Vous êtes sur la page 1sur 234
Vulnerability Technological Cultures New Directions in Research and Governance Edited by Anique Homme!s, jessica M esman,


Vulnerability Technological Cultures New Directions in Research and Governance Edited by Anique Homme!s, jessica M esman,

Technological Cultures

New Directions in Research and Governance

Vulnerability Technological Cultures New Directions in Research and Governance Edited by Anique Homme!s, jessica M esman,

Edited by Anique Homme!s, jessica M esman, and Wiebe E. Bijker

Vulnerability Technological Cultures New Directions in Research and Governance Edited by Anique Homme!s, jessica M esman,

The M IT Press Cambridge, Massachusetts london, England



  • 9 Entrainment, Imagination, and Vulnerabil ity-lessons from large­

Scale Accidents

in the Offshore I nd ustry


Ger Wackers



Vulnerable Practices: Organizing through Bricolage in Railroad




johan M. Sanne

Part Ill

The Governance of Vulnerabil ity



Governing a Vulnerable Society: Toward a Precaution-Based





Gerard de Vries, lmrat Verhoeven, and Martin Boeckhout



Regu lating Risks by Rules: Compl iance and Negotiated Drift in the

Dutch Chemical industry under the Seveso Regime


Anique Hommels, Esther Versluis, Tessa Fox, and Marjolein van Asselt



Dealing with Vul nerabil ity:

Balancing Prevention with Resilience as a

Method of Governance


Ger Palmboom and Dick Willems



A Pragmatist Approach to the Governance of Vulnerability


]ozef Keulartz and Maartje Schermer



F rom Sustainabil ity to Transformation: Dynamics and Diversity in

Contents 9 Entrainment, Imagination, and Vulnerabil ity-lessons from large­ Scale Accidents in the Offshore I nd

Reflexive Governance of Vulnerabil ity

Andy Stirling







Contents 9 Entrainment, Imagination, and Vulnerabil ity-lessons from large­ Scale Accidents in the Offshore I nd


Contributing Authors

Andy Stirl ing is the research director for the

Science and Technology Policy

Research (SPRU) department at the University of Sussex, and co-director of the ESRC STEPS Centre (on social, technological and environmental pathways to sustainability). He is an interdisciplinary researcher, focusing on challenges around "opening up" more democratic governance of knowledge and inno; vation and has served as a UK and EU policy advisor on many related issues. :

Contributing Authors Andy Stirl ing is the research director for the Science and Technology Policy Research

lmrat Verhoeven is an assistant professor at the Department of Political " ' Science of the University of Amsterdam. From 2009 to 2012, he worked, as a postdoc researcher at the Department of Sociology at the same uni­ versity. Before moving to the university, Verhoeven was a staff member at the Wetenschappelijke Raad voor het Regeringsbeleid (WRR}, the Scien­ tific Council for Government Policy, in The Hague. His current research examines the processes of contentious politics, the impact of institutions on active citizenship, developments in volunteering within the changing welfare state, and the role of emotions in politics and democracy.

Contributing Authors Andy Stirl ing is the research director for the Science and Technology Policy Research

Esther Versluis is an associate professor of European regulatory governance at the Department of Political Science, Maastricht University. Her research concentrates on questions related to European regulatory governance. She specializes in the European policy process, and, among other pursuits, she analyzes the implementation of European risk regulations on the national level. In 2011, she worked as a Fulbright visiting scholar at Cornell Univer­ sity to further explore the role of experts and expertise in EU risk regula­ tion.

Shiv Visvanathan is a professor at O.P. Jindal Global University, Sonipat, Haryana, India. He was a professor at the Dhirubhai Ambani Institute of In­ formation and Communication Technology, Gandhinagar, India, and has held the position of senior fellow at the Center for the Study of Develop­ ing Societies (CSDS) in Delhi. He has held visiting professorships at Smith College, Stanford, Goldsmiths, Arizona State University, and Maastricht

University. He is the author of Organizing for Science (1985) and A Carnival

for Science (1997), and co-edited Foul Play: Chronicles

of Corruption (1999).

Gerard de Vries is a member of the Council at the Scientific Council for Government Policy (Wetenscha p pelijke Raad voor het Regeringsbeleid; WRR) in The Hague, The Netherlands; a professor of philosophy of sci­ ence at the University of Amsterdam; and a fellow of Wolfson College (at Cambridge, England). De Vries's academic work is concerned with the so­ cial, political, and ethical aspects of contemporary science and technology. De Vries was chairman of the WRR project team that prepared the report

Contributing Authors


Contributing Authors Uncertain Safety that acts as the basis for chapter 11, "Governing a Vulner­ able

Uncertain Safety that acts as the basis for chapter 11, "Governing a Vulner­ able Society: Toward a Precaution-Based Approach."

Ger Wackers received a Ph.D. in science and technology studies from Maas­ tricht University. While he held a position at the Maastricht University Science, Technology, and Society (MUSTS) center, much of the empirical research on large-scale accidents in the offshore industry was conducted in collaboration with the Center for Technology, Innovation, and Culture at the University of Oslo. Currently, he is an associate professor at Narvik University College in Norway.

Dick Willems is a professor of medical ethics at the Amsterdam Medical Center (AMC) at the Department of General Practice in the Division of Public Health.

Contributing Authors Uncertain Safety that acts as the basis for chapter 11, "Governing a Vulner­ able
Contributing Authors Uncertain Safety that acts as the basis for chapter 11, "Governing a Vulner­ able
  • 1 Studying Vulnerability in Technological Cultures

Wiebe Bijker, Anique Hommeis, and jessica Mesman

Vulnerability in Technological Cultures Wiebe Bijker, Anique Hommeis, and jessica Mesman Vulnerability in technological cultures

Vulnerability in technological cultures is all around us.1 Such vulnerabil­ ity may be caused by the techno-scientific character of our societies, as when a power failure stops all electricity-driven activities. Or it may result from natural events, such as hurricanes, but then the vulnerability is often shaped by technological circumstances, such as unexpectedly breaking levees. Societies also seek to defend against vulnerabilities by using science and technology, as in the case of high-tech storm surge barriers or innova­ tive forms of non-pesticide crop management. The boundary between the technical, social, and natural can be fuzzy, though: for instance, was the breakdown in international business in the spring of 2010 due to the erup­ tion of the Icelandic volcano Eyjafjallajokull which caused ash to cover large areas of northern Europe, the high-tech character of air transporta­ tion, or our culture of global mobility? Vulnerabilities are not usually given in an unambiguous way. Depend­ ing on your perspective, you may deem the neonatology ward of a hospital a strategic research site to study vulnerability (because so many babies there hover on the edge between life and death) or a very safe place (considering that there is a low level of failures given the socio-technical complexities of the situation). Vulnerability, finally, is not only or purely negative. A certain degree of vulnerability is necessary to create space for learning and adaptation in a society. Vulnerability, in this sense, is equivalent to openness and flexibil­ ity. Once properly addressed, such vulnerability with accompanying cop­ ing mechanisms may yield a more flexible and resilient society than one that tries to avoid all vulnerabilities. People living in cities that are vulner­ able to frequent power failures have developed flexible ways of coping with such electricity shutdowns. This means that such cities are probably bet­ ter able to cope with an even bigger blackout, while a less vulnerable city

Vulnerability in Technological Cultures Wiebe Bijker, Anique Hommeis, and jessica Mesman Vulnerability in technological cultures
Wiebe Bijker, Anique Hommels, and Jessica Mesman with less experience in dealing with small shutdowns
  • 2 Wiebe Bijker, Anique Hommels, and Jessica Mesman

Wiebe Bijker, Anique Hommels, and Jessica Mesman with less experience in dealing with small shutdowns

with less experience in dealing with small shutdowns might be completely paralyzed by that same big blackout. This is the vulnerability that we seek to investigate in this volume: het­ erogeneous and varied by being related to science and technology, by being context-dependent, and by having not necessarily only negative connota­ tions. Why study vulnerability at all? The primary goal of this volume is to make a bold argument that scholars should use vulnerability as a frame to understand our world and contribute to a better world, rather than losing

with less experience in dealing with small shutdowns might be completely paralyzed by that same big
with less experience in dealing with small shutdowns might be completely paralyzed by that same big

themselves in detailed technical risk analyses and rhetoric. (As we shall argue below, we do not mean to criticize risk studies, but to propose a more encompassing and compassionate frame for these questions.) We want to call attention to issues of justice and livelihood in a globalizing world. Studying the relations between science, technology, and society (which we call "STS studies") has yielded a broad range of empirical and theoretical insights. This now offers a solid base to address "big" questions of justice, peace, and democracy. We argue, and the authors in this volume dem­ onstrate, that asking questions about vulnerability is a promising way to address such "big" issues in technological cultures. Today's modern societies are thoroughly scientific and technological. They rely heavily on scientific advice, scientific-technical standards, and technological systems. Science and technology are not neutral: they shape our activities and our sensemaking. When a new technique or instrument is adopted in medical practice, it transforms not only what doctors and their patients do, but also the way they think; this may also imply shifts in the meaning of health, illness, and medical care. The tank irrigation sys­ tems in South India are shaped by the social and cultural relations between Hindu and Muslim farmers (Shah, 2003). And coastal defense (we mean dunes, dikes, and levees) in the Netherlands and the United States mirror the differences in risk culture in both countries (Bijker, 2007). If technolo­ gies are socially shaped, societies are technologically built (Bijker & Law,

themselves in detailed technical risk analyses and rhetoric. (As we shall argue below, we do not
themselves in detailed technical risk analyses and rhetoric. (As we shall argue below, we do not


New, science-based technologies such as the Internet, genetic engineer­ ing, nanotechnologies, and new forms of sustainable energy production create numerous opportunities for modern societies. But at the same time, none of these novel technologies is without risks. Citizens, governments, universities, and businesses face challenges when trying to take advantage of the seemingly infinite opportunities that scientific and technological developments offer, and there may be unintended consequences, with disastrous effects on society. The Challenger space shuttle explosion, the Chernobyl nuclear accident, the Bhopal chemical disaster, and the Exxon

New, science-based technologies such as the Internet, genetic engineer­ ing, nanotechnologies, and new forms of sustainable
Wiebe Bijker, Anique Hommels, and Jessica Mesman with less experience in dealing with small shutdowns

Studying Vulnerabil ity in Technological Cultures


Valdez oil spill in the 1980s reminded us that large-scale systems are vul­ nerable to human error and technical malfunctioning, with far-reaching


Accidents, as Charles Perrow ( 1984/1999) argued, are "nor­

mal" in large, complex, and tightly knit technological systems. Modern societies, as Ulrich Beck (1986, 1 992) diagnosed, are "risk societies." Science and technology not only help to tame risk and vulnerability; they also create new risks, vulnerabilities, and possibilities for accidents. Once we have large technological systems, accidents are inevitable; because we live

in modern scientific and high-tech societies, risks are inevitable. In other

words, scientific and technological developments do not only support and strengthen societies; they also make societies vulnerable at the same time. Such vulnerability is an inevitable characteristic of today's technological cultures. In this book, we will explore vulnerabilities in technological cultures. Our focus on vulnerability is not pitted against the existing work in risk studies. The authors in this volume have actively contributed to and are drawing on research in risk studies and related fields. As we will argue in this introduction, however, we hope that an agenda in terms of vulner­ abilities will allow for a novel and broader understanding of key questions related to the risks and benefits of science and technology in modern soci­ eties. Broadening the analysis from risk to vulnerability, we argue, will draw attention to issues of justice, solidarity, and livelihood. It will also help us to include comparative research on societies in the global north and south. This book engages with a number of crucial issues of the role of vulner­ ability in current societies. First of all, we focus on the issue what vulner­ ability can mean and how it affects cultures and societies. Both Bangladesh

and the Netherlands experience a risk of flooding. Yet, evidently, the peo­ ple in the Netherlands are vulnerable in a different way than the people

in Bangladesh, and they both have different ways of coping: differences

in geography, climate, and social and technical infrastructures make for a very different impact of an eventual flooding disaster. Different value systems, community relations, and knowledge and technology systems induce different defense strategies. Do societies in the global south and north have different vulnerabilities and different coping strategies? Show­ ing the multiplicity of vulnerability-the idea that different actors give different meanings to it and that vulnerability is the outcome of specific circumstances-is one of the key arguments of this book. Second, this book concerns the conceptual frameworks for empirical description and theoretical analysis of vulnerability. These frameworks elu­ cidate the complex and interwoven dimensions of the vulnerability issue.


Bijker, Anique Hommels, and jessica Mesman


A study of vulnerability, therefore, requires insight in the analytical and political potential of such concepts. We aim to explore theoretical concepts that illustrate the ambiguity, context-dependency, and constructed charac­ ter of vulnerability. An investigation of concepts such as bricolage, dissent, and imagination reveals that deviation from rules, standards, and protocols will not lead automatically to more vulnerability in all cases. Rules and practices that produce resilience in one context may result in vulnerability in another. Third, this book deals with the governance implications of living in vulnerable technological cultures, discussing, for instance, how the gover­ nance of vulnerabilities can be organized. Which practices, which politics, and which ethics are needed to cope with vulnerability in technological cultures? Discussing national and international risk policies, risk regula­ tion in the chemical industry, and health prevention programs, new frame­ works are proposed for a more reflexive way of dealing with vulnerabilities in technological cultures. Importantly, this book argues for building on the positive meaning of vulnerability in devising strategies for coping with it:

Wiebe Bijker, Anique Hommels, and jessica Mesman 4 A study of vulnerability, therefore, requires insight in

by giving space to uncertainty, doubt, and error, or by drifting away from established rules, we may enhance our capacity to cope successfully with vulnerability.

Vulnerabi lity in Technological Cultures: Approach

In this volume, we develop an approach to key questions in current soci­ eties about the risks and benefits of modern science and technology. To do this, our entry point is technological cultures, rather than modern societ­ ies; and the key concept is vulnerability, extending the analysis of risk. In this section, we briefly introduce these choices by explaining the focus on technological culture, giving a preliminary definition of vulnerability, and reviewing the differences between vulnerability and risk.

Technological Culture: A New Level of Analysis

If we take the concept of culture to mean "a group's shared set of mean­ ings, its implicit and explicit messages encoded in social action" (Traweek, 1988), the concept of "technological culture" is used to denote the shared set of assumptions that govern the interactions in modern societies, which are so pervasively constituted by science and technology. The concept of culture is typically used to highlight whatever humans learn, in contrast to whatever is innate. Thus, it is contrasted with nature:

Wiebe Bijker, Anique Hommels, and jessica Mesman 4 A study of vulnerability, therefore, requires insight in
Wiebe Bijker, Anique Hommels, and jessica Mesman 4 A study of vulnerability, therefore, requires insight in

where nature provides humans with base needs and desires, culture provides

Studying Vulnerabil ity in Technological Cultures


for content and meaning. Therefore, "culture refers to community-specific ideas about what is true, good, beautiful, and efficient" (Shweder, 2001, p. 3153). These cultural meanings are socially inherited and customary; they constitute a way of life. A cultural account helps to explain why the members of a particular culture say and do things to each other with words, acts, and artifacts. As such, it accounts for the goals, values, and worldviews embraced by a particular group. The border between the concepts of culture and society is sometimes quite thin. To relate the two, we follow the characterization of culture by Han­ nerz (1992) in three dimensions. The first dimension is formed by ideas, values, goals, and worldviews. His second dimension is the well-known ele­ ment that is sometimes called material culture: forms of externalization of ideas, values, and goals in the form of artifacts. Hannerz's third dimension of culture is the social distribution of meanings and artifacts. This third dimension gets us close to the concept of society as constituted by relations between humans, groups, and-from an STS perspective-technologies. Thus, our technological culture largely overlaps in its ontology with modern

society as constituted by science and technology (Bijker, 2006). The important

difference lies in the implicit research heuristics: studying culture implies following an ethnographic approach that maps not only social structures, but also goals, values, and worldviews.2 The term technological culture thus primarily invokes another level of analysis: adding the cultural to the soci­ etal (Bijker, 2010). In this volume, therefore, we use a research heuristic that combines sociological analysis of society's structures and power relations with an ethnographic analysis of the values, goals, and worldviews that govern the interactions, and also provides an STS analysis of the relations between technology and society, between science and politics, and between ecosys­ tems and human communities.

Studying Vulnerabil ity in Technological Cultures for content and meaning. Therefore, "culture refers to community-specific ideas
Studying Vulnerabil ity in Technological Cultures for content and meaning. Therefore, "culture refers to community-specific ideas
Studying Vulnerabil ity in Technological Cultures for content and meaning. Therefore, "culture refers to community-specific ideas

Vul nerability: A Preliminary Defin ition

The term vulnerability can be found in many fields, such as economics (e.g., poverty and livelihood studies), climate studies, studies of technological systems (e.g., critical infrastructures), disaster studies, development stud­ ies, and anthropology. All these disciplines have their own concepts of vulnerability.3 For social scientists, for instance, vulnerability is "the set of socio-economic factors that determine people's ability to cope with stress or change, " while climatologists talk about vulnerability in terms of "the likelihood of occurrence and impacts of weather" (Brooks, 2003, p. 2). Alwang and Siegel (2001) review the many kinds of vulnerability in

Studying Vulnerabil ity in Technological Cultures for content and meaning. Therefore, "culture refers to community-specific ideas


Wiebe Bijker, Anique Hommels, and jessica Mesman

6 Wiebe Bijker, Anique Hommels, and jessica Mesman academic literature, including technological vulnerability, livelihood vul­ nerability,

academic literature, including technological vulnerability, livelihood vul­ nerability, structural and proximate vulnerability, nutritional vulnerability, and socioeconomic vulnerability. Some definitions focus on the risks and the stress that people are subject to, or on their powerlessness to respond or cope with these stressors. The focus can also be on the outcome in terms of (welfare) loss (Brooks, 2003), on the limitation of an individual's decision­ making capacity (biomedicine), or on the risk of being wronged (bioethics) (Levine, 2004, p. 397). Problems arise when one tries to compare different measurements of vulnerability. Both the object and method of measurement differ across disciplines and thus makes interdisciplinary comparison very difficult. Due to its multiplicity in meaning and measurement, vulnerability as a concept runs the risk of losing its analytical and practical value. To deal with this problem, some clarifying frameworks have been proposed, using classifi­ cations of vulnerability by type of technology, scale of disaster, and type of cause (Adger, 2006; Alwang & Siegel, 2001; N. Brooks, 2003; Gallop:in, 2006; Hilhorst & Bankoff, 2004; Martin, 1 996; Murray & Grubesic, 2007). Humans can be vulnerable, as can plants, animals, ecosystems, groups, communities, institutions, artifacts, technical systems, and organizations. Vulnerability is an emergent property of systems. Those systems, in our analysis, may be social and physical, cultural and natural, small and big; we will return later to a discussion of the various units of analysis and their implications for the study of vulnerability. We shall also argue below that vulnerability, as an emergent system's property, should not be considered as given, intrinsic, and essential, nor as purely negative.4 Rather, we will propose to study it from a constructivist perspective. At this point, it suf­ fices to highlight that a comprehensive characterization of vulnerability requires a description of natural, social, and technical aspects. To charac­ terize the vulnerability of New Orleans asks for a description of hurricane statistics, class and ethnicity distributions in the lower quarters, levees' qualities, and the ecosystem of the marshes (Bijker, 2007). These aspects are embedded seamlessly in the system. Vulnerability, however, cannot be analyzed by looking only at the inside of a system. It also relates to that system's environment. The same person, clad in many layers of wool, can be invulnerable up in the Himalayas but vulnerable in Hyderabad, India. A characterization of a system's vulner­ ability thus requires attention to its environment, including the risks that it runs as well as the resources that it may draw upon. Finally, vulnerability is not purely negative. It is, as we suggested previ­ ously, at least inevitable in our highly developed "technological cultures"

Studying Vulnerability in Technological Cultures


and "risk societies." In some contexts, it may even assume a slightly posi­ tive connotation. Some degree of vulnerability, for example, goes hand in hand with openness to change and willingness to engage with the uncer­ tainties that are associated with learning and innovating. Our concept of vulnerability thus is hybrid in three important ways. First, it is natural, sociocultural, and technical. Second, it relates to the unit of analysis, as well as to its environment. Third, it can have both negative and positive connotations. This threefold hybridity is important because it raises new questions and thus complements risk-based approaches.

Studying Vulnerability in Technological Cultures 7 and "risk societies." In some contexts, it may even assume

Vulnerabi l ity and Risk: A Review

Vulnerability is a problematic concept-why use it at all? Risk has always played an important role in technological system theories, while vulner­ ability is a relatively recent addition to the way of thinking about the risks and failures of technological systems. And indeed, the distinction between the two concepts has not always been clear. Several attempts have been made to clarify the relation between risk and vulnerability. In more recent literature, it is generally assumed that vulnerability is a broader concept than risk (Bijker, 2006; Leach, 2008; Sarewitz, Pielke, & Meykhah, 2003). Risk is often associated with quantitative approaches (risk = probability of hazard x the impact of hazard); vulnerability, in contrast, is linked to situ­ ations that are less specified and more difficult to predict because it results from social processes that are partly outside the system. This assessment, however, pertains to only one category of risk analy­ sis approaches: the traditional quantitative approach of risk estimation, dominated by engineers and natural scientists. Recent work in risk studies has incorporated more qualitative aspects. These studies can be grouped roughly into four categories: 1) risk perception research, 2) cultural and symbolic studies of risk, 3) risk-society theories, and 4) governance theories

Studying Vulnerability in Technological Cultures 7 and "risk societies." In some contexts, it may even assume


to risk. 5

Studying Vulnerability in Technological Cultures 7 and "risk societies." In some contexts, it may even assume

Risk perception research emerged since the 19 70s, when concerns about new hazards such as the use of pesticides and nuclear power became more prominent in public debates (Slovic, 2000). Risk perception scholars Paul

Studying Vulnerability in Technological Cultures 7 and "risk societies." In some contexts, it may even assume

Slovic and Baruch Fischoff conducted large-scale surveys in which they asked how people perceived risks (Lofstedt & Frewer, 1 998), their basic question being "Who fears what, and why?" The risk perception approach builds on the assumption that different people perceive risk in different ways, depending on their experience, position, and agency to change the situation. The risk of traffic is perceived to be less than that of cancer, partly

Studying Vulnerability in Technological Cultures 7 and "risk societies." In some contexts, it may even assume

because we can choose not to be part of traffic. Results from studies of

  • 8 Wiebe Bijker, Anique Homme!s, and Jessica Mesman

risk perception have been used to explain and predict why people would oppose specific technologies (Slavic, 1998, 2000). Recently, risk perception scholars have argued for a more democratic control of risk assessment by a closer involvement of laypeople. What started as a study of individual psychology (often cast in quantita­ tive analyses, subsumed under the heading of "psychometry") developed into a more constructivist view on risk perception in the late 1980s and early 1 990s. Risk perception was no longer seen as a purely individual mat­ ter, but as something that strongly interrelates with social, cultural, and political factors, such as the role of worldviews, gender, and trust (Lofst­ edt and Frewer, 1 998; Slavic, 2000). Douglas and Wildavsky (1982) argued, for instance, that the factors that determine which risks are perceived as acceptable cannot be explained by individual psychology or objective real­ ity. Instead, risk perceptions develop out of social and cultural processes that are unique to each society and are affected by values, ideologies, and the sociopolitical economic situation (Beck, 1986, 1992; Jasanoff, 1993; Johnson & Covello, 1987; Krimsky & Golding, 1992; Nelkin, 1989; Renn, 2005; Wynne, 1 989). The key scholar representing these cultural and sym­ bolic studies of risk is the anthropologist Mary Douglas. She was among the first to argue that risk judgments have a clear political, moral, and aesthetic dimension and that risks are socially constructed (Douglas, 1992; Lupton,

Wiebe Bijker, Anique Homme!s, and Jessica Mesman risk perception have been used to explain and


The third category, risk society theories, is represented by scholars such as Ulrich Beck and Anthony Giddens. They are primarily interested in how conceptions of risk are related to modernity. According to Beck (1992), we are experiencing a transition from an industrial society to a "risk society." In a risk society, risks are uncertain and difficult to calculate and predict and therefore can have a devastating impact on society. Rather than the "logic of wealth distribution, " which structured earlier societies, the "logic of risk distribution" is the key mechanism in a risk society. The normative aim of the risk society is achieving safety (rather than equality, as in the industrial society) and this goal is framed in negative and defensive terms:

Wiebe Bijker, Anique Homme!s, and Jessica Mesman risk perception have been used to explain and

"one is no longer concerned with attaining something 'good' but rather with preventing the worst" (Beck, 1992, p. 49). In a similar vein, Giddens stresses that risks are always present in the abundant choices we have, in the technologies we use, and in the experts we trust. Just as Beck does, Giddens emphasizes that we live in a "risk culture," where risks are created by humans and have a more far-reaching effect because of the globalized nature of these risks (Giddens, 1 990, 1991). However, unlike Beck, Gid­ dens stresses not only the risks and dangers, but also the opportunities of

Studying Vulnerabil ity in Technological Cultures


Table 1. 1 Different vocabularies associated with risk and vulnerability










Opening up





















Studying Vulnerabil ity in Technological Cultures Table 1. 1 Different vocabularies associated with risk and vulnerability

prevention-precaution, where the first has spurred decades of science, tech­ nology, and practices, while the latter still invokes deep political and theo­ retical debate. Risk vocabulary is prosaic in offering clear descriptions and definitions, often with quantitative means, whereas vulnerability vocabu­ lary is more poetic and qualitative, adding emotions and openness to alter­ native imaginations. In this new dialect that we seek to develop by adding vulnerability vocabulary to risk vocabulary, the cultural is added to the societal. Gemein­ schaft is added to Gesellschaft, and solidarity to security. To an important emphasis on legal structures and relations, an open-ended and less focused attention to justice is added. To complement a probabilistic analysis of problems and decision making, a qualitative and discourse-based atten­ tion to ethical questions is called for. Proper procedures are important and should be explicated as much as possible to build sound democratic frame­ works for dealing with hazards, but a broader notion of prudence needs to be added in recognition of the inherent unpredictability of science and technology and openness of cultural developments. Humility in recogniz­ ing the limitations of the human capacity to harness the world should complement sophistication in finding social-scientific and technological means to control risks Qasanoff, 2007).

12 Wiebe Bijker, Anique Hommels, and j essica Mesman Studying Vulnerability: Units of Analysis After comparing
  • 12 Wiebe Bijker, Anique Hommels, and j essica Mesman

Studying Vulnerability: Units of Analysis

After comparing the concepts of risk and vulnerability, we will now con­ tinue our review of the literature by concentrating on vulnerability. We will structure this review by discussing three units of analysis: the human world, the human-made world, and the natural world. Humans, both as individuals and as social ensembles, constitute the first unit of analysis; socio-technical systems and organizations are the second; and ecosys­ tems are added to compensate for the anthropocentric biases of the first two. After thus reviewing the literature on the vulnerability of humans and organizations, as well as of ecosystems, we will turn to a discussion of research strategies and heuristics.

