ethos and Experiment

smallpox eradication became part of WHO’s global campaign, not Zaire‘s local one. What Zaireans wanted from the eradication campaign were quickly pushed aside. Zaireans desperately wanted a census and WHO agreed to take one as it vaccinated; finding out what percentage of the populated was immunised was sound medical practice. Moreover, Zaireans wanted vaccination certificates, documents with official stamps that the government demanded, insisting their citizens expected them: these more than scars or health proved they had been vaccinated. The census was the first to go, as it went much more slowly than a day’s vaccinations and once jet-injectors were in daily use it became impossible to do as quickly as inoculations. The vaccination certificates were harder to get rid of: this postcolonial state did not immunise, but it certified. WHO desperately produced project-specific ways of explaining why they were prohibitively expensive: ‘For every jet injector in operation, on average 6 clerks were required to prepare the certificates’ (Fenner et al. 1988: 918).11 In the end, Zairean demands for paper won out, and a new certificate was introduced on which the vaccinee could write his or her name and a member of the vaccination team could stamp it. This was perhaps the only assertion of national ideas about health and power in the campaign in Zaire (Fenner et al. 1988: 913–29). In Sudan, a country whose own very effective vaccination programme had fallen apart a few years before, eradication went relatively easily, despite a huge country, bad roads, and a war: the depth of medical expertise in the country was actively engaged in the project, often coming up with better and more creative solutions than WHO staff could do. But the southern part of the country was at war, although government incursions into territory held by the Anyanya Resistance Movement were common. In 1970 representatives of the Anyanya approached the WHO in Geneva and asked for smallpox vaccine. They claimed the disease was not present but the constant flux of soldiers and refugees might create an epidemic at any time. They explained how they would get vaccine into the south without using routes in the Sudan. Besides, they could get around in the south; government and WHO vaccinations teams could not. WHO by its charter could only provide aid to member nations and the Anyanya’s request ‘was a quandary because ... it was in everyone’s best interest for this area to be better protected’. The solution was to give vaccine to the Anyanya representatives and recording the amount as ‘lost from inventory’. When the war ended in 1972, it appeared that a vaccination campaign had been conducted (Ibid., 918) The very language of the official anecdote (this ap454

Differences in Medicine, Differences in Ethics

pears in a text box in the WHO history), ‘everyone’s best interest’, glosses Sudanese sovereignty in favour of a humanitarianism. Under what circumstances did African states act as sovereign bodies and immunise their citizens’ bodies? And when did they choose not to? The actions of African sovereignty took place outside of official agencies, among those nations or non-nations that were not members of WHO. For most of the period of global smallpox eradication, Portugal was a member nation of WHO and its African colonies – all engaged in protracted and eventually successful warfare against Portugal – were not. In Guinea-Bissau, in West Africa, Portugal reported no cases of smallpox throughout the period, and did not allow anyone from the AID/CDC smallpox programme in the colony. The PIAGC, the rebel group that controlled most of the countryside, was concerned about smallpox and sent health workers to be trained in Guinea, where they were given bifurcated needles and smallpox vaccine. The PIAGC conducted vaccinations in the areas they controlled and reported smallpox cases to WHO (ibid.: 880). Presented as another text box anecdote, this tale would reveal the ways that modern nationalism seeks to mark and immunise the bodies of its subjects except for one important point: the vaccine and needles obtained in Guinea were the CDC’s. The nationalism of marking West African bodies was mediated by the internationalism with which the drugs and the technology to mark those bodies were obtained. This raises another question: who is allowed to immunise whose population? The Anyanya maintained that neither government forces nor WHO personnel could or should vaccinate southern Sudanese, and in the interests of expediency and ‘everyone’ the WHO concurred. The CDC quietly supported rebels in Guinea-Bissau. In war-torn southern Africa, however, the question of who could be immunised – and whose inoculation was in whose ‘best interest’ – seemed much more ambiguous. In Malawi (formerly Nyasaland) resistance to smallpox vaccination did not end with independence. In 1968, a national, Peace Corps assisted mass vaccination campaign in the south could only inoculate 50 per cent of the population. But included in Malawi’s 1968–1974 population was an unknown and fluid number of refugees from neighbouring Mozambique, most of whom had entered the country illegally and many of whom travelled back and forth to Mozambique frequently. To prevent smallpox the Malawian government often sent police cordons to surround these refugee villages, and held the populations captive until all had been vaccinated (ibid.: 977–78). 455

Evidence, Ethos and Experiment

Struggles over smallpox reporting from Guinea-Bissau or the coerced inoculation of refugees in Malawi even as its citizens hid from vaccinators raise another issue: to what are reports of disease and reports of immunised citizens a part of a larger presentation of social success to the wider world? Forced inoculations of refugees in Malawi raises the question of to what lengths states can go to protect their population. The coercion that no newly independent government would have dared apply to its citizens could be applied to refugees, as they were simultaneously protected from their own states’ failure to immunise and from harming the uninoculated citizens of Malawi. In South Africa and Southern Rhodesia, issues of who is said to have which disease and how severe that disease is, may reveal more about how states (especially those outside the WHO orbit) present their own civilizing missions than it does about epidemiology.

IV How medical was the smallpox eradication campaign? That very question needs to be historicised: how medical was vaccination in Africa, ever? Clearly the people who hid from smallpox vaccinators or chased them away did not see the procedure they resisted as medical. But what about the people who actively sought vaccination? The medical literature is somewhat torn in reporting this – representing Africans as eager consumers of medicine never fully allowed for any resistance to those procedures – but where there are detailed reports they are revealing. In 1951, for example, heath officials in Nairobi, Kenya noted that more than 43,000 inoculations were given to Africans at the Inoculation Centre at Town Hall and of these, 40,041 were typhoid and paratyphoid inoculations, even though these diseases were rare in Nairobi. Moreover, less than 10,000 came back for a second dose. A higher single dosage was introduced, but medical department officials sought to find out why Africans were so eager to have this injection. A day’s survey of 510 Africans waiting for typhus inoculations revealed that only 2 or 3 per cent associated the inoculation with protection against typhoid, while a high proportion associated the inoculation with malaria (and described their symptoms), respiratory ailments, skin diseases, wounds, stomach problems, and for safe journeys. The anonymous authors suggested that Africans sought T.A.B. injections because each inoculation stimulated the recipient’s antibodies and cured a number of minor infections.12 However, the long lines of men and women describing 456

Differences in Medicine, Differences in Ethics

their coughs and aches and travel plans indicates something else: that the vaccine itself and the uses to which western medicine put it had become immaterial, what was important was receiving the injection the meaning and importance of which was as broad as the range of needs of the people lining up to receive them. WHO observed the same phenomena in Rwanda. By the time smallpox eradication began in eastern Africa, variola major had all but disappeared. Only Tanzania regarded smallpox as any kind of threat; Kenya and Uganda had fewer than ten deaths from smallpox between 1961 and 1964 and the last case of smallpox in Kenya was reported four months before the eradication programme began. Rwanda had little smallpox and a low rate of immunisation; freeze-dried vaccine had been manufactured there since 1965. It was distributed in vials that when diluted contained enough vaccine for 600 people. Sent to rural clinics, most without refrigerators, this vaccine lost its potency in a day or two, but clinic staff continued to give a few smallpox vaccinations every day until the vial was empty. In 1968 a WHO smallpox eradication advisor noted that a hospital was giving smallpox inoculations from a vial reconstituted three months before (Fenner et al. 1988: 945, 947, 954–55). What could such practices have meant? These inoculations could not have prevented smallpox, but then, smallpox was not such a threat that Africans would have sought the vaccination to prevent it. The recirculation of inactive vaccine – in the absence of smallpox – in which clinic worker and patient imply that what is important is the inoculation itself, not what it does. And what it does may have, in the minds of those seeking smallpox vaccination, nothing to do with smallpox or even disease: Rwandans may have come to clinics in 1966 or ‘67 to protect themselves from harm when travelling, to cure malaria, or to cure a cough or skin rash. In the absence of variola major, this must have seemed a harmless use of medical facilities. In the gaze of worldwide smallpox eradication, however, this became another example of why Africans should not be allowed to manage this campaign, rather than evidence that vaccination was not medical. That gaze did not take into account resistance, reinterpretation, or the many meanings smallpox may have had. The self-hagiography of smallpox eradication simply did not allow for accounts of peoples’ terror at anything so massive and so commanding of political support. I fear that what I have written can be interpreted as suggesting that there are no problems in medical ethics: that since Africans have interpreted disease the way no westerner would there can be no issue of intervention, 457

Evidence, Ethos and Experiment

capture, or forcible containment. My goal here is to argue, with as much force as medical intervention, the opposite: that the broader context of medical intervention, and the prevention of epidemic diseases, originates so far from African concerns and African consent that it cannot be called ethical at all. And indeed, making the question of curative procedures one of medical ethics simply avoids the larger question. Scholars and health professionals concerned about Africa should not worry if this or that individual consented to a specific injection. On the contrary, we should be attentive to why some injections are available and others are not, and who, from where, has decided on that availability. By focusing on the smallpox eradication campaign of the 1970s I have tried to show how African ideas about curing and healing, and about western biomedicine, have remained separate and distinct, in dialogue rather than in conflict. More important, perhaps, is that the history of smallpox eradication in Africa reveals something decidedly not curative: the ability of international, humanitarian agencies to arrange ceasefires and negotiate with rebels and loyalists alike, so they might quickly mark the bodies of millions of Africans, despite bad roads and rudimentary health services. The Africans so marked are indeed saved from the scourge of smallpox, but they may have no reason to assume – let alone believe – that vaccination against smallpox, whether good or bad, had anything to do with the prevention of contagious disease. My point in this article is not that the vaccination campaigns conceived and executed as political acts are unethical; instead I have tried to show that the ethics with which vaccination campaigns are conducted have little overlap with the medical understandings and meanings of the men and women vaccinated with such speed and precision.

Notes 1. For Baganda variolation, see Roscoe (1911: 102–3) and White (1995: 1392–93). 2. For a very partial list, there is Ajose (1957); Singleton (1976); Morgan (1979); M’Bokolo (1982); isaBuckley (1985); Hunt (1999). 3. I am collapsing the differences between the more lethal variola major and the less lethal and commonplace in East Africa variola minor. This is in part a writing strategy that allows me to write with a level of generalisation that attention to varieties of smallpox would hinder, and in part a way for me to access the gaze of 1960s epidemiologists, who, in their eagerness to eradicate smallpox worldwide, only noted the difference in passing. For a straightforward medical history that argues that variola major entered east Africa through socioeconomic changes, see Dawson (1992: 90–103). 458

Differences in Medicine, Differences in Ethics

4. For this history, see Delaporte (1991); Farley (1991); Lyons (1992). 5. See Zeller (1971) and Dawson (1992: 100) for historians who failed to make the connection as well, but see Herbert (1975) for one who did. 6. It is interesting to locate these studies in the recognition of variolation that occurred in the wake of worldwide smallpox eradication in South Asian history, see Greenough (1980) and Arnold (1993: 116–58). 7. For a study of how some postcolonial traditional healers labelled variolation ‘African injections’ and offered them for sale for any number of personal and marital woes, see Frankenberg and Leeson (1976: 223–57). 8. A few key works in this literature are Janzen and Akinstall (1978); Feierman (1979); Ranger (1981); Vaughan (1994). 9. WHO mentioned the involvement of the CDC in passing, and reserved its strongest criticisms, bordering on contempt, for UNICEF. The CDC noted that it’s greatest and most persistent conflicts were with AID missions in West Africa, and made a point of stating that WHO’s main contribution was to foot the bill for smallpox eradication. Years later a CDC veteran – and a co-author of the WHO official smallpox history – was to say that smallpox was eradicated in spite of WHO, ignoring that it was WHO’s clout with non-member nations and Washington that allowed for a greater internationalisation of smallpox eradication than the CDC’s epidemiology could have mustered. But such a statement reveals the extent to which international agencies were described as autonomous, separate and distinct from the individuals who worked in them. International agencies, WHO in particular, represented themselves as transcending the individual and national. 10. A Biafran university lecturer in paediatrics and a physician from the Institut Pasteur organised the vaccination programme, which was funded by the French as was much in the Biafran secession. 11. For the history and meaning of such certification, see Hunt (1999: 237–80). 12. Anonymous (1951: 476).