12 Wiebe Bijker, Anique Hommels, and j essica Mesman Studying Vulnerability: Units of Analysis After comparing

Vul nerability of People

Human beings can be harmed, they can die, and their communities may be disrupted or demolished. However, not all individuals are equally vul­ nerable. Much of the scholarly attention to the vulnerability of people has focused on developing countries (Bankoff, 2001; Bhatt, 1998; Hilhorst and Bankoff, 2004). Often, researchers in this area use the notion of vulner­ ability to move away from a risk-based approach, which they consider too limited in scope and explanatory power. Hilhorst and Bankoff (2004), for instance, claim that a vulnerability approach allows for more adequate measurement of exposure to risk because it incorporates social factors. Disasters, they argue, are affected by social power structures that generate unequal exposure to risk. In particular, they suggest that quantitative risk­ based approaches are too limited because they do not incorporate factors such as class, gender, and ethnicity. Their vulnerability approach, on the other hand, "seeks to combine the risk that people and communities are exposed to with their social, economic, and cultural abilities to cope with the damages incurred" (Ibid., p. 2). John Twigg adds that such vulnerabili­ ties may change over time: "More and more people have become vulner­ able to hazards because of changes in their social, economic, cultural, and political environment and circumstances." (Twigg, 1 998, p. 2) An analysis of the social dynamics of international and economic rela­ tions may contribute to our understanding of how development projects generate vulnerability, according to Hilhorst and Bankoff (2004). Thus, Martin (1996) proposes to investigate whether corporate or political inter­ ests contribute to maintaining the causes of vulnerability or the vulnerabil­ ity itself. Bankoff (2001, p. 1 9) criticizes vulnerability approaches in which tropical areas or "developing countries" are designated as "disease-ridden,

Wiebe Bijker, Anique Hommels, and jessica Mesman property of an individual or a group; rather,
  • 14 Wiebe Bijker, Anique Hommels, and jessica Mesman

property of an individual or a group; rather, it is socially constructed and political by nature. (We will return to this when discussing our research heuristics in the next section.) This literature allows us to broaden the con­ cept of vulnerability by paying attention to issues of development, poverty, justice, and life stories. Most existing research on these questions, however, generally ignores the role of technology in the construction of vulnerabil­ ity. In this volume, we seek to remedy that neglect.

Wiebe Bijker, Anique Hommels, and jessica Mesman property of an individual or a group; rather,

Vulnerability in Technological Systems and Organizations

Wiebe Bijker, Anique Hommels, and jessica Mesman property of an individual or a group; rather,

The second unit of analysis that we want to distinguish is non-human. Here, we do not want to limit ourselves to artifacts but analyze socio-technical systems and organizations as well. In plain parlance, this includes objects such as aircraft, large technical systems such as nuclear power plants, and organizations such as a military defense system. Much scientific discourse on vulnerability is rooted in a conceptual­ ization of society as increasingly complex and constituted by science and

technology. In the risk society posited by

Beck ( 1986, 1992), the implica­

tions of modern science and technology are more important determinants of societal development than the production relations that described the development of industrial societies. In the "network society" proposed by Manuel Castells (1996), infrastructures have become more and more intertwined. This intertwinement has important implications for the vul­ nerability of this society. According to Bekkers and Thaens (2005), "our dependence on ICT [Information and Communication Technology] and the fact that they are all interconnected makes our societies vulnerable. · · · disturbance in one of these networks could spread relatively easily to the networks" (p. 37). Lukasik (2003) argues that dependencies between systems are often unplanned and unknown. Winner (2004) agrees that dependency on technological systems is a source of vulnerability, but that, paradoxically, people in the global north are quite happy with it because they consider it a prerequisite for progress and freedom. The complexity of systems increases as a result of these interdependencies, and no single person or organization is fully knowledgeable about the whole system. As a result of the increasing complexity and intertwinement of these systems, Lukasik argues, the impacts of system failures are aggravated. Moreover, if the system fails, no single organization can fix the problems. The relation between complexity and vulnerability is a key focus in the Normal Accidents Theory (NAT) of Perrow ( 1984/1999). He demonstrated

Wiebe Bijker, Anique Hommels, and jessica Mesman property of an individual or a group; rather,

that organizations do not behave as rationally as is usually assumed: deci­ sions may not be clear and straightforward, priorities do not always match,

Studying Vulnerabil ity in Technological Cultures


all people to live in their own ecological circumstances. Embracing such a principle of ecological justice would, for example, fundamentally change the discourse on the vulnerability of the adivasi people in India, who are displaced to make way for big infrastructural works such as dams. Classic forms of justice would call for compensation measures: new land or some money. Our broad principle of ecological justice would question the right to move these people from their ecological environment in the first place, recognizing that their technological culture is inextricably bound to that specific ecology. When broadened in this way, it is similar to Shiv Visvana­ than's concept of cognitive justice: the right of every people to have their own style and system of knowledge. Our broad principle of ecological justice also resembles Oliver-Smith's concept of vulnerability as a "political ecological concept." "Vulnerability is the conceptual nexus that links the relationship that people have with their environment to social forces and institutions and the cultural values that sustain or contest them" (2007, p. 10). By locating vulnerability at the intersection between nature and culture, Oliver-Smith enables us to focus on the totality of relationships in a given situation to prevent environ­ mental determinism. So, even though the ecological system as the third unit of analysis does not figure explicitly in this volume, recognizing it as part of the people-technology-nature triad is an important building block in conceptualizing risk and vulnerability, and as such, it is reflected in the thinking and researching of the authors of this volume.8

Heuristics of Studying Vulnerabil ities

We have now sketched our general approach to vulnerability, reviewed the relationship with the concept of risk, and discussed three important units of analysis. In addition, we have identified three different forms of heterogeneity. The first heterogeneity of vulnerability is that vulnerability pertains to the natural, the social, and the technical; the second is that vulnerability is a system's property, and at the same time, it is context dependent; and the third is that vulnerability is inevitable, but it is not automatically negative-it can be positive. These forms of heterogeneity can be translated into research heuristics for studying vulnerability. In this final section of this introduction, we shall elaborate the research heuristics that inform the chapters in this book. The first heuristic is an ethnographic approach to studying vulnerability by choosing as a research entry point technological culture rather than modern society. The second is to employ a constructivist approach; and the third is to transcend an exclusively

Wiebe Bijker, Anique Hommels, and J essica Mesman negative connotation of vulnerability and also look
  • 20 Wiebe Bijker, Anique Hommels, and J essica Mesman

negative connotation of vulnerability and also look for more charitable aspects.

An Ethnography of Vulnerability in Technological Cultures

We already argued in the beginning of this introduction that technolo­ gies are used to increase reliability and enhance safety, but they also cause vulnerability. Medical equipment in intensive care units allows a better monitoring of newborn babies, and higher dikes help to prevent flooding. But being dependent on electrical networks makes us vulnerable to power outages, and if you do not have dikes, they cannot break. Thus, vulnerabil­ ity is a characteristic of technological cultures. Therefore, we propose to study vulnerabilities that happen in techno­
logical cultures. The first implication is to trace vulnerabilities' relation to technologies. This can be a lack of technologies (as when the lack of simple water purification technologies causes a high mortality by cholera), or it can be the unintended effects of the use of technologies (such as the increased financial debts of Indian farmers because of their need to buy chemical pes­ ticides). Almost all instances of vulnerability in today's societies are shaped
by technologies. At the same time, our default defense mechanism against vulnerability is to call upon technology too. To improve patient safety in a high-risk and technology-intensive environment such as an intensive care unit, we typically invest in more technologies: electronic technolo­ gies for monitoring and social technologies of protocol to discipline doc­ tors and nurses. When the livelihoods of hand-weaving communities in India are threatened by globalizing markets because their customers start buying mass-produced synthetic and brightly printed fabrics, one reaction is to invest in the technology of the power loom and embrace the admin­ istrative technologies of centralized marketing. When the Netherlands is threatened by rising sea levels due to climate change, the country responds by investing in more technology (such as levees). The second implication of studying vulnerabilities happening in tech­ nological cultures is to study vulnerability in vivo. This requires the use of ethnographic research methods and the investigation of these technologies with a special sensitivity to the social and cultural dimensions of technol­ ogy by drawing on the work in STS of researchers such as Hackett et al. (2008). Studying vulnerabilities in technological culture will help to bring out the interactions between the various dimensions of vulnerability: tech­ nological, scientific, social, economic, political, ethical, and cultural. We will elaborate on this in the next section.

Wiebe Bijker, Anique Hommels, and J essica Mesman negative connotation of vulnerability and also look

Studying Vulnerability in Technological Cultures


A Constructivist Perspective on Vulnerabil ity

Taking a constructivist perspective means that we analyze vulnerabilities in their historical, political, economic, and geographical context. We pro­ pose to view vulnerability not as an intrinsic and static characteristic, but rather as an emergent property that will depend on and result from specific circumstances. A failure of an elevator in a hospital may have more severe consequences than a similar failure in an apartment building. Thus, we assume that vulnerabilities are not fixed properties of systems; rather, they are context dependent and usually change over time. The chapters in the first part of this book provide detailed examples of this emergent character of vulnerability. In some cases, people's vulnerability might not be resolved, but the form of vulnerability may change over time. In the case of emergency commu­ nication, for instance, a major problem can be lack of coordination. In the Netherlands, for example, various different emergency units (police, fire brigade, ambulance, etc.) had trouble communicating with each other because they had different radio frequencies. After the introduction of ICT and a new technical standard, new vulnerabilities emerged in the form of an increased dependence on a small number of network suppliers (Hommels & Cleophas, 2008). Jessica Mesman gives another example in her chapter on critical care medicine in this book. She shows how the practice of double­ checking medication by nurses (intended to reduce patient vulnerability) may lead to a disturbance of the concentration of the nurses who are asked to do this double-check. This may cause a nurse to commit errors, resulting in new forms of vulnerability for the patient. Also, the way of coping with vulnerabilities differs in different contexts (e.g., the way of dealing with water management in the United States and in the Netherlands), depending on the differences in technological culture (Bijker, 2007). Several authors in this volume scrutinize these "local" specificities of vulnerability. Furthermore, a focus on vulnerability means taking seriously the role of vulnerable groups or people in technological cultures. People may be vul­ nerable, and their situation may seem inescapable, but applying a construc­ tivist perspective may offer new perspectives and coping strategies. Where, how, and when does vulnerability become a problem, and to whom? Which vulnerabilities do different social groups identify? In many high­ risk environments, safety and vulnerability are aspects that are negotiated among the actors involved. In railway maintenance, for example, trade­ offs have to be made between productivity (getting the job done on time) and the safety of the maintenance workers. How are safety and reliability


Wiebe Bijker, Anique Hommels, and Jessica Mesman

negotiated? What is the role of risk perceptions or estimations of a system's vulnerability for actions that were not taken? A constructivist perspective on vulnerability means that we take a close look at the negotiations among actors about the safety or risks involved in the systems and organizations in which they work. A constructivist approach also opens up alternative strategies. "Things could have been otherwise, " as one of the key dictums of constructivism goes. So, instead of the default response of expecting technology to defend against vulnerability, other choices could be made, and in some cases, they have been made. Safety in intensive care units could be enhanced by other forms of social interaction and training; hand-loom weavers could try to innovate their designs and logistics so as to cater to high-end international niche markets; rather than raising its river dikes, the Netherlands has embraced a new program called "more space to the rivers, " which aims to increase the water storage capacity of the rivers by allowing selected farm lands next to the river to flood.

22 Wiebe Bijker, Anique Hommels, and Jessica Mesman negotiated? What is the role of risk perceptions

Vulnerabil ity as Potentially Positive

When we say that a system is vulnerable, we typically want to say that it is susceptible to harm. Vulnerability thus is often characterized as a prop­ erty or characteristic of systems, though a socially constructed one. Mostly, vulnerability is used as a specific rather than a generic characteristic: a city may be vulnerable to damage by specific disturbances such as floods and not be vulnerable to damage by other disturbances such as earthquakes. In such contexts, vulnerability is generally taken as something negative to be avoided, and harnessed if it cannot be avoided. The third heuristic that we used in this volume is that vulnerability need not be considered exclusively negative. Vulnerability of a culture can even be considered a necessary condition for its survival: only when a culture is capable of learning, innovating, and reacting to external threats in a flexible manner will it be sustainable in the long run. To be innovative, Schumpeter (1942) has argued, one has to be creative and take risks, and that implies some degree of vulnerability. This idea is taken further by the notion of "creative dissent, " discussed in the chapter by Shambu Prasad in this volume, which argues that dissent is necessary to innovate and that innovation, in turn, is necessary for coping with the vulnerabilities that India faces today. Snook's theory on practical drift describes how gradual deviation from protocols can lead to accidents, and it highlights the role that rules can play in making an organization safer. His example of the friendly-fire accident

Studying Vulnerability in Technological Cultures


in northern Iraq shows how a gradual deviation from the rules results in an increasingly vulnerable organization. This might lead some to the conclu­ sion to apply rules more strictly. Others have argued that flexible applica­ tion of the rules can be beneficial or even necessary for normal functioning of technological systems and organizations. It has also been argued that for a smooth functioning of technical systems, it is sometimes necessary to take risks. When air traffic controllers want to upset the air traffic sys­ tem (for example, because of a labor conflict), they do not need to strike­ working to the rule is enough to obstruct the system. John Law's analysis of the Ladbroke Grove train accident in 1 9999 provides a good example of the positive effects of flexible rule following for the "normal" functioning of technological systems. The signalers who had to stop a train if it passed a red light had developed a practice of waiting because they knew from experience that this signal would normally be turned to green within 20 seconds. This practice (although against the rules) usually guaranteed a quick and accident-free train passage-but not in this case: "waiting and seeing" on that fatal day caused a major accident.) In other words, a system that always follows strict rules will probably not function well for any pro­ longed period. A culture that is not also a bit vulnerable will not survive. Another concept that highlights a positive aspect of vulnerability by stressing the need for flexibility and improvisation is bricolage. The notion of bricolage (Ciborra, 2000) starts from the idea that improvisation and tin­ kering are necessary strategies when dealing with occupational risks under time constraints. In chapter 10 of this volume, "Vulnerable Practices: Orga­ nizing through Bricolage in Railroad Maintenance," Johan Sanne argues that if railway maintenance technicians would be less productive and less efficient-and perhaps even less safe-if they did not act as "bricoleurs." After all, rules do not apply to all situations under all circumstances, so some creativity is necessary. At the same time, Sanne acknowledges that brico­ lage can be risky because procedures and resources will not be improved so long as the maintenance technicians manage to do their jobs successfully (by using bricolage), thus "hiding" potential weak spots in their practices. This can lead to increasing negative vulnerability. This argument about positive vulnerability mirrors the warnings that historians and sociologists of technology have regularly made against a belief in "technological fixes." As sociologist Benjamin Sims observed when introducing a special panel on Hurricane Katrina at the 2005 Soci­ ety for Social Studies of Science ( 4S) meeting: "One danger of relying too much on technological fixes is that these typically create yet more lay­ ers of infrastructure, which can make emergency response organizations

Studying Vulnerability in Technological Cultures in northern Iraq shows how a gradual deviation from the rules


Wiebe Bijker, Anique Hommels, and j essica Mesman

Sally Wyatt's provocative statement that "our guilty secret in SIS is that really we are all technological determinists" (Wyatt, 2008, p. 1 75), as we do take seriously the obduracy that emergent properties of technology can acquire over time.

  • 5. This categorization is (loosely) based on the one by Deborah Lupton (1999).

  • 6. We are aware that others, like Ben Wisner and Piers Blaikie (e.g. Wisner, Blaikie,

Cannon, & Davis, 1994), have identified points of dialogue with risk theories and

risk society theory in particular.

  • 7. Perrow (2007) refined the distinction between dependence and interdependence

of social organizations in this context. Whereas dependence is something that should be avoided, interdependence should be promoted: "Power involves depend­ encies; safety and security in the infrastructure involves interdependencies" (p. 296) . Interdependence is a situation in which there are many actors involved, producers, suppliers, and multiple customers: "if anyone of these fails, there are alternative sources of producers, suppliers, and customers" (p. 300).

  • 8. If nature were to be explicitly studied, we would of course have conceptualized

it as socially constructed: as informed by recent SIS work in general and work in the sociology of scientific knowledge (SSK) specifically.

  • 9. John Law, "Ladbroke Grove, or How to Think about Failing Systems," published

by the Centre for Science Studies, Lancaster University, at http://wvvw. lancs.ac.uk/


10. The distinction between parts 1 and 2 is one of emphasis: all chapters in the first part employ sophisticated conceptual frameworks, and all chapters in the second part are empirically grounded.


Part I

To discuss vulnerability in technological cultures requires that we first identify what to study.Theref ore, the first part of this volume is devoted to gaining a better understanding of what vulnerability is and does.Based on historical as well as contemporary accounts, these empirical chapters pro­ vide a detailed exploration of the problem of vulnerability in technological cultures and demonstrate how vulnerability is an emergent and relational property that pervades everyday life.Each of the five chapters shows when, where, and how vulnerability affects people's lives and the socio-technical structures they live in. The set of case studies in part I spans different technological domains, such as agriculture, medicine, and the chemical industry. In addition, the chapters display a great variety of research materials and styles. Empiri­ cal material includes in-depth interviews with policy makers, ethnographic observations in hospitals, and the highly charged testimonies of victims. Moreover, the style of telling these stories differs with each author, ranging from an evocative and politically engaging style in chapter 6, "Narratives of

Vulnerability and Violence:

Retelling the Gujarat Riots, " by

Shiv Visvana­

28 Part I To discuss vulnerability in technological cultures requires that we first identify what to

than and Teesta Setelvad, to a more detached academic tone in other chap­ ters.The styles of presentation and analysis also differ strategically between the authors in order to give full credit to the varied character of the empiri­

cal material and the issues at stake. If we want to engage seriously with the vulnerability of people in technological cultures, and more specifically with what has been happening in Gujarat, a different language and indeed imagination are both called for. Despite the fact that the majority of the chapters in this part focus on Indian examples, we claim that the argu­ ments developed in this part-just as in the other two parts-transcend the local contexts from which they originated.For example, the idea of the multiplicity of vulnerability is demonstrated both by Julia Quartz in chap­ ter 2, "Agricultural Change in a South Indian Village: An Account of the Multiplicity of Vulnerable Livelihoods, " on Indian farmers; and by Jessica Mesman in chapter 4, "Relocation of Vulnerability in Neonatal Intensive Care Medicine, " on neonatology in a Dutch hospitaL Part I begins with the chapter by Quartz, who studies the livelihood vulnerabilities of farmers in India. Building on developments in the field of livelihood studies, she starts her analysis by applying a systemic per­ spective to vulnerability that stresses the interrelatedness of nature-related factors and social structural factors.To avoid an essentialist notion of vul­ nerability, she uses a diachronic perspective to analyze vulnerable liveli­ hoods. Using this frame of analysis, she describes how the perception of who is to be considered as vulnerable changes diachronically between the

28 Part I To discuss vulnerability in technological cultures requires that we first identify what to

framing the

Vulnerabil ity Issue


Green and Gene revolutions. Moreover, after applying a multiplied syn­ chronous frame of analysis to policymakers, industries, and civil society groups, Quartz demonstrates how the meaning of vulnerable livelihood also differs synchronically. In this way, the chapter exemplifies not only the multiplicity of vulnerability, but also the fact that vulnerability is an issue of negotiation, and therefore also an issue of power structures. This issue of power structures is also the central theme of chapter 3, "Cultural Politics of Vulnerability: Historical-Ethnography of Dearth and Debt, and Farmers' Suicides in India, " in which Esha Shah pays attention to (hi)stories about vulnerability and focuses on collective memories in India. However, whereas Quartz focuses on the modes of reasoning of poli­ cymakers, Shah studies the farmers. Similar to Quartz, Shah demonstrates how vulnerability is not so much shaped by a deficit of food and produc­ tivity, but by structural sociocultural inequalities, which in turn form the basis for a structurally unequal distribution of resources. Shah describes how liberal techno-institutions equate rural vulnerability with economic scarcity. In this rational economic approach, according to Shah, vulner­ ability is used as a political instrument to legitimize the introduction of new technological regimes like the Green Revolution and the Gene Revo­ lution. Shah challenges this economic association between vulnerability and scarcity by employing the notion of "cultural politics." This cultural approach helps her to study the sociocultural connotations of scarcity in relation to farmers' suicides. Drawing on a detailed analysis of everyday conversations and historical narratives of famine, this chapter demon­ strates how cultural imaginaries about scarcity and pauperization lie at the basis of vulnerability, and how cultural narratives play a crucial role in this matter. On the basis of these findings, she rejects the limited scope of the rational economic framework that refers to the dynamics of globalization and liberalization in order to explain farmers' suicides. Instead, the experi­ ence of agrarian vulnerability in India, Shah argues, is closely connected to a deeply ingrained cultural ideology of hierarchy and to Indian notions of the good life. Shah's chapter underlines the multiplicity of vulnerability. In a simi­ lar vein, Jessica Mesman describes in chapter 4 how, when, and where these multiple vulnerabilities coexist, compete, and complicate practices. Whereas Quartz and Shah have stressed that vulnerability is an everyday experience of particular social groups, Mesman elaborates on the every­ dayness of vulnerability to underline that vulnerability is an emergent and relational property. To do so, her case study is situated in the high­ tech environment of an intensive care unit for newborns. By analyzing

framing the Vulnerabil ity Issue 29 Green and Gene revolutions. Moreover, after applying a multiplied syn­
framing the Vulnerabil ity Issue 29 Green and Gene revolutions. Moreover, after applying a multiplied syn­
framing the Vulnerabil ity Issue 29 Green and Gene revolutions. Moreover, after applying a multiplied syn­

Framing the Vulnerabil ity Issue


Visvanathan and Setelvad also show that the notion of vulnerability is closely tied to that of citizenship. In their chapter, citizenship is couched as a claim to normalcy, welfare, and well-being, while vulnerability is the breakdown of this expectation of normalcy. Studying vulnerability, they argue, should not only focus on susceptibility for disruption, but also should include the ability to recover. To examine people's ability to restore normalcy, they focus on the aftermath of the Gujarat riots in India in 2002. Visvanathan and Setelvad point to the analytical value of narratives: If one wants to understand vulnerability, one should pay attention to the chang­ ing nature of the stories about it. Normalcy, they argue, requires rituals of storytelling. If these narratives cannot be told due to collective denial or indifference, there is no closure. Without these narratives of truth and jus­ tice, there is no return to normalcy, and vulnerability continues. Without ignoring the existence and analytical strength of vulnerability as an objec­ tive, systemic, and scientific concept, Visvanathan and Setelvad prefer a phenomenological approach in which the subjective experience of the vic­

tims of the riots is at the center of attention. In giving victims a voice, the authors aim to provide an antidote against objectification, indifference, and ruthlessness with respect to experiences of victims of collective vio­ lence. In this way, survivors become subjects of history instead of mere

obj ects of violence.

While doing this, the role of technology is not denied;

rather, it is taken into account on different levels: as symbol (the train to Pakistan), as materiality (the artifacts that were used in the killing), and as a metaphor (the technocratic style of managing society). In this way, the authors show how technologies and the technological culture that comes with lt work together to create vulnerabilities through framing the domi­ nant discourse. All the chapters in this first part of the book address in their own way what it means to study vulnerability in technological cultures. Vulnerabil­ ity, these chapters show, cannot be considered as one stable entity; instead, it is firmly embedded in the context in which it emerges. By opening up a diversity of domains ranging from legal, political, medical, and agrarian practices, part I demonstrates the multiplicity of vulnerability. It can, for example, be related to people's experiences, to technological equipment, or to societal institutions like law and democracy. Furthermore, vulner­ ability is explicated in many different forms, like narratives or conditions of bodies and machines. The broadness in scope of these chapters testifies to the wide range of issues and perspectives that we hope to address via the concept of vulnerability.

Framing the Vulnerabil ity Issue Visvanathan and Setelvad also show that the notion of vulnerability is


Vulnerability and Deveiopment-Bhopal's lasting


Sheila jasanoff

On December 2, 2007, 23 years after the gas leak incident in India that is still regarded as the world's worst industrial disaster, the multimedia news conglomerate Asia News International published a story headlined "Bhopal Gas Tragedy Victims Battle the Odds Sans Adequate Compensation." 1 Not long after, the International Campaign for Justice in Bhopal (IC]B), a coali­ tion of nongovernmental organizations and activists advocating for greater corporate responsibility, reported that 50 survivors of the disaster had com­ pleted the 800-kilometer trek from Bhopal to Delhi on foot; the aim of these padyatris, or marchers, was to hold Prime Minister Manmohan Singh to what they said were broken promises of rehabilitation, environmental cleanup, and clean water in Bhopal,Z Also in 2007, Animal's People, a novel by Indra Sinha that was based loosely on the Bhopal disaster, was short­ listed for Britain's prestigious Man Booker Prize for literature (Sinha, 2007). The author credited part of the book's inspiration to his conversations with Sunil Kumar Verma, a mentally disturbed Bhopal survivor whose suicide in 2006 exemplified for many the failures of relief for gas-affected residents of the city. These gestures captured the undying force of memory in a deeply aggrieved and injured population, but as Shiv Visvanathan and Teesta Setelvad suggest in chapter 6, "Narratives of Vulnerability and Violence:

Retelling the Guj arat Riots, " history has a way of squeezing the most excru­ ciating narratives of vulnerability into banal and comfortable storylines, most pleasing to the winners. In a possibly final writing of history's pages, a district court in Bhopal in June 2010 convicted eight Indian Union Carbide executives (including one that was then deceased) of criminal negligence for their parts in the disaster. The modest fines and two-year prison sen­ tences that the court imposed struck the surviving victims and their fami­ lies as too little, ludicrously too late, and lacking any proportionality to the enormity of the disaster. A leading spokesman for the victims, Satinath


Sheila jasanoff

Sarangi, denounced the verdicts, comparable in severity to sanctions for vehicular homicide, as "the world's worst industrial disaster reduced to a traffic accident" (Polgreen & Kumar, June 7, 2010). Almost a quarter cen­ tury after the tragic events that ravaged Bhopal, and decades after a $470 million financial settlement that was by far the largest of its kind in India, these episodes offer poignant but puzzling evidence of a human tragedy that refuses to end. The scale of the Bhopal disaster and the persistence of loss and suffer­ ing would be enough in themselves to make this a paradigm case of vul­ nerability in technological societies. But the story has added dimensions that make it especially germane to the themes of this volume, including that of transnational vulnerability. Bhopal was an instance of technology transfer gone terribly awry in the relatively early days of industrial global­ ization. What happened there offers sobering insights when globalization promises to make transfers of technology more pervasive and less respect­ ful of epistemic, political, and cultural borders than ever before. From the likely proliferation of nuclear power, with its attendant questions of waste disposal, to the spread of new agricultural biotechnologies, nanotechnol­ ogy, synthetic biology, and their potential health and environmental con­ sequences, we confront an era when hazards flow freely around the world, although responsibility for risk governance, enforced by conservative legal systems,3 still remains largely contained within national boundaries. Some have termed this a time of incalculables, of unknown unknowns more suited to preparedness than control (Lakoff & Collier, 2008). A reflection on Bhopal, informed by history but with an eye to the future, may offer diagnostics for dangers that lurk beneath the surface in such a world. The institutional fissures and incommensurabilities that Bhopal laid bare offer a ghostly X-ray image of the deep structures of global vulnerability. Many systems failed in Bhopal, but the primary focus in this chapter is on the law. The accident posed special challenges to the law, whose insti­ tutional purposes preeminently include the settlement of disputes, the remediation of harm, and the reimposition of order in contexts that have been subjected to abnormal strain or disruption. Whether disputes arise in the most intimate of social relations, such as the family, or from widely dispersed and loosely connected transactions, such as transnational tech­ nology transfer as in Bhopal, the law's function is not only to remedy prob­ lems but to make sure, as far as possible, that similar breakdowns will not occur again-at least not within that particular type of relationship or from those specific causes. In this respect, the law is one of society's premier mechanisms for reducing vulnerability while also ensuring social cohesion,

Sheila jasanoff Sarangi, denounced the verdicts, comparable in severity to sanctions for vehicular homicide, as "the

Vulnerabil ity and Development-Bhopal's Lasting Legacy


stability, and justice. Law and lawyers were amply represented in attempts to repair the social and material fabric destroyed by the Bhopal disaster. It is therefore especially interesting for analysts of vulnerability to ask why, despite the law's intense engagement with this case, the victims continue to believe that there was no satisfactory closure, no proper redress, and hence no effective deterrent for the future. Much of the answer, as we shall see, lies in the collision of cultures that occurred in Bhopal: between India and the United States, layperson and expert, and activist and corporation, among others.