References Ajose, O.A. 1957. ‘Preventative Medicine and Superstition in Nigeria’, Africa 27(3): 269– 71. Anonymous. 1951. ‘Dawa ya Sindano’, East African Medical Journal 28(11): 476. Arnold, D. 1993. Colonizing the Body: State Medicine and Epidemic Disease in Nineteenth Century India. Berkeley: University of California Press, pp. 116–58. Buckley, A.D. 1985. ‘The God of Smallpox: Aspects of Yoruba Religious Knowledge’, Africa 95(2): 187–99. Cook, A.R. 1945. Uganda Memories (1897–1940). Kampala: Uganda Society. Dawson, M.H. 1992. ‘Socioeconomic Change and Disease: Smallpox in Colonial Kenya, 1880–1920’, in S. Feierman and J. Janzen (eds), The Social Basis for Health and Healing in Africa. Berkeley, University of California Press, pp. 90–103.

459

Evidence, Ethos and Experiment

Delaporte, F. 1991. A History of Yellow Fever: An Essay on the Birth of Tropical Medicine. Cambridge, MIT Press. Farley, J. 1991. Bilharzia: A History of Imperial Tropical Medicine. Cambridge: Cambridge University Press. Feierman, S. 1979. ‘Change in African Therapeutic Systems’, Social Science and Medicine 13B: 277–84. Fenner, F., D.A Henderson, I. Arita, Z. Jezek and I.D. Ladnyi. 1988. Smallpox and Its Eradication. Geneva: WHO. Frankenberg, R. and J. Leeson. 1976. ‘Disease, Illness and Sickness: Social Aspects of the Choice of Healer in a Lusaka Suburb’, in J.B. Louden (ed.), Social Anthropology of Medicine, ASA Monographs 13. London: Academic Press, pp. 223–57. Greenough, P. 1980. ‘Variolation and Vaccination in South Asia, c. 1700–1865: A Preliminary Note’, Social Science and Medicine 14D: 345–47. Hackett, C.J. 1980. ‘Yaws’, in E.E. Saben-Clare, D.J. Bradley and K. Kirkwood (eds), Health in Tropical Africa during the Colonial Period. Oxford: Clarendon Press, p. 87. Headrick, R. 1994. Colonialism, Health and Illness in French Equatorial Africa, 1885– 1935. Atlanta, GA: African Studies Association Press. Herbert, E.W. 1975. ‘Smallpox Inoculation in Africa’, Journal of African History 16(4): 539–59. Herskovits, M. 1938. Dahomey: An Ancient West African Kingdom, vol. 2. New York: J.J. Augustin. Hunt, N.R. 1999. A Colonial Lexicon of Birth Ritual, Medicalization and Mobility in the Congo. Durham, Duke University Press. Imperato, P.J. 1968. ‘The Practice of Variolation among the Songhai of Mali’, Royal Society of Tropical Medicine and Hygiene 63: 869–74. ———. 1974. ‘Observations on Variolation Practices in Mali’, Tropical and Geographical Medicine 26: 429–39. ———. 1975. A Wind in Africa: A Story of Modern Medicine in Mali. St Louis, MO: Warren H. Green. Janzen, J. and W. Akinstall. 1978. The Quest for Therapy in Lower Zaire. Berkeley: University of California Press. Leakey, L.S.B. 1977. The Southern Kikuyu before 1903, vol. 3. New York: Academic Press, pp. 889–91. Lyons, M. 1992. The Colonial Disease: A Social History of Sleeping Sickness in Northeastern Zaire, 1900–1940. Cambridge, Cambridge University Press. M’Bokolo, E. 1982. ‘Peste et societe urbaine a Dakar: l’epidemie de 1914’, Cahiers d’Etudes africaines 22(1–2): 85–86. Morgan, R.W. 1979. ‘The Sopono Cult and Smallpox Vaccinators in Lagos’, Working Paper 11, African Studies Center, Boston University. Ogden, H.G. 1987. The CDC and the Smallpox Crusade. Washington, DC: US Public Health Service. Pankhurst, R. 1965. ‘The History and Traditional Treatment of Smallpox in Ethiopia’, Medical History 9(4): 347, 351–52. Ranger, T.O. 1981. ‘Godly Medicine: The Ambiguities of Mission Medicine in Southeast Tanzania, 1900–1945’, Social Science and Medicine 15B: 864–70. Roscoe, J. 1911. The Baganda: Their Customs and Beliefs. London: Cass. 460

Differences in Medicine, Differences in Ethics

Singleton, M. 1976. ‘Smallpox in Person: Personification or Personalization (Africa)?’, Anthropos 71: 169–79. Vaughan, M. 1991. Curing Their Ills: Colonial Power and African Illness. Stanford: Stanford University Press. ———. 1994. ‘Curing and Healing: Issues in the Social History and Anthropology of Medicine in Africa’, Social History of Medicine 7: 284–95. White, L. 1995. ‘“They Could Make Their Victims Dull”: Genders and Genres, Fantasies and Cures in Colonial Southern Uganda’, American Historical Review 100(5): 1392–93. ———. 2000. Speaking with Vampires: Rumor and History in Colonial Africa. Berkeley: University of California Press. Zeller, D. 1971. ‘The Establishment of Western Medicine in Buganda’, Ph.D. dissertation. Columbia University, New York.

461

Notes on Contributors

Hansjörg Dilger is a Junior Professor of Social and Cultural Anthropology at the Freie Universität Berlin. Dilger has carried out long-term fieldwork on HIV/AIDS and social relationships in Tanzania, focusing on the dynamics of kinship and Neo-Pentecostalism in the context of rural– urban migration, as well as on the responses of national and nongovernmental actors to the HIV/AIDS epidemic. He is author of Living with Aids: Illness, Death and Social Relationships in Africa (2005; in German) and co-editor of Morality, Hope and Grief: Anthropologies of AIDS in Africa (2010; with Ute Luig). His current book project is about Christian and Muslim Schools in Dar es Salaam, exploring how religious actors’ involvement in the field of secondary education in Tanzania correlates not only with longstanding histories of religious engagement in the field of education, but also with the potential for exclusion and conflict in the highly diverse urban landscapes of Eastern Africa. Between 2004 and 2010, Dilger was the chair of the work group Medical Anthropology within the German Anthropological Association. Steven Feierman is Professor of History and Sociology of Science, and Professor of History, at the University of Pennsylvania. He is a historian and ethnographer of medicine, religion, politics and science in Africa, holding a Ph.D. in history from Northwestern University and a D.Phil. in social anthropology from Oxford University. He has done many years of research in Africa. Among his books are Peasant Intellectuals: Anthropology and History in Tanzania, and the Shambaa Kingdom. He has won fellowships from the Guggenheim Foundation, the Institute for Advanced Study at Princeton, the Center for Advanced Study in the Behavioral Sciences at Stanford, the Wissenschaftskolleg zu Berlin, the National Endowment for the Humanities, the Fulbright Program and the National Science Foundation. P. Wenzel Geissler teaches social anthropology at the Institute of Social Anthropology at the University of Oslo and at the London School of Hy463

Evidence, Ethos and Experiment

giene and Tropical Medicine , where he started a research group dedicated to ‘Anthropologies and Histories of African Biosciences’ (see http://aab.lshtm.ac.uk). He is also an associate member of the LOST project at the Max Planck Institute for Social Anthropology. He studied first medical zoology in Hamburg and Copenhagen (Ph.D. 1998) and social anthropology in Copenhagen and Cambridge (Ph.D. 2003). Since 1993 he has worked in western Kenya, conducting first medical research and then several years of ethnographic fieldwork, which contributed to the monograph (with Ruth Prince) The Land is Dying (2010). He has also co-edited (with Felicitas Becker) Aids and Religious Practice in Africa (2007). Currently he is conducting a four-year ethnographic study of transnational HIV research conduced by the Kenyan Medical Research Institute and the US government’s Centres for Disease Control and Prevention around Kisumu, Kenya, and he is writing a historical ethnography of postcolonial medical research and scientific collaboration in the same geographical area. Suzette Heald is a Social Anthropologist with extensive field experience in Africa. Her early research concentrated on issues of violence, masculinity and ritual among the Gisu of Uganda (see Controlling Anger: The Anthropology of Gisu Violence ( 1989/9); Manhood and Morality: Sex, Violence and Ritual in Gisu Society (1999)). Recently, her interests have moved into two other areas, HIV/AIDS and vigilantism in East Africa. Work on AIDS began during the two years that she spent teaching at the University of Botswana (1997–1999). Since then, she has also returned to her long-term fieldwork among the Kuria of Kenya, begun in 1985 (see Praise Poems of the Kuria (1997)). Under the aegis of the Crisis States Research Centre of the London School of Economics, she has been doing research on the sunsugungu vigilante movement, tracing its history in Tanzania and its later penetration into Kuria District, Kenya. This has resulted in a series of articles and, most recently, returning to another interest, an ethnographic film titled Law and War in Rural Kenya. See further: www.crisisstates.com/ aboutus/healdbiog.htm. Currently, she holds a fellowship at the Institut d’Études Avancées (IEA) in Nantes. Patrick Kachur is Deputy Chief for Science for the Malaria Branch at the US Centers for Disease Control and Prevention, where he established the Strategic and Applied Sciences Unit. He has conducted public health research and advised intervention programmes in Africa, southeast Asia and Haiti. Originally trained as a Physician at Northeastern Ohio Universities 464

Notes on Contributors

College of Medicine, he received postgraduate training at the University of Ilorin (Nigeria) Faculty of Health Sciences, Columbia University College of Physicians and Surgeons, and Johns Hopkins University School of Hygiene and Public Health with specialised training in community medicine and public health. From 2002 to 2006 he was assigned to a field post at the Ifakara Health Institute in Tanzania where he was Co-Principal Investigator of the Interdisciplinary Monitoring Project for Antimalarial Combination Therapy. Ann Kelly is a Research Fellow with the Anthropologist of African Biosciences (AAB) group at the London School of Hygiene and Tropical Medicine. She holds a Ph.D. in Anthropology from Cambridge University, and since her doctorate has been engaged in research on how the community, polity and society are conceptualised by international public health. Her ethnographic fieldwork in Tanzania, Benin and The Gambia explores the production of scientific facts in Africa, with special attention to the built-environments, material artefacts and practical labours of experimentation. A second focus is on malaria control interventions, how rural and urban spaces are configured to manage the flow of human, parasite and mosquito populations. She is currently developing a project exploring the forms of mobility and genres of memory among municipal health workers in African cities. Guillaume Lachenal is Lecturer in History of Science at the Paris Diderot University, Paris, and a researcher at the CNRS (UMR 7219 group). After studying biology at the Ecole Normale Supérieure in Paris, he received his doctorate in history of science and technology at Paris Diderot University in 2006. His dissertation examined the history of public health and medical research in Cameroon in the context of late-colonialism and decolonisation. His current research explores the history of medical research in Africa, especially in former French colonies, from the post-Second World War development era to the neoliberal transformations of the late twentieth century. Drawing from the methods of microhistory, science studies and historical ethnography, his research takes medical institutions, technologies and discourses as sites of political domination, contestation and imagination. He has authored several articles on the history of science and medical research in Africa and is currently finishing a book project on the biography of Pentamidine, a drug used in the 1950s in Africa during sleeping sickness campaigns . He is coordinating several col465

Evidence, Ethos and Experiment

laborative research projects in Africa, on the history and anthropology of blood transfusion (funded by the French National Agency for Aids Research), on medical research and memory (in collaboration with the London School of Hygiene and Tropical Medicine), and on the historical epidemiology of HIV and hepatitis C virus in Africa. Stacey Langwick, M.P.H., Ph.D., is an assistant professor in the Department of Anthropology at Cornell University. Dr Langwick also holds an associateship with the LOST project at the Max Planck Institute for Social Anthropology, in Germany. Her concern with health and development issues in Africa began in 1992, when she worked with the International Population Council and the Zimbabwean Ministry of Health evaluating maternal and child health programmes. Since 1998, she has conducted ethnographic research on healing practices in Tanzania. Dr Langwick received her Ph.D. from the University of North Carolina, Chapel Hill, and served as an assistant professor of Anthropology and Women’s Studies at the University of Florida from 2002 to 2006. Her research and writing address moments of cultural and scientific translation that occur during encounters between biomedical and indigenous practitioners, technologies, knowledges and medicines. She is interested in diverse (and, at times, seemingly conflicting) claims about the nature of the body and its threats, as well as the ways that these claims come to be coordinated at community, national and global levels. Her book is entitled Bodies, Politics and African Healing: The Matter of Maladies in Tanzania . She is also co-editing a volume entitled Transnational Medicines, Mobile Experts: Globalization, Health and Power in and beyond Africa. Melissa Leach is a Professorial Fellow at the Institute of Development Studies (IDS) Sussex, and Director of the ESRC STEPS Centre. She originally trained as a Geographer (MA Cambridge) and Social Anthropologist (Ph.D. London). Over the last 25 years she has been closely involved both in ethnographic fieldwork, speaking four African languages, and in extensive interdisciplinary research. She has engaged anthropology with historical, ecological and science and technology studies approaches, as well as worked with foresters, agricultural and medical scientists. Her work has forwarded new perspectives in several fields, including science, technology and development; the anthropology of environment and development; gender-environment relations; environmental and forest history and challenging ‘received wisdoms’ about environmental change. Her 466

Notes on Contributors

recent work has explored the politics of science and knowledge in policy processes linked to environment and health, especially vaccination, medical research trials and emerging infectious diseases. With STEPS Centre colleagues, she has recently co-led an international process to produce a ‘New Manifesto’ for Innovation, Sustainability and Development (www.anewmanifesto.org). Her recent book publications include: Dynamic Sustainabilities: Technology, Environment, Social Justice (with I. Scoones and A. Stirling, 2010); Epidemics: Science, Governance and Social Justice (ed. with S. Dry, 2010); Vaccine Anxieties (with James Fairhead, 2007). John Manton is a Leverhulme Research Fellow in the research group Anthropologies and Histories of African Biosciences at the London School of Hygiene and Tropical Medicine. He holds a D.Phil. in history from Oxford. His research deals with the interactions between medical research, clinical practice, and welfare and development in Africa, and he has written on the relations between missionary and state leprosy control services in colonial Nigeria, and on global developments in the treatment of leprosy in the midtwentieth century. Catherine Molyneux, Ph.D., is employed by Oxford University and has been working as part of the KEMRI-Wellcome Trust Research Programme in Kilifi, Kenya, since 1994. She currently co-leads the Social and Behavioural Research (SBR) group in Kilifi. Catherine’s research work has focused on two main areas: the interface between health systems and households, and the interface between health researchers and communities. Her current main focus is on community accountability; producing new thinking, evidence and recommendations around strengthening community involvement in biomedical research and health delivery in sub-Saharan Africa. Despite the prominence of community involvement in policy and practice in this part of the world, there is little recent published experience, particularly regarding community involvement in biomedical research. Strengthening social science capacity in East Africa is a priority of SBR and is built into all activities of the group. Babette Mueller-Rockstroh, a Midwife and Anthropologist of Medical Science and Technology (Ph.D. in Science and Technology Studies, Maastricht), currently holds a post-doc position at the LOST project at the Max Planck Institute for Social Anthropology in Halle, Germany. Her research interests lie with the transfer of medical technology from 467

Evidence, Ethos and Experiment

North-South and vice versa, as well as South-South, in particular with its impact on the field of reproductive healthcare. She has carried out fieldwork on the transfer of imaging technology (obstetric ultrasound) from Europe to Ghana and Tanzania, focusing on the mutual shaping of both technology and the recipient societies. Mueller-Rockstroh is the author of the monograph Ultrasound Travels: The Politics of a Medical Technology in Ghana and Tanzania (2007). Her current book project is about the intertwinement of reproductive healthcare practices with practices and policies of HIV/AIDS management in Mwanza Region, Tanzania. This explores how lay people and healthcare professionals reorganise reproductive health practices in the presence of antiretroviral treatment (ART) and how ART, in turn, may be shaped by locally existing practices to deal with pregnancies, childbirth and child rearing as well as with a desire for children. Vinh-Kim Nguyen is a practicing HIV and Emergency Physician in Montréal, an Associate Professor in the Department of Social and Preventive Medicine at the Université de Montréal, and the author of The Republic of Therapy: Triage and Sovereignty in West Africa’s Time of AIDS (2010), and the co-author of An Anthropology of Biomedicine (2009). He is an associate member of the LOST project at the Max Planck Institute for Social Anthropology. Kenneth S. Ombongi was awarded his Ph.D. in African history at Cambridge University. He holds an MA and M.Phil. from the University of Nairobi, Kenya, and Jawaharlal Nehru University, New Delhi, respectively. He taught history at the Catholic University of Eastern Africa in Nairobi and is currently a Lecturer at the Department of History, University of Nairobi. From December 2003 to April 2004 he was a visiting fellow in the African Studies Centre/Wolfson College, Cambridge University, in which he engaged in research on the urban malaria control policy in Kenya. He is now on leave of absence from the University of Nairobi to serve as the Principal of Kenya Utalii College, a leading hospitality and tourism training institution in Africa. His broad areas of research interest and publications focus on how the state has shaped the understanding and utility of bioscience, medicine and the execution of public health policy in Africa. He is also interested in the dynamics of development and electoral politics in Africa. At the moment Dr Ombongi is a partner in the World Health Organization (TDR) sponsored project on the designing of the ‘Best practice guidance on the deployment of genetically modified mosquitoes as a disease control measure in developing countries’ (MOSQGUIDE Project). 468

Notes on Contributors

Lyn Schumaker has a Ph.D. in history and sociology of science, technology and medicine from the University of Pennsylvania, with previous training in anthropology, history of science and African studies from Michigan State University. Her monograph entitled Africanizing Anthropology: Fieldwork, Networks, and the Making of Cultural Knowledge in Central Africa was published in 2001. She worked at the Wellcome Unit for the History of Medicine, University of Manchester, 1994–2009 and received a two-and-a-half-year University Award in 1999 for research on the history of indigenous, mission and mining company medicine on Zambia’s colonial Copperbelt. Recent work considers antiretroviral therapy in the context of the history of western pharmaceuticals and indigenous medicines in Zambia, as well as African perspectives on privately funded medical philanthropy, such as the Bill and Melinda Gates Foundation’s funding of malaria, tuberculosis and HIV/AIDS research and mining magnate A. Chester Beatty’s funding of early malaria research. Recent articles published in 2008 include ‘Slimes and Death-dealing Dambos: Water, Industry and the Garden City on Zambia’s Copperbelt’, Journal of Southern African Studies 34(4): 8230–40; and, with Virginia Bond, ‘Antiretroviral Therapy in Zambia: Colours, “Spoiling”, “Talk” and the Meaning of Antiretrovirals’, Social Science and Medicine 67(12): 2126–34. Marilyn Strathern, Emeritus Professor of Social Anthropology, Cambridge, has recently been made Life President of the (UK and Commonwealth) Association of Social Anthropologists. Her interests have been divided between Melanesian and British ethnography. She is probably most well known for The Gender of the Gift (1988), a critique of anthropological theories of society and gender relations applied to Melanesia, which she pairs with After Nature: English Kinship in the Late Twentieth Century (1992), a comment on the cultural revolution at home. Her most experimental work is an exercise on the comparative method called Partial Connections (1991). The debates around legislation on human embryology and fertilisation stimulated her interest in reproductive technologies, leading to a collaborative research project (1990–91) that examined some of the issues in the context of kinship. Projects over the last 20 years are reflected in publications on reproductive technologies, and intellectual and cultural property rights, while ‘critique of good practice’ has been the umbrella under which she has written about audit, accountability and interdisciplinarity. Some of these themes are brought together in Kinship, Law and the Unexpected (2005). She currently chairs the Nuffield Council of Bioethics’ Working Party on Human Bodies in Medicine and Research. 469

Evidence, Ethos and Experiment

George Ulrich has been Rector of the Riga Graduate School of Law since August 2009. Prior to this, he served as Secretary General of the European Inter-university Centre for Human Rights and Democratisation (EIUC, based in Venice, Italy) from 2003 to 2009, and as Academic Coordinator and Programme Director of the European Master’s Degree in Human Rights and Democratisation from 2001 to 2004. From 1999 to 2001 he was employed as Senior Researcher at the Danish Centre for Human Rights. And from 1996 to 1998 as Research Fellow at the Institute of Anthropology, University of Copenhagen, and visiting researcher at Makerere University, Kampala, Uganda. In the latter context he conducted a comprehensive study, assigned by the Danish development cooperation agency, Danida, on The Ethics of Donor-sponsored Research in Developing Countries. He obtained his Ph.D. as well as an M.A. in philosophy from the University of Toronto, Canada, and holds the degree of Cand. Mag. in social anthropology and history of ideas from Aarhus University, Denmark. Among his current research interests are issues related to: the philosophy of human rights, health and human rights, international medical ethics, professional ethics and ethics for human rights professionals. Luise White is Professor of History at the University of Florida. She is the author of The Comforts of Home: Prostitution in Colonial Nairobi (1990), Speaking with Vampires: Rumor and History in Colonial Africa (2000) and the Assassination of Herbert Chitepo: Texts and Politics in Zimbabwe (2003), and co-editor, with David William Cohen and Stephan Miescher, of African Words, African Voices: Critical Practices in Oral History (2001) and, with Douglas Howland, of The State of Sovereignty: Territories, Laws, Populations (2009). Susan Reynolds Whyte, Professor at the Department of Anthropology, University of Copenhagen, has carried out fieldwork in Tanzania, Kenya, and especially Uganda, where she has been working intermittently since 1969. For 14 years she had responsibility for a long-term collaboration with Makerere University, supporting the training of Ugandan Ph.D. and Masters students, and engaging in joint research projects. Presently she is participating in two other enhancements of research-capacity programmes in Uganda. Her research falls mainly in the areas of family, generation and gender, misfortune, health, disability, medicines and transformations in local moral economies. She has co-edited four books: The Social Context of Medicines in Developing Countries (1988), Disability and Culture (1995), 470

Notes on Contributors

Disability in Local and Global Worlds (2007), and Generations in Africa: Connections and Conflicts (2008), written a monograph Questioning Misfortune: The Pragmatics of Uncertainty in Eastern Uganda (1997), and co-authored Social Lives of Medicines (2003).