Vulnerabil ity and Development-Bhopal's Lasting Legacy stability, and justice. Law and lawyers were amply represented in

Accident and Aftermath

For a brief moment in the late 20th century, Bhopal, the sprawling, unre­ markable capital of central India's province of Madhya Pradesh, became a staging ground for the ambiguous and contested emergence of global neo-liberalism. The histories of states and markets, medicine and law, social activism, and corporate power violently collided in that quiet pro­ vincial city. Though the loudest reverberations from that encounter have died down, the echoes will last as long as anyone is left to remember and retell Bhopal's story. What happened there in the Orwellian year of 1984 says much about the human costs of globalization. It also illustrates the law's incapacity to restore order when radically different cultures of knowl­ edge, experience, and justice come together in a deadly embrace. Just such a clash occurred in Bhopal. The results are important not only for post­ colonial historians and students of legal history, but also for students of technological globalization and policymakers concerned with the rise of transnational vulnerabilities. Bhopal's tragedy was as much about the fail­ ure of powerful institutions to generate and act upon timely knowledge of risks as it was about maimed lives and justice delayed or denied. The resulting double failure, of law and of public knowledge making, deserves a lasting memorial. Minutes after midnight on December 3, 1 984, an industrial accident of unprecedented violence shattered Bhopal.4 A cloud of heavy, deadly methyl isocyanate (MIC) gas escaped from a pesticide plant belonging to Union Carbide Corporation (UCC) and spread across some of the poorest sectors of the city. Particularly hard hit was the colony of]. P. Nagar, located just across the street from the plant and inhabited mostly by Muslims and low-caste Hindus. The acutely toxic agent caused lethal fluid buildup in the lungs, irritated the respiratory system, damaged the eyes, and induced a wide vari­ ety of reproductive disorders in the victims. An estimated 3,800 of Bhopal's


Sheila jasanoff

poorest citizens died instantly or within days of the accident, suffocated by the acutely toxic fumes. Hundreds of thousands more were injured, many suffering intense physical and mental distress years, even decades, after the accident.5 One such delayed victim was Sunil Kumar Verma, a 34-year-old activist who lost his father, mother, and five siblings in the disaster. As an adult, Verma ceaselessly campaigned for justice for the survivors until he hanged himself on July 26, 2006 (Shaini, 2006). He had been under treatment for paranoid schizophrenia, but whether is mental condition was related to the gas exposure will never be known. What we do know, how­ ever, is that no planned response, before or after the disaster, contained within it adequate or even imagined remedies for Verma's damaged life. The psychic toll of the disaster is not the only unknown that lingers over Bhopal long after a tragedy that took more lives than did the terrorist attacks of September 1 1 , 2001, in the United States. What, for example, caused the disaster? We know that water introduced into a storage tank that contained highly reactive, liquid MIC unloosed an explosive reaction, but how the water got into the tank remains a mystery. At least four expla­ nations have been offered for what went wrong, ranging from contingent, case-specific, and personal to systemic, political, and global. UCC and its corporate successors insist that it was an act of intentional sabotage by a disgruntled employee. Students of technology and society call attention to the failure of the Indian government and the exporting company to take seriously the social components of a well-functioning safety system, such as trained personnel, effective monitoring, and adequate budgetary support Oasanoff, 1994a). Policy analysts fault the Indian government's lax regulatory standards and refusal to heed early warning signs of poor management and maintenance at the plant (see, in particular, Bowonder, Kasperson, & Kasperson, 1 994). Finally, the victims and their international supporters blame a post-colonial economic order that recklessly puts lives in developing countries at risk, creating the preconditions for what Indian activists have termed genocide, or the selective annihilation of people on ethnic grounds (see, for instance, Bhopal, Industrial Genocide?, 1985). Adhering to the sabotage story, UCC, a wholly owned subsidiary of Dow Chemical since 2001, listed this entry on its website in 1 984: "In December, a gas leak at a plant in Bhopal, India, caused by an act of sabotage, results in tragic loss of life" (Union Carbide Corporation, 2010). A link led to a con­ sultant's report, prepared for the Arthur D. Little company in Cambridge, Massachusetts, by an analyst of South Asian origin, that ruled out accident and blamed the government of India for preventing an immediate, all-out investigation (Kalelkar, 1 988; Union Carbide Corporation, 2010).

Sheila jasanoff poorest citizens died instantly or within days of the accident, suffocated by the acutely
Sheila jasanoff poorest citizens died instantly or within days of the accident, suffocated by the acutely
Sheila jasanoff poorest citizens died instantly or within days of the accident, suffocated by the acutely


94 Sheila j asanoff magnitude and diversity. Galanter surveyed a decade of Indian tort cases from

Sheila j asanoff

magnitude and diversity. Galanter surveyed a decade of Indian tort cases from 1975 to 1984 and concluded that delays of Dickensian Bleak House proportions were routine, even for cases of no great complexity, and that in India (unlike in the United States), there had been no tie-in between industrial disasters and progressive developments in tort law (Galanter, 1 994, pp. 1 45-146). Galanter's work was quite possibly the first systematic study of a developing nation's lack of legal instruments, and the associ­ ated professional experience, to deter the careless operation of extremely hazardous, imported industries. India, in his uncompromising judgment, did not possess the home-grown legal competence to handle the disastrous consequences of an eminently non-homegrown technology. Galanter, like many Indian activists, concluded that the only reasonable remedy was to move the process of adjudication back to the United States, where UCC was legally incorporated and where its lethal production process was designed. Yet, despite Galanter's efforts to shift the legal venue to the United States, the claims of Bhopal's victims never went to trial, either there or in India. All claims filed in the US courts were consolidated into the South­ ern District of New York, where Judge John F. Keenan dismissed the case in May 1 986 on the legal ground of forum non conveniens (inappropriate choice of forum for a lawsuit).7 For Keenan, an empiricist common law jurist rather than a theorist of global capital, it was plainly most reasonable to litigate where the facts could be easily investigated and witnesses sum­ moned. Because the disaster had occurred in India, and the greatest num­ ber of claimants were also in India, Keenan concluded that the Indian legal system was best positioned to determine the causes and fix liability for the injuries. To support this position, he constructed an image of India's sov­ ereignty that included legal capacity as its very backbone. Invoking India's colonial past, Keenan sanctimoniously observed, "To deprive the Indian judiciary of this opportunity to stand tall before the world and to pass judg­ ment on behalf of its own people would be to revive a history of subservi­ ence and subjugation from which India has emerged. "8 Curiously, it fell to this US federal judge to defend the competence of the Indian courts against the government's own position, although his ringing defense worked in UCC's favor because the company was determined to keep the conflict con­ tained within India. In retrospect, Keenan's reasoning reveals a blockage of the imagination that, even back then, hampered any deep reflexivity about the social and political aspects of technological systems; translated into today's terms, such attitudes would equally impede creative thinking about the law's capacity to cope with vulnerability. At stake in Bhopal's aftermath was

94 Sheila j asanoff magnitude and diversity. Galanter surveyed a decade of Indian tort cases from
94 Sheila j asanoff magnitude and diversity. Galanter surveyed a decade of Indian tort cases from

Vulnerabil ity and Development-Bhopal's lasting legacy


not mainly the question of India's national sovereignty and self-respect, themes that Keenan elaborated in terms that made his judgment sound all the more patronizing. What the case broke open were the black-boxed rela­ tionships of co-production that link the development of material technolo­ gies with contextual social practices such as the law, in ways that define a culture's ways of creating and living with risk and vulnerability.9 US law's relatively high receptivity to liability claims, for instance, as well as late-20th-century US developments such as the mass tort action, synchronize well with the needs of an advanced industrial state with weak welfare provisions, in which litigation frequently, if imperfectly, stands in for social insurance. In Europe, stronger welfare states and a culture of pre­ caution have gone hand in hand with substantially less litigation than in the United States. India's history of industrialization and liability law was radically different. Both before and after Bhopal, India did indeed experi­ ence innumerable small-scale industrial accidents, but these brought mis­ fortune to an afflicted few whose sufferings seldom made their way into law courts. Bhopal entailed a scaling-up of harm that existing Indian legal technologies were entirely unprepared for. It was like fighting an AIDS epidemic without antiretrovirals, or a rampant wildfire with an ancient garden hose. The right social tools did not exist-from the means of regis­ tering, documenting, and monitoring mass injuries to the expertise needed to argue a complex lawsuit across national legal, linguistic, and cultural borders. These deficits had nothing to do with national sovereignty; they had everything to do with vastly different trajectories of industrialization, embedded in divergent histories of capital formation and colonialism. Keenan's decision, in any event, left India with little choice but to settle. In May 1989, the government of India accepted UCC's offer of $470 mil­ lion-far less than the $3.3 billion that the plaintiffs had initially sought in the Bhopal district court. That settlement not only put an end to all outstanding claims against UCC resulting from the gas leak, but also ended any further official inquiry into the facts. Formally, the case was closed. For the survivors, though, this was merely the end of the beginning of a saga that brought them neither cognitive closure nor a sense of justice achieved.

Vulnerabil ity and Development-Bhopal's lasting legacy not mainly the question of India's national sovereignty and self-respect,

Asymmetries of Knowledge and Power

In the survivors' accounts, both UCC and the Indian state were impli­ cated in equally reprehensible acts of denial-of knowledge as well as legal responsibility. UCC, for example, denied knowing both medical facts about the toxicity of MIC and management facts about what had been happening

Vulnerabil ity and Development-Bhopal's lasting legacy not mainly the question of India's national sovereignty and self-respect,


Sheila jasanoff

Sheila jasanoff at the plant under the supervision of its partly owned subsidiary, Union Carbide India,

at the plant under the supervision of its partly owned subsidiary, Union Carbide India, Limited (UCIL). The first denial left victims without effec­ tive antidotes or treatment options in the wake of the accident and later; the second denial undercut their efforts to hold UCC, the parent company, responsible for UCIL's negligence. Much of the history of Bhopal activism consists of successive attempts to counter these official, top-down denials with knowledge, arguments, and novel forms of legal and political action generated from below. But none of this wealth of social creativity altered the terms of a settlement that many regarded as fundamentally flawed and unjust. In the victims' repeated attempts to reframe and reopen the contro­ versy, one can see resurfacing the asymmetries of power-between states and corporations on one side and gas-affected people on the other. Global­ ization, it appears from the Bhopal example, is anything but even-handed in its flows and frictions. In any production system, the manufacturer inevitably knows more about what is being made than does the user-let alone those inadver­ tently exposed to explosions, toxic releases, and other failures of safety systems. To correct this imbalance, most modern regulatory systems place on the producers of hazardous substances, such as industrial chemicals, the burden of generating and disclosing information about the characteristics of their products, as well as ultimate liability for harm. In the case of MIC, however, very little was known about the chemical's long-term effects, or even about its immediate acute effects on human lungs and eyes, before the gas cloud descended on Bhopal. In effect, a huge, uncontrolled field experiment was conducted on unsuspecting human subjects with a substance that industrial researchers had found too toxic to test properly in laboratory settings. In retrospect, the decision to store large amounts of MIC in on-site tanks in Bhopal turned out to be a grotesque reversal of the precautionary principle: ignorance, far from prompting extra care, underwrote the export of a production process that was wholly unsuited to its new social context and that proved, in the end, catastrophic for tens of thousands of people. Much of what is known today about MIC, therefore, was learned after the fact, from the bodies of those killed or exposed in December 1 984 and, eventually, from the survi­ vors' children (Varma & Varma, 2005, pp. 42-43). Gathered under extreme and stressful conditions, and with large stakes hanging on the findings, post hoc information about deaths and injuries was bound to be suspect in many respects. The Indian state, in particular, deployed its medical and administrative resources with sovereign disregard that undercut the victims' claims of illness and, in one noteworthy instance,

Sheila jasanoff at the plant under the supervision of its partly owned subsidiary, Union Carbide India,

Vulnerabil ity and Development-Bhopal's lasting legacy


their subj ective experiences of healing. Relief workers who arrived on the scene soon after the disaster reported that only one treatment seemed to afford exposed persons any comfort. This was the chemical sodium thio­ sulfate, an antidote to cyanide poisoning. Government doctors, however, denied that cyanide gas had been released along with the MIC, and in an action whose arbitrariness victims' groups never forgave, forcibly put an end to efforts by a people's clinic to administer the antidote. The Indian Council of Medical Research (ICMR), once among the most trusted of the government's scientific units, forfeited its credibility and proved inca­ pable of resolving the controversy. Activists cited ICMR meeting minutes that appeared to give credence to the cyanide poisoning theory, but that were not acted upon by medical and other officials who were, as activists believed, intent on managing the news and covering up the most disturb­ ing aspects of the gas leak.10 An August 1985 interview conducted with Mira Sadgopal, a doctor from one of the people's clinics active in Bhopal, neatly summed up the differ­ ences in perception that fed distrust between the medical establishment and the victims. On one side was the weight of official knowledge, backed by class confidence and claims of objectivity; on the other, the uneducated beliefs of poor laypeople, depreciated by the powerful as merely subjective. Which side would control policy? Establishment physicians, Sadgopal said, "have got utter disregard and utter disrespect for people whom they well­ they don't like them, these poor people" (Friends of the Earth Malaysia, 1985, p. 44). And yet, she observed, the patients' personal reports were all there was to go on in the absence of formal clinical knowledge:

Vulnerabil ity and Development-Bhopal's lasting legacy their subj ective experiences of healing. Relief workers who arrived

Now when we are trying to measure exactly how much benefit people are getting from sodium thiosulphate we have very little clinical evidence at this point, I mean changes in signs. But people are talking about distinct symptomatic relief for certain symptoms like headache, muscular pain, sleeplessness, chest pain, breathlessness, palpitations, blurring of vision-the responses are not uniform but certainly they are very distinct evidence of relief of the symptoms, but these results have not been accepted so far by almost all of the doctors in Bhopal, because they say there is no improvement in signs. (p. 44)

Vulnerabil ity and Development-Bhopal's lasting legacy their subj ective experiences of healing. Relief workers who arrived

In the tug-of-war between people's "subjective" claims and physicians' "objective" knowledge, objectivity won, and the disputed antidote was withheld from most victims. The casualty, however, was trust, and no offi­ cial medical accounting of the disaster's immediate toll ever gained wide­ spread public acceptance. Knowledge of MIC's long-term effects remained similarly contested. The ICMR inexplicably ended its follow-up studies of Bhopal victims in 1 994, a

Vulnerabil ity and Deveiopment-Bhopal's Lasting Legacy


Vulnerabil ity and Deveiopment-Bhopal's Lasting Legacy but also the social factors that made it so much

but also the social factors that made it so much more difficult to initiate bottom-up policy change around the ruinous exported plant than around the merely risky homegrown facility. Bhopal's knock-on effects in the United States reconfirm much that is known from comparative studies of science, technology, and law. Specifi­ cally, the community-empowering TRI program, built on "made in the United States" legal concepts and practices, highlights how initial dispari­ ties of expertise and political power can influence even the ways in which people learn from disasters. Changes in sociotechnical systems play out against prior settled expectations about what constitutes adequate knowl­ edge, what counts as justice, and how the two should be integrated. These expectations, moreover, are consolidated and continually reperformed by powerful actors and institutions, creating the stable elements of political culture that I have elsewhere termed " civic epistemologies" : that is, shared understandings about credible knowledge claims, and how they ought to be articulated, represented, and defended in public domains Oasanoff,


If Bhopal gave rise to right-to-know laws in the United States, it was in part because technical information and "trust in numbers"17 already occu­ pied a privileged place in American activists' civic epistemologies. The EPA's description of the TRI program reaffirms the close connection in US politi­ cal culture between faith in the accessibility of obj ective information and confidence in the possibility of justice. As the EPA website further states:

"The goal of TRI is to empower citizens, through information, to hold com­ panies and local governments accountable in terms of how toxic chemicals are managed. " 18 The citizen imagined here is the classic Jeffersonian agent, capable of meaningful political action so long as she or he is adequately informed. That is the kind of citizen who agitated for more information on toxic releases; in turn, that is also the citizen whose needs the TRI was designed to meet. It hardly needs adding that such a policy response makes good sense in a nation with a long history of meeting people's welfare needs by facilitating individualized self-help-especially through informa­ tion provision and ease of access to the courts. That concept of citizenship loses all meaning in the face of Indian reali­ ties, however. For Indian activists, authenticity of experience, horizontally transmitted within communities, has proved a more reliable foundation for action than perennially suspect corporate claims or the assurances of dis­ tant, possibly influenced, state experts (on this point, see jasanoff, 2005b). The accounting systems needed to validate impersonal public knowledge, to open it up to collective witnessing and skeptical cross-checking, are


Sheila Jasanoff

not yet in place. Those institutional absences, along with the structures of administrative responsibility that legitimate knowledge enables, can exacerbate other forms of vulnerability. In Bhopal, as we have seen, the Indian government's clumsy denial of the victims' subjective, physically attested experiences heightened the local distrust of official, disembodied knowledge. Those initial missteps made any subsequent agreement on just resolutions between the state and its citizens, based on shared accounts of reality, all but impossible. From the Bhopali viewpoint, it is only through personal suffering that anyone in this stricken city has acquired the stand­ ing to speak truth to power.

A Visit to Bhopal

Much of what has been said above will be available to the talented histo­ rian of the late 21st century who wishes to look back on Bhopal as a water­ shed in the long march of globalization; and minds trained by the dialogue between science and technology studies and the social studies of risk will be particularly attentive to the co-productionist aspects of the story-that is, to the inseparable connections between social and techno-scientific orders. But are there elements of Bhopal's meaning that no archive can store and that we, as witnesses and narrators of our own present, have a special obli­ gation to record? What goes missing when thick experience is put away in document boxes, slowly gathering dust or (more often in India) being destroyed by termites and weather? It was in search of answers to these questions that I visited Bhopal with a friend in the summer of 2004, nearly 20 years after the disaster. I had written about the tragedy intermittently throughout that period (see particularly ]asanoff, 1994b) and had inter­ viewed or interacted with most of the leading figures involved in the early years of litigation, but I had never been to the city itself, and I wanted to hear from the Bhopal activists how they felt their cause had fared. Coincidentally, it was during our August visit that the Indian govern­ ment paid the final installment of the relief fund to the victims, and news­ papers were full of accounts of money misspent on consumer goods, such as motorcycles. Leaders of two major survivor organizations, Abdul ]abbar of the Bhopal Gas-Affected Woman Workers' Organization and Satinath Sarangi of the Sambhavna Trust (also known as Bhopal People's Health and Documentation Clinic), told a different story.19 Hospitals had been built, they said, but they were mismanaged and did not provide adequate care even for the gas victims, let alone for those afflicted by chronic pollution near the long-abandoned UCC plant. Relief funds, amounting to no more

Vulnerabil ity and Development-Bhopal's lasting legacy


than $ 1 ,000 per person, fell hopelessly short of covering the real costs of lifetimes of chronic illness. In a poisoned city, the chronicle of unredressed and unacknowledged medical cases had in any case swollen to include, by 2004, a generation of children born to gas-exposed parents. Their claims had not even entered the picture when UCC and the government of India agreed in 1989 to close their books on the case. jabbar and Sarangi followed different paths in their efforts to secure the moral closure that legal negotiations between UCC and the Indian state had denied to victims. jabbar's family had lived 2 kilometers from the plant at the time of the accident, in the direct line of exposure. He coughed throughout our meeting and spoke English awkwardly as he told us stories of early deaths and enduring illness. His activism, too, had stayed close to home. Jabbar's strategy was to empower local residents-not so much with knowledge as with other skills-while maintaining constant pressure on the Indian government in the hope that justice will prevail someday. His center offered women training in craft skills such as sewing and embroidery, teaching them eco­ nomic and social self-sufficiency. The women had come a long way, he told us, since the 1980s, when most still wore the traditional, all-concealing burka of Indian Muslims. They could stand up for themselves now and had learned to ask for reasons and identification if police or other officials came to question them. But jab bar's most passionate wish seemed unlikely ever to be granted: to see Warren Anderson, former chairman of UCC, j ailed "for one day, but at least for one day." Anderson had visited India once, just after the accident, and was briefly detained but then released. Subse­ quently, he was charged with culpable homicide in an Indian court and declared a fugitive from justice, but he never returned to India, and the Indian government did not press for his extradition. The criminal convic­ tions of June 2010, like other developments in Indian activism and juris­ prudence, remained applicable only to Indian nationals. Sarangi's strategy looked outward, enlisting an international network of experts and activists and soliciting attention from the media to keep pressing on the world's forgetful conscience. Workers at his clinic had documented significant growth abnormalities among sons of gas-affected parents and published their results in JAMA (Ranj an et al., 2003). In 2004, Sarangi was spearheading attempts to hold Dow Chemical legally respon­ sible for environmental damage resulting from UCC's allegedly negligent operations at the plant before the gas release. Health claims resulting from that earlier contamination, Sarangi and his supporters argued, could not have been extinguished by the 1989 settlement, which covered only the


Sheila Jasanoff

gas victims. But to pursue those collateral claims before a court-ordered time limit ran out, Bhopal activists had to persuade the government of India to declare, quickly and publicly, that its parens patriae role, as articu­ lated in the 1 985 Bhopal Act, did not extend to injuries caused by pre­ disaster environmental conditions in the city. Under severe time pressure, the activists could find no more effective instruments of persuasion than their own bodies. The language of suffering bodies still possesses in Indian political space an authenticity and a power to convince that cannot be matched by any number of published scien­ tific articles. In June 2004, Sarangi, accompanied by two of Bhopal's most famous activists-Rashida Bee and Shahid Noor-and hundreds of volun­ teers, began a time-honored Indian ritual of protest: they fasted on the sidewalk outside the Jantar Mantar, Delhi's famous 1 8th-century outdoor observatory. But for bodies to speak truth to power, they had to be seen to be wasting, and time was in very short supply. The Bhopalis decided

gas victims. But to pursue those collateral claims before a court-ordered time limit ran out, Bhopal

on a waterless fast, so that the effects would be more speedily visible, but their goal was to move the state, not to engage in pointless heroics of self­ sacrifice. Their attending physicians saw to it that they did not risk kidney failure from rapid dehydration in Delhi's punishing summer heat. A well­ coordinated campaign by phone, fax, and e-mail kept up the pressure on government officials, who bowed to the £asters' demands within a week.

on a waterless fast, so that the effects would be more speedily visible, but their goal

It cost the state little to give in to the protesters this time, other than some temporary embarrassment; all the government did was clear a legal hurdle to letting the collateral action against Dow go forward. For the Bho­ palis, however, the state's move was more consequential because it opened

It cost the state little to give in to the protesters this time, other than some

a new legal front, along with the chance to generate yet more scientific representations of hurt and injury that might circulate in worlds outside the axes of Bhopal and New Delhi. Once again, however, the creativity of the pressure tactics in India ran up against immovable legal barriers in the United States. Called into the Bhopal conflict once again, Judge Keenan held, and the Second Circuit Court of Appeals affirmed, that the plaintiffs had no legal claim to the supposedly contaminated land or groundwater, that these belonged to the Indian government, and that it was impracti­ cable in any event for a US court to ask for remediation of land owned in its own country by a foreign sovereign.20 Once again, India's sovereignty, as constructed and defended by a US judge, foiled India's damaged citizens in their search for transnational justice. It is worth stressing that for Bhopal activists, medical science did not prove to be an "immutable mobile" that adequately captures or translates

a new legal front, along with the chance to generate yet more scientific representations of hurt

Vulnerabil ity and Development-Bhopal's Lasting Legacy


to distant venues the reality of their suffering.21 Science was one speak­ ing voice among many, a tool that some in Bhopal used with indifferent success for talking across institutional boundaries and trying to enhance credibility. But what mode of representation could possibly express the activists' unending quest for proper compensation? In what discourse of persuasion could a settlement be forged, or even contemplated, between radically incompatible experiences of suffering and ideas of accountability? American public reason, formally articulated by experts in law, science, or economics? Or Indian public morality, with its vibrant, symbolic demands for social justice and the assumption of responsibility by the state for a grievously harmed community? Or neither of these? It is noteworthy that only a work of fiction, Indra Sinha's novel Animal's People, seems to have succeeded at all in capturing the Bhopalis' claims on the conscience of a forgetful world. Vulnerability reduction, and its more optimistic twin, increasing resil­ ience have emerged as terms of art in the global sustainability discourse of the 21st century. They signal important shifts in thought because they turn the analyst's attention away from the physical, natural, and even social determinants of risk to the needs of the humans and habitats most likely to be harmed. Timely engagement between factory operators, state officials, and the residents of ]. P. Nagar in the analytic framework of vul­ nerability might have mitigated the enormity of what happened in that place in December 1984. The Bhopal story alerts us, however, that changes in language alone will never be enough to increase the resilience of the world's most at-risk communities. Deeper structures of vulnerability and power need to be uncovered and corrected for. In this case, those structures came to light only through catastrophic loss and protracted struggle. They included the hidden relations of dependence between an unsuspecting community and an industrial facility with unrecognized and undisclosed hazards; between a developing country government and an economically powerful multinational; between a post-colonial imagination of progress and ideas of technological advancement imported from abroad; between India's low-risk, low-yield economy and the US high-risk, high-yield model of growth; between an untested and a mature system of liability laws; between the subjective experiences of the poor and authorized medical knowledge; and between some of India's poorest, most marginal citizens and a state as yet unprepared to care properly for them through medicine or law. Foresight, protection, and justice-under these conditions? Bho­ pal's victims never had a chance.

Vulnerabil ity and Development-Bhopal's lasting legacy


have been offered in the course of the Bhopal litigation, giving rise to a subpolitics of numbers. See note 14.

  • 6. At one extreme, a set of estimates friendly to the victims' cause holds that nearly

20,000 people died and 200,000 were exposed to the poisonous gas (Varma & Varma, 2005, p. 3 7). By contrast, the Supreme Court of India, in explaining its settlement order in the Bhopal litigation, stated, "as a rough and ready estimate, this Court took into consideration the prima facie findings of the High Court and estimated the number of fatal cases at 3000" (Union Carbide Corporation vs. Union of India Etc., 1990 AIR 273). It was, of course, in the Indian government's interest to cap the number of dead and injured at levels that supported the reasonableness of the set­ tlement amount.

Vulnerabil ity and Development-Bhopal's lasting legacy have been offered in the course of the Bhopal litigation,
  • 7. In re: Union Carbide Corporation Gas Plant Disaster at Bhopal, India, 634 F. Supp.

842 (SDNY 1986).

Vulnerabil ity and Development-Bhopal's lasting legacy have been offered in the course of the Bhopal litigation,
Vulnerabil ity and Development-Bhopal's lasting legacy have been offered in the course of the Bhopal litigation,
  • 8. In re: Union Carbide Corporation Gas Plant Disaster at Bhopal, India, at 866.

  • 9. For detailed discussion of the co-production of science, technology, and norms,

see Jasanoff (2004).













sodium thiosulfate can be found in a report prepared for the Asia-Pacific's

sodium thiosulfate can be found in a report prepared for the Asia-Pacific's People's Environment Network (APPEN)

People's Environment Network (APPEN) (See Friends of the Earth Malaysia, 1985,

pp. 4-1 1).


Such commissions have been set up after other national-scale disasters, such as

the mad cow crisis in Britain in the 1990s and following the 9/1 1 terrorist attacks in the United States in the 2000s. For an analysis of styles of knowledge making revealed in these episodes and in the Bhopal case, see Jasanoff (2005b).