471

Index

aboriginals see first-nations peoples access to healthcare HIV/AIDS treatments 22, 333, 431–432, 436–438 by participation in medical research 43–44, 236–237, 441 accountability 149 and audit cultures, criticism of 146, 147–150 ethical 146–147, 152–154, 167 obstacles to realization of 157, 158–162 and ethics of medical research 145–146 legal 151 situated 13, 159, 166 Achmat, Zackie 437 ACTs see antimalarial combination treatments Adalja, K.V. 369 n.9 Africa biopolitics in 26 n.16 environment of, blamed for malaria 406 healthcare in 24 n.9, 34, 184–185, 187–188, 342 Africanization of 361–362 costs of 350 n.14 HIV/AIDS treatment and care 333 networks of care 178, 194 n.7 postcolonial 450 privatization of 103, 104 ultrasound technology use 246–248 variolation practices in 448–449 medical aid to 189 medical research in 1–2, 25 n.15, 173, 175–176, 257, 299–300 clinical trials 14, 299–300, 441, 442, 443

of ART 432–433 in colonial era 20, 357–360, 367, 421, 448 and ethics 445–446, 458 imperialist tendencies in 1, 22, 354–355, 356, 357, 359, 360, 367 on medicinal plants 264, 266, 268, 272, 281–284 NGOs/non-state actors in 7, 365– 367, 368, 370 n.20 Pasteur Institutes’ role in 373, 390 postcolonial 363–364 RCTs 258, 412 rumours associated with 159 nationalism in 361, 362, 363–364, 455 neoliberalism in 13, 80–81, 298 postcolonial relations with France 373–374, 394 public health in 354 in colonial era 20, 356–357, 360, 447–448 garden city ideas 413–414 measles vaccinations 450–451 postcolonial 13, 20, 21, 365–367, 450 smallpox eradication campaign 452–456, 457, 458 typhoid vaccinations 456–457 social sciences in 42 state in colonial 358 crisis/weakness of 301, 353 and public health policies 13, 20, 365–367 and smallpox eradication campaign 452–453, 455–456 postcolonial 361 see also Africans; East Africa The African Civilizations (Thairu) 364

473

Index

African Union (AU), Decade of Traditional Medicine (2001–2010) 265 Africanization and cultural stereotypes 388–389 of healthcare 361–362 of medical research 363, 364 Africans avoidance of vaccinations by 446 doctors 193, 361–362 evidence-based medicine practised by 14, 172–173, 178 information needs of 175 insecurities of 173, 174 low-tech medicine criticised by 14–15, 24 n.10, 172 medical research contributions by 17, 30–32, 43, 44–47, 50, 172, 191, 201, 202 training of 192, 361–362, 379 miners 404, 414 communications with management 415–416 researchers 42, 43 urban living believed to be inappropriate for 411 views of protective capacities of vaccinations by 449, 456–457, 458 agency of ancestors, in traditional healing practices 288 of women 93–94 agricultural practices, endangering malaria control 417, 418, 420–421 Ahidjo, Ahmadou 382 Akerele, O. 291 n. 16 Allday, Elizabeth 131, 139–140 Amayo, Erastus 192 American Anthropological Association (AAA), ethical code of 111, 154–155 AMREF (African Medical and Research Foundation) 366–367 ancestors, role in traditional healing practices 288 Angell, Marcia 161 animals, testing of drugs on 129

anthropology by colonial administrators 411 engaged 119 n.1 field research in ethics of 11–12, 99–101, 111, 117, 119 n.2, 120 n.6, 154–155 consent procedures 105, 109, 111, 120–121 n.10 and epistemologies of illness and healing 111–112, 113–115, 118 HIV/AIDS research 99–103, 105–106, 107–110, 113–115, 116–118, 119–120 n.3, n.4 fieldnote records 32 knowledge production by 118–119 social relations in 11, 13–14 medical research contributions by 6–7 on ethics of medical research 2–3, 11, 146, 165 on transformational capacities of knowledge 73 n.14 see also ethnography antimalarial combination treatments (ACTs) research into 25 n.14, 203–218 use of 211–218, 220, 221 antiretroviral medicines/treatment (ARV/ART) in Africa access to 22, 436–438 clinical trials 432–433 effects of 440 in Kenya 336, 343, 349 n.7 in Uganda 35, 350 n.14 Antoine, Abondo 274 Antweiler, C. 119 n.1 Anyanya Resistance Movement (Southern Sudan), participation in smallpox eradication campaign 454–455 applied research 119 n.1 appropriate technology approaches 24 n.9, 172, 190, 193 n.3, 245 see also low-tech medicine arboviruses, research into 390 474

Index

Arendt, Hannah 396 n.3 artesunate (AS, antimalaria medicine), research into 208–210 artimisinin 204 ARV/ART see antiretroviral medicines/treatment atebrin (antimalaria medicine), research into 407, 412, 424 n.8 audit cultures 145 criticism of 146, 147–150 Audit Cultures (Strathern) 147–148, 149, 153 autonomy 151 of research 234, 235 awareness of HIV/AIDS in Botswana 350 n.11 in Kenya 337–338 prevention methods 340, 342 Awori, Nelson 363, 364 B.283 clinical trials of leprosy and tuberculosis treatment with 132–135, 138 discovery of 129–130 B.663 see clofazimine Bambali (Gambia) 229–231 Barnes, Joseph 131, 132–134, 139–140 Barry, Vincent 128–129, 130, 135, 137, 141 n.8 Bayer pharmaceutical company 424 n.8 Baylies, C. 121 n.15 Beatty, A. Chester 405, 406 Beck, A. 356 beer brewing, illicit 417, 424–425 n.18 beerhalls, as a means to control miners 416 being, opposed to knowing 36, 63 beliefs/theories of causation, and information provision 57 benefits of medical research 175, 188, 235–236, 239–240, 241–242 for participants in medical research 237–238, 239–240, 241 Benjamin, Walter 172 Berg, M. 291 n.18

Berguer, R. 201 Bhat, Dr 273 Biafra conflict (Nigeria), and smallpox eradication campaign 453, 459 n.10 bilirubin levels in newborns, research into 180–181 bioavailability 442, 443 bioethics see ethics, medical/of medical research biological citizenship 440, 441 biomedical policies see public health, policies biopolitics, African 26 n.16 biopower 443 birth asphyxia, use of term 183–184 birth certificates 38 births spacing of 86–87 use of ultrasound scans after 251–252 blood donations/tests eagerness to participate in 46–47 resistance to/distrust of 84, 87, 89, 90, 92–93, 236, 243 n.13, 446 fieldworkers dealing with 94–95 bodies 65–70 body parasites, research into 297–298, 305–310, 329 n.8, n.10–n.14 Botswana, HIV/AIDS awareness in 350 n.11 boundaries, destabilisation of 6 Bradford Hill, Anthony 232, 233, 234 breastfeeding practices, in Gambia 86–87 Brottes, Henri 380, 388 Browne, Stanley 126, 127, 136, 137, 138, 142 n.25 Burkitt, Dr 369 n.6 Cameroon independence of 378 Pasteur Institute/Centre Pasteur of (IPC/CPC) 375–378, 383, 384–385 Cameroonization of 21, 377, 378–380, 382–384, 385, 386 and stereotypes of Cameroonian culture 388–389 475

Index

Etoundi controversy at 380–382, 386–388, 391 French expatriates at 377, 380, 382–383, 384, 385 morality/Cameroonization of 393–394 professional hierarchies at 377, 385, 392–393 research activities of 373, 390 capacity building in collaborative medical research 374 ethical 162 career mobility of African researchers 43 of fieldworkers 45–46, 330 n.16 Carothers, J.-C. 388–389 Cassell, Jackie 97 n.1 causation, beliefs/theories of, and information provision 57 CBOs see community-based organizations CDC (Centers for Disease Control, United States) 164 HIV/AIDS research in Africa by 432–433 involvement in smallpox eradication campaign in Africa 447, 450–451, 452, 453, 455, 459 Tanzanian medical research project funded by 212 cerebral palsy, causes of 176 Chafer, Tony 396 n. 2 Chambon, Louis 381, 382–383 Chang, Y.T. 136 Charlton, B.G. 202 chaulmoogric acid 129 childcare in Gambia 85–88 influence of infant cohort studies on 78 Children’s Hospital of Philadelphia (CHOP), care for newborns in 177, 193 n.5, 194 n.6 Christianity of Kenyan fieldworkers 312, 319 sexual morality ideas in, and HIV/AIDS 340, 341–342

Christophers, S.R. 423 n.3 citizenship biological 440, 441 genetic 441 in Kenya 309–310, 317 therapeutic 78, 88–89, 96, 103–104, 438, 440, 441 Clifford, J. 32, 118 clinical practices medical technology influences on 257 of traditional healers 271 clinical practitioners see doctors clinical trials in Africa 14, 299–300, 441, 442, 443 of ART 432–433 Good Clinical Practice (GCP) in 23 n.2 of leprosy treatments 126, 127, 132–135, 136–137, 138–140 of malaria vaccinations 229–231, 236–237 benefits of 239–240, 241 end of 238 participation in 11 relations with researchers 139–140 pragmatic/localisation of 15–17, 202, 231 of tuberculosis treatments 132, 134–135 see also randomised controlled trials clofazimine (B.663), development and testing as leprosy treatment drug 125–126, 127, 135–137, 138, 139 co-authorship of scientific papers, by traditional healers 278, 284 Cochrane, Archie 243 n.10 Cochrane, Robert 133 coercion mechanisms of rule compliance 151 in smallpox vaccinations 456 Coetzee, A.J.P. 417 Cohen, Lawrence 442 Cohen, Patricia 167 n.4 collaboration between African and Western researchers 15, 21, 24 n.7, 190–191, 299, 374–375 476

Index

non-state actors in 365–367, 368, 370 n.20 between researchers and healers 267, 270, 273, 274, 277, 278–280, 289–290 in HIV/AIDS research and treatment 285–288 Collier, S.J. 7 colonial era hierarchies from, continued in postcolonial settings 391 medical research in 20, 357–360, 367, 421, 448 public health in 20, 356–357, 360, 447–448 garden city ideas 413–414 malaria control programmes in 403, 404–410 colonialism see imperialism communication of African miners with management 415–416 domination-free 160 as flows and blockages of substances 69–70 on HIV/AIDS, local 100 in medical research 57, 84, 95 communities anthropological studies of 6 knowledge owned by 276–277 therapeutic 429, 434 in HIV/AIDS care 439–440, 443 leprosy care 429–430 reproduction of 443 trial 6–7, 24 n.5, 231 knowledge gained by 10 community-based organizations for HIV/AIDS care 429, 438–439 access to treatment through 436–438 confessional counselling techniques used by 431, 434–436 and corruption in Kenya 335 complementarity, of traditional medicine and biomedicine 275, 277–280 compliance, coercive mechanisms of 151 condoms, mistrust and suspicion of 350 n15

confessional technologies access to HIV/AIDS treatment through abilities in 437–438 in HIV/AIDS counselling 431, 434–436 confidentiality issues in HIV/AIDS research 105–106, 117, 121 n.15 in HIV/AIDS testing and diagnosis 110, 113, 114–116, 121 n.14, 344, 347 see also disclosure of HIV/AIDS positive status consent procedures in anthropological field research 105, 109, 111, 120–121 n.10 in medical research 35–36, 83, 93–94, 163, 220–221, 222 contexts creation of 73 n.11 of knowledge 65 of traditional medicines 281, 289 and perspectives 66 significance of 59, 60 social, of medical research 11, 58–59, 162–165 of technology 259, 260, 261 coopération 373, 396 n.2 in Cameroon 378–379 end of 394–395 in medical research projects of 374–375 renegotiation of 382 Copperbelt (Zambia) malaria control programmes in 403–404, 422–423 by mining companies 404–410, 420 malaria research in 417–418 mining companies in 410, 411–412 deaths in 405, 407–408, 409, 424 n.11 racial segregation in housing of miners at 414–415 welfare services provided by 410–411, 412, 414 worker control at 411, 414, 415, 417 corruption, in Kenya 335, 365

477

Index

counselling confessional techniques used in 431, 434–436 access to treatment through abilities in 437–438 mandatory element of VCT 347 culturalism 388 cultures audit 145, 146, 147–150 research 421, 422 stereotypes of 388–389 DALY system (Disability Adjusted Life Years), use of 187–188 Dalzell, A.C. 407, 409, 412 Dapsone treatment for leprosy 125, 131, 137 DDT 329, 410, 429 De Cock, K. 333, 348 deaths AIDS-related 338–339 anthropological research on, implications for researcher 113 foetal, ultrasound technology used for diagnosis of 252 neonatal 86, 180 witchcraft blamed for 338–339, 340, 350 n.11 in Zambian coppermines 405, 407–408, 409, 424 n.11 democratic public sphere, erosion of 12–13, 24 n. 6 The Design of Experiments (Fisher) 232 developing countries applicability of medical research results from industrialised countries in 180–181 healthcare in 189–190 care for newborns 178–180, 183–184 technology needs 177–178 research/medical research in 52–53, 172–173 ethics of 147, 158–159, 161–162, 166–167, 190–191 local knowledge needs in 176, 178–180, 181, 185–187

need for integration with local social processes 162–165 Nuffield Council on Bioethics on 58–59, 72 n.2 political economy of 7 see also Africa, medical research in development aid integration of HIV/AIDS into 430 medical, priorities in 189 medical research being similar to 52 see also coopération Dewey, John 41, 55n .4 diagnoses of ectopic pregnancies 250–251 of leprosy 133 of malaria 409 Digoutte, Jean-Pierre 383 Dilger, Hansjörg 11–12, 463 diploicin 129 disability, measurement of burden of 187–188 disclosure of HIV/AIDS positive status 433–434 disease see illness Disease Control Priorities (World Bank) 189 divination and magic 35, 408 doctors African 193, 361–362 evidence-based medicine practised by 14, 172–173, 178 information needs of 175 insecurities of 173, 174 low-tech medicine criticised by 14–15, 24 n.10, 172 medical research contributions by 17, 30–32, 43, 44–47, 50, 172, 191, 201, 202 training of 192, 361–362, 379 American, insecurities of 173–174 French, at Pasteur Institutes in Africa 389–390 Doumbo, O. 59, 64 Dozon, Jean-Pierre 396 n.3 drainage, as malaria control measure 407, 408, 409