The demands of the Bhopal survivors to Indian government and the interna­

tional community can be found at http://www.bhopal.net/2006demands.html.



For further elaboration of this point, see Jasanoff (2004).



Examples include such rights-shaping concepts as self-determination, human

rights, crimes against humanity, environmental justice, affirmative action, and equal

pay for work of equal worth.



On some European developments, see van Eijndhoven (1994).



US Environmental Protection Agency, Toxics Release Inventory (TRI) Program;

http://www. epa.gov/tri/tri program/whati s.htm.



I use this phrase to refer to American political culture's particular devotion to

the objectivity of numbers, as recounted in works like Theodore M. Porter (1995) and Sheila Jasanoff (1986). For an account that contrasts civic epistemology in the United States, Britain, and India, see Sheila Jasanoff (2005b).


Narratives of Vulnerability and Violence: Retelling the

G ujarat Riots

Shiv Visvanathan and Teesta Setelvad

On February 27, 2002, the Sabarmati Express passenger train was attacked at Godhra, a town in the Indian state of Gujarat. A total of 59 people were burned alive in the carnage. Almost all of them were Hindus, many return­ ing after performing religious services at Ayodhya. The rioting that fol­ lowed after the burning was virtually unprecedented in India. Neither the state machinery nor the ruling party headed by Chief Minister Narendra Modi attempted to control the violence. With few exceptions, even the administrative apparatus seemed to remain indifferent. At the same time, the media largely interpreted the violence in a standard fashion as a prod­ uct of secular and communal rivalries (Engineer, 2003). This chapter is a study of the Gujarat riots in 2002. It is an attempt to use the concept of vulnerability to increase our understanding of the riots. Memory and time, we will argue, are crucial to understand vulnerability as a central category of experience. In particular, such focus allows victims to be not merely the object of violence but the subject of history as well. This approach actually makes room for them to tell their stories within a schol­ arly context. Our argument, then, concentrates more on the aftermath of the disaster than on the rioting itself. To arrive at a better understanding of the nature of the violence committed, we argue, one should carefully con­ sider its effects and the a posteriori narrative constructions too. The process of the return to normalcy should be as much part of the analysis as the breakdown of normalcy. But let us first provide some basic context. Gujarat in 2002 was ruled by the Bharatiya Janata Party (BJP), headed by Modi, who in 2001 had taken office as the state's chief minister. To sustain its power, the BJP had initiated a series of revivalist processions called Yatras. Yatra means "jour­ ney, " and in a political sense, the journey is symbolic, a plea to remember significant events from the past and use them to create an invigorating framework for action in the present. As such, these Yatras were designed

6 Narratives of Vulnerability and Violence: Retelling the G ujarat Riots Shiv Visvanathan and Teesta Setelvad

1 10

Shiv Visvanathan and Teesta Setelvad

to travel to sites of pilgrimage such as Somnath (Guj arat) and various sites in Kashmir, which represent epic memories in the Hindu imagination and trigger a mnemonic of stories about violence inflicted on the Hindu major­ ity. Modi rode to electoral power on this revivalist agenda. In the early years of the twenty-first century, the state of Gujarat was extremely sensitive to social tensions and rivalries. Every year during Holi, Basant Panchami, and other festivals, riots would occur as a matter of rou­ tine, claiming a few lives on every occasion (Engineer, 2003) . Elections were coming up in 2002, and the BJP was rather uncertain about its future after its defeat in assembly elections in the states of Uttar Pradesh and Pun­ jab. The controversies that arose in the wake of the earthquake of 2001 had also created a sense of unease. On February 28, 2002, one day after the gruesome attack on the train, the Vishwa Hindu Parishad, a rightist fundamentalist group financed pre­ dominantly by the diaspora, announced a Guj arat bandh, a general strike resulting in the closure of the state agencies as a symbol of protest. This gave rise to an explosion of violence on a large scale. On the first day alone, the death toll rose to about a hundred. The chief minister explained it away as the normal reaction of an angry mob. In fact, he described it as a New­ tonian phenomenon, with every action triggering an equal and opposite reaction. The rioting would turn out to be one of the longest in Indian history, lasting for some four months and claiming thousands of victims.

Shiv Visvanathan and Teesta Setelvad to travel to sites of pilgrimage such as Somnath (Guj
Shiv Visvanathan and Teesta Setelvad to travel to sites of pilgrimage such as Somnath (Guj
Shiv Visvanathan and Teesta Setelvad to travel to sites of pilgrimage such as Somnath (Guj

The Question of Technology

Shiv Visvanathan and Teesta Setelvad to travel to sites of pilgrimage such as Somnath (Guj

Can we understand these riots better when we see them as happening in a technological culture? What role did technology play in the riots? Tech­ nology enters the riots at three levels in Gujarat-as symbol, as materiality, and as a metaphor for the change of nature erasure and memory. In a collective sense, the most haunting memory of collective violence is the train. The train symbolizes the massification of violence, the fact that its cargo can include thousands of bodies. The trope of the train marks the beginning of many riots. The train to Pakistan is the unforgettable sym­ bol of the Partition. The train enacted the everydayness of genocide, the reciprocity of violence between Lahore and Amritsar. Saadat Hasan Manto immortalized it with a simple line: "The train to Amritsar was seven hours late." Just that glancing reference to the lost time was sufficient to convey the mayhem that accompanied the event. And the train reared its head again with Godhra. The Sabarmati Express in 2002, carrying Kar Sewaks from Ayodhya, was burned at the station, leaving 59 dead.

Narratives of Vulnerabil ity

and Violence

Narratives of Vulnerabil ity and Violence Killing as a collective act needs the materiality of technology.


Killing as a collective act needs the materiality of technology. First, there was the train, and then two lethal instruments of murder followed. There were three everyday objects. The dharyu, an agricultural implement used by farm hands, became a tool to disembowel bodies, especially of women. More dramatic was the use of the humble gas canister to blow up houses. Cooking gas became a destroyer of homes. But the systematic nature of the riots is most evident in the use of the third simple technology: computer printouts, which were used to pick out Muslim homes and shops. The 1 984 riots in Delhi saw the first use of school registers to identify Sikh homes, but 2002 was the first time computer records were used to systematically identify minority homes. Yet what made the riots even more macabre was the use of two other technologies. One was the use of the mobile phone to create connectivity among the rioters. The second was the deliberate use of chemicals in arson. Survivors speak of the use of numerous tiny bottles of foreign import, whose contents not only ate into the skin but scorched the walls indelibly. Traces of these chemicals were seen during the Gulberg Housing Society riot, where the Congress MP Ehsan jafri and 69 others were murdered. Riots often are too simplistically viewed as a communal problem. They are then seen as occasional bursts of emotion against a particular commu­ nity. But the Gujarat riots, and probably many others in India, appear to be more systemic threats to minority citizens when we highlight the role of technologies. If we recognize how they are embedded in a technological culture, the Gujarat riots seem to be part of a planned urbanization. They did set the stage for an urban cleansing, equivalent to an ethnic cleansing. When homes are emptied, real estate is born. The sudden upsurge of urban­ ization in various parts of Guj arat, like Naroda Patia, makes one wonder if riots consciously or unconsciously are a part of a deeper logic. Studying the materiality of loss made us wonder if riots are, in fact, a form of economic warfare. Riots, along with dam projects, have become a major cause for large-scale, collective displacement today in India. While riots create urban real estate in one part of the state, they also serve to exile minorities ruthlessly to another part. Anyone who doubts this should visit the camp at Citizen Nagar in Ahmedabad. Ironically dubbed a "transit" camp, it clings to a huge garbage site. The dump was small in 2002, but today it is a gigantic structure over a mile in circumfer­ ence and seven stories in height, a mountain of waste smelling of garbage and chemicals, acrid with smoke, the delight of birds of prey. We see a jux­ taposition of two allied forms of waste-urban waste and urban survivors wasted by riots.

Narratives of Vulnerabil ity and Violence Killing as a collective act needs the materiality of technology.

Narratives of Vulnerabil ity and Violence

1 13

dominant castes. These dominant castes produced the traditional response of a mechanical act of brutality. What created the extra vulnerability was the expectation of justice in a democratic society. The vulnerability of a social system, in other words, increases with the sense of democracy that it is perceived to embody. There is a "baroquization" of the system because even freedom perpetuates and reinforces inequity. The interesting concept of baroquization was systematically used by Mary Kaldor in her classic study The Baroque Arsenal (1981). Kaldor employed the idea to study the involution of innovation to understand the production of weapons in which more and more investment produces less and less impact. One example of this could be the tank after World War II. The weapon reportedly increased in complexity without producing a corresponding increase in efficiency. Baroquization subverts the system in counterproductive ways. In our study of violence, we witness a baroquiza­ tion of the rule of law. The systems and institutions of democracy, when applied to these systems of violence, make the delivery of justice even more difficult. The baroquization of the system increases the vulnerability of the victim. Instead of being a guarantee of safety, the law has become a ritual of waiting and a parallel system of violence, which compounds the traditional violence. We can see a similar erosion of justice through the baroquization of the rule of law in other instances of genocide and ecocide. Genocide does not always stem from collective violence or brutality. The collective elimina­ tion or displacement of a people can take place with the best intentions. Modern India has created a nation of 40 million internal refugees as a result of the country's efforts to develop economically, and thus it has created more refugees through development than through the wars that it has fought. In all these cases, the system of democracy becomes self-subverting, or anti-citizen. In this perspective, riots may come to serve as an extension of electoral politics. The majority, tired of some minority, uses its majoritarianism as a vehicle to eliminate this minority. Exterminism becomes a property of majoritarian electoral systems. The chances of a survivor from a minority to successfully appeal to the rule of law increasingly grow smaller, causing vulnerability to become part of the logic of the system. This gives rise to the question of what is resistance and the route to survival in such a system, or what is the nature of agency, recovery, and resilience? Can we still apply Gramsci's (1994) model of the "pessimism of intellect and the optimism of will" to such a system? Our narratives are a search for an ethnography of survival and hope.

Narratives of Vulnerabil ity and Violence dominant castes. These dominant castes produced the traditional response of

1 14

Shiv Visvanathan and Teesta Setelvad

Specifically, we explore how the concept of vulnerability can be applied to survivors of the Gujarat riots. Every structure has its story. It is a game of possibilities, specified synchronically and diachronically. Any notion of normalcy or status demands its rituals of storytelling. Vulnerability threat­ ens storytelling by creating a state of being where narratives are too fragile to be completed. Every life as a story demands closure, whatever the variety of interpretations. Vulnerability as a state of being designates the incom­ plete, aborted story. Order is not restored. Justice is not complete. It is a perpetual disruption of expected narratives. An incomplete, liminal story has sociological and philosophical consequences. A sense of expectations is distorted. Personhood, which biographically demands a collection of sto­ ries and the availability of timetables, is thwarted. Time is fundamental to the idea of narrativity.

Vulnerability and the Violence of Riots

The Gujarat carnage of 2002 claimed over 2,000 lives. Ironically, it also helped Chief Minister Narendra Modi and the BJP to retain power. The Gujarat riots strongly differed from earlier ones, in terms of both the soci­ ology of rioting and the governance of its aftermath. First, rioting is gener­ ally an urban phenomenon, but the Gujarat carnage spread to over 200 villages. Second, tribal members in Gujarat rarely engage in rioting, but this time they constituted a critical core, which raises issues with respect to political economy or the spread of Hinduization. Third, riots tend to be spontaneous events that peak quickly, but the Gujarat riots lasted for months. Fourth, order tends to be restored quickly after a riot: the state assumes responsibility for the aftermath, and civil society engages in acts of mourning and solidarity. However, after the Guj arat carnage, the state refused to accept responsibility for the victims. Fifth, there was a sense of exterminism, which normally is not part of rioting. In exterminism, one attempts to annihilate the population in demographic terms, denying it the possibility of reproducing itself. There is no space for negotiation, com­ promises, or the "other" to be part of a future neighborhood. The violence perpetrated has a zero-sum quality in that it aims to eliminate the "other." Exterminism seeks erasure, while most riot narratives suggest some sense of adjustment, interaction, and even instances of compassion vis-a-vis the other. A zero-sum relationship, in terms of both power and orientation, destroys the possibility of reciprocity. Finally, violence, which typically has a random and spontaneous component, here appeared to be planned, and acts of rape were performed by neighbors.

Shiv Visvanathan and Teesta Setelvad Specifically, we explore how the concept of vulnerability can be
Shiv Visvanathan and Teesta Setelvad Specifically, we explore how the concept of vulnerability can be
Narratives of Vulnerabil ity and Violence Although the bestiality of the event is obvious, the indifference

Narratives of Vulnerabil ity and Violence

1 15

Although the bestiality of the event is obvious, the indifference of the aftermath was more profoundly eerie. It was here that the conventional riot narratives lost their footing. A riot as a form of violence is always tem­ porary. It presupposes a wider social domain in terms of state and law. The community in a larger sense intervenes to restore peace. In this sense, riots presuppose a return to normalcy. This includes not merely relief and reha­ bilitation, but also a sense of repair, where society intervenes to heal and to rebuild the normative order. With riots, there is always an expectation of healing, of some sense of justice. Riots, no matter how violent, have a sense of embodying the ethics of moral repair. Riot narratives, for example, include stories of how friends helped each other despite their ethnic divide. In a riot narrative, the temporary enemy returns as a permanent neighbor; violence appears as an episodic disruption, never as a permanent state of affairs. Regardless of the violence, riots also come with the promise of some notion of truth-telling, some idea of justice, and some return to the nor­ mal. These three elements guarantee that citizenship for marginals and minorities has some solidity. Citizenship needs order and normalcy, as part of the social contract. Normalcy guarantees that life in a Hobbesian sense ceases to exist and that life is no longer outright solitary, poor, nasty, short, and brutish. Conversely, vulnerability that results in a breakdown of expected narratives and a disruption of rituals and timetables denies citi­ zenship. Such vulnerability is the end of citizenship because it is the end of storytelling as a narrative of return. In the Gujarat events, we are facing an absurd drama of fragments. The expected narratives of citizenship as a set of narratives of law and order are not available. The idea of citizenship is a claim to normalcy, welfare, and well-being within a specified territory. What vulnerability emphasizes is that the standard timelines and the expectations of normalcy may not happen. The victim remains in a state of liminality, without hope of reha­ bilitation, reciprocity, or repair. It is a breakdown of the standard narratives of the life cycle of a riot, which always reiterates its temporariness. The riot is always a fragment of a society gone wrong. The usual expectation is of a return to law and order, to some idea of truth and justice. Vulner­ ability emphasizes that this wait for normalcy may be a long one, even an incomplete one. The conventional idea of vulnerability sees it as an episodic event. The questions we want to raise are: What happens if vulner­ ability is continuous? What happens when crisis is an everyday situation, when there is perpetual fear and threat, though not always with the thick­ ness of terror?

Narratives of Vulnerabil ity and Violence Although the bestiality of the event is obvious, the indifference

1 1 6

Shiv Visvanathan and Teesta Setelvad

The anthropologist Clifford Geertz (1973) proposed the idea of "thick description" as the task of ethnography. Citizenship similarly involves lay­ ers of thick description. It is a skin of narratives, events, episodes. Narratives create the chain of being that we call the "citizen." It is when narratives collapse or truncate and time and timetables became so disorderly that vulnerability is born. If citizenship is as predictable as a bourgeois novel, vulnerability is a skin of broken short stories, of a person surviving on fragments of narrative. We have organized the following fragments of testi­ mony into two sets of extracts-one about the riots and the other about the

  • camps. The interviews in their entirety were recorded by Teesta Setelvad, first for Communalism Combat and then for the nongovernmental organi­ zation (NGO) Citizens for Justice and Peace. [In some cases, we have also included direct quotes from the police's First Information Report (FIR), filed by the victims.] Testimonies Set 1: Events in February-April 2002, I nterviewed at the Godhra Relief Camp, March 22) Place: Randhikpur, Panchmahal district Witness: Bilkees (age 19; Rabia, her neighbor and relative, was with her at the time of the i nterview) On the highway just outside the village we were set upon by a mob and 14 persons from my family were butchered and killed-7 from my father's

  • family and 7 from my in-laws' side. ( ...) All had lethal weapons in their hands - swords, spears, scythes, sticks, daggers, bows, and arrows. They started screaming, "Kill them! Cut them up!" They raped my two sisters and me and behaved in an inhuman way with my uncle and aunt's daughters. They tore our clothes and raped eight of us. Before my very eyes they killed my 3Vz-year-old daughter. The people who raped me are Shailesh Bhatt, Lala doctor, Lala Vakil and Govind Navi, all of whom I know very well. After raping me, they beat me up. Having been injured in the head, I fainted. They left, assuming I was dead. ( ...) My aunt, my mother, and my three sisters all met with the same fate. I am S-6 months pregnant. My husband and in-laws were away for Id. My husband came to meet me yesterday. All the other villagers, including Rabia's family, had fled the day before, but we stayed behind because my

1 1 6 Shiv Visvanathan and Teesta Setelvad The anthropologist Clifford Geertz (1973) proposed the idea
1 1 6 Shiv Visvanathan and Teesta Setelvad The anthropologist Clifford Geertz (1973) proposed the idea

aunt's daughter was about to deliver. That delay has cost us everything. I

have filed a complaint with the police but I don't know whether I will get

have filed a complaint with the police but I don't know whether I will get


have filed a complaint with the police but I don't know whether I will get justice.2

1 1 8

Shiv Visvanathan and Teesta Setelvad

I was an eyewitness to the shameful rape of Khairunnisa, daughter of Mahrukh Bano. It was an animal-like mob of 11 who gang raped her. I was hiding in the toilet of my house at that time. After this, they burnt the entire family alive, one by one. The head of Khairunnisa's mother was cut off. I saw them mixing some solvent in the petrol. The bodies found later were in a horrifying condition. I saw with my own eyes, petrol being poured into the mouth of 6-year­ old Imran. A lit matchstick was then thrown into his mouth and he just blasted apart.4

Set II: Afterl ife

Place: Talimul lslam, Nandasan, Gandhinagar district Witness: 5yed Nasir, Manager, relief camp,

We stopped getting any relief from the government after May 27. Earlier we used to get wheat, sugar, rice, and oil. For the government there is no camp, no refugees, now. But there are still 419 persons from 95 families in the camp. They are from the districts of Gandhinagar, Mehsana, Patan, and Ahmedabad. We had some grain that lasted up to June 13; after that it has been very difficult. Some Ahmedabad-based and Mumbai-based organizations have helped with grain. Today, feeding the persons is very difficult. These are not persons with any land. They used to work in fields. Their homes have been completely destroyed but they have received barely Rs 5,000-15,000 in compensation. On February 28 itself, MLA Sureshbhai Patel was named and identified by many survivors as leading the mob in village Paliyar. One month ago, a two-day meeting was held in the village where it was decided that "Miyab­ hai gaam rna nahin joyiye" ("We do not want Muslims in the village"). The total population of the village is around 3,000. The Muslims who have fled just do not want to go back.5

Witness: Mozaam Khan Place: Pansar village, Kalol taluka, Gandhinagar district

Shiv Visvanathan and Teesta Setelvad I was an eyewitness to the shameful rape of Khairunnisa,

Twenty-five persons belonging to four families from our village are now in the Nandasan camp. In Pansar village, there are about 450-500 Muslims. We have been holed up here for four months, because there is a boycott by the villagers who say they do not want a single Muslim in the village. The government is not helping at all either. The sarpanch of our village, Gopalbhai Maganbhai Patel, belongs to the B]P. The mamlatdar had directed him to get us back. He flatly refused. He

Narratives of Vulnerabil ity and Violence


Vulnerability demands a multiplicity of lived times that the state and the majority community try to truncate. Vulnerability is created twice:

first through violence and then through denial or truncation. It becomes a denial of access to one's right to have a narrative. In violence, storytelling is one entitlement that the survivor demands. The suppression of the story is the second circle of vulnerability. If violence as invasion is the first circle, then silence and repression comprise the second circle. Waiting forms the third circle of this Dantesque world. Waiting appears as a corpus of rituals by which the social signals a return to normalcy. Waiting creates a sociol­ ogy of expectations marked by a timetable. However, waiting does not appear only as an extended timetable, but also as an extended geography. One begins with the simple displacement from home. Life in a relief camp marks liminality. This liminality is doubly underlined because the camps are not the creation of a concerned state. After the 2002 riots, for the first time Gujarat refused to establish camps, fearing that it would be read as an acknowledgment and confession of guilt. Intelligence reports of the Guj arat state government presented in affidavits before the Nanavati Commission of Enquiry (reporting in 2008) indicate that in 13 districts for which reliable data is available, 70,000 people have not returned home. "Waiting" means waiting to return to work, and then, when you do return, finding that local things have taken over the place where you had your shop. Neighbors come to you with morsels of hospi­ tality, denying their role in the violence. It is an invitation to normalcy, a signal to forgive and forget. The rule of law often becomes a ritual of delay. More particularly, and despite the best constitutional intentions, the forces of justice might them­ selves delay justice. Vulnerability becomes ironic because the very institu­ tions that promise to maintain justice defer the rule of law and contribute to more vulnerability. When justice is performed as farce, vulnerability intensifies because the survivor feels helpless. History becomes positivist and frozen in print when it is majoritarian and transformed from being plural and unofficial into being hegemonic and official: it becomes a simple narrative of conquest and defeat. Such history reduces memory to one strand of experience, denying validity to other interpretations. Such history leaves minority groups with only two options. First, it legitimates exterminism by converting riots into zero­ sum games: electoral majorities now want to eliminate the opponent. The second option is a notion of citizenship as defined by the majority. The ghettoized Muslim seeking his identity is now seen as either a Malthusian

Narratives of Vulnerabil ity and Violence Vulnerability demands a multiplicity of lived times that the state
Narratives of Vulnerabil ity and Violence Vulnerability demands a multiplicity of lived times that the state


Shiv Visvanathan and Teesta Setelvad

threat or an object of emancipation. Either way, history legitimizes vio­

lence. This positivist, majoritarian history-which denies plural, personal, and narrative histories-also demands the erasure of memory.
lence. This positivist, majoritarian history-which denies plural, personal,
and narrative histories-also demands the erasure of memory. "Let's forget
and move on, " as argued by the new middle class audience on TV. Justice
demands memory, but it is precisely this memory and this right to story­
telling that the vulnerable in Gujarat are asked to abandon.
In fact, this explains the confusion around what has been called the
Bandukwala argument. juzar Bandukwala, a physicist and professor at Bar­
oda University, was a civil rights activist who faced threats and harassment
for his stands. He repeatedly fought for rights, justice, and memory, but
he did so to little avail. One could argue that his act might be misunder­
stood as weakening the sense of struggle among survivors; but we feel that
it needs a sympathetic understanding. After a while, Bandukwala argued
that he, as a practicing Muslim and a citizen, unilaterally forgave the per­
petrators of the riots and wanted to move on. He explained that he did
not want to carry the curdled memories of violence with him any longer.
Memory becomes a double burden when justice becomes elusive. In this
respect, vulnerability is critically related to memory and time. The structure
of expectation is crucial, as is true of the politics of memory. It introduces a
drama of choices. Does the victim forgive and move on? Does the majority
apologize? Is some form of forgiveness possible? One must add, however,
that on rethinking, Bandukwala emphasized that justice is primary, but
waiting for justice can induce passivity in the victims.
Vulnerability opens the question of ethical repair, and this goes beyond
the question of physical rehabilitation emphasizing a restoration of habi­
tat and occupation. To convert a habitat into a neighborhood requires a
nomos, a normative sense of understanding, of truth-telling. In this sense,
justice is an unfolding of normalcy in time. Waiting for justice, for rec­
ognition of what happened, is a waiting in time. Many activists attacked
Bandukwala for ignoring the plea for justice, seeing his pronouncements
as signs of evasiveness or amnesia. But what Bandukwala argued was that
as vulnerability searches for justice, it gets bogged down in stereotype. It
needs to recover agency. Vulnerability caught in stereotypes loses its sense
of innovation. Bandukwala wanted to tell his own story differently. Vul­
nerability, by reaching for an alternative narrative, finds a sense of agency.
The actor forgives the perpetrator unilaterally and moves on: if justice is
unavailable, forgiveness is the only creative option.
A minority always faces a ghettoization in space and time. If it is easy
to grasp a ghetto as a spatial enclave, time creates its own ghettos. Is a
minority to shut itself off, or can it modernize with the rest of society while

Narratives of Vulnerabil ity and Violence


rather than an immaculately hatched term. And there is a change in the tacit construction of vulnerability. Stories that had been anecdotal, autobiographically discrete, and articu­ lating grief, anger, and pain now become a collection of statements, com­ plaints, and accusations. As the individual survivor becomes community, the concept of vulnerability also grows to network-like proportions. The survivor turns from voice to theory, from narrative to ethnoscience. For Chief Minister Narendra Modi, the survivor represents a failure of the social and a challenge to governance. Modi differentiates tactically between the Muslim as ethnic and the idea of citizenship as secular. He then argues, while emphasizing his secular claims to governance, that Mus­ lims are welcome as citizens. There is a hint here that vulnerability of the Muslim community is a mask for ethnicity. The Muslim is portrayed as a reluctant citizen. As a populist strategy, this immediately appeals to the dominant majority, which is tired of human rights pleas and critiques from the liberal press. Oddly, claims of vulnerability here increase vulnerability. Vulnerability is read in jurisprudential terms as a self-imposed injury of an ethnic group caught in a time warp and reluctant to embrace citizenship and development. Realizing that vulnerability is negotiable, the survivor-working with social scientists and activists-builds narratives into testimonials. The sur­ vivor thus creates his own social science as a form of testimony, narrative, and resistance. The survivor tuned by the activist realizes that narratives without a sense of concepts might be self-defeating. They sense that con­ cepts are filters that block certain arguments and allow others. Citizenship for the survivor becomes a learning game in law and social science. The government lawyers, in turn, have been peeved at this countermove. They have accused the Citizens for Justice and Peace of "influencing victims." The survivor learns citizenship, learns law, learns the art of legal testimony. The process of learning itself creates a community of pride and compe­ tence. Initial indifference of the group now opens into a sense of normalcy and citizenship as a learning curve. These become rituals of confidence building that are forms of healing. One of us confessed at that time that it was a reinvention of citizenship.9 What we see here is an appropriation of narratives, of the very act of storytelling. Any recovery of justice must begin with an act of retelling, through the rituals of discovery and investigation that make another form of story possible. The poetry of ethnography that activist-scientists engage in is precisely this-it prevents the possibility of forgetting, of erasure, of

Narratives of Vulnerabil ity and Violence rather than an immaculately hatched term. And there is a


Shiv Visvanathan and Teesta Setelvad

amnesia. Dissent thus is a set of mnemonical aids for a society prone to forgetting and erasure. If vulnerability is pathology in time, only a retelling of times lost and expected can redeem a people. This could be the most important lesson of the South African Truth and Reconciliation Commis­ sion (TRC) (Soyinka, 1999). But one must add a caveat. In India, memory a s oral telling is important. It demands listening, and the patience and silence of listening. The TRC was a spectacle, a demonstration, a show, a visual act. What India culturally might need is a style of storytelling that allows for multiplicity and retelling. By conceptualizing vulnerability as a problem of time and a problem of storytelling, one captures the everydayness of vio­ lence in the power of language to define reality. If the real accompaniments to terror are silence, indifference, and anonymity, the therapy for vulner­ ability is voice, narrative, and identity. A timetable for recovery needs to include a ritual of storytelling. Such an approach to vulnerability does not deny a more scientific defi­ nition of vulnerability. It seeks to thicken it. It calibrates and defines. As an ethical term, "vulnerability" is far more textured than "well-being. " It is not just about the other; it is about our openness to the other's being. What the middle class in Gujarat is pushing for is that extra bit of indifference, of forgetting. The solitude of the Muslim is the absence of a story. That is where the strategy and hope of the Special Investigation Team (SIT) were. This team was created by the Supreme Court of India and chaired by the former head of the Central Bureau of Investigation (CBI). Its rituals of visits actually were an invitation to a huge retelling of the story and a plea to listening. It was an invitation to a Katha (a storytelling) and a Sunvai (a hearing), where the victims record history again. Unfortunately, the SIT failed to meet expectations; it fell back on the work of the local police, showing little initiative of its own. The court had to appoint an amicus curae to look into the SIT's investigation. This report is currently unavailable to the public. In a previous section, we discussed how politics sustains storytelling and how activist and survivor sustain memory. In the final section, we shall confront technology and its challenge to the politics of memory.

126 Shiv Visvanathan and Teesta Setelvad amnesia. Dissent thus is a set of mnemonical aids for

Vulnerability and the Discourse of Technology

126 Shiv Visvanathan and Teesta Setelvad amnesia. Dissent thus is a set of mnemonical aids for

In a previous section, we listed elements of technology that entered the ethnography of the riots. We have given technology a secondary place in our narrative so far because it was not primarily technology that haunted the riots. Technology-or, more precisely, the technological culture that

Narratives of Vulnerabil ity and Violence


forms around it-most significantly played a role by creating vulnerabili­ ties through framing the discourse. The first example is the remaking of Narendra Modi from a communalist to a technocrat; the second is the fate of the Muslim survivor as citizen; and the third example is the role of riots in the overall shaping of the urban discourse. Modi understood that the core competence of a politician must be built around different identities, or-to switch metaphors-he understood that he needed a second skin. He realized that it was the Hindutva man in him that had to be deconstructed and recomposed. Like Eliza Doolittle, he had to project a new way of speaking. He disaggregated elements of his Hindutva identity to create a new one. Hindutva evoked the state as the god of society. Organizational skills, asceticism, a sense of competence as machismo, and a clear idea of history: these constituted a technocratic style of managing society. Modi presented himself as the Vivekanand in politics, a reformer using technology as a cultural metaphor for a new era. Modi transformed Hindutva into a more neutral but aggressive techno­ cratic idiom. Management became a form of masculinity, and the idea of Hindutva, first seen as local and parochial, now became globalized. Modi's Gujarat behaved like a city-state, a combination of Singapore and Shanghai on a larger scale. What came to his aid was the language of the World Bank. His expert handling of the earthquake the year before the riots was crucial. Modi used the World Bank rules and vocabularies to transmute his ideolo­ gies into methodologies of audit and standards creation as the new form of accountability. Modi was responsive to World Bank idioms and norms, and he could preen himself with numbers. The new aura of accountability found a connection to an obsession with security. Security became the technocratic idiom of nationalists. Security was also the machismo that would fight terror. Guj arat's handling of terror was presented to the rest of India as exemplary. The brilliance of it was that security and accountability became positivist terms, measured by degrees of control. In Modi's thesaurus, they substituted for the ethics of respon­ sibility. The "old" meaning of responsibility was encompassing in its phi­ losophy. It was inclusive by involving minorities. Responsibility was a way of life that implied conversation. Security and the new, associated form of accountability were handled with forceps. They were distancing terms. If responsibility sounded soft, security was hard. It exuded power, control, and hierarchy. Guj arat was secure under Modi, while Delhi was vulnerable to terror under an effete Congress. The myth of efficiency, epitomized as security and stability, needed investments as a continuing barometer of success. Modi played the


Shiv Visvanathan and Teesta Setelvad

self-styled magnet for investments brilliantly. In this new age of liberaliza­ tion, investments are manna, the gift from heaven all tyrants seek. Invest­ ments can silence shrill critics wailing about riots. Gujarat was to beco m e the Camelot of investments, and its first center was Sanand, a city in Ahmedabad district. In Sanand, Modi created a dreamland for the automo­ bile industry by inviting Ford, Tata, and Maruti to establish plants there, and contouring this hub with a stunning array of ancillary industries that would add to employment. Modi's message to the corporations was clear, and Ratan Tata was among the first to sense it when he said, "It would be stupid not to be in Gujarat at this stage." Modi had become Guj arat's best political salesman, and his clients were the corporations and the diaspora. He enacted his vision of shining Gujarat, thus impressing the diaspora Indians, who starved for a sign of efficiency and decisiveness in their homeland. For them, as for Time magazine later, here was an Indian who could stand up to the Chinese. It was a helpful aura to have, especially with the US government. A nuclear plant or two would become an apt mutual token of esteem. Modi has injected the idea of development as a credo deep into the middle class of Gujarat. For them and him, development is a process that cannot wait, that is inevitable as Darwin's survival of the fittest. The Modi credo then suggests that those who act tangential to or are recalcitrant about development are reluctant citizens. The idea of development creates a double demand on ethnics, marginals, and minorities. It presents them two specific challenges. First, it asks them to de-ghettoize and de-ethnicize themselves and erase identity and memory. It asks them to forget the riots:

why wait for justice when we are offering you development? It argues that development can be more distributive than justice. The second challenge is that they have to become citizens, and that citizenship is defined as join­ ing the mainstream. Here Modi's discourse also suggests that minorities did hide behind their ethnicity and behaved like reluctant citizens. Modi's is the truly secular option. The majoritarian electoral democracy of Congress plays to religious sentiments, while BJP's offer of development is an invi­ tation to secular citizenship. With this fascinating salvo, Modi claims the higher moral ground. Many Muslims find this suggestion tempting. They realize that they need to join the mainstream, but they also sense the crafti­ ness of the Modi option. He is asking them to abandon memory and justice to accept entry into development, yet development might be a zero-sum game too. They sense that the new urbanization in the aftermath of the riots may disempower them further.

Narratives of Vulnerability and Violence


Gujarat has always been the most urbanized part of India, with at least 57 major towns. Modi is now building a new wave of urbanization on top of this. Modi articulates urbanization as process and promise. As a pro­ cess, there is logic to its demands that necessitate certain decisions. Instant cities, unlike instant coffee, are complex entities. Modi realizes that cities are coalitions of opportunities. The city caters to a middle class, to the corporations hungry for land, to a network of fixers who create opportuni­ ties around a city. Each act of Modi invokes a corporation and urbanizes Gujarat. Modi has allied himself with newly emerging entrepreneurs like Advani, Mittal, Nirma, and Tata. He has offered the Japanese, always hun­ gry for land abroad, the opportunity for development in two cities. He has hypothecated the coastline to the corporations, like Adani Enterprises, whose control of pipelines and ports makes it a formidable force. Corpora­ tions that are desperate for land find Modi amenable. The middle class, seeing in these investments the prospect of employment, is also content. The Indian diaspora sees in him an almost American competence, quick­ ness, and decisiveness that is rarely evident in Indian politicians. Modi has become the new urban hero. Yet one senses unease about these new cities. There is wonder whether they are a kind of enclosure movement, a new way to displace nomadic and pastoral populations. Gujarat has been the home of these great nomadic and pastoral civilizations. The speed of Modi's policies of urbanization makes one wonder whether marginal minorities are doomed in this feat of citizenship that we call the city. The local activist has transformed himself into a development hero, with the city as his script. Modi as a development statesman now projects messages at three levels. Locally, he is a BJP chief minister; nationally, he is a future candidate for prime minister; globally, he is a player articulating the rhetoric of climate change. He is handling three parallel identities, with the help of the assets of the technological culture that is being created in Guj arat. In discussing Modi, we have shown that majoritarian vulnerability resurfaced as security, and that technology reinvented the chief minister as a managerial hero. The vulnerability of the majority trumps the vul­ nerability of the minority. In the new development discourse, minorities are presented as less-than-modern, reluctant citizens who are resistant to development. The new discourse on development erases the riots, recon­ structing them as a philistine and barbarian prelude. The new chorus of progress makes the survivors' protests sound like an outmoded script in a new era of technology. The authors of this chapter wish to emphasize hope.

Narratives of Vulnerability and Violence 119 Gujarat has always been the most urbanized part of India,

Part II

Exploring the Ambiguity of Vulnerability

Exploring the Ambiguity of Vulnerabil ity


how resilience can be achieved in local practices. They show a discrepancy between the micro-, meso-, and macro-level accounts of resilience in the scholarly literature and how achieving resilience actually works locally­ this latter process being in need of coordination across all three levels. The important issue of how to cope with vulnerabilities resulting from "gaps" between local and global rationalities in local practices is taken up and further elaborated by the last two chapters of this part. The notion of "bricolage" (Ciborra, 2000; Sanne, 2007) starts from the idea that in order to work safely and deal with occupational risks in time, improvisation and tinkering is a necessary strategy. Johan Sanne (2007) argues that if railway maintenance technicians did not act as "bricoleurs, " they would be less productive and less efficient in their work, and perhaps even less safe. After all, rules do not apply to all situations under all circumstances, and there­ fore some creativity is necessary. At the same time, Sanne (in chapter 10, "Vulnerable Practices: Organiz­ ing through Bricolage in Railroad Maintenance") acknowledges that brico­ lage can be risky because procedures and resources will not be improved so long as the maintenance technicians manage to do their work (by using bricolage) and thus "hide" potentially weak spots in their practices. This can thus lead to (negative) vulnerability. Whereas Sanne mainly stresses the riskiness of allowing bricolage, Ger Wackers emphasizes the importance and beneficial effects of such kinds of improvisation in chapter 9, "Entrainment, Imagination, and Vulnerabil­ ity-Lessons from Large-Scale Accidents in the Offshore Industry. " Analyz­ ing a case study of a near-catastrophic accident ou a Norwegian oil platform (Snorre A), he concludes: "The successful recovery from the gas blowout on Snorre A confirms that 'resilience of this system resides in its people rather than in its technology'" (citing Woods & Cook [2006, p. 70]) . And "for the success of the recovery [of the oil platform] , improvisation in one domain was of particular importance." This point also shows that the vulnerability or resilience of a technological system is something that is achieved rather than given. Vulnerability is socially constructed: "The same system can be deemed relatively invulnerable-when unruly behavior is interpreted as people taking their responsibility, using their experience, improvising to accommodate to changing conditions; or it is deemed vulnerable-when such unruliness is defined as violating the regulations and creating risky situations." (Bijker, 2006, p. 58) It is not a straightforward matter to decide whether deviating from rules, standards, and protocols will always lead to more vulnerability. Rules and practices that produce resilience in one con­ text may result in vulnerability in another context.

Exploring the Ambiguity of Vulnerabil ity how resilience can be achieved in local practices. They show


Resilience: Contingency, Complexity,

and Practice

Stephen Healy and jessica Mesman

There are different concepts that contribute to theorizing about vulner­ ability. "Creative dissent, " as the previous chapter shows, is one of them. Similarly, the analysis of "resilience" can contribute to our understanding of vulnerability as it directs our focus on its positive aspects. As such, resil­ ience should not be considered as the oppositional counterpart of vulner­ ability, but as a relational notion. Vulnerability, as has been argued in the introduction of this volume, is an emergent system's property. Likewise, resilience is a system's property too. All systems, whether ecosystems, technological systems, economies, corporations, or nations, possess resilience-an ability to recover from disruptions-because without it, they perish.1 Resilience, a fundamental principle of systems ecology, has gained prominence in many fields since the early 1970s, including physics, psychology, organizational studies, sociology, aviation studies, and health care management. This has resulted in the identification of many different kinds of resilience, such as institu­ tional resilience (Pidgeon, 1997), coastal resilience (Klein, Smit, Goosen, & Hulsbergen, 1998), ecological resilience (Gunderson & Holling, 2001), mental resilience Oohnson & Wiechelt, 2004), disaster resilience (Many­ ena, 2006), engineering resilience (Gallopin, 2006), and socioeconomic resilience (Klein et al., 1 998), to name a few. These involve characterizing resilience as an outcome or process characteristic, as intrinsic or contin­ gent, as a matter of system structure or function, and as a paradigm or "just an expression" (Manyena, 2006). This pervasive ambiguity regard­ ing resilience remains a notable feature of many accounts, particularly popular ones, even though the use of the term has increased markedly in recent years. This imprecision is reinforced by a tendency to overlook the normative dimensions of resilience. While certain kinds of resilience, like those of

Resilience: Contingency, Complexity, and Practice Stephen Healy and jessica Mesman There are different concepts that contribute
Resilience: Contingency, Complexity, and Practice Stephen Healy and jessica Mesman There are different concepts that contribute


Stephen Healy and jessica Mesman

totalitarian regimes or polluted water supplies, are unarguably undesir­ able (Carpenter, Walker, Anderies, & Abel, 2001),2 others, such as that of coal-fired electricity generation, are subject to political contestation and technical uncertainty. Questions such as the resilience of what, for whom, and at what cost or tradeoff, have been thrown into sharp relief by recent insights regarding tensions between how resilience is conceived in research on socio-technical transformation relative to that in systems ecology.3 While analogous notions of sustainability frame both, it has become clear that resilience that is conceived in socio-technical terms undermines resilience that is conceived socio-ecologically. The attention (and clarity, given normative considerations) tend to reflect the scale (or "level") at which resilience is characterized. Many accounts of resilience focus on the level of person-to-person, or institutional, interaction, where there is commonly some consensus regarding what is, and what is not, desirable. As the level of analysis moves up to that of societies and/or other meta-level systems (such as ecosystems and, particularly, when mat­ ters of governance are broached), other considerations take center stage, placing the burden of identifying and accounting for matters of priority and significance on the analyst. The many varieties of resilience, and the ambiguities to which they give rise, are echoed in how the relationship between resilience and closely related concepts such as adaptability, transformability, robust­ ness, resistance, and reliability are variously interpreted. This chapter will not attempt to settle these differences or provide a comprehensive review of the many accounts of resilience.4 Our aim, rather, is to analyze the varieties of resilience so as to illuminate the empirical description and theoretical understanding of vulnerability issues. This is achieved by structuring this chapter in terms of the scale, or level, of the domain of application of the accounts of resilience examined. We characterize these in terms of: (1) the micro-level of person-to-person and person-to-thing interaction; (2) the meso-level of institutions, socio-technical systems, and workplaces (among others); and (3) meta-level system perspectives. We start by examining the undertheorized micro-level and showing how resilience at this level can be understood in terms of situational specif­ ics. In conclusion, we discuss the continuities, tensions, and potential synergies across and between these "level" -bound analyses, and the implications of these for both the theoretical understanding of vulner­ ability issues and for constituting resilience effectively in real-world contexts.

Stephen Healy and jessica Mesman totalitarian regimes or polluted water supplies, are unarguably undesir­ able (Carpenter,

Resi lience



M icro-level: From Rational Actors to

Situated Ensembles

The Rational Actor Paradigm

The "rational actor paradigm" (RAP; see, for example, Jaeger, Renn, Rosa, & Webler, 1998), which dominates the field of risk management, is the best-known micro-level approach. RAP centers upon the idea that society consists of autonomous individuals who use their rational capacities to optimize their existence in a material world divorced from them by cal­ culating the consequences of a potential course of action and choosing the option that maximizes their individual welfare. While more nuanced approaches have been developed (which are discussed later in this chap­ ter), RAP focuses upon the maximization of instrumentally conceived indi­ vidual preferences, which are then, implicitly, taken to confer resilience. Notwithstanding the evident shortcomings of this caricature of human actors, "preferences" are typically characterized as relatively unwavering individual dispositions and commitments, marginalizing contextual influ­ ences. This is a result of RAP's uncritical assimilation of the Cartesian divide between the mind and body/matter, with minds conceived as little more than "cognitive containers" for static "preferences, " which are assumed to represent the world at large. This disregard for contextual circumstance is reinforced by a tendency to reduce the external world to generic abstractions focused on individual welfare, such as "benefits" and "costs." As a result, the means by which minds obtain these representations-observation and contemplation-are for this "canonical rationality, " the primary corporeal practices of interest with action, and the engagement with materiality that a focus upon corporeal action demands, which is largely overlooked. RAP is thus poorly equipped to attend to situational specifics and their dynamics. This is problematic in situations in which contextual specifics play a signifi­ cant systemic role conferring vulnerability rather than resilience. The tendency to obscure the situated specifics distinguishing particular contexts has resulted in moves to bolster the RAP model via the incorpora­ tion of particular social and contextual factors. Such moves are evidenced by how psychological and sociological approaches to risk have been assimi­ lated to this view, resulting in further attention to sociocultural consid­ erations traditionally overlooked by RAP. This has led to more nuanced approaches that, for example, recognize that subjective factors influence rational actors, in addition to a rational estimation of the likelihood of adverse outcomes from their actions. However, the underlying calculative model of self-optimizing "rational actors" remains essentially unchanged.


Stephen Healy and Jessica Mesman

Corporeality, Materiality, and Action

Stephen Healy and Jessica Mesman Corporeality, Materiality, and Action The characterization of resilience in terms of

The characterization of resilience in terms of corporeality, materiality, and action makes a fundamental break with "canonical rationality." Such rationality overlooks bodily practices other than those by which cognitive abstractions are understood to take up residence in minds-observation and contemplation. However, "real practices" count because we do not simply absorb cognitive abstractions from observation and contemplation; we develop them as we interact with the many people and things that make up our world. In the account developed here, we are therefore primarily concerned with interrogate actions, the dynamic practices constitutive of them, and the many corporeal and material considerations that these encompass. We summarize these using the term "practices, " intended to mean the dynamic, intersubjective, discursively mediated, yet inherently socio-mate­ rial interactions between people, and between people and things, rather than simply repetitive bodily actions. It is these "practices and skills which produce people, selves, and worlds" (Thrift, 2000a, p. 216) and which, as a result, constitute both vulnerabilities and resilience. Our account, there­ fore, focuses upon such practices and the details, and context, of their per­ formance. This view thus replaces the determinate possibilities suggested by RAP with a window onto the contingencies of practice(s), potentially illuminating how they may be better catered for. Of particular relevance to this emphasis is research suggesting that our species may be "hardwired" for anticipation, movement, and action, rather than contemplation and observation (Thrift, 2000b). Many contemporary contexts center upon skillfully enacted performances requiring rapid, high­ stakes decision making and complementary suites of complex practices involving interactions between people and, commonly, machines. Hospi­ tals are good examples, especially those involving critical care practices, like emergency wards, neonatal intensive care units, and operating the­ aters. The remainder of this section illustrates these arguments by focus­ ing upon how vulnerabilities due to fragile lives, risky contingencies, and emotions are managed through fine tuning, improvisation, inborn charac­ teristics, and material and intersubjective interactions in these contexts.5 In a critical care practice, like a neonatal intensive care unit (NICU), the unexpected and the uncertain reign supreme. One phone call about an admission coming up from the delivery department can change the whole state of affairs on a ward. The admission of a baby sets in motion an ordering process geared toward the infant's incorporation into the every­ day practice of the NICU. Within a few minutes, the staff have to stabilize



the baby as swiftly and effectively as possible. To be able to define the con­ dition of the baby as precisely as possible, several blood tests are conducted and electrodes positioned on the baby's body. In this way, the staff ensure nonstop information about vital functions, such as heart rate, blood pres­ sure, respiratory rate, saturation, and temperature. An X-ray will help the doctors to take a good look at the baby's lungs and check if the ventilator tube is in the right position. All these efforts result in several lines and tubes that go in and out the incubator and connect the baby's body to a set of machines. While carefully coordinating their activities, hospital staff focus upon solving the immediate problems that they encounter and, given the fragility of the patient population in critical care areas, this is never simply a matter of solving a patient's medical problem. The required medical interventions necessitate a dynamic, intricate, and ongoing fine tuning of actions and reactions between people, as well as between people and machines, and it is this dynamic web that constitutes both vulnerabilities and resilience. For instance, the vital f u nctions of a premature baby are quite unstable. These kinds of children are insufficiently developed and lack the capacity to self­ regulate their body systems effectively. While critical care staff are highly trained and their activities are focused by well-developed and sophisticated protocols, these can never f u lly account for contingencies. If these proto­ cols are sometimes adequate, when they fail or are inadequate to the chal­ lenge on hand, new practices deploying available resources-social and/ or material-have to be rapidly improvised (Conradson, 2003; Harrison, 2000; Mesman, 2008). Such "dynamic webs, " constituted by individuals and the others, including the things with which they interact, emerge from the relational perspective employed here in which practices, or more specifically the details of their performance and those of the contexts in which these per­ formances are embedded, are the primary concern. The entire human and non-human "ensemble, " including the practices constitutive of it, thus focus this account. Thrift (2000b) discusses these in terms of a "distrib­ uted ecology of thought" noting that " [n]ot only do objects make thought do-able but they also very often make thought possible" (p. 38). Intersub­ jectivity, then, becomes more than simply a human-to-human affair that also encompasses interactions with objects-whether they be tools, com­ plex technological systems, or merely mundane, everyday objects. Such ensembles can be understood and analyzed on many levels-subjective, intersubjective, technological, and/or institutional, for example-and in dimensions ranging from spatial, temporal, through to affective ones.

Resil ience


Table 8. 1 The primary resilience concepts developed for micro-level analysis



Focus on




Calculative, rational


Maximization of




individual welfare.



The performance of

The performative


attention to the co-constitution of corporeality, materiality, affect, & anticipation.

situated practices and contingencies.

character of circumstance. Interdependencies with other levels

It is important to reemphasize that our concern is with practices inter­ preted as socio-material ensembles in which human and non-human enti­ ties are understood to be constituted by and through the relations between them, rather than existing a priori of them. Effective insights regarding situa­ tional resilience, similarly, require the illumination of the relations between the circumstantial elements in which resilience emerges, including, specifi­ cally, embodied and affective considerations of the form discussed above. However, it needs to be stressed, as will be underscored in the rest of this chapter, that the achievement of resilience in practice depends upon the structuring and alignment of these circumstantial elements with impera­ tives arising at higher levels (organizational and societal, for example). Table 8.1 (after Folke, 2006) provides an overview of the primary resil­ ience concepts developed for micro-level analysis in this section. This started by examining the RAP, which provided a point of departure for us to propose a relational perspective on resilience focused by situational con­ siderations overlooked by RAP. This situational perspective encompasses not only the sociocultural character of person-to-person interactions, but also the embodied and material characteristics of the interactions between people and people and people and things. We argued, using hospital work as an example, that these are all constitutive of the resilience specific to particular circumstances.

The Meso-level: Institutions and Socio-Technical Systems

Quantitative Risk Assessment

Quantitative Risk Assessment (QRA) evolved after World War II as a response

Resil ience Table 8. 1 The primary resilience concepts developed for micro-level analysis Resilience Characteristics Concepts


Stephen Healy and Jessica Mesman

system incorporates, the more it can deal with unanticipated change (p.

59), warning that a focus on achieving reliable systems under normal con­ ditions is "a far more dangerous objective than seeking flexibility and resil­



inevitably lead[ing] to short term gains


at the expense of

long term losses" (p. 58). For Foster, the reliability argument is built on an illusion, and the size, complexity, and control structures of today's social and technological systems make them inherently vulnerable. This concern has recently been reflected in the Resilience Engineering approach/ which argues that systems should not only be reliable but resilient (Dekker, 2005). In this perspective, safety is not a static system feature or an outcome, but rather an emergent property conceived as "a dynamic, interactive, com­ municative act that is created as people conduct work, construct discourse and rationality around it, and gather experiences from it" (Dekker & Holl­ nagel, 2006, p. 6). Safety cannot then be predicted or sustained on the basis of control over certain parts of the system but must involve learning geared to increase a system's adaptive capability. Resilience engineering focuses on the individual human role in socio-technical systems, concentrating on the capacity of practitioners to anticipate and adapt to changes and recover from problems (Dekker, Hollnagel, & Woods, 2005; Dekker & Hollnagel, 2006; Hollnagel et al., 2006). It aims to improve practice in such a way that it becomes "better able to recognize, adapt to, and absorb disruptions that would otherwise fall outside the base it was designed to handle" (Dekker & Hollnagel, 2006). Resilience engineering develops explicit guidance on coping with variation and other challenges that involve trade-offs between situational pressures and safety by monitoring and assessing the adaptive capability of an organization and, therefore, considers anticipation and adaptation core organizational activities. Resilience engineering presents itself as panacea for the management of safety problems in all kinds of domains, and at all levels. However, it neglects "culture/' understood in an anthropological sense "as deeply held complex systems of shared and often tacit meanings and values ... acquired through many years of socialization [that] inform social action, identity, and belongings" (Rowley & Waring, 201 1). Resilience engineering is still, thus, essentially a "measure and manage" approach to safety that

tends to gloss over the cultural fabric of socio-technical systems.

Organizational Culture

Barry Turner's Man-Made Disasters (Turner,

1 9 78) provided a pioneering

approach to organizational and, more broadly, contextual dimensions to


Stephen Healy and J essica Mesman

Stephen Healy and J essica Mesman

errors" (Sharrock, Young, & Faulkner, 2005; Young, Sharrock, & Faulkner,


Turner's analysis in Man-Made-Disasters became a significant contributor to later studies emphasizing organizational culture (see, e.g., Gherardi, 2004;

Gherardi, Nicolini, & Odella, 1 998; Hopkins, 2005). Those advocating this view argue that adequate attention to organizational culture/practices can counteract the technological "inevitability" diagnosed by NAT and make organizations safer. Hopkins re-analyzes the Three Mile Island nuclear plant

accident to conclude that not only "was [it] not a normal accident


also that]

blame on

historical ethnography of the Challenger disaster (1996), which showed the importance of historical knowledge of the context in which accidents occur, also emphasized the significance of organizational culture. Vaughan under­ lines that a healthy organizational culture is dynamic, and that an organiza­ tion can become vulnerable when unduly focused on rules and regulations

or when other pressures are allowed to create institutional rigidities. Turner's attention to situated specifics went beyond narrowly organiza­ tional concerns to anticipate matters outlined in the previous section, with observations such as the following:

the origins of unintended consequences can be located at the physiological level

... where physical slips and simple typing and speech errors originate, at the psycholog­ ical or social psychological level from which many individual accidents attributed to "human error" spring, or at the level of organizational processes. (Turner & Pidgeon,

1997, pp. 149-1 50)

Man-made disaster (MMD) is thus an analysis of disaster as collective failure, in which attention is redirected away from a narrow focus upon either individual failure or poor management toward broader collective, organizational considerations. Nick Pidgeon (1997) elaborated these ideas to develop the notion of "institutional resilience, " emphasizing the "shared cognitions and administrative structures and resources which support, rather than constrict, the development of organizational understandings regarding risk and danger" (Pidgeon & O'Leary, 2000, p. 1 8). While over­ lapping with HRO and the work of Reason (through emphases on manage­ ment commitment and effective rules and norms, for example), MMD is distinguished by an embrace of contingency and a stress upon effective organizational learning to cater to it. As a result, recent work has examined barriers to organizational learning, in particular informational problems and organizational politics (Pidgeon & O'Leary, 2000).



Since 2000, there has been a growing recognition that human actors deserved more attention, echoing the analysis of the previous section, with some advocating "a reappraisal of human work within fallible systems" (Mesman, 201 1 ; Owen, et al., 2009). This became the focus of a group from different scholarly communities-including cultural history, science and technology studies, human factor analysis, cognitive ergonomics, and complexity theory-who claimed that

there is a need to better understand safety in systems firstly, by analyzing the posi­ tive role of human actions with as well as in those systems, and secondly through developing better understandings of the ways in which humans successfully accom­

plish work in those environments

and to highlight individual, collective and

. . . systemic resources that are used to enhance resilience in acting with complex socio­ technical systems. (Owen et al., 2009, p. 1)

A key insight of these studies is the identification of coordination between the individual and the collective and systemic resources that enhance prac­ tices within a complex work environment. Analysis of day-to-day activities shows how those involved rely on different ways of ordering to deploy their knowledge, experience, and technology to preserve adequate levels of safety (e.g., Mesman, 2008, 2009).