478

Index

DVDB (Division of Vector Borne Diseases, Kenya) 297–298, 325 in Kisumu research activities of 305–310, 329 n.8 resources of 298–299, 300–301, 307 staff of 302–305, 310–326 earth, eating of, and illness 253–254 East Africa HIV/AIDS in, anthropological field research on 99–103 smallpox in 458 n.3 campaign for eradication of 457 Eastern Africa Network on Medicinal Plants and Traditional Medicine 265, 275, 279 ectopic pregnancies, diagnosis of 250–251 education of fieldworkers 310–312 value of 322 see also training effectiveness of ARV 440 of malaria control programmes 407, 420 studies 26 n.16, 199, 243 n.10 Effectiveness and Efficiency (Archie Cochrane) 243 n.10 elites, local, bypassing of 24–25 n.10 Elmi, Prof. 273 emergent ethics notion 146 engaged anthropology 119 n.1 epidemiology 185–186, 202 epistemologies of illness and healing, and ethics of anthropological field research 111–112, 118 equality see inequalities ethical accountability 146–147, 152–154, 167 enforcement of 155–156 obstacles to realization of 157, 158–162 ethics 17, 151–152, 259 of anthropological/ethnographic field research 11–12, 99–101, 111, 117, 119, n.2, 120 n.6, 154–155

consent procedures 105, 109, 111, 120–121 n.10 and epistemologies of illness and healing 111–112, 113–115, 118 on HIV/AIDS 99–103, 105–106, 107–110, 113–115, 116–118, 119–120 n.3, n.4 emergent 146 of intimacy/familiarity in French postcolonial relations 375 medical/of medical research 2–3, 5, 13, 23 n.2, 156, 235, 243 n.15 and accountability 145–146 in Africa 445–446, 458 anthropological contributions to 2–3, 11, 146, 165 contextual issues 11, 58–59, 162–165 debates about 12, 23 n.1, 96 in developing countries 147, 158–159, 161–162, 166–167, 190–191 on HIV/AIDS 102, 190–191 local peoples’ attitudes towards 159, 344, 348–349 and power relations 22–23, 160–162 of randomization of treatment 234 privatised 12, 149–150, 151 professional audit cultures applied to 147–148 codes of 154–155, 168 n.7, 235 taxonomy of categories of ethical issues 155–156 public 12–13, 146, 150, 152, 154, 156–157 unhinged 157–165 see also consent procedures ethnic clashes, in Kenya 330 n.17 ethnobotanical research 281, 282–283 ethnography critical 8–9 historical 375 of medical research 257 research contribution to medical research 53, 54, 96

479

Index

quasi-kins used in 40 in Sukuta (Gambia) 79 see also anthropology ethos 3–4, 24 n.3 and politics 5–6 of science/medical research 3, 4, 5 field research 306–308, 313, 329 n.13 in French coopération projects 375 Etoundi (Prosper Etoundi Essomba Ntsama) 379–380, 383–384, 390, 397 n.20 controversy over 380–382, 386–388, 391 eugenics studies, in Africa 357–358 evangelical idioms, used in HIV counselling 434–435 evidence in medical research/evidencebased medicine 14, 15, 16, 18, 42–43, 171, 197, 197–198, 235, 243 n.10 in developing countries 172, 173 hierarchies of evidence 200 hybrid forms of evidence 289 in public health 202–203 RCT approach used in 200, 235 statistics used for 234 traditional medicine integrated into 276, 288–289, 291 n.18 in United States 14, 173–174 universal standards for 14, 25 n.15 exceptionalism, of AIDS 19, 333, 348 exchange transfusions for newborns, dangers of 176, 180–181 experiments quasi 199 versus observational studies 15, 25 n.12, 200–202, 222 see also clinical trials; randomised controlled trials failed states 328 n.4 Fairhead, James 11, 72 n.6, 97 n.1, 237 families, professional 313–316, 317–318, 329 n.15 Farmer, Paul 100, 116, 193 n.3 Feierman, Steven 14–15, 24 n.10, 256, 257, 463

Ferguson, J. 7, 24 n.9 field research/field work anthropological/ethnographic centrality of social relations in 11, 13–14 consent procedures 105, 109, 111, 120–121 n.10 ethics of 11–12, 99–101, 111, 117, 119 n.2, 120 n.6, 154–155 and epistemologies of illness and healing 111–112, 118 of HIV/AIDS research 99–103, 105–106, 107–110, 113–115, 116–118, 119–120 n.3, n.4 knowledge production by 118–119 as male domain 321 medical 41 heroics/ethos of 306–308, 313, 329 n.13 on parasitic diseases 305–310, 329 n.8, n.10-n.14 writings of 32, 33, 39–40 fieldworkers in medical research career mobility of 45–46, 329 n.16 life stories of 302–305, 307–311, 318–319, 330 n.18 in malaria control programmes 410, 418–420, 422 perspectives of 83–85, 94–95 recognition/acknowledgment of 50–52, 53–54, 55 n.6 recruitment and training of 45–46, 310–312, 314–315, 316–317 social relations with local people 309–310, 317 financial accounting, idiom of 148 first-nations peoples, perspectives on human remains in museum collections 62–63, 73 n.9 Fisher, Ronald Aylmer 232, 233–234, 235 fluids see liquids focus-group discussions 242 n.1 foetal death, ultrasound technology used for diagnosis of 252 forwardness, notions of 327, 328 Foucault, M. 442 480

Index

fractal persons 69, 74 n.18, n.19 France expatriates at Pasteur Institutes in Africa from 389–390, 394–395 in Cameroon 377, 380, 382–383, 384, 385 morality/Cameroonization of 393–394 foreign policy in postcolonial Africa 373–374, 394 see also coopération Frank, Dr see Kaharuza, Frank Freeman, Denis 132–133, 135, 139–140 Freeman, Edward 134 Full Gospel Bible Fellowship Church (FGBFC, Dar es Salaam), HIV/AIDS approaches of 106 funeral books 36 Gambia 78 childcare practices in 85–88 ethnographic research in 79 healthcare in 80, 81, 82 medical research in infant cohort studies 78–79, 81–85 malaria vaccine trials 229–231, 236–237 benefits of 239–240, 241 end of 238 political and economic situation in 80–81 Gamet, André 378–379 garden cities/Garden City movement 413, 424 n.9 application in Africa of 413–414, 419 GCP see Good Clinical Practice Geddes, Patrick 413 Geigy (pharmaceutical company) 135–136, 137 Geissler, Paul Wenzel 18–19, 34, 40, 49, 54 n.1, 73 n.14, 159, 230, 254, 271, 420, 463–464 gender roles, in Kenya 319, 321 generalisability, of medical research results 198–199, 219 generation 302

and growth 330 n.21 of science 317–318 genetic citizenship 441 Gerrets, Rene 291 n.21 Ghana healthcare in 188 care for newborns 180 Giddens, Anthony 163 Giri, Ananta 149, 150 global ethical standards 157 global public goods 175, 188–189 Good Clinical Practice (GCP) 23 n.2 good technology 17–18 Goodman, Nelson 256 government housing of employees of 321–322 and science 19, 318, 322–323, 327–328, 330 n.21 in medical research 7, 20, 49, 168–169 n.12 see also state government-citizen bonds, in Kenya 309–310 greeting conventions, in Uganda 40 growth 325–326, 327 and generation 330 n.21 Guinea-Bissau, smallpox eradication campaign in 455 Gupta, A. 7 Habermas, Jürgen 151–152, 160, 168 n.8 Habicht, J.P. 203 Hacking, Ian 168 n.10 Harare Central Hospital, care for newborns in 177, 194 n.6 Harrington, J.A. 120 n.7 Harrison, C.R. 407, 409, 421 Harvey, D. 24 n.6 Headrick, D.R. 356, 369 n.4 Heald, Suzette 19, 350 n.16, 464 healing prayers 107 touch in 271 traditional see traditional medicine/healing writing as component of 34–35 481

Index

healthcare in Africa 24 n.9, 34, 184–185, 187–188, 342 Africanization of 361–362 in colonial era 356 costs of 350 n.14 networks of care 178, 194 n.7 postcolonial 450 privatization of 103, 104 ultrasound technology use 246–248 variolation practices in 448–449 in developing countries 189–190 care for newborns 178–180, 183–184 technology needs 177–178 and medical research access to healthcare for medical research participants 236–237, 239–240, 441 integration of facilities 81–82, 91 participants perceiving research as healthcare provision (therapeutic misconception) 34, 43–44, 46–48, 77–78, 91–92, 93, 95–96, 238–239, 243 n.14 touch in 49–50, 55 n.5, 271 traditional healers/medicine recognised/integrated in 264, 266–267, 268–281, 289, 290 n.14, 364 writings in 34–35, 37–38 see also evidence in medical research/evidence-based medicine healthworkers relations with rural populations 49–50 traditional healers, role/position of 266–267, 268–280, 290 n.14 training of 214 Hecht, Gabrielle 377 Helsinki Declaration of Ethical Principles for Medical Research involving Human Subjects 243 n.15 Henderson, Ralph 451

Herbert, Eugenia 448 herbs see plants hierarchies colonial, continued in postcolonial settings 391 of evidence in medical research 200 flattening of, in HIV/AIDS therapeutic communities 440 of knowledge 289 professional at Pasteur Institute (Cameroon) 377, 385, 392–393 in medical collaborative research 374–375 among medical fieldworkers 315–316 racialization of 391–393 Higenyi, Pabire 36 history ethnographic 375 of HIV/AIDS epidemic 430–432 medical research in Africa shaped by 1 of pharmacological research 12, 126–127, 132 of public health 18–19, 20–21 HIV/AIDS activism 103–104, 434, 439, 440 approaches towards human rights based 19, 103, 105, 348, 431, 433–434 as humanitarian emergency 432 of neo-Pentecostalism 106–108 of NGOs 103, 105, 106, 111–112, 121 n.11, 346, 350 n.18 awareness of 337–338, 340, 342, 350, n.11 counselling and testing 35, 55 n.5, 109, 121 n.13, 432 confessional technologies used in 431, 434–436 disclosure of positive status 433–434 by healers 287–288 secrecy/silence around positive HIV/AIDS status 100, 110, 112–113, 114–116, 121 n.14, 344, 347 482

Index

voluntary (VCT) 333–334 availability of 335, 343 consent procedures in 35 mandatory counselling element of 347 resistance to/mistrust of 334, 343–346, 348 training of staff involved in 346–347 deaths related to 338–339 exceptionalism of 19, 333, 348 history of epidemic 430–432 in Kenya 334–338, 349 n.2, n.3, n.9 prevention 334 awareness of methods of 340, 342 neglected in Kenyan policies 335 and VCT procedures 334, 343–346, 348 research in developing countries/ Africa on 429, 432–433 anthropological, ethics of 99–103, 105–106, 107–110, 113–115, 116–118, 119–120 n.3, n.4 to control mother to child transmission 190–191, 432–433 disqualification of participants in 433 ethical criticism of 158–159, 161 treatment and care of access to treatment 22, 333, 431–432, 436–438 community-based organizations involved in 335, 429, 438–439 in Kenya 334, 335, 342–343 in Tanzania 104–105, 106, 112–113, 121 n.12, 285–288 therapeutic communities/ citizenship in 103–104, 439–440, 443 traditional medicine in 106, 121 n.12, 274, 285–288, 342–343 Hoeyer, K. 36 Hogerzeil, Lykle 126, 127, 136 hosting visitors 40 housing of government employees 321–322

of miners, racial segregation in 410, 412, 413–415 human remains in museum collections, disputes about 10, 61–63, 72 n.8, 73 n.9 human rights and anthropological research 118 HIV/AIDS approaches based on 19, 103, 105, 348, 431, 433–434 humanitarian emergencies, HIV/AIDS as 432 Hunt, N.R. 37–38 Hutchinson, Sharon 29, 39 identities, multiple 68 idioms evangelical 434–435 of financial accounting 148 Ifakara Health Institute (IHI, Tanzania) 212–213 Iliffe, John 192, 361 illness metaphors used for 405, 408 perceptions/notions of 447, 457 of childhood illnesses 87 flow/blockage symbolism in 70, 87 on HIV/AIDS 111–112, 113–115, 118 in Tanzania 253–254 speaking about 113 Imkamp, F. 137 Imperato, James Pascal 448 imperialism ethical 161 in research activities in Africa 1, 22, 304–305, 354–355, 356, 357, 359, 360, 367 IMPM (Institut de Recherche Médical et des Plantes Médicinales, Cameroon) 382 incorporation, forms of 68 individual rights, and public health interests 5, 19–20, 348 industrialised countries evidence-based medicine in 171, 172 medical research in 176