The meso-level view of resilience described in this section illustrates a shift from calculative approaches such as QRA toward a focus on institutional and contextual considerations, resulting in a recognition of the importance of the resilience of socio-technical systems over their reliability. Contem­ porary risk and safety studies now place collective, and more broadly, orga­ nizational considerations at the forefront of analysis. We argue that the latter approaches-HRT, the work of Reason, Vaughan, Weick, Turner, and Pidgeon, and, most notably, the more recent work sketched above-that give additional, and more sophisticated, attention to organizational and contextual matters provide some continuity with the situational perspec­ tive developed in the previous section. Table 8.2 (after Falke, 2006) presents an overview of the resilience concepts developed for meso-level analysis.

The Meta-level: From Ecosystems to Societies

Engineering Resi lience

Ecology has traditionally understood resilience as the capacity to resist disturbance and change and, concentrating on near-equilibrium behavior,

Resi lience

  • 171

shift. Social-ecological systems have become a particular contemporary focus of attention, concerned with adaptive capacity understood in terms of the ability to self-organize and learn. However, this literature commonly overlooks the normative character of "desirable" change trajectories with ecological integrity privileged as the basis for them.13 This tension between an emphasis on ecological imperatives and the more recent concern with social and institutional processes remains unresolved. The ecological focus on system outcomes involves a meta-scale view of contingency, and of its management, that is at odds with the necessity for adequate attention to the situated specifics of relevant circumstances and of meso-scale processes outlined in the previous two sections of this chapter. While work on the adaptive management of social-ecological systems has become increasingly sophisticated in examining broader contextual and organizational matters, these tend to be subordinated to imperatives emerging from systems ecology without acknowledgment, or apparent cog­ nizance, of the broader relevant literature on contextual and/or organiza­ tional matters. The challenge that these difficulties pose has recently been thrown into sharp relief by the identification of tensions between the way resilience is conceived in adaptive management and in correlating work on transitions management (e.g., Foxon, Reed, & Stringer, 2008; Smith & Stir­ ling, 2008). The latter, focused upon facilitating a socio-technical transi­ tion (e.g., to a low-carbon economy) is also informed by complex adaptive systems theory (Foxon et al., 2008, p. 7) and on achieving sustainability, but conceived in terms of different outcomes (i.e., technological transi­ tions) and different processes (Foxon, Reed, & Stringer, 2008) to adaptive management. So, whereas learning to build resilience to ecological change focuses adaptive management, transition management focuses on facilitat­ ing radical, systemic, socio-technical transformation, for which adaptive resilience can be an impediment.

Resi lience 171 shift. Social-ecological systems have become a particular contemporary focus of attention, concerned with


Resi lience 171 shift. Social-ecological systems have become a particular contemporary focus of attention, concerned with

This section examined meta-level perspectives on resilience developed in systems ecology, most notably engineering resilience, ecological resilience, and adaptive environmental management. The tensions between this work and that on achieving socio-technical transformation were briefly discussed earlier in the chapter. An admission in a recent discussion of this tension that "the devil is in the detail" (Smith & Stirling, 2008, p. 3) underscores both the significance of the situational considerations that we identify as a key to the facilitation of resilience in practice and of synergies between the various level-bound analyses of resilience examined. These arguments are

Resil ience


  • 12. Relatively rapid changes in ocean circulation patterns are thought likely to

be central to this, the key concern being that the resultant change will prove

less convivial to human civilization (and the maintenance of other flora and fauna) than current conditions (see, e.g., http://www. aip.org/history/climate/oceans. htm).

  • 13. The Resilience Alliance (www.resalliance.org) and its journal Ecology and Society

(www.ecologyandsociety.org) are the primary vehicles for this perspective.

Resil ience 12. Relatively rapid changes in ocean circulation patterns are thought likely to be central


Johan M. Sanne

manageable, and necessary to keep the trains running, and because its widespread practice tends to reproduce its causes or because there might be no other options for the technicians. Bricolage that includes trading occupational safety for increasing production is taken for granted by front­ line management, which has also practiced it and which has to maintain its budgets while accomplishing contracted tasks. So bricolage seems nec­ essary to fulfill local goals with given, restricted resources. Bricolage that involves adding protection measures might not be globally rational, but it is locally rational in the absence of adequate managerial action. Unfortu­ nately, when technicians continue to make do with an inadequate set of resources, they also help to reproduce that mind-set; after all, they prove they can do it, and they also reproduce the virtues of bricolage in their community of practice. As I have shown, bricolage is only dangerous in certain situations, but it is often advantageous both for the organization and for technicians' enactment of a desired advantage. For this reason, only dangerous bricolage needs to be reduced. Management needs to find out in what situations bricolage is risky, inform technicians about these situations, and change whatever conditions that necessitate it (Peterson & Wingqvist, 1982). This will also serve to signal that bricolage, if endorsed as such, does not need to endanger technicians' lives.

Johan M. Sanne manageable, and necessary to keep the trains running, and because its widespread practice


Part Ill

When the first part of this book foregrounded the question of what we study by asking about vulnerability in technological cultures, and the sec­ ond part explored how to study vulnerability, this third part questions why to do it at all: what do these vulnerability studies produce, how do we benefit from them, and can they help to address questions of governance of risks and benefits of science and technology? In part I, we explored various forms of vulnerability by investigating a broad spectrum of empirical domains, ranging from violence in Gujarat and the chemical disaster in Bhopal to neonatal intensive care units in the Netherlands and farming communities in Andhra Pradesh. This resulted in a multifaceted concept of vulnerability: vulnerability is natural and social and technological; it may pertain to people, technological systems, and ecosystems; it has negative and positive connotations; and it adds to risk analysis and management, rather than substituting it. In part II, the con­ cept of vulnerability was further elaborated and operationalized by mount­ ing specific theoretical approaches to its investigation. Concepts thus introduced were precaution, creative dissent, resilience, and bricolage. The purpose of this third and last part of the book is to turn to practice. How could the broad conception of vulnerability in technological cultures, as introduced in part I and embedded in theoretical frameworks in part II, be used to address current questions of safety and democracy? Thus part III asks the following question concerning governance of vulnerability: which practices, which policies, which politics, and which ethics are needed to cope with vulnerability in technological cultures? Chapters in this part illustrate our answers to this question, with discussions of national and international policy, chemical risk regulation, medical prevention pro­ grams, pragmatist ethics, and an integrated framework of reflexive gover­ nance of vulnerability. In chapter 1 1 , "Governing a Vulnerable Society: Toward a Precaution­ Based Approach, " Gerard de Vries, Imrat Verhoeven, and Martin Boeckhout put uncertainty center stage, build on the new conceptualization of vulner­ ability, and then address the normative dimensions of policy making in the domain of safety issues. Once it is understood that uncertainties will persist, despite all efforts to limit them, uncertainty deserves pride of place over risk. To account for the dynamics of uncertainty, risk, and vulnerabil­ ity, we need to consider "uncertainty" as the default option and "risk" as a special condition-a condition that may come about when efforts to trans­ late uncertainty into calculated risk have been achieved successfully. They review the "classical risk approach" and its current challenges, and then subscribe to the International Risk Governance Council (2005) approach

Part Ill When the first part of this book foregrounded the question of what we study

The Governance of Vulnerability

The Governance of Vulnerability that distinguishes simple, complex, uncertain, and ambiguous risk prob­ lems. However, as


that distinguishes simple, complex, uncertain, and ambiguous risk prob­ lems. However, as De Vries, Verhoeven, and Boeckhout argue, it remains unclear how the question of allocating responsibilities is addressed. While risk management strategies for "simple" and "complex" risk problems are matched by explicit, legally defined frameworks that assign responsibili­ ties to government and private parties, a similar legal system is absent for "uncertain" or " ambiguous" risk problems. In these cases, risk management strategies rely on trying to reach reasonable agreement among concerned parties, rather than on using the force of law. Finally, these authors make a plea for a fundamental change in perspective: from "outcome responsibil­

ity" to "obligation responsibility." They then show how the Dutch Scien­

The Governance of Vulnerability that distinguishes simple, complex, uncertain, and ambiguous risk prob­ lems. However, as

tific Council for Government Policy (WRR) has translated this change in perspective into a new precaution-based approach to safety issues into the Dutch governance of vulnerability. Anique Hommels, Esther Versluis, Tessa Fox, and Marjolein van Asselt explore how vulnerabilities are dealt with in a regime for regulating risks in chemical industries, the so-called Seveso regime, in chapter 1 2, "Reg­ ulating Risks by Rules: Compliance and Negotiated Drift in the Dutch Chemical Industry under the Seveso Regime." This set of two directives and three amendments in the European Union attempts to regulate risks in the chemical industry and thus avoid accidents. Tracing the history of the Seveso regime, they demonstrate that the regulators aimed at continu­ ously tight(er) regulation as the best way to control the vulnerabilities of the chemical industry. They criticize this approach as an illusion of con­ trol. By investigating the local practices of Dutch companies and inspec­ tors in applying the Seveso rules and regulations through the conceptual lenses of "compliance" and "drift, " Hommels et al. show how rules are negotiated in interactions between regulators and regulated. This practice of "negotiated drift, " combined with some measure of compliance, is a very productive way of dealing with the vulnerabilities of the chemical industry. The authors argue for a transition from risk regulation based on quantitative risk calculation, more standardized rules, and strict compli­ ance, to a strategy of vulnerability governance that is more reflexive about risks and uncertainties, prioritizes learning from accidents, gives attention to a company's safety culture, and acknowledges and supports practices of negotiated drift. Ger Palmboom and Dick Willems, in chapter 13, "Dealing with Vulner­ ability: Balancing Prevention with Resilience as a Method of Governance, " discuss the prevention strategy to deal with vulnerability in medicine and argue that the success of this approach is threatened by its tendency to


Part ill

limitless expansion. In health care, the governance of patients' vulnerabil­ ity can take two forms: either making people less vulnerable by strength­ ening their resilience, or avoiding the risks and threats that they may be vulnerable to. The latter strategy is called "prevention" and is the most prevalent in actual medicine, but there are serious doubts as to its viability to deal with the increasing vulnerabilities in a technological (medical) cul­ ture. Palmboom and Willems show that the infiniteness of detectable risk will undermine prevention as a strategy of governance for vulnerability in the long run. This is related to a tendency that has been observed in many chapters in this book: the number of known risk factors gets higher every day, while the probabilities of these risks actually occurring seem to get lower. They exemplify this by studying the relation between health screening and risk knowledge among patients. Their alternative is to stress the partially positive side of vulnerability: they argue for the possibility to enhance the resilience of patients by regarding vulnerability as a source of, rather than a threat to, life. Making patients actively think about their usual coping strategies will help to maintain a level of "sense of coherence"-the medical operationalization of resilience. In chapter 1 4, "A Pragmatist Approach to the Governance of Vulner­ ability, " ]ozef Keulartz and Maartje Schermer demonstrate how pragma­ tist ethics provides the philosophical and ethical background to the shift from a risk-based to a vulnerability-based perspective. They distinguish two approaches to the governance of uncertain and possibly hazardous emerging technologies. The first approach focuses on controlling risk and reducing vulnerability by reducing the exposure and sensitivity of socio­ technological systems to stress. This strategy relies on the implementation of rules and regulations to contain hazardous conditions. This strategy of risk avoidance can be counterproductive due to the imminent danger of overregulation and overcontrolling. The second approach stresses the resil­ ience and robustness of socio-technical systems and their capacity to adapt to emerging vulnerabilities. Keulartz and Schermer especially examine what pragmatism can contribute to this second approach. They elaborate two pragmatic tools: dramatic rehearsal and reconstructive thinking. Dramatic rehearsal as a means to explore various possible futures can be helpful for governing and guiding new technological developments and their social context. It offers us images of possible futures and so enables us to discover new vulnerabilities and risk, as well as adaptive capacities and opportuni­ ties. It can reveal the multitude of actors and agencies that have a stake in the developments, as well as their different values, views, and emerging

220 Part ill limitless expansion. In health care, the governance of patients' vulnerabil­ ity can take

The Governance of Vulnerability


and shifting responsibilities. Reconstructive thinking is aimed at facili­ tating and fostering communication and cooperation across borderlines that separate these diverse actors and agencies. It requires the capability to look at problems from various angles simultaneously. Strategies such as reframing, gradualization, common ground dialogue, and boundary work can help bridge the rift between different views of risks, vulnerabilities, options, and opportunities, and thus open new and unexpected possibili­ ties for exchange and fruitful collaboration. Dramatic rehearsal and recon­ structive thinking are important tools for successful adaptive management of (emerging) socio-technical systems. The final chapter, "From Sustainability to Transformation: Dynamics and Diversity in Reflexive Governance of Vulnerability, " by Andy Stirling, concludes this book with a comprehensive framework to describe various styles and strategies of governance of technological culture. After review­ ing the dominant ways of framing technological vulnerabilities in terms of contrasting notions of sustainability, Stirling foregrounds the need to include contested normativities in our analyses. Stirling can now discern systematically between temporalities of vulnerability expressed as "shock" or "stress, " and between styles of action undertaken as "control" or "response." He then articulates the dynamic properties of stability, durabil­ ity, resilience, and robustness, as well as their relationships to sustainability and technological vulnerability. Underlying the rich spectrum of possible strategies for a reflexive governance of vulnerability in technological cul­ tures, Stirling significantly highlights the importance of epistemic, norma­ tive, and ontological diversity. This book thus ends with the following line of argument. The baton is taken over from part II by De Vries et al., when they frame governance of vulnerability around the central role of uncertainty and the need to address the normative dimensions of vulnerability governance. They answer this call by proposing specific roles for obligation responsibility and precaution in the governance of vulnerability. Then Hommels et al. underline this plea by demonstrating the illusion of control by regulation, and continue to propose a much more reflexive style of governance that uses negoti­ ated drift as its core governance strategy. Palmboom and Willems continue the same focus on reflexivity by proposing an active-thinking attitude by which patients can strengthen their resilience. This strategy builds on the positive aspect of vulnerability: giving space to doubt, uncertainty, error, and ambivalence may lead to strengthening resilience when geared toward enhancing the patient's sense of coherence. The foundation of such a


Part Ill

strategy can be found in pragmatist ethics, as Keulartz and Schermer dem­ onstrate. The patients' active thinking is here generalized into the pragma­ tist's tools "dramatic rehearsal" and "reconstructive thinking. " They thus further operationalize the shift from control of risks to adaptation to vul­ nerabilities. Finally, Stirling combines all this into a comprehensive model of reflexive governance that reconnects an ethics of sustainability, stability, and resilience to strategies of normative and ontological diversity.

Part Ill strategy can be found in pragmatist ethics, as Keulartz and Schermer dem­ onstrate. The

Governing a Vulnerable Society


new normative basis for governing a vulnerable society and to the policy implications that may follow.1

The Classical Risk Approach

Policies to help govern a vulnerable society may be conceived as compris­ ing three components. In the first place, as "vulnerability" is a broad and rather vague term, they require a specific conceptualization-and prefer­ ably operationalization in measurable, quantitative terms-of the issues at stake. The second component involves norms that address how responsi­ bilities for taking preventive or precautionary measures, and now when disaster strikes should be assigned and distributed. The third component is the translation of the former two elements into legal, institutional, and organizational terms. For several decades, the dominant approach has been based on a concep­ tualization of vulnerability in terms of risk (i.e., the product of the chance of an adverse effect and the extent of damage). Predominantly focusing on preventing particular hazardous outcomes of actions, risk policy comprises the double task of risk assessment (the identification and evaluation of risks in terms of probabilities and extent of damage) followed by risk man­ agement (taking measures to limit or control risks that are deemed to be unacceptable). Because of its canonical status, we refer to this arrangement as the "classical risk approach." The classical risk approach's first component, the conceptualization of vulnerabilities in terms of risk, has a long prehistory. The concept of risk dates back to the 1 6th century (Asselt, 2000). In the early 20th century, it was introduced in the social and agricultural sciences and in the world of finance and insurance. Risk analysis as a deliberate and formal approach to safety issues moved to the center stage of safety policy in both the pub­ lic and private sector after World War II. In the Netherlands, the move was triggered by a major incident, the flood of 1953. The design of the "Delta Works"-an immense system of dams, dikes, storm surge barriers, and locks-that was started after the catastrophe was based on detailed cal­ culations of the likelihood of flooding and the expected amount of damage in various regions (Milieu- en Natuurplanbureau and Rijksinstituut voor Volksgezondheid en Milieu, 2004). Similar approaches were introduced elsewhere (e.g., in the nuclear industry in the 1960s, and a decade later, in the chemical industry). In 1975, the WASH- 1 400 Reactor Safety Study (Ras­ mussen Report), which assessed the risks of a serious accident at a nuclear

Governing a Vulnerable Society new normative basis for governing a vulnerable society and to the policy
  • 226 Gerard de Vries, lmrat Verhoeven, and Martin Boeckhout

Gerard de Vries, lmrat Verhoeven, and Martin Boeckhout reactor, set the international standard for estimating

reactor, set the international standard for estimating the consequences and probabilities of a major industrial calamity. By the 1 980s, risk analysis had become an established discipline in engineering schools (Hale, 2006). Over the past decades, the risk-based approach to safety issues has evolved gradually into what now is a widely accepted framework. In this framework, risk assessment is primarily delegated to science. Based on the scientific identification and evaluation of risks, authorities decide whether to take measures or not. Risks assessed as unacceptable call for risk man­ agement, which comprises regulation, oversight, monitoring, and enforce­ ment. Actual practice, however, is more complicated than the analytic division of labor between risk assessment and risk management suggests. An extensive amount of literature tries to find a middle ground between ideals and practicalities; cf. Fischhoff et al. (1981) for an early entry and International Risk Governance Council (2005) and Renn (2008) for the state of the art. The current popularity of the risk-based approach toward governing vul­ nerability can hardly be overstated (Power, 2007). Risk assessment and risk management have become common practices in a wide variety of domains where safety issues are perceived to be at stake-from new technologies to infectious diseases, from hazardous substances to traffic, information­ communication technologies, and the financial world. Because risk management requires allocating responsibilities for preven­ tion to specific actors, it has to be complemented by a second, normative component. Based on insight into the nature of the risk at hand and analy­ ses of where and by whom effective and efficient action might be under­ taken to prevent adverse effects to occur, as well as considerations of costs and justice, a balance must be found between the principle of individual responsibility on one hand and the principle of solidarity or public interest concerns that may guide collective action on the other. To a large extent, individual responsibility is thereby also reinterpreted in terms of risk, with actions conceptualized as bets on outcomes and responsibility as a special relation that we have with the outcomes of our bets (Honore, 1 999; Steele, 2004). This concept of responsibility is known as " authorship responsibil­ ity" or "outcome responsibility" (Anderson, 2008; Honore, 1 999). When it is decided that collective action is needed and that government should be involved, public agencies still may play various roles. In some cases, they will accept full "operational responsibility. " For example, in most coun­ tries, the operational responsibility for flood prevention and vaccination programs is assigned to public authorities. However, governments may decide to limit themselves to taking on only "final responsibility." In these


Gerard de Vries, lmrat Verhoeven, and Martin Boeckhout

decades, flood prevention policies today face the uncertain consequences of global warming. Risk assessment is also confronted with a growing awareness that the role of experts is more complicated than originally conceived. Already in the early 1980s, it was understood that risk assessment involves normative judgments. Although performed by scientists, it is not a value-free affair (Fischhoff et al., 1981). Even where the same laboratory equipment is used worldwide, styles of accountability and notions of objectivity may vary strongly among nations Qasanoff, 1 986, 2005b). Experts with different national, normative backgrounds may derive opposite assessments of risk while agreeing on scientific data [cf. Gillespie et al. (1982) for environmen­ tal risks; Bodewitz, Buurma, and De Vries (1987) and Abraham and Davis (2007) for pharmaceuticals; and Whiteside (2006), Chapter 3, for GMOs] . In a globalized world, this again is a source of uncertainty; when issues become politicized, risk assessment leaves the relatively quiet corner that the classical risk approach had reserved for experts. The same happens when the role of experts in risk assessment is ques­ tioned from outside the scientific world. Experts were late in picking up "early warnings" from journalists and individual or organized citizens on the pesticide dichlorodiphenyltrichloroethane (DDT; Carson, 1 962), radia­ tion hazards near the Sellafield nuclear site in England (Wynne, 1996), and air pollution in the Rotterdam area in the 1 960s (Dijstelbloem, 2007). Initially, they also substantially underestimated the hazards of radiation, polychlorinated biphenyls (PCBs), diethylstilbestrol (DES), asbestos, and BSE (European Environment Agency, 2001). Like risk assessment, risk management increasingly faces issues of uncer­ tainty. A first challenge is rooted in success. By addressing safety issues time and again with the same approach, risk policy has evolved into an immensely complex system of national and international regulation. This raises profound questions about the efficacy and efficiency of the risk regimes (Hood, Rothstein, & Baldwin, 2001). Inconsistencies often occur. Moreover, when fate strikes and a major accident occurs, the complexity of regulations and their implied distribution of responsibilities often leads to erratic behavior of authorities and to distasteful blame games in political arenas and in courts. In either case, trust in safety policies and the legiti­ macy of future policies erode. Beck ( 1999) contends that this indicates the emergence of a new phase in the development of modernity, which he calls the "risk society." By successfully managing ever more minute harms, more intractable forms of vulnerability escape consideration to the point where "organized irresponsibility" emerges.

Gerard de Vries, lmrat Verhoeven, and Martin Boeckhout decades, flood prevention policies today face the uncertain

Governing a Vulnerable Society


Risk management is also put to the test because citizens have become increasingly critical about how political choices, including ones that per­ tain to safety issues, are being made (Hajer, 2003, 2009). Moreover, they have both the assertiveness and the political and legal means to effectively bring to a standstill policies defended by experts as rational and necessary. For example, in the Netherlands, emergency overflow areas designed to cope with high water levels of rivers have met intense protests that forced the government to withdraw its plans (Verhoeven, 2006), while Swedish citizens have lodged complaints against Universal Mobile Telecommuni­ cations System (UMTS) aerials that led to the postponement of the full rollout of 3G mobile telephony (Soneryd, 2007). The UK opposition to the measles, mumps, and rubella (MMR) vaccination was succeeded by a Dutch movement that encouraged young girls not to be vaccinated against cervi­ cal cancer. Long conceived as an issue that could be dealt with along the established lines of expert-based decision making, vaccination suddenly became perceived outside scientific circles, both in mainstream media and on the Internet, as a problem involving substantial uncertainties. The experts' view that the proposed measures were based on the best available scientific evidence was suddenly reduced to being just one opinion among others. In all these cases, the unexpected opposition forced both experts and policymakers to shift gears or to engage in public discussions about the merits and possible side effects of the policies that had been proposed. Similar problems have emerged in the private sector. In 1 995, on the basis of extensive risk assessments (and to the satisfaction of UK safety authorities), Shell claimed to have found a rational way to discard of the Brent Spar, an oil storage and tanker loading buoy, only to discover that the public, alarmed by Greenpeace activists, thought differently. The con­ sumer strike that followed and the blow to Shell's reputation caused sub­ stantial financial losses for the company (Fombrun & Rindova, 2000).

Governing a Vulnerable Society Risk management is also put to the test because citizens have become
Governing a Vulnerable Society Risk management is also put to the test because citizens have become

Accommodating Uncertainty: Amending the Classical Risk Approach

The abovementioned incidents illustrate the kind of challenges that the classical risk approach has to face. They have caused the risk policy com­ munity to amend the framework of the classical risk approach. To allow for dealing with a wider range of policy situations, taxonomies of risk problems have been introduced with degrees of incertitude serving as the main distinguishing mark (International Risk Governance Council, 2005; Klinke & Renn, 2002; Milieu- en Natuurplanbureau, 2003; Renn, 2008; Renn & Walker, 2008). The purpose of these taxonomies is to chart how


Gerard de Vries, lmrat Verhoeven, and Martin Boeckhout

each category of risk problems calls for a specific set of strategies in risk management. The various challenges to the classical risk approach are thus primarily accommodated by an amendment of the way vulnerability is conceptualized. The IRGC white paper "Risk Governance-towards an Integrative Approach" may serve as an authoritative example [International Risk Gov­ ernance Council (IRGC), 2005] . Based on the different states of knowledge about particular risks, the IRGC distinguishes between "simple, " "com­ plex, " "uncertain, " and " ambiguous" risk problems. " Simple" risk problems present themselves with little or no uncertainty. In these cases, relevant information is available in textbooks and handbooks; past experience or laboratory research allows reliable estimates of frequencies of occurrence of adverse effects and the extent of damage. In other cases, however, risk assessment is complicated by the presence of still unknown intervening variables, inter-individual variety, or long-term delays between cause and effect. Now we have entered the category of "complex" risk problems. Knowledge about complex risks has not yet reached handbook status; details are still discussed in scientific journals and are likely to be revised when more studies become available. The third category, "uncertain" risk problems, applies to threats about which little or no knowledge about their chances, extent of possible damage, and/or cause-effect relations is avail­ able. For instance, such problems may present themselves in relation to new technologies because evidence of the long-term impact of full deploy­ ment of a technology in real-life situations is not yet available. Finally, society may be confronted with "ambiguous" risk problems, characterized by conflicting views about the interpretation of a risk and its tolerability. The IRGC leaves the task of deciding which category of risk is at stake in a particular case to a "screening board" consisting of members of the risk assessment team, risk managers, and key stakeholders, such as industry, nongovernmental organizations (NGOs), and representatives of related reg­ ulatory or governmental agencies [International Risk Governance Council (IRGC), 2005, p. 5 3]. Originally developed in academic circles, the taxonomies have by now reached policy circles and have begun to be adopted as the basis of poli­ cies of both national governments and the European Union (EU; Klinke, Dreyer, Renn, Stirling, & Van Zwanenberg, 2006). However, when the pro­ posed taxonomies are exported to policy practice, several problems may emerge. First, although simple and complex risk problems may seem to be the least exciting categories from an academic point of view, they may still be

Governing a Vulnerable Society


ridden with substantial difficulties for policymakers. Time and again, poli­ cymakers discover that while dealing with risk problems that are conceived to be simple or complex ones, they cannot rely on established decision­ making procedures and have to invoke other strategies, such as getting involved with a much wider range of stakeholders. Opposition to UMTS aerials is a case in point. According to most experts, UMTS technology pres­ ents a simple, or at most a complex, risk problem. However, in the face of growing public opposition, policymakers may find it to be ambiguous. Second, the reverse may-and indeed does-occur sometimes. Under the pressure of such factors as economic competition, policymakers may want to cut off prolonged scientific debate, with the result of mistaking uncertain or ambiguous risk problems for simple ones. Third, they may be confronted with new natural conditions, which cause risk problems to migrate from one category to another. For example, in flood prevention, established policies based on long-term time-series of sea level fluctuations are by now challenged by the uncertain local effects of sea level rise caused by global warming. These uncertainties present severe problems not only for hydrologists, but also for policymakers, who may have to convince citizens and local authorities that far-reaching and pain­ ful measures have to be taken on the basis of an assessment of "uncertain" risks. In all these cases, risk problems almost overnight may turn out to be of a different type than policymakers and their advisers originally thought. The various issues that emerge when risk taxonomies are used as a guide for policy practice point to a major weakness of such classifications. Intro­ ducing risk categories based on the degree of incertitude fails to account for the fact that in society, knowledge and incertitude are distributed over time and place. Risk taxonomies are essentially contestable. What may appear to some to be a simple problem may turn out to be an uncertain or ambigu­ ous risk problem elsewhere, or at a later moment, for various reasons. The screening boards that the IRGC white paper proposes are poor instruments to address the problem that perception of vulnerabilities varies in time and social position. The idea that such boards can cut short public discussions about the proper way to categorize a risk problem will often prove to be an illusion. Assertive citizens show little respect for a consensus reached in closed board meetings. And dissidents within the scientific community will find access to the media when they encounter little understanding for their views in board meetings. Governing vulnerabilities is a more unruly affair than the amendments to the classical risk approach that have been introduced in the past allow. To anticipate this unruliness, a more radical revision of the governance of vulnerability will be required.

Governing a Vulnerable Society 231 ridden with substantial difficulties for policymakers. Time and again, poli­ cymakers

Governing a Vulnerable Society

Governing a Vulnerable Society 233 However, the reluctance to translate the amendments into normative consequences is


However, the reluctance to translate the amendments into normative consequences is understandable. Responsibility is at odds with uncertainty. In the dominant Western legal (and ethical) tradition that emphasizes out­ come-responsibility, unforeseeable adverse consequences are conceived as a matter of force majeure. The law operates under the understanding that nobody can be held responsible for consequences that reasonably couldn't be foreseen. In other words, uncertainty serves as an excuse for inaction. To draw normative and legal consequences from the challenges to the clas­ sical risk approach, the complex relations between risk, uncertainty, and responsibility need to be considered first.

Governing a Vulnerable Society 233 However, the reluctance to translate the amendments into normative consequences is

The Dynamics of Uncertainty and Risk

Risk and uncertainty are complex concepts indeed. "Uncertainty" refers to limited knowledge (Asselt & Petersen, 2003). The nature of this limitation may vary from inexactness (statistical problems, such as substantial margins of error in measurements) and unreliability (problems at the level of models or constructs) to ignorance (nothing is known about likelihoods, effects, their relations, or substantial lack of insight into the underlying processes (Milieu- en Natuurplanbureau, 2003, p. 22). Klinke and Renn (2002) and the IRGC (2005, p. 30) use a similar approach to designate variations in uncertainty. Uncertainty may refer to both "known unknowns" and to "unknown unknowns. " Obviously, risk implies uncertainty, as the concept refers to the pos­ sibility or chance that adverse effects occur, not to a catastrophe that will strike with certainty. However, we also may be more or less certain about the existence of a risk. A further complication of the concept of risk derives from what Klinke and Renn (2002, p. 6) have called the "dual nature" of risk. Risk seems to have two faces. On the one hand, risks present them­ selves as real, objectively given entities. Smoking cigarettes is related to health risks that unmistakably exist. On the other hand, "risk" does not refer to the catastrophe itself-the disease that one may contract because of heavy smoking-but to an adverse effect that is expected with some prob­ ability. Expectations, however, are mental or social constructs that belong to the domain of perception and subjectivity, rather than to the objective world. Hence, Klinke and Renn conclude that risk has a dual nature: it is both a feature of the objective world and a subjective construct. The dual nature of risk has incited debates between realist and social­ constructivist interpretations of risk. Much confusion, however, can be


Gera rd de Vries, lmrat Verhoeven, and Martin Boeckhout

avoided by dropping the heavily loaded philosophical notions of "objec­ tivity" and "subjectivity" in favor of a new approach suggested in recent science and technology studies (Latour, 1 987, 1 999, 2005). How should a risk problem be categorized? It all depends on when and where we enter the scene. Today, we find the health risks of ciga­ rette smoking to be a well-established fact. But if we travel back in time, we will find that even in the 1 940s, established medical experts did not expect smoking to cause a threat to public health (Patterson, 1987). So indeed, risk has two faces; but these two faces are not just a matter of analysts quarrelling about the realist and social-constructivist interpreta­ tions of risk. Risk as a social or mental construct and risk as an objectively given entity are connected by the long trajectory of efforts of numerous epidemiologists and medical scientists that transformed contested and uncertain conjectures about the effects of cigarette smoking into hard facts. Framing the two faces of risk as its "dual nature" neglects this time­ dimension; the complicated work that goes into translating the one face of risk into the other is taken for granted. Accounting for the dynamic relation between risk and uncertainty leads to a number of conclusions. In the first place, we may observe that risks are complex constructions. To translate uncertainties into risks, data must be collected and cause-effect relations need to be established. All of this will take time, effort, and organization. Only after this work has been achieved can a risk function as an objectively given entity (i.e., as something that exists independently of the person who makes the observation). The trans­ formation of uncertainty into risk, however, may remain incomplete. In other words, uncertainties may linger. Sometimes the path will even be covered in reverse order when new studies raise doubts about what for a long time had been accepted as objective fact. Second, once it is understood that risks are constructed in a long process that starts with uncertainties, and that uncertainties may persist in spite of efforts to reduce them, we may observe that analytically, uncertainty deserves pride of place over risk. To account for the dynamics of uncer­ tainty and risk, "uncertainty" should be considered as the default option and "risk" as a special condition-namely, as a condition that may come about when efforts to translate uncertainty into calculated risk have been achieved successfully. For this reason, the view that informs current risk problem taxonomies, where uncertainty is conceived as an attribute that applies only in exceptional cases, has to be reversed. Uncertainty has to be considered as the default situation; only after substantial efforts, uncertain­ ties may be translated in more or less uncertain risks. The idea is nicely

Governing a Vulnerable Society 235 phrased by Nowotny et al.: "Risks are calculable islands ... in

Governing a Vulnerable Society


phrased by Nowotny et al.: "Risks are calculable islands




sea of

uncertainty" (cited in Asselt & Vos, 2006, p. 314). Third, this reversal leads to a fundamental shift in thinking about risk assessment and risk management. Stressing the work and organization that is needed to translate uncertainty as far as possible into risk-rather than taking these efforts for granted-implies a fundamental change in perspec­ tive on risk assessment and risk management. Accepting uncertainty as the default option, and risk as a special condition that may come about when efforts to translate uncertainty into calculated risk have been achieved suc­ cessfully, may be considered the first component of a new, precaution­ based approach to safety issues. Similar to the case of the classical risk approach, the first component of the precaution-based approach to safety issues has to be complemented by a second, normative component, as well as with institutional and organi­

zational arrangements and laws to put the approach into practice. Giving pride of place to uncertainty over risk, however, raises profound normative questions. We will discuss these first, before addressing some of the insti­ tutional and organizational issues that are implied by the precaution-based approach.

Changing the Perspective: F rom Outcome Responsibility to Obligation Responsibil ity

In the dominant Western legal and ethical tradition, actors are held respon­ sible only for the consequences of their actions that they reasonably could have foreseen. The classical risk approach-and laws that are based on this approach-is grounded in this conception of ("outcome") responsibility. Conceiving actions as bets on outcomes and responsibility as a special rela­ tion an individual or collective actor has with the outcomes of his or her bets, an actor is held accountable for adverse consequences that a reason­ able person could have foreseen. If a man has made a wrong bet, he has to bear the consequences. He has to compensate victims, to step down from office, and if he has broken the law, he may even be prosecuted. However, for consequences that reasonably couldn't be foreseen, nobody can be held responsible. Such consequences are conceived as a matter of force majeure. It should not come as a surprise that company lawyers actively play this card to avoid liability claims (McGoey, 2009). Of course, the line between "consequences that a reasonable person could have known might occur" and "unforeseeable consequences" is hard to draw. The law may however specify conditions that apply. In the EU

Governing a Vulnerable Society 235 phrased by Nowotny et al.: "Risks are calculable islands ... in
  • 236 Gerard de Vries, lmrat Verhoeven, and Martin Boeckhout

directive for product liability, for example, a producer is excused from lia­ bility if he can prove that "the state of scientific and technical knowledge at the time when he put the product into circulation was not such as to enable the existence of the defect to be discovered" (85/3 74/EEC, art. 7e). In the end, a court may have to decide whether the actor should be held responsible for the consequences of his or her actions or be excused. Limitation of the concept of responsibility to foreseeable effects makes good sense. In a democratic state, under the rule of law, citizens may expect lawmakers to stipulate what the state imposes on them in precise terms. Without specifying precise conditions, legal security may easily be j eop­ ardized. It may also obstruct the important ex ante effect of attributing responsibilities. The specification of conditions invites actors to reflect on the harmful effects that their actions might have prior to their acting. An actor who is held responsible for all effects, including unforeseeable ones, may be tempted not to act at all-which may have adverse effects as well­ or, confronted with an endless array of possible effects for which he or she may be held accountable, decide to simply refuse to look, to go ahead and to wait and see what will happen. However, in opposition to the dominant view that focuses on out­ come responsibility, another longstanding legal and ethical tradition links responsibility to an actor's social position, rather than the outcomes of actions. Stressing obligation responsibility (Anderson, 2008), this tradition discards the link between responsibility and foreseeable effects of actions to focus on the responsibilities that are implied by occupying a specific position on society-e.g., public administrator, physician, or citizen. Each of these social positions comes with obligations to act in the face of threats that may endanger not only the actor, but also others. According to this tradition, "it is not the origin of injury, but the possibility of preventing and reducing its costs, that allows us to judge whether there was or was not unjustifiable passivity in the face of disaster" (Shklar, 1 990, p. 81). While outcome responsibility focuses attention on actions and the damage that they may inflict, obligation responsibility focuses on the vulnerability of the social and natural environment of someone who holds a certain posi­ tion. Obligation responsibility requires a person to be attentive toward vulnerability and to act when there are reasons to think that vulnerable humans, society, or the natural environment might become endangered. Obligation responsibility thus shifts the perspective. Whereas outcome responsibility requires a legal person to prevent foreseeable adverse effects, obligation responsibility requires attentiveness toward vulnerability and to react proactively when there are reasons to think that harm might occur

  • 238 Gerard de Vries, lmrat Verhoeven, and Martin Boeckhout

left untouched; uncertainty is treated as an exceptional condition for risk management, and the precautionary principle will be invoked only at the stage when decisions have to be made. In contrast, in the WRR formulation to safety issues, "precaution" means taking uncertainties seriously tout court; that is, as the default option (cf. also chapter 15, "From Sustainability to Transformation: Dynamics and Diversity in Reflexive Governance of Vulnerability"). It does not just demand private parties and public authorities to provide proof of safety, but to accept the responsibility to be attentive to the vulnerability of the human, social, and environmental context in which they operate, and to undertake the efforts to translate uncertainties as far as possible into risks, to allow better-founded judgments. Under this reading, the criticism that the precautionary principle will hamper innovation and progress is also unfounded. The WRR formulation of the precautionary principle demands parties to move, to start an investigation, or to do research. Confronted with the threat of serious and irreversible infliction of vulnerability, inertia would be precisely the wrong reaction. More activities and innovation are required than would be expected if this principle is not applied. The WRR principle of precaution calls for both public authorities and businesses to move, to act, to take steps that might help to translate uncertainties into articulated risks that allow for an evaluation of possible courses of action. A panoply of measures may be considered ( cf. White­ side, 2006, p. 52ff.). The principle may call for gathering further informa­ tion, for performing new research, for monitoring and tracing effects, for promoting a view that deliberately includes a wide variety of disciplinary expertise, for developing new technologies, for using larger safety margins, and for creating backup systems. Finally, precaution can mean imposing a moratorium, or a ban on activities. But first and foremost, "precaution" means that actors have to accept the responsibility to take uncertainties seriously into account by default. In the face of uncertainty, new activities are required. In contrast to the fear that is often expressed, precautionary policies are more likely to advance new research rather than bringing inno­ vation to a stop. This new precaution-based approach moves issues of responsibility to another plane than the classical risk approach. Actors are not just held responsible for the actions they undertake and held liable for damage that they may inflict on others; in light of the vulnerability of humans, society, and the natural environment, their primary responsibility is to organize the efforts that a proactive approach to uncertainty requires. This may involve a variety of specific obligations-from having adequate research

Gerard de Vries, lmrat Verhoeven, and Martin Boeckhout left untouched; uncertainty is treated as an

Governing a Vul nerable Society


facilities needed to translate uncertainties as far as possible to calculable risks, to adequate post-marketing monitoring facilities, multidisciplinary technology assessment, adequate whistleblower procedures, and consulta­ tion with stakeholders. The obligation to focus on vulnerability may also entail further obligations (e.g., compliance with sustainability-oriented policies). Of course, which specific duties are implied depends on the kind of activities that are undertaken and the position of the actor in society. Obligation-responsibility is position-specific. A company that plans to get involved with GMOs or nanotechnology will have to meet other demands besides one that produces products based on well-established technologies.

Governing a Vul nerable Society: Putting the Precaution-based Approach into Practice

A number of developments both in the private and the public sector have already anticipated core elements of the precaution-based approach pro­ posed by the WRR. For example, after the Brent Spar affair, Shell delib­ erately changed its risk assessment and management procedures to systematically include the expectations of a wide variety of stakeholders to enhance alertness for vulnerabilities that company experts may overlook (Fombrun & Rindova, 2000). In pharmacy, physicians are systematically invited to collect and report anecdotal information about side effects of drugs, to signal possible-but as yet uncertain-risks at an early stage. A worldwide monitoring system for early detection of threats of new epidem­ ics organized by the World Health Organization (WHO) is in place. EU food safety issues are discussed in terms consonant with the approach outlined previously (Dreyer & Renn, 2009). It is by now widely understood that tak­ ing uncertainties seriously will require organizations to have the flexibil­ ity to deal with the unexpected (i.e., to stimulate organizational learning). This includes enabling processes of stakeholder consultation, stimulating creative dissent, and to have provisions for early warners to come to the fore, as well as the presence of early listeners and a preparedness to bypass existing procedures if necessary (cf. also chapter 7, "Creative Dissent: Link­ ing Vulnerability and Knowledge in India"). The precautionary approach, however, cannot rely only on volun­ tary measures. To give the approach the full force required for governing a vulnerable society responsibly, legislation will be required that obliges both public and private parties to organize effectively the alertness to pos­ sible infringements of vulnerability that the precaution-based approach demands. It requires them to have research facilities available that not only

Governing a Vulnerable Society


allows taking up the full consequences of the uncertainties that society must confront. Once this step is taken, a new, precaution-based approach to safety can be formulated that allows for legal codification and institu­ tional and organizational provisions. For decades, the classical risk approach has been a successful framework for dealing with safety issues. However, because of the new threats that global warming, new technologies, and still unknown infectious diseases (to cite just a few examples) present, but also-as the BSE crisis made clear­ because of the increasing complexity of supply chains, a new approach to safety issues is urgently needed to govern our vulnerable society. The precaution-based approach that we have outlined in this chapter provides a normative guideline to take up these challenges.


  • 1. This chapter is based on the report Uncertain Safety, which we wrote for the WRR,

the Scientific Council for Government Policy in The Hague, the Dutch cabinet's think tank for long-term policy issues. The WRR has adopted the report as official advice to the Dutch cabinet.

Governing a Vulnerable Society 241 allows taking up the full consequences of the uncertainties that society


Jozef Keulartz and M aartje Schermer the resilience and adaptive capacities of these systems. "Resilience" is

Jozef Keulartz and M aartje Schermer

the resilience and adaptive capacities of these systems. "Resilience" is the capacity of a system to cope with disturbances without shifting into a qual­ itatively different state. A resilient system has the capacity to withstand shocks and surprises and, if damaged, to rebuild itself (for more, see chapter 8, "Resilience: Contingency, Complexity, and Practice"; chapter 13, "Deal­ ing with Vulnerability: Balancing Prevention with Resilience as a Method of Governance"; and chapter 15, "From Sustainability to Transformation:

Dynamics and Diversity in Reflexive Governance of Vulnerability"). The management paradigm of the vulnerability approach is that of adaptive management. Adaptive management can steer dynamic systems toward more resilient and productive states and avoid undesired regime shifts through a learning-by-doing approach (i.e., through continuous monitoring and adapting responses to feedback from the system in col­ laboration with a diversity of stakeholders). This management approach rejects the assumptions of the "modernist" management approach, includ­ ing its restrictive emphasis on disembedded knowledge, linearity, and con­ trol. Adaptive management as a strategy for natural resource management can be traced to the seminal work of Holling (1978), Walters (1986), and Lee ( 1993). However, the ideas that inspired adaptive management can be traced back earlier to philosophical pragmatism, particularly to John Dew­ ey's approach to social learning and experimental activism in democracies (Lee, 1 993; Norton, 1999). Adaptive management recognizes that diversity of stakeholders with dif­ ferent types of knowledge and from different institutional settings is a criti­ cal factor in maintaining and enhancing resilience of socio-technological systems. Inclusion of a broad diversity of actors within processes of delibera­ tion and decision making provides risk distribution and a greater range of options for responding to stress and surprise. Redundant systems can com­ pensate for human errors and unpredictable changes and the risks associated with "putting all eggs in one basket" (see chapter 15). However, the inclu­ sion of an ever-growing group of actors and agencies with different interests, views, and values will more often than not lead to conflicts over new and emerging socio-technological systems. To avoid fragmentation, adaptive management has to provide an arena for trust building, social learning, conflict resolution, and collaboration across the boundaries between stake­ holder groups and their different social worlds. In this chapter, we will turn to the pragmatist roots of adaptive man­ agement and examine the extent to which pragmatist methods of inquiry can contribute to the successful governance of vulnerability of our socio-

A Pragmatist Approach to the Governance of Vulnerabil ity technological systems (i.e., to the enhancement of

A Pragmatist Approach to the Governance of Vulnerabil ity


technological systems (i.e., to the enhancement of their resilience and adaptive capacity). In the first section of this chapter, we will describe the nature of prag­ matist ethics. This approach differs from traditional ethics in at least two respects. First, pragmatism has always been interested more in the process of (moral) inquiry than in its ready-made products. Second, in pragmatism, the discovery of new moral vocabularies that can be deployed for tackling emergent problems is of as much importance as the justification of moral judgments. These shifts have led pragmatists to develop new methods of ethical inquiry. Two methods deserve special attention because they can contribute to the adaptive management of new and emerging socio­ technical systems: "dramatic rehearsal" and "reconstructive thinking." In the second section, we will show that dramatic rehearsal can be a helpful tool for ethical inquiry into new technologies, as well as a tool for governing and guiding new technological developments. The use of dra­ matic rehearsal shows that many actors are involved in socio-technological developments, each with their different interests and points of view. This implies that in dealing with the vulnerabilities and adaptive capacities of new socio-technological systems, a wide variety of views, interests, and per­ spectives must be taken into account. Dewey's idea of "reconstructive thinking, " discussed in the third sec­ tion, has inspired the development of methods for coping with different risk perceptions and the underlying value pluralism and value conflicts. These methods share an emphasis on fostering and facilitating commu­ nication and cooperation across the boundaries that separate communi­ ties with different and sometimes conflicting interests, views, and values. They are meant to counteract black-and-white thinking, which more than once has been a stumbling block to the advancement of a flexible attitude toward and an open discussion of possible futures-a prerequisite for posi­ tive vulnerability (see chapter 1, "Studying Vulnerability in Technological Cultures").

Pragmatist Ethics

Following a relatively long incubation period, in recent decades pragma­ tist philosophy has made a remarkable comeback, not only in the United States, but also in Europe. The names of Wilfrid Sellars, W. V. Quine, and Hillary Putnam mark the main stages in the transition from a more ana­ lytical to a more pragmatist approach to philosophy. The final stage was

· A Pragmatist Approach to the Governance of Vulnerabil ity

· A Pragmatist Approach to the Governance of Vulnerabil ity From justification to Discovery The second

From justification to Discovery


The second shift in attention is one from the "context of justification" to the "context of discovery" (to use the language of the philosophy of sci­

ence). Pragmatists attach as much importance to heuristics, to the " art of invention, " as they do to the justification of moral judgments. Pragmatism emphasizes the importance of novel constructs and hypotheses with which

emergent moral problems can be tackled. "Pragmatist ethics [


is not

only interested in the application of pregiven normative rules, but in the construction of new possibilities for moral action" Ooas, 1993, p. 253) In more traditional ethics, on the other hand, the emphasis lies mostly upon the justification of moral judgments and scant attention is paid to discov­ ering new ways of understanding. Pragmatists argue that in our daily lives, we have enormous faith in the compass of fixed moral beliefs and self-evident customs. But due to the openness of the world, which is enhanced by technological innovations, we are regularly confronted with unexpected and novel problematic situ­ ations. We are then forced to revise our beliefs and customs. This requires

moral inquiry and discovery-in the sense of creative construction-of new concepts and vocabularies.

Three Antitheses

Together, these two shifts define what moral inquiry and learning mean in situations of incomplete knowledge, unpredictable events, and constant change. But in order for the processes of inquiry, deliberation and deci­ sion making to be successful, pragmatists argue, we should abandon some theses that have a long tradition in Western philosophy. As Rorty has con­ tinually emphasized, the main pragmatist theses are in fact antitheses; i.e., theses in opposition to particular basic philosophical principles that form obstacles to productive problem solving and satisfactory conflict resolu­ tion. The main antitheses are anti-foundationalism, anti-skepticism, and anti-dualism.


Pragmatists reject every form of foundationalism and have embraced falli­ bilism instead. The "quest for certainty, " which philosophy has celebrated since Descartes, ought to be given up permanently because it is illusory. Without exception, all our convictions are provisional and open to repeal or review. The preoccupation with general and abstract truths is often coun­ terproductive because it distracts attention from concrete problems and conflicts, which actually call for tact, flexibility, and context sensitivity.

A Pragmatist Approach to the Governance of Vulnerabil ity


good can be expected from these new technological powers. It is inter­ esting that next to the vulnerability of specific technological systems, the main type of vulnerability that is addressed by the bioconservatives in this debate is the vulnerability of our morality and our way of life. It is pointed out that key moral values may be under threat from these new technological developments; they may be forgotten, or they may change-and change for the worse. The reaction to this perceived vulner­ ability of morality is a firm "no" to new developments, or attempts to stop or control these new technologies. Very little credit is given to the resilience and adaptive capacity of our moral system, our way of life, or even our humanity. Whereas in the utopian scenario our present vulnerabilities are con­ sidered a reason for swift further technological development, without any recognition of possible risks and new vulnerabilities that may arise, in the dystopian scenario the possible risks and vulnerabilities are con­ sidered a reason to ban the feared technological developments altogether. This dichotomous way of discussing the future, based on inflated images of technological possibilities, is not very fruitful in dealing with actual uncer­ tainties and risks. The focus is one-sided, involving either progress or risk, and there is no middle ground where adaptation and learning by doing could be considered as fruitful options. In summary, current methods of exploring the future suffer from an oversimplified-linear, one-dimensional-picture of technological devel­ opment, and they are subject to the dynamics of hype, leading to sterile speculation and black-and-white thinking.

Dramatic Rehearsal

In order to avoid the pitfalls and problems discussed above, we propose another approach to imagining future and emerging technologies and their social and cultural ramifications. Dramatic rehearsal offers an alterna­ tive for speculative ethics and polarized debates about new technologies. Whereas speculative ethics is one-sided, focusing solely on future possibili­ ties, dramatic rehearsal is meant to open up latent possibilities for growth, meaning, and fruitful action in the present. Dewey used the notion of dramatic rehearsal as a way to explain what (moral) deliberation is in individual cases:

Deliberation is actually an imaginative rehearsal of various courses of conduct. We give way, in our mind, to some impulse; we try out, in our mind, some plan. Fol­ lowing its career through various steps, we find ourselves in imagination in the pres­ ence of the consequences that would follow; and as we then like and approve, or

A Pragmatist Approach to the Governance of Vulnerabil ity good can be expected from these new


Jozef Keulartz and Maartje Schermer

dislike and disapprove, these consequences, we find the original impulse good or bad. (Dewey, cited in Caspary, 2000, p. 1 12)

In line with his experimentalism, Dewey also described such dramatic rehearsals as "thought experiments" (Caspary, 2000, p. 1 14) . We propose to use this notion of dramatic rehearsal in a broader sense, not restricting it to individual intrapsychic deliberation but including pub­ lic, shared deliberation. Moreover, we will take it to include deliberations about broad developments, not merely about specific situations of choice. This involves thick descriptions of complex scenarios, of various compet­ ing possible lines of action and courses of conduct. To avoid the pitfalls of speculative ethics, dramatic rehearsal should stay very close on the heels of emerging technologies and be cautious of wandering off into science fic­ tion futures. Moreover, technological determinism should be avoided and ample attention should be given to interaction of social and technological developments (i.e., to the co-creation of symbolic and material worlds). Finally, dramatic rehearsal should incorporate the basic tenets of a resil­ ience approach by focusing not merely on the risks and benefits of new technologies, but also on the adaptive possibilities and resilience of emerg­ ing socio-technological systems. At this point we can take some inspiration from the material hermeneu­ tics of Don Ihde (Ihde, 1993). Material hermeneutics aims at the exploration of possible future worlds that are disclosed and shaped by new technolo­ gies, taking into account the lessons from science, technology, and society (STS) regarding the ways in which technologies co-shape new entities, new vocabularies, and new relationships, as well as new vulnerabilities, new responsibilities, and new forms of resilience. Material hermeneutics helps us to investigate whether our common moral vocabularies and our cur­ rent institutional arrangements are still suitable or have to be revised or replaced in the light of these possible future worlds.

Jozef Keulartz and Maartje Schermer dislike and disapprove, these consequences, we find the original impulse good
Jozef Keulartz and Maartje Schermer dislike and disapprove, these consequences, we find the original impulse good

The Example of In Vitro Fertilization

Jozef Keulartz and Maartje Schermer dislike and disapprove, these consequences, we find the original impulse good

An example of material hermeneutics can be found in Schermer and Keulartz (2002). There, we took the case of the new technology of in vitro fertilization (IVF) to explore, in retrospect, the consequences and ramifications that a new technique can have on society. We focused on the novel character of IVF and showed how this created new entities, prac­ tices, and relationships. Along with these came new vulnerable parties and new risks. In turn, shifts in moral responsibilities and social roles can be recognized; moreover, these developments also urged a differentiation and

A Pragmatist Approach to the Governance of Vulnerabil ity


rethinking of various central moral concepts, such as that of "person" or "parent." One of the great novelties of IVF was that it separated the embryo from the body of its mother, and in doing so, it created a new, distinct entity­ the "early human embryo." The embryo as a new subject raised quite a lot of discussion. What kind of moral status should this new creature have? According to some it was just a "blob of cells, " whereas others considered it the most vulnerable form of human life, worthy of moral concern. The conceptualization of the embryo has a great impact on the way in which technologies involving embryos are regulated. Another effect of IVF technology was the fragmentation or differentia­ tion of parenthood. For example, biological motherhood (which could already be distinguished from social motherhood) was broken into genetic and gestational motherhood. Moreover, IVF has increased the number of parties involved in the creation of a child. Not only the commissioning parents, but also sperm and egg cell donors, physicians, lab technicians, surrogate mothers, sperm and embryo banks, infertility centers, laborato­ ries, and surrogacy agencies can be involved. This creates new responsi­ bilities of the various parties involved in creating, sustaining, raising, or discarding new life. Finally, IVF technology is one of the technologies that contributed to the emergence of a new practice within medicine-a practice that is productive rather than curative-which helps people fulfill desires rather than cure diseases or alleviate symptoms. This new emerging practice creates new roles and responsibilities for doctors and patients (who are transformed into service providers and customers) and requires the development of new guiding rules and moral routines. The point of this example of retrospective material hermeneutics is to show how new technologies can have a profound effect on risks, on social roles and responsibilities, on the central concepts we use to understand and explain the world, and on the moral framework that we use to justify our moral judgments. Technology does not merely change the material and social world, but also our "symbolic order, " including the moral order. Material hermeneutics can be used as a prospective tool to help imagine possible futures, and is an attempt to include the socio-moral-conceptual dynamics into these images.2 As such, it shows the openness and adaptabil­ ity of socio-technological systems and helps identify windows of experi­ mentation and possibilities for learning, for enhancing resilience and for meaningful deliberation and choice.