483

Index

applicability in developing countries 180–181 inequalities in collaborative research 2, 21, 189, 379, 382, 387, 394 infant cohort studies, in Gambia 78–79, 81–85 information epidemiological, need for 185–186 evaluation of, as assessment of technology 257 needs of African doctors 175 provided to participants in medical research 57, 60–61, 64 by fieldworkers 84, 95 sources of, traditional healers as 282 informed consent see consent procedures insecticides, used in malaria control programmes 410, 419–420, 421 insecurity, medical, of African and American doctors 173–174 Institute of Traditional Medicine (Tanzania) 265, 267, 276, 289 research activities of 280–285 intellectual property rights, of traditional healers 278, 286–287 interaction, ethical 157 interests, and perspectives 61 International Conference on Traditional Medicinal Plants (Tanzania, 1990) 268–274 Ireland Medical Research Council of (MRCI) 139 research and development of leprosy treatment by 125–126, 130, 135, 137 tuberculosis research by 128–130, 132, 141 n.10 tuberculosis in 127–128 Islam, healing practices in, writing as component of 35 James, William 41 jina see spirits

Kachur, S. Patrick 15–16, 25 n.11, n.14, 26 n.16, 191, 464–465 Kaharuza, Frank (Dr Frank) 30–32, 43, 44–47, 50, 51, 52, 54–55 n.2 Kakobe, Zachary 106 Kaleebi, Noerine 99 Kalitzkus, Vera 113 Kambarami, R.A. 194 n.6 Kant, Immanuel 151 Kasili, Edward George 363, 364 Katahoire, Anne 49 Kayombo, Edmund 263, 266, 279, 282, 285, 287, 289, 290 Kelly, Ann 15, 16, 230, 465 KEMRI (Kenya Medical Research Institute) 298, 362, 364, 366 Kenya citizenship in 309–310, 317 corruption in 335, 365 ethnic clashes in 330 n.17 gender roles in 319, 321 healthcare/public health policies in 304, 353–354, 354–355, 356 Africanization of 361–362 in colonial era 355–357, 367 racial segregation for health reasons in 356, 359–360 postcolonial 361–364, 362–363 and weakening state 365, 365–367, 368 HIV/AIDS in 334–337, 349 n.2, n.3, n.9 awareness of 337–338 treatment availability 334, 335 VCT in availability of 335 resistance to/mistrust of 334, 343–346 training of staff involved in 346–347 independence of 305, 355, 361 malaria in, colonial policies on control of 358–360 medical research in 168 n.12, 304, 370 n.21 in colonial era 357–360 484

Index

on parasitic diseases 297–298, 305–310, 329 n.8, n.10–n.14 and postcolonial state 353 recruitment and training of fieldworkers in 310–311, 314–315, 316–317 modernity in 304–305, 362 youth unemployment in 322 Kenya and East Africa Medical Journal (KEAMI) 360 Kenyatta, Jomo 361, 362 kernicterus, causes of 176–177 Kibaki, Emilio Mwai 349 n.2, 364 Kilifi District (Kenya), medical research in 366 kinship relations, among medical fieldworkers 314 Kisumu (Kenya) District hospital 297 DVDB in research activities of 305–310, 329 n.8 resources of 298–299, 300–301, 307 staff of 302–305, 310–326 knowing different ways of 60 opposed to being 36, 63–64 knowledge 63, 70, 71 local/traditional medical 174–175, 270–271, 276–277 de-and re-contextualizing of 281, 289 on HIV/AIDS 338 need for 14, 15–17, 176, 178–180, 181, 185–187 protection of 277–279 use of 183 versus global public goods 189 need for, about HIV/AIDS status 346 practices 58 production of 42, 53, 326–327 in anthropology 118–119 in medical research 61, 281, 289 role of paper in 29 and social relations 10, 54 transmission of 67, 73 n.14, 309

types of 9–10, 30 Koumare, Prof. 270, 274 Kuria District (Kenya) healthcare in 336, 349 n.6 HIV/AIDS in 349 n.9 and charismatic Christianity/ Pentecostalism 340–341 research into knowledge and opinion of 337–342 testing and treatment availability 335–336, 343, 349 n.7 laboratories mining companies as 412, 417, 421–422 as sites of innovation for medicinal plant research 284 labour relations, in Zambian copper mines 411–412 labour shortages, at Zambian copper mines 410 Lachenal, Guillaume 21, 465–466 Lamba people (Zambia), refusal to work in mines 416–417 Lamprene see clofazimine Langwick, Stacey A. 18, 466 Latour, Bruno 257 law, coercive aspects of 151 Law, J. 66, 69, 72 n.3, 73 n.11, n.13 Leach, Melissa 11, 72 n.6, 73 n.14, 97 n.1, 237, 466–467 legal accountability 151 leprosy 139 care 429–430 diagnoses of 133 medicines for treatment of, development and testing of 125–126, 127, 130–137, 138–140, 141 n.23 Lewis, C.I. 150 life stories, of researchers/fieldworkers 302–305, 307–311, 318–319, 330 n.18 liquids 69–70, 74 n.20 localisation, of medicine/medical research 14, 15–17, 176, 178–180, 181, 185–187, 202, 231

485

Index

Long, Norman 52–53 looping effects 163, 168 n.10 low-tech medicine 16 criticism of 14, 24–25 n.10 see also appropriate technology approaches Luanshya river (Zambia) 409 Luanshya town (Zambia) 407, 424 n.16 beerhalls in 416 Lysenko, 233 magic see divination and magic magnification processes 70 Mahunnah, R.L.A. 276, 278, 279, 286–287 malaria diagnosis of 409 research and control of 21 ACT research 25 n.14, 203–218 in colonial era 358–360, 367, 403–404, 417–418, 421, 422 fieldworkers in 410, 418–420, 422 by mining companies 21, 404–410 effectiveness/success of 407, 420 local agricultural practices as obstacle to 417, 418, 420–421 racial segregation methods 420, 423 n.3 Ross method/species sanitation 405, 408–409, 412, 420, 423 n.3 in Sierra Leone 405, 423 n.3 vaccine trials in Gambia 229–231, 236–237 benefits of 239–240, 241 end of 238 treatment of 187, 409 with ACTs 211–218, 220, 221 Malawi, smallpox eradication campaign in 455 Manton, John 12, 467 Mapundo, Dr 273 Marcus, G.E. 118 Marks, H.M. 243 n.6 marriages, of Kenyan fieldworkers 319, 330 n.19

Marsden, P.D. 81 Mbembe, Achille 387–388 Mbidde, Edward 161 Mboya, Tom 364 Mbwambo, Pastor 107 measles, vaccination against, trials in Africa 450–451 medical records, importance of 36, 37 medical research 41 in Africa 1–2, 25 n.15, 173, 175–176, 257, 299–300 in colonial era 20, 357–360, 367, 421, 448 imperialist tendencies in 1, 22, 354–355, 356, 357, 359, 360, 367 NGOs/non-state actors in 7, 365–367, 368, 370 n.20 Pasteur Institutes’ role in 373 postcolonial 363–364 anthropological/ethnographic contributions to 6–7, 53, 54, 96 on ethics of medical research 2–3, 11, 146, 165 benefits of 175, 188, 235–236, 239–240, 241–242 collaborative 15, 21, 24 n.7, 190–191, 299, 374–375 inequalities in 2, 21, 189, 379, 382, 387, 394 non-state actors in 365–367, 368, 370 n.20 communication in 57, 84, 95 in developing countries 52–53, 172–173 on HIV/AIDS 158–159 integration with local social processes 162–165 Nuffield Council on Bioethics on 58–59, 72 n.2 political economy of 7 by doctors/clinical practitioners 17, 30–32, 43, 44–47, 50, 172, 191, 201, 202 end of 51–52, 238, 240 ethics of 2–3, 5, 23 n.2, 235, 243 n.15 and accountability 145–146

486

Index

anthropological contributions to 2–3, 11, 146, 165 contextual issues 11, 58–59, 162–165 debates about 12, 23 n.1, 96 in HIV/AIDS research 102, 190–191 and local peoples’ attitudes towards 159, 344, 348–349 and power relations 22–23, 161–162 of randomization of treatment 234 ethos of 3, 4, 5, 375 evidence in 14–15, 16, 18, 42–43, 171, 235 hierarchy of 200 hybrid forms of 289 statistics used for 234 universal standards for 14, 25 n.15 generalisability of results in 198–199, 219 and government/state 7, 20, 49, 168–169 n.12 and healthcare access to healthcare for participants 236–237, 239–240, 441 integration of facilities 81–82, 91 participants perceiving research program as healthcare provision (therapeutic misconception) 34, 43–44, 46–48, 77–78, 91–92, 93, 95–96, 238–239, 243 n.14 on HIV/AIDS in developing countries/Africa 429, 432–433 to control mother to child transmission 190–191, 432–433 ethical criticism of 158–159, 161 in industrialised countries 176 applicability for developing countries 180–181 localisation/local knowledge in 14, 15–17, 176, 178–180, 181, 185–187, 202, 231 on malaria 21, 403–404, 417–418, 421, 422 487

on medical technology 258 assessment of technology 255–256, 257 ultrasound technology 245–246, 250–253, 260–261 networks of actors engaged in 6, 8–9, 126, 390 on parasitic diseases 297–298, 305–310, 329 n.8, n.10-n.14 participants in 72 n.1 benefits for 237–238, 239–240, 241, 441 consent procedures 35–36, 83, 93–94, 163, 220–221, 222 disqualification of 433 experiences of being written about 34–37 information provided to 57, 60–61, 64 integration into modernity of 441–442 knowledge of 54, 58 perceptions of research programs by 46, 60, 67 as government interventions 49 as healthcare provision (therapeutic misconception) 34, 43–44, 46–48, 77–78, 91–92, 93, 95–96, 238–239, 243 n.14 recognition/acknowledgment of 50–52, 55 n.6, 231, 241 recruitment of 82–83, 90 therapeutic citizenship of 78, 88–89, 96 willingness of 46–47, 310 withdrawal/refusal of 84, 90, 236, 412 women/mothers 79, 88–94, 433 perspectives on 10, 67, 96, 256 by fieldworkers 83–85, 94–95 quasi-experiments in 199 social relationships/social practices of 10, 49, 54, 58, 67–68, 71, 230–231 post-trial situations 238–239, 240–241

Index

on traditional healing/medicines 18, 263–265, 266, 267, 272–274, 276–285, 288–289, 291 n.18 validity of 135, 198, 200, 219, 223 on variolation 448 writings of 32, 33, 39–40, 49, 53 and career mobility of researchers 43 see also clinical trials; field research, medical; observational studies; pharmaceutical research Medical Research Council of Ireland (MRCI) 139 research and development of leprosy treatment by 125–126, 130, 135, 137 tuberculosis research by 128–130, 132, 141 n.10 medical technology assessment of 255–260 influences on clinical practices 257 see also ultrasound technology medicines chests 368–369 n.3 distribution of, and power structures 22 for leprosy treatment, development and testing of 125–126, 127, 130–137, 138–140 local, clinical trials for testing of 14 resistance to 186, 191, 204 safety in pregnancy of, research into 25 n.14, 216–218, 220, 221 traditional ownership of/rights to 274–275, 278–279, 286–287 research into 18, 263–265, 266, 267, 272–274, 276–285, 288–289, 291 n.18 and role/position of healers 266–267 standardization of doses of 272 use of 272–273 in HIV/AIDS treatment 106, 121 n.12, 274, 285–288 meningitis 187

Mercer, Catherine 97 n.1 metaphors, used for illness 405, 408 Meyer, Harry 450 midwives, as ultrasound operators 249–250, 251–252 migration, therapeutic 437 miners African 404, 414 communications with management 415–416 control of 411, 414, 415, 417, 421 health of 409 housing of, racial segregation in 410, 412, 413–415 white, in Zambia 406, 412, 414 mining companies control of miners at 411, 414, 415, 417, 421 labour relations at 411–412 malaria control programmes by 404–410, 420 welfare services provided by 410–411, 412, 414 in Zambia 410, 411–412 deaths in 405, 407–408, 409, 424 n.11 worker control at 411, 414, 415, 417 miscarriages, spirits (jina)/satanic forces responsible for 253–254 Mnaliwa, Dr 278 modernization/modernity 304 embracement of lifestyles of, by Kenyan fieldworkers 321–323, 324–325 ethos of 4 in Kenya 304–305, 362 participants in medical research integrated in 441–442 and public health 20, 21, 403–404 Mogensen, Hanne 34, 49–50 Moi, Arap 334 Mol, A. 65–66, 69, 72 n.6 Molyneux, C.S. 43–44, 67, 238–239, 467 morality of French expatriates at Pasteur Institutes 393–394 488

Index

sexual, and HIV/AID transmission 340, 341–342 see also ethics mortality rates see deaths mosquitoes 329 n.13, 410, 418 see also malaria mothers childcare practices by 85–88 perspectives and experiences of infant cohort studies 79, 88–94 research into transmission of HIV/AIDS to children from 190–191, 432–433 mourning practices, writings in 36 Mshui, E.N. 275 Mueller-Rockstroh, Babette 17–18, 467–468 Mukiama, Prof. 276–277 Mulenga, Benard 419, 420 Mulenga, James 418 multidrug therapies (MDT), for leprosy treatment 137 multiple bodies 69 multiple identities 68 multiple world views 66 Mungai, Joseph Maina 354, 361, 364, 369 n.10 museums, human remains in collections of, disputes about 10, 61–63, 72 n.8, 73 n.9 NACC (National AIDS Control Council, Kenya) 334–335 Nairobi Medical School (NMS) 361–362 Nambozo (midwife) 44, 45–46, 47, 50, 51, 52 The Nation (newspaper, Kenya) 158–159 nationalism in Africa 361, 362, 363–364, 455 Cameroonization 21, 377, 378–380, 382–384, 385, 386, 388–389 neoliberalism 24 n.6, 154, 164 in Africa 13, 80–81, 298 and evidence-based medicine 291 n.18 and rise in audit cultures 149

neonatals see newborns neoPentecostalism/Pentecostalism, approaches to HIV/AIDS 106–108, 340–341 networks of actors engaged in medical research 6, 8–9, 126, 390 anthropological studies of 6 of care, in Africa 178, 194 n.7 newborns bilirubin levels of, research into 180–181 care of 176–177 in developing countries 178–180, 183–184 in United States 176, 177, 182, 193 n.5, 194 n.6 deaths of 86, 180 resuscitation of 182–183 NGOs/non-state actors approaches to HIV/AIDS of 103, 105, 106, 111–112, 121 n.11, 346, 350 n.18 see also community-based organizations for HIV/AIDS care medical research involvement in Africa by 7, 365–367, 368, 370 n.20 Nguyen, Vinh-Kim 21–22, 78, 88, 103–104, 468 Nigeria civil war in, and smallpox eradication campaign 453, 459 n.10 clinical trials of treatments of leprosy in 126, 127, 131–135, 136–137, 138–139 Njoroge, Mr 278 Nkunya, M.H.H. 272–273 Nodia, G. 328 n.4 non-state actors see NGOs/non-state actors Novas, C. 440 Ntenje, John 419 Nuffield Council on Bioethics on information provision to participants in medical research 57, 64

489

Index

on medical research in developing countries 58–59, 72 n.2 Nuremberg code of ethics 235 Nyanza Province (Kenya), HIV/AIDS in 334 nzoka, responsible for miscarriages 253–254 observational studies 15, 16, 199, 221–222 plausibility evaluation in 203–206, 218–219 versus RCTs/experiments 15, 25 n.12, 200–202, 222 in public health research 202–203, 221–222 Office of Traditional Medicine (Tanzania) 265, 268 Ogoja (Nigeria), clinical trials of treatments of leprosy in 126, 127, 131–135, 138–139 Ojiambo, Hilary 363, 364 Ombongi, Kenneth S. 20, 168 n.12, 468 Ong, A. 7 Orwa, Dr 277, 278 Oski, Frank 181 overseas research see Africa, medical research in; developing countries, medical research in ownership, of medicinal plants 274–275, 278–279, 286–287 paper roles of 9 in knowledge production 29 and social relations 29, 37–40, 55 n.3 paradigms of research 59, 73 n.10 parasitic diseases, research and control of 297–298, 305–310, 329 n.8, n.10–n.14 participants in medical research 72 n.1 benefits for 237–238, 239–240, 241 consent procedures 35–36, 83, 93–94, 163, 220–221, 222 disqualification of 433 experiences of being written about 34–37

information provision to 57, 60–61, 64 integration into modernity of 441–442 knowledge of 54, 58 perceptions of research programs by 46, 60, 67 as government interventions 49 as healthcare provision (therapeutic misconception) 34, 43–44, 46–48, 77–78, 91–92, 93, 95–96, 238–239, 243 n.14 recognition/acknowledgment of 50–52, 55 n.6, 231, 241 recruitment of 82–83, 90 therapeutic citizenship of 78, 88–89, 96 willingness of 46–47, 310 withdrawal/refusal of 84, 90, 236, 412 women/mothers 79, 88–94, 433 see also clinical trials, participation in participatory approaches in research 60, 72 n.6 Pasteur Institutes (Africa) 373, 374, 389, 394 of Cameroon/Centre Pasteur (IPC/CPC) 375–378, 383, 384–385 Cameroonization of 21, 377, 378–380, 382–384, 385, 386 and stereotypes of Cameroonian culture 388–389 Etoundi controversy at 380–382, 386–388, 391 French expatriates at 377, 380, 382–383, 384, 385 morality/Cameroonization of 393–394 professional hierarchies at 377, 385, 392–393 research activities of 373, 390 Cours Pasteur (Paris) 379 French expatriates at 389–390, 394–395 Patterson, J.H. 356

490

Index

Peirce, Charles 41 Pell, J.P. 201 Pels, Peter 148, 149, 154 Pentecostalism/neoPentecostalism, approaches to HIV/AIDS 106–108, 340–341 perspectivalism 10, 59, 66, 72 n.3 perspectives and contexts 66 on human remains in museum collections 61–63, 73 n.9 and interests 61 on medical research 10, 67, 96, 256 by fieldworkers 83–85, 94–95 of researchers 59, 61–62, 65–66, 67 Petryna, Adriana 440 pharmaceuticals see medicines pharmacological research 125 on drugs for tuberculosis treatment 128–130, 141 n.10 historical studies of 12, 126–127, 132 of medicinal plants 283 transnationalism of 126, 139 see also clinical trials Philadelphia, care for newborns in hospitals of 177, 193 n.5, 194 n.6 Philips (company), ultrasound technology developed by 246–248 phototherapy, use of 180 physicians see doctors phytochemical studies of medicinal plants 283 Piot, Charles 48–49, 68, 69 placenta, delivery of 251–252 plague 356, 369 n.9 plants, medicinal dosages of 272–273 local knowledge of 270–271 ownership of/rights to 274–275, 278–279, 286–287 research on 264, 266, 268, 272, 281–284 plausibility evidence/evaluation 203–206, 218–221 pneumonia, deaths caused by 424 n.11 Poirier, André 381, 387

poisoning, as element of witchcraft 339 policies see public health, policies political economy in Gambia 80–81 of medical research in developing countries 7 politics bio 26 n.16 and ethos 5–6 and medical ethics 22 Pool, Robert 34, 159, 230 postcolonial era in Africa French foreign policy in 373–374, 394 public health 13, 20, 21, 360, 361–367, 450 postcolonial management, culturalism in 388–389 postmetaphysical societies 168 n.8 power structures/relations in audit processes 148, 149, 153 bio 443 and distribution of medicines 22 and ethics/medical ethics 22–23, 160–162 and paper writings 37–40, 55 n.3 between traditional healers and medical researchers 277, 284 pragmatic clinical trials 15–17, 202, 231 prayers, healing through 107 pregnancy delivery of placenta 251–252 ectopic, diagnosis of 250–251 measuring time of 261 n.3 safety of medicines used in 25 n.14, 216–218, 220, 221 prevention of HIV/AIDS transmission awareness of methods for 340, 342 in Kenyan policies 335 between mother and child, research into 190–191, 432–433 and VCT procedures 334, 343–346, 348 Prince, R.J. 49, 73 n.14, 271 privatised ethics 12, 149–150, 151 privatization, of healthcare in Africa 103, 104 491

Index

public ethics 12–13, 146, 150, 152, 154, 156–157 public goods exclusion from 164 medical research contribution to 175, 188–189, 239, 240, 241 public health 13 in Africa 354 in colonial era 20, 356–357, 360, 447–448 garden city ideas 413–414 malaria control programmes 403 postcolonial 13, 20, 360, 365–367, 450 smallpox eradication campaign 452–456, 457, 458 typhoid vaccinations 456–457 and health science, history of 18–19, 20–21 and individual rights 5, 19–20, 348 and modernization 20, 21, 403–404 policies on HIV/AIDS 19, 430–431 in Kenya 353–354, 356 in Sierra Leone 367 and weakness/crisis of African state 13, 20, 365–366 research 7, 222–223 observational studies 202–203, 221–222 public sphere 153–154 quantative research data 30–32 quasi-experiments 199 quasi-kins, used in ethnographic research 40 quinine, distribution of, in Kenya 358–359 racial segregation for health reasons 412, 420, 423 n.3 in Kenya 356, 359–360 in Sierra Leone 367 in housing/welfare services provided by mining companies 410, 412, 413–415

racialization, of professional hierarchies 391–393 radiographers, ability to use ultrasound technology 248, 249 randomised controlled trials (RCTs) 15, 25 n.12, 193 n.2, 197–198 in Africa 258, 412 dominance of 235 localisation of 16 origins of 232–234, 243 n.6 safety issues in 217 versus observational studies 15, 25 n.12, 200–202, 222 in public health research 202–203, 221–222 randomization, value of 232–233, 234 Ravisse, Piere 381–382 RCTs see randomised controlled trials Reade, Charles Compton 413 reality, in medical research 15, 16, 17 recognition/acknowledgment of participants and fieldworkers in medical research 50–52, 53–54, 55, n.6, 231, 241 of traditional healers 264, 266–267, 268–280, 289, 290 n.14 recruitment of fieldworkers in medical research 310–312 of participants in medical research 82–83, 90 rehabilitation technicians, training of 188 relevance, of research 163 research applied 119 n.1 autonomy of 234, 235 benefits of 175, 188, 235–236, 239–240, 241–242 covert/secrecy of 152, 167 n. 5, 284 cultures 421, 422 in developing countries ethics of 147, 158–159, 161–162, 166–167 need for integration with local social processes 162–165 492

Title

rumours associated with 159 international, as global public good 175, 188–189 paradigms 59, 73 n.10 participatory approaches in 60, 72 n.6 Western methods of 59–60 see also field research; medical research ‘researched’ see participants in medical research researchers 72 n.1 African 42, 43 ethical accountability/involvement of 113–115, 122 n.16, 152–153 knowledge practices of 58–59 life stories of 302–305, 307–311, 318–319, 330 n.18 local, lack of authority of 190–191 perspectives of 59 on human remains in museum collections 61–62 on medical research 67 unitary world view 65–66 recognition/acknowledgment of 52, 54 relations of with research participants in clinical trials 139–140 information offered to 57 with traditional healers 267, 270, 272–273, 277–280, 282–283, 284–285, 289–290 in HIV/AIDS research and treatment 285–288 resuscitation, neonatal 182–183 rights see human rights; individual rights; intellectual property rights Roan Mine (Zambia) control of miners’ movements by 415 malaria control programme by 404–410, 420 research culture of 421 racial segregation in housing of miners at 414–415 welfare services provided by 410–411, 412, 414 worker control at 411, 414, 415, 417

Rockstroh see Mueller-Rockstroh Rose, N. 440 Ross, Ronald 367, 405 malaria control method developed by 408–409, 423 n.3 Ross Institute (London) 406 expedition to Northern Rhodesia (Zambia) 406–407, 410 malaria control methods of 405, 412, 420 malaria research by 422 Rosser, Walter W. 193 n.2 Ruel, Malcolm 350 n.13 Rufiji District (Tanzania) ACT introduction in 211–216 research into antimalarial combination treatments in 205–218 rumours anthropological theories on 112 associated with medical research in Africa 159 role in HIV/AIDS treatment and care 112–113 rural areas healthworkers working in 49–50 living in 319 Rwanda, smallpox eradication campaign in 457 safaris, fieldwork seen as 306–307 safety of medicines used in pregnancy 25 n.14, 216–218, 220, 221 of traditional medicines 272–273 Sarungi, P.M. 275 Sayi, J.G. 275 Schramm, K. 119 n.2 Schumaker, Lyn 21, 285, 469 science ethos of 4, 5 experimental model in 41 generation of 317–318 and government/state 19, 318, 322–323, 327–328, 330 n.21 research methods in, Western 59–60 493

Index

social study of 17, 163 see also research scientists see researchers secrecy around HIV/AIDS testing and diagnosis 100, 110, 112–113, 114–116, 121 n.14, 344, 347 of research 152, 167 n.5, 284 of traditional healers 277–279, 287 Sekagya, Dr 278, 279 self-discipline, in medical field research 312–313 Sequeira, James H. 354, 355 sexual morality, in Christianity, and HIV/AIDS 340, 341–342 sexual partners of HIV/AIDS infected persons, risks of infection for 347–348 Shauri, Mr 270, 271–272 Shore, Chris 147, 148 Sierra Leone public health policies in 367 malaria control programmes 405, 423 n.3 Silla, Eric 429–430 Simpson, W.J. 369 n.13, 406–407, 413 situated accountability 13, 159, 166 Small, Mary 79, 97 n.1 smallpox in Africa 458 n.3 African perceptions of 447 campaigns for eradication of 450, 451, 459 n.9 in Africa 452–456, 457, 458 vaccinations 448, 451 in Africa 451, 454 coercion used in 456 variolation used against 445, 448–449 Smith, G.C. 201 snakes, as metaphor for illness 405, 408 social goods 18 social processes, local, integration of medical research in 162–165 social relations 48–49 in anthropological field research 11, 13–14

of health workers 49–50 in medical research 10, 49, 54, 58, 67–68, 71, 230–231 post-trial situations 238–239, 240–241 and paper/writings 29, 37–40, 49, 55 n.3 switching between 66–67, 73 n.12 social sciences, in Africa 42 South Africa, HIV/AIDS testing by healers in 287–288 Spearpoint, C.F. 405, 416 species sanitation/Ross method of malaria control 405, 408–409, 412, 420, 423 n.3 spirits (jina)/satanic forces protection against 85, 86 responsible for AIDS/HIV infection 108 responsible for miscarriages 253–254 Stakeholders Workshops on Traditional Medicine and Medicinal Plants (2003 & 2004) 275–277 standards ethical 145, 150, 156, 157 universal, for evidence in medical research 14, 25 n.15 state 301–302 African in colonial era 358 crisis/weakness of 301, 353 and public health policies 13, 20, 365–367 and smallpox eradication campaign 452–453, 455–456 postcolonial 361 failed 328 n.4 rise of 442–443 and science 19, 318, 322–323, 327–328, 330 n.21 in medical research 49, 168–169 n.12 see also government statistics, use in medical research 234 Stephens, John W.W. 423 n.3 494

Index

stereotypes, cultural, and Africanization 388–389 stigmatization of HIV/AIDS sufferers 106, 107 see also secrecy Storke, Arthur 406 Strathern, Marilyn 10–11, 30, 36, 85, 122 n.16, 147, 193, 256, 469 streptomycin trials 232, 234 Sudan, smallpox eradication campaign in 454–455 suffering, anthropological research on 113 Sukuta (Gambia) 79 childcare practices in 85–86 ethnographic research in 79 healthcare system in 80, 81, 82 medical research in 78–79, 81–85 sulfadoxine/pyrimethamine (SP) treatment of malaria, research on 207, 208–210 sulphetrone treatment of leprosy 131–132 surveys 32–33, 49, 92, 97 n.1 demographic and health in Kenya (2003) 347–348, 349 n. 3, n.9 of HIV testing and counselling 42, 47 as source of ethnographic evidence 79, 80 used in malaria research 207, 209 Swellengrebel, Nicholas 408 syphilis 252 talismans, use of 86 Tanzania healthcare in privatization of 104 traditional healers recognised/ integrated in 264, 268–280, 290 n.14 ultrasound technology use in 250–253 for after-birth scans 251–252 equipment procurement 248–250 pregnant women’s views of 253–255

HIV/AIDS in accepted ways of speaking about/culture of silence around 110, 113, 115–116, 121 n.14 anthropological field research on 101–103, 120 n.4 neoPentecostal approaches to 106–108 NGO approaches towards 105, 106, 121 n.11 treatment and care of 104–105, 106, 112–113, 121 n.12, 285–288 medical research in on ACT introduction 203–218 plausibility evaluation in 203–206, 218–221 on traditional medicines 264–265, 267, 280–288 on ultrasound technology use 245–246, 250–253, 260–261 assessment of technology 255–256 pregnant women’s views of 253–255 Taylor, C. 70, 74 n.19 teaching practices 59–60 technography 246, 261 technology appropriate 24 n.9, 172, 190, 193 n.3, 245 assessment of 255–260 contexts of 259, 260, 261 good 17–18, 255, 256, 258, 259–260 ultrasound as 257 needs for, in healthcare in developing countries 177–178 new 176 transfer of 145–146, 248 temporality/time and ethos 4 in medical research projects 51–52 in medical writings 41–42 Thairu, Kihumbu 364 therapeutic citizenship 78, 88–89, 96, 103–104, 438, 440, 441

495

Index

therapeutic communities 429, 434 in HIV/AIDS care 439–440, 443 in leprosy care 429–430 reproduction of 443 therapeutic migration 437 therapeutic misconception 34, 43–44, 46–48, 77–78, 91–92, 93, 95–96, 238–239, 243 n.14 time see temporality/time Timmermans, S. 291 n.18 tobacco farming, in Kenya 350 n.12 touch, in healthcare/healing capacities of 49–50, 55 n.5, 271 traditional medicine/healing ancestors’ role in 288 Chinese (TCM) 121 n.12, 265 in HIV/AIDS treatment 106, 121 n.12, 274, 285–288, 342–343 integration into healthcare of 264, 266–267, 268–281, 289, 290 n.14, 364 research into 18, 263–265, 266, 267, 272–274, 276–285, 288–289, 291 n.18 secrecy of 277–279, 287 variolation used in 449 training of African doctors 192, 361–362, 379 of fieldworkers in medical research 45–46, 314–315, 316–317 of healthworkers 214 of rehabilitation technicians 188 of VCT staff 346–347 see also education transcription 32, 33 transnationalism, of pharmacological research 126, 139 Treatment Action Campaign (South Africa) 437 triage 436 used by community-based groups for HIV/AIDS sufferers 437–438 trial communities 6–7, 24 n.5, 231 knowledge gained by 10 tropical diseases 448 tuberculosis

in Ireland 127–128 treatment of multi-drug resistance 191 research on 128–130, 132, 134–135, 141 n.10, 232 Twomey, D. 141 n.9, n.22 typhoid, vaccinations against, popularity in Africa 456–457 Uganda greeting conventions in 40 healthcare in 34, 46 HIV/AIDS treatment and testing 35, 350 n.14 Ulrich, George 12–13, 17, 24 n.10, 470 ultrasound technology 247 in Africa 246–248 in Tanzania after-birth use of 251–252 procurement of equipment 248–250 research on use of 245–246, 250–253, 260–261 assessment of technology 255–256, 257 pregnant women’s views of 253–255 unhinged ethics 157–165 unitary world views 65–66 United Kingdom medical research in Gambia funded by 78, 81, 229 National Health Service (NHS) in 235, 243 n.10 United States care for newborns in 176, 177, 182, 193 n.5, 194 n.6 evidence-based medicine in 14, 173–174 universal standards, for evidence in medical research 14, 25 n.15 universalism, rejection of 66 urban areas living in 318–319, 321–322 believed to be inappropriate for Africans 411 496

Index

Garden City movement’s ideas about 413 racial segregation in 359 Uzuakoli (Nigeria) clinical trials of treatments of leprosy in 136–137, 139 vaccinations against malaria trials 229–231, 236–237 benefits of 239–240, 241 end of 238 against measles, trials in Africa 450–451 in postcolonial Africa 450 protective perceptions of 88, 446 in Africa 449, 456–457, 458 against smallpox 448 in Africa 451, 454 coercion used in 456 and variolation 449–450 against typhoid, popularity in Africa 456–457 validity of ethical judgements 17 of medical research 135, 198, 200, 219, 223 see also evidence in medical research/evidence-based medicine universal standards of 25 Vallvé, Carles 202 values social goods as 18 Western 156, 166 variolation 445, 448–449 and vaccination 449–450 Vaughan, Megan 288, 357 VCT (Voluntary Counselling and Testing) for HIV/AIDS 333–334 availability of 335, 343 consent procedures in 35 mandatory counselling element of 347 resistance to/mistrust of 334, 343–346, 348 training of staff involved in 346–347

Verran, H. 59, 66 Vint, F.W. 357 Vischer, Wolfgang 136 visitors, hosting of 40 Viveiros de Castro, E. 65 Voluntary Counselling and Testing for HIV/AIDS see VCT Waane, S.A.C. 290 n.14 Waddy, B.B. 389–390 Wagner, R. 69 Watchko, Jon 181 Watson, Malcolm 405, 406, 407, 408, 409, 410, 412, 414, 417, 421, 423 n.3 welfare services, provided by mining companies 410–411, 412, 414 welfare state, dominance of RCT approaches in 235 Wellcome Trust, activities in Kenya 366, 370 n.22 West scientific research methods of 59–60 unitary world view of 65–66 White, Luise 22–23, 112, 470 WHO (World Health Organization) role in smallpox eradication campaigns 452–456, 459 n.9 Traditional Medicine Unit of 265 Whyte, Michael 36, 325 Whyte, Susan Reynolds 9–10, 67, 241, 291 n.20, 325, 350 n.14, 470–471 widows, vulnerability towards HIV/AIDS 341 Wilson, Godfrey 411 witchcraft, deaths blamed on 338–339, 340, 350 n.11 witnessing, by anthropologists 100 women agency of 93–94 and HIV/AIDS disclosure of HIV positive status 433–434 participation in HIV/AIDS research 433 therapeutic communities 439 vulnerability for infection 341 497

Index

married to Kenyan fieldworkers 319, 321, 330 n.19 participation in medical research by 79, 88–94, 433 pregnant, views on use of ultrasound technology 253–255 World Bank, on medical development aid priorities 189 Wright, Susan 147, 148 writings experiences of subjects of 34–37 of field research 32, 33, 39–40 healing through 34–35 in healthcare 34–35, 37–38 of medical research 43, 49, 53 in mourning practices 36 and social relations 29, 37–40, 49, 55 n.3 Yates, D. 200 Yoder, P.S. 42, 47 ‘Yoruba logic’ 72 n.4

Zaire, smallpox eradication in 453–454 Zambia copper mines in 410, 411–412 deaths at 405, 407–408, 409, 424 n.11 racial segregation in housing of miners at 410, 412, 413–415 welfare services provided by 410–411, 412, 414 worker control at 410–411, 412, 414 malaria control programmes in 403–404, 422–423 by mining companies 404–410, 420 malaria research in 417–418, 421, 422 Zgambo, Jackson 418 Zgambo, Joseph 405, 408 Zimbabwe, healthcare in 177, 183, 187–188 Zurovac, D. 219

498