Jozef Keulartz and Maartje Schermer

of overproduction and overconsumption. Hence the slogan "Limits to Growth," as the famous Club of Rome called its first report from 19 72. In the global south, environmental degradation is framed quite differently­ not as a consequence of too much material wealth, but of too much poverty. Hence there were fierce protests of developing countries at the

first United Nations Conference on the Human Environment the same year

(19 72) against the possible limits on their industrial development and their exploitation of natural resources. The gap in problem and risk perception between industrialized and developing countries also comes to the fore in the relationship between the environmental movement and the Third World movement. While the environmentalists accuse the Third World movement of having a blind spot regarding the environmental damage of economic development, they are in turn accused of giving priority to plants and animals over human beings in the Third World. This gap in problem and risk perception became a stumbling block for international cooperation that in the 1980s came to be considered more and more vitally important for coping with the environmental crisis. Environmental problems like climate change or loss of biodiversity have effects that transcend national boundaries: they cannot be solved by the unilateral decisions of individual states; rather, they require international cooperation. Moreover, these problems are interconnected by multiple feedback loops and cannot be solved in isolation-they require an inte­ grated approach. On the one hand, international cooperation is essential, but on the other, this cooperation is frustrated by a gap in the perception of environmental problems. The notion of "sustainable development" was introduced to close this gap in problem perception. This notion gained popularity after the Brundt­ land report Our Common Future (World Commission on Environment and Development, 1987) linked two political issues-environment and devel­ opment-which were issues that had been handled separately in the past. The notion of sustainable development meets both the industrialized and developing countries halfway. It acknowledges the necessity of transform­ ing the economy, while at the same time recognizing the need for poverty alleviation and social equality. The notion of sustainable development, therefore, is a good example of successful reframing. It brought two com­ peting frames together into a new frame that balanced the various risk assessments and problem perceptions and opened up a new window of cooperation and reorganization of responsibilities.

300 Jozef Keulartz and Maartje Schermer of overproduction and overconsumption. Hence the slogan "Limits to Growth,"

A Pragmatist Approach to the Governance of Vulnerabil ity

G radual ization


Gradualization is a special case of reframingo As we have already seen, prag­ matism is an anti-dualistic movement of thought. Most serious moral con­ flicts are not conflicts between good and evil, black and white, but between things that can be situated somewhere in between these extremes. One method to break up dualisms is gradualization: thinking in terms of degrees instead of boundaries. An example of this method is the debate between animal protectionists and nature conservationists about the moral problems associated with the introduction of large grazing animals in Dutch nature reserves (Klaver, Keulartz, van den Belt, & Gremmen, 2002). These animals are basically domesticated species that are derived from hoofed animals that were once wild, such as cattle, horses, sheep, and goats. These animals are subjected to a process known as "de-domestication" and have to learn to fend for themselves. The management policies of de­ domestication, which entail minimizing supplementary feeding and veteri­ nary assistance, have been most controversial. There is a great deal of debate over the question of whether these animals should be seen and treated as domesticated or as wild. While the majority of animal protectionists view the released horses and cattle as domesticated animals to be cared for as indi­ viduals, most nature conservationists prefer to treat them as wild animals. Here, again, we are confronted with different risk and problem perceptions:

whereas animal protectionists stress the vulnerability of individual animals, nature conservationists are focused on the vulnerability of ecosystems that need grazing herbivores to become self-sustaining. As a result of this discord, people exhaust themselves in unproductive boundary disputes in which both sides claim an exclusive moral jurisdiction over large grazing animals. This deadlock can be overcome by replacing the notion of a clear-cut borderline between nature and culture with the idea of a broad continuum or scale. Then the status of grazing animals introduced into nature reserves is no longer a question of "either-or" but of "less or more." These animals do not simply cross a distinct dividing line between culture and nature; they do not walk from domestication into the wild (that is, from a moral domain of individual care to one of concern for the ecological whole). They gradually move from a thoroughly cultural context to one that is increas­ ingly natural. In this de-domestication process, both animal protectionists and nature conservationists will be indispensable. Thus, the gradualization strategy can help to bridge the rift between these groups and can open new possibilities for communication, cooperation, and the reorganization of responsibilities. Unlike a strategy of control and containment, it enables


Jozef Keulartz and Maartje Schermer

both sides to see possibilities for adaptation and to appreciate the resilience of both the individual animals and the ecosystems.

Common G round Dialogue

The method of gradualization shows some similarity with the so-called Common Ground Dialogue, in which people try to depolarize conflicts by shifting the focus from their disagreements to background agreements and deeper shared values. An example of this strategy concerns the efforts to create a common ground dialogue between pro-life and pro-choice representatives in the abortion debate. These efforts were motivated by dismay at the escalating incivility in clinic protests and public debates, and they also reflected a perceived stalemate in the courts and legislatures. Pro-life and pro-choice representatives met and discovered shared concerns about, for example, the expediency to reduce abortion by preventing unwanted pregnancies­ teaching young people to resist peer pressure for early sexual activity proved to be acceptable to both sides. The common ground dialogue approach assumes that in a polarized conflict, people's views fall on a continuum. When people identify them­ selves on a particular side of an issue such as "pro-choice" or "pro-life," they are only placing themselves somewhere on a continuum other than the exact center:

302 Jozef Keulartz and Maartje Schermer both sides to see possibilities for adaptation and to appreciate

The idea of a continuum encourages awareness of how little we can assume about another person's set of beliefs if all we know about them is that they choose one label over another. It opens us to look for diversity on both sides, and to imagine that two given people with different labels may be as similar as two other people with the same label. It fosters curiosity about the views of the particular individuals on the "other side" encountered within the common ground approach. Gacksteit & Kaufmann, 1 999)

Selecting a point on a continuum for one's position allows for nuances and distinctions in a way that a label alone does not. By assuming a continuum in a polarized conflict, hard issues can be addressed without leading to bad feelings, which is a precondition for reaching jointly agreeable solutions.

Boundary Work

Dewey's inspiration is also at work in STS, mainly under the heading of "boundary work. " This term was launched in 1981 by Steven Woolgar, and further developed by Thomas Gieryn (1983) in the context of the discus­ sion on the demarcation of science and non-science. Gieryn studied how ; wtm� <<�rvP out a domain of co2:nitive authority for their discipline.

A Pragmatist Approach to the Governance of Vulnerabil ity

Boundary Objects


Susan Leigh Star and james Griesemer have shifted the focus from compe­ tition over cognitive claims and cultural capital to cooperation across the lines that separate communities. On the basis of a case study of the historical development of the Museum of Vertebrate Zoology at the University of Cali­ fornia, Star and Griesemer have shown how heterogeneity and cooperation can coexist in the field of science. Scientific work is heterogeneous, requir­ ing many different actors and viewpoints, but at the same time, it requires cooperation "to create common understandings, to ensure reliability across domains and to gather information which retains its integrity across time, space, and local contingencies" (Star & Griesemer, 1 989, p. 387). Star and Griesemer have introduced the notion of "boundary objects" to explain how people handle the tension between the heterogeneous nature of scientific work and its requirement for cooperation in practice. Bound­

ary objects are "both plastic enough to adapt to local needs


yet robust

enough to maintain a common identity across sites. They have different meanings in different social worlds but their structure is common enough to more than one world to make them recognizable" (Star & Griesemer,

1 989 p. 393).

In their case study of the Museum of Vertebrate Zoology, Star and Gri­ esemer distinguish various types of boundary objects, such as repositories, ideal types, coincident boundaries, and standardized forms. Another exam­ ple of a boundary object is the Geographical Information System (GIS) technology. This technology connects and mediates multiple social groups with divergent and even contradictory values (Harvey & Chrisman, 1 998). Yet other examples include prototypes, models, and maps.

A Pragmatist Approach to the Governance of Vulnerabil ity Boundary Objects Susan Leigh Star and james

Trading Zones

Another concept to explain how heterogeneity and cooperation can coexist in the field of science is "trading zone, " a concept introduced by Peter Galison. He has shown how members of different scientific subcul­ tures, such as theoreticians, experimentalists, and instrument makers, are able to cooperate in spite of their differences when they create trading zones. In such an intermediate domain, "the trading partners can hammer out a local coordination, despite vast global differences" (Galison, 1997, p. 783). Galison's notion of trading zones includes not only boundary objects but also interlanguages. Participants from different scientific subcultures communicate using a stripped-down interlanguage, much as traders cre­ ated pidgins and creoles. Pidgins are contact languages usually created spontaneously out of a mixture of various languages; they can evolve into


Andy Stirling typically concern much more fundamental ontological commitments that underlie and crosscut all these kinds

Andy Stirling

typically concern much more fundamental ontological commitments that underlie and crosscut all these kinds of analytical constructs (Ison, Maiteny, & Carr, 1997). Even where there is agreement over specific normative char­ acterizations of human well-being, social equity, or environmental integrity, contrasting perspectives and engagements may each yield radically diver­ gent consequences for the design, implementation, and interpretation of policy interventions aimed at reducing vulnerability (Schon & Rein, 1994). Divergent framings not only exist of " actors, " "technologies," or "vulner­ abilities, " but also of the "systems" and "trajectories" of these assemblages (Leach, Scoones, and Stirling, 2010b). In this sense, the "context" of any given Sustainable response to vulnerability includes not only a multiplicity of "objects, " but also a diversity of observing "subj ects"-and the mutually co-conditioning relations between the two (see figure 15.1). It is against this background of divergent framings of vulnerability and Sustainability that we may best appreciate the imperative to "open up" contrasting implications under different perspectives and contexts (Stir­

ling, 2008a) . Although the parameters of action and temporality, with CONTEXT (includes subjective and objective
2008a) . Although the parameters
of action and temporality, with
CONTEXT (includes subjective and objective aspects)

Figure 1 5.1

Some key concepts: a context of divergently framed vulnerabilities, technologies, and actors for a given socio-technical trajectory


Andy Stirl ing

reflect long-run climate change, investment in new agronomic practices is likely to yield a more sustainable counter to vulnerability than would the provision of enhanced storage in existing food systems. Similarly, coastal defenses that are effective against episodic storm surges will nonetheless be overwhelmed by a progressive secular sea-level rise-against which a more sustainable long-term response might be managed retreat (Adams, 1997). However construed, addressing what is actually a "shock" a s if it were a "stress" (and vice versa) can be a serious mistake. Under any given normative framework, then, Sustainability measures that address vulnerabilities framed in terms of shocks will be relatively conservative in their implications for relevant institutions, practices, and infrastructures (Meadowcroft, 1999). This is because the aim is simply that of sustaining vulnerable qualities through intermittent disruptions. When the temporality of disturbance is interpreted as stress, however, sustainable reductions in vulnerability can (even under the same normative frame­ work) present much more radical political challenges. Instead of achieving improving quality under essentially continuous conditions, the objec­ tive is to achieve this under what are held to be fundamentally changing conditions. An additional key dimension to governance for Sustainability (no matter how it is viewed) concerns the style of action envisaged. In short: is the aim to reduce vulnerability by attempting to control change, or by respond­ ing to it (Stirling, 2007b)? Once more, the complexities of intentionality, agency, causality, and intractability in the real world lie along a spectrum between these differentiating poles. And this continuum is itself typically constituted in very different ways, depending on context and perspective. Again, however, the purpose here is not one of ontological taxonomy but heuristic analysis: to highlight an important distinction that is otherwise frequently elided. Alongside shock and stress, figure 15.2 represents sche­ matically the essential difference between styles of action aiming at control or response. As with the earlier contrast between shock and stress, this dis­ tinction between controlling and responsive action holds important practi­ cal implications for governance strategies aimed at Sustainable reductions in vulnerability. A style of action oriented toward control rests on a conviction that the drivers of the changes in question (whether shocks or stresses) are in princi­ ple (at least to some threshold extent) tractable to deliberate intervention. This requires a number of conditions to be fulfilled. First, there must be confidence in the quality of the understandings of the relevant causal rela­ tionships. For instance, the etiological structures underlying the relevant

From Sustainabil ity to Transformation

3 1 3

vulnerabilities must be seen as determinate and predictable. This resonates strongly with risk-based epistemologies, downpla ying more indeterminate uncertainties, ambiguities, and ignorance (Stirling, 2010). Representations of knowledge thus may typically embody conventional restrictive regu­ latory assumptions (Stirling and Scoones, 2009). A second condition for tractability is that drivers must not only be seen as generally susceptible to action, but also must offer specific moments, modes, and loci for man­ ageable interventions, which are identifiable and achievable with available instruments, time, and resources. Third, there must be confidence that the consequences of such interventions will (to some acceptable extent) be restricted to those that are desired and predicted-with no significant adverse collateral effects. Any qualifications on these conditions for tracta­ bility will undermine claims (or understandings) that shocks or stresses are "subj ect to control" (see figure 1 5 .2). Rather than aiming to control drivers of change, a responsive style of action aims directly to address the emergent consequences. Even under the most bullish of perspectives, this is the residual course of action where the phenomena in question are held to be intractable-perhaps because

style of action

control respond temporal ity of change Key (see also figure 1 5.1): intended effect on driver
temporal ity
of change
(see also figure 1 5.1):
intended effect on
driver of vulnerabil ity
i ntended effect on
vulnerable qual ity

Figure 1 5.2

Contrasting temporalities and actions in addressing technological vulnerabilities


Vulnerabil ity and Strategic Diversity

Andy Stirling

So far, the implications for the governance of vulnerability of sustainabil­ ity and resilience have been discussed in very general terms. The heuristic framework developed here holds practical strategic significance too (Leach, Scoones, and Stirling, 2010a; Stirling, 2007b). Strategies aiming at stability focus on trying to control what are held to be tractable drivers of shocks to any given system. Strategies aiming at durability focus on controlling trac­ table drivers of stress. Strategies aiming at resilience focus on responding to intractable drivers of shock (represented as being external to the trajectory). Strategies aiming at robustness focus on responding to intractable drivers of stress. I shall elaborate on these strategies by discussing a few examples. Recall the case of spikes in the costs of global commodities (like primary fuels, agricultural inputs, and industrial feedstock), which were mentioned earlier as "shocks" that can affect vulnerable groups like subsistence farm­ ers or other economically marginal households. Here, control-oriented governance interventions may aim directly to achieve stability through regulating the relevant market prices, thus directly manipulating the cause of the shock as experienced by those who are vulnerable. In cases where governance entertains greater ambitions to agency, efforts at controlling for stability may take the form of trade sanctions against the responsible producers, thus forcing a lowering of the prices. Either way, such control actions contrast with response-oriented strategies aiming at resilience, which take the prices to be intractable to control and instead focuses on capacities or measures that ameliorate the effects on the vulnerable groups. Resilience-oriented interventions undertaken in anticipation of global price spikes might include the fostering of flexible supply chains (Costello, 2004) or redundant alternative production capabilities or resources and capacities (Farrell, Zerriffi, & Dowlatabadi, 2004) to substitute for the commodities in question. The aim is to absorb the impact of global price spikes affecting particular resources, without actually controlling the market prices them­ selves. If it turns out that the price spikes are intermittent and of short duration, then control-oriented strategies for stability may appear most effective. However, if the disruptions are more frequent and long-lasting, then resilience-oriented strategies will likely be judged preferable. A failure to distinguish whether action is in the style of control or response, there­ fore, may obscure an important practical consideration in the governance of vulnerabilities to shock. Another example mentioned previously was the vulnerabilities of resource-poor farmers to shocks from drought. These might be addressed

318 Andy Stirling style of action control respond temporality of change stress
Andy Stirling
style of action
of change

Figure 1 5.4

Contrasting strategies for stability, durability, resilience, and robustness

strategies, which would have involved greater investment in the structures on these same flood plains from which robustness strategies require with­ drawal. Again, then, there emerge strong practical implications for gover­ nance action in the face of vulnerability. Figure 15.4 takes this discussion a step further by highlighting in a more systematic way further circumstances where the present distinctions between stability, durability, resilience, and robustness may be crucial to strategies for reducing vulnerabilities in technological cultures. With respect to technological infrastructures, for instance, there are potentially important distinctions to be made between the ways in which strategies for reliability may provide for stability, while strategies for flexibility may offer a better basis for resilience (Collingridge, 1 980). Electricity systems provide a concrete example. Here, strategies for increased reliability may aim to stabilize the system by increased maintenance, overengineering transmission and distribution equipment, or overprovision of generating capacity. Resilience, by contrast, may be better achieved through flex­ ibility, by improving staff training, ensuring redundancy between facili­ ties dependent on different fuels, or building dual-firing capabilities that can accommodate multiple fuels as inputs. Yet these infrastructure strate­ gies for both reliability and flexibility may each also contrast with (and

326 Andy Stirl ing transilience eg: Greenpeace targets nuclear poWer after Chemobyl e.g.: Dutch government "Energie
Andy Stirl ing
eg: Greenpeace
targets nuclear
poWer after
e.g.: Dutch government
"Energie Transitie"
e.g,; social movements
"climate camp"
transition tOwns'

Figure 1 5.7

Contrasting framings of target properties in governance action against vulnerability

complements of stability, resilience, durability, and robustness discussed thus far. Instead of describing the different ways in which socio-technical trajectories may be maintained by control or response in the face of shock or stress, these properties describe a range of corresponding ways in which such trajectories are susceptible to reorientation through disruption. For reasons that will be given shortly, I will refer to these properties as "trans­ duction, " "transition, " "transilience, " and "transformation." Of course, this further general dimension in the possible dynamics of socio-technical trajectories is already widely recognized, not least in the social ecology literature. Folke et al. (2005) point out, for instance, that "[a] resilient social-ecological system may make use of crisis as an opportunity to transform into a more desired state" (p. 441). Broad roles for transition and transformation are often pointed to elsewhere in this field (Evans, 2008; Gunderson, 1999; Holling, 2004; Ludwig, 2001; Olsson, Folke, & Hahn, 2004). But, especially when articulating ecological functions and socio-tech­ nical structures, this kind of juxtaposition of "resilience" and "transforma­ tion" can compound the ambiguity inherent in the fundamental conflation of structure and function in this literature reviewed earlier (Smith & Stirling, 2010). If so much is lumped together under the single dynamic property of resilience, it can become unclear exactly what is being maintained and what

F rom Sustainabii ity to Transformation is being disrupted. This provides avoidable opportunities for legitimation strategies

F rom Sustainabii ity to Transformation

is being disrupted. This provides avoidable opportunities for legitimation strategies by powerful interests of the kinds also noted previously. It is to address this kind of potential ambiguity-and the resulting confusions or strategic manipulations in governance-that it may be seen as necessary and helpful to adopt the more nuanced, three-dimensional representation shown in figure 15.7. As with the two-dimensional grids employed thus far, the purpose is not to assert some even more elaborate ontology. The point is that only by envisaging these dynamics of framing in three interlinked dimensions of normativity, temporality, and agency is it possible to encap­ sulate properly the multivalent relationships between different kinds of dynamic properties and associated governance interventions. In these terms, then, the dynamic property referred to in the foreground of figure 15.7 as "transduction" is one where control-style actions seek to disrupt a socio-technical trajectory, sending it in a specific new direction, by exercising control to condition a short-term shock. In this sense, "trans­ duction" is an antonym of "stability, " in that stability describes the pro­ pensity to maintain trajectories in the face of shock, while transduction relates to the role of shock in facilitating controlled disruption away from these same trajectories and toward some particular intended alternative. The term "disruption" itself does not fully describe this because the focus is on the dislocation, rather than the intended controlled ends. But, widely used in fields like genetics, psychology, and signal engineering, the term "transduction" is well suited for this purpose. It has for centuries meant "a removing from one place to another" (Simpson & Weiner, 1989). Here, the specificity of destination underscores the presumptively controlled nature of the defined process. Either way, we see here again the dual face of vul­ nerability, in that pursuit of stability is driven by concerns over vulnerabili­ ties experienced by a trajectory, while strategies for transduction are driven by particular envisaged solutions to vulnerabilities caused by that same tra­ j ectory (Bijker, 2006; Stirling, 2010a). An example of a situation in which control-style actions were aimed at disrupting an existing sociotechnical trajectory in very specific and tar­ geted ways can be seen in the interventions of the UK government against coal power in the 1980s. The overriding purpose then was to exploit the shock of a specific instance of industrial action by miners to affect a more far-reaching transduction in the trajectory of the electricity supply industry as a whole. In particular, a key aim was to achieve a shift to a significantly different (and to the government of the day, less vulnerable) socio­ technical trajectory based more on nuclear power. Evidence that this was so, includes memoirs published later by the then Secretary of State for Energy,


Andy Stirling

admitting that "[t]he need for 1diversificationr of energy sources, the argument I used to justify the [nuclear] programme, was code for freedom from [National

Union of Miners] blackmail" (Lawson, 199, p. 68; italics added). This exam­ ple underscores the point made above: that it is not only marginal perspec­ tives that may favor transduction. As observed by Rahm Emanuel, former chief of staff to US president Barack Obama, "you never want a serious crisis to go to waste" (Seib, 2008). In this case, the UK government was also a very powerful political actor armed with many options for control-style strategies. But it is a reflection of the perennial predicaments of control that even a government should nonetheless find itself, in some important sense, outside the particular incumbent framing (in this case, one associ­ ated with the trajectory of the coal-based electricity supply industry itself), and wishing to see this transducted (disrupted to particular ends) . Likewise, there is a further dynamic property that is of corresponding importance to resilience under vulnerability, but which arises only where outside perspectives are taken into account. This is referred to in figure 15.7 as "transilience"-a word has been employed recently in related lit­ erature (Abernathy et al., 1985), and more generally for hundreds of years (Simpson & Weiner, 1 989) . To be precise, the dynamic property of transil­ ience is a susceptibility of a trajectory to be disrupted by responsive actions in the face of shock. The difference with transduction is that transilience involves responsive actions toward open-ended outcomes in the face of shock, rather than efforts to control the precise orientation of the shocks or their consequences. For consistency, an example here may also be taken from the energy sec­ tor. As the author knows from personal experience, the major international environmental nongovernmental organization (NGO), Greenpeace, main­ tained its campaigning focus against the global civil nuclear industry dur­ ing the 1 970s and 1 980s on the specific target of irradiated fuel reprocessing (and associated waste management). Greenpeace did so because these parts of the nuclear infrastructure were analyzed to be most vulnerable to actions intending to disrupt the nuclear trajectory as a whole. The more specific part of the industry constituted by nuclear power was then held to be less vulnerable to radical political action. However, the Chernobyl accident in April 1986 presented a shock to exactly this part of the industry. For the first time, there arose the opportunity for responsive actions by outside actors to effect disruption to the core trajectory of the contested industry. This particular form of vulnerability on the part of a global energy infra­ structure illustrates a counterpoint to resilience, for which it is interesting that there already exists the (neglected) term "transilience."


Andy Stirling

diffuse and marginalized positions provide less of a basis for responding opportunistically in timely and targeted ways to some particular contin­ gent shock. Civil society organizations like "climate camps" and "transition towns, " therefore, seek more exclusively to create change around discourse concerning long-term, large-scale secular stresses. Examples of this include climate change or "peak oil, " referring to presumed imminent mismatch between continuously rising oil demand and feared leveling off of global supply. The potentially transformative consequences are also more distrib­ uted, diffuse, and diversely oriented than a single transition, opening a more indeterminate array of possible alternative trajectories. It is in these ways that the symmetrical representation of outside as well as inside framings in figure 15.7 highlights the importance of a range of more diverse and nuanced dynamics than are conventionally made explicit in the governance of vulnerability. The context, style, and orientation of governance interventions can all be seen to vary quite radically, depend­ ing on the prevailing circumstances. The conventionally undifferentiated dynamics of sustainability can-under inside framings-be resolved into distinct arenas for governance action in pursuit of stability, durability, resil­ ience, and robustness (as shown in figure 1 5 .3). These each have counter­ parts under outside framings in transduction, transition, transilience, and transformation. These latter conditions for disruption constitute necessary but individually insufficient conditions for the neglected counterpart of sustainability: changeability. If we wish to attend fairly to the full range of implications of vulnerabilities in a technological culture-arising both from and for sociotechnical trajectories (Bijker, 2006; Stirling, 2010a)­ then we need to address both changeability and sustainability when debat­ ing and designing governance interventions. Figure 15.7 thus also illuminates the dynamics of vulnerability in more plural, complex, nuanced, and recursive-but also precise-forms than is often acknowledged. While the present heuristic framework may offer very concrete ways to prompt questions and test hypotheses, its main value may lie in helping to catalyze a generally more open and reflexive appreciation of the governance challenges posed by these multivalent dimensions of vulnerabilities in technological cultures.

330 Andy Stirling diffuse and marginalized positions provide less of a basis for responding opportunistically in
330 Andy Stirling diffuse and marginalized positions provide less of a basis for responding opportunistically in

Conclusions: Toward Reflexive G overnance of Technolog ical Vulnerabil ity



highlighted the





dynamics of vulnerability (associated with shock and stress) and styles











332 Andy Stirling
Andy Stirling

Figure 1 5.8

Toward more progressive and reflexive governance of vulnerability

change, as well as stasis and sustainability. So may the ordering dynamics shaping our technological commitments help nurture more plural ontolo­ gies, returning us to the left side of figure 15.8. It is in this way that the recursive interactions between epistemic, normative, and ontological plu­ ralism may help to maintain more reflexive-and effective-governance of vulnerability (Stirling, 201 1). Rather than depending on cognitive ideo­ logical or procedural qualities (or exhortations) of individual social actors, this institutional notion of reflexivity arises in distributed unstructured political space. By helping to reduce pressures for closure in every domain, this enhanced pluralism may also afford greater political scope for those marginal voices and interests otherwise suppressed by more structured institutions and discourse. It is in these ways that the dynamics of diver­ sity, reflexivity, and more equitable governance of vulnerability may be seen to be profoundly interlinked-and have important practical policy implications.



Allen, E. C., Manuel, ]. C., Legault, C., Naughton, M. ]., Pivor, C., & O'Shea, T.
Allen, E. C., Manuel, ]. C., Legault, C., Naughton, M. ]., Pivor, C., & O'Shea, T. M.
(2004). Perception of child vulnerability among mothers of former premature
1 1

Alvares, C. A. (1980) . Homo faber: Technology and culture in India, China, and the West (rom 1 500 to the present day. The Hague: M. Nijhoff Publishers.

Alvares, C. A. (1992). Science, development, and violence: The revolt against modernity. Delhi: Oxford University Press.

Alwang, ]., & Siegel, P. B. (2001). Vulnerability: A view from different disciplines. Wash­ ington, DC: The World Bank.

Amin, S. (1995). Event, metaphor, memory: Chauri Chaura, 1 922-1992. Berkeley: Uni­

versity of California Press.

Ancona, D., & Chong, C. L. (1996). Entrainment: Pace, cycle, and rhythm in organi­ zational behavior. Research in Organizational Behavior, 18, 25 1-284.

Andersen, P. H., & Drejer, I. (2008). Systemic innovation in a distributed network:

The case of Danish wind turbines, 1 9 72-2007. Strategic Organization, 6(1), 13-46.

Anderson, ]. (2008). Verantwortung. In S. Gosepath, W. Hinsch, & B. Rossler (Eds.),

References Allen, E. C., Manuel, ]. C., Legault, C., Naughton, M. ]., Pivor, C., & O'Shea,

Handbuch der politischen Philosophie und Sozialphilosophie.

Berlin: De Gruyter.

Anspach, R. R. (1993). Deciding who lives: Faithful choices in the intensive-care nursery. Berkeley: University of California Press.

Antonovsky, A. (1987). Unraveling the mystery of health: How people manage stress and stay well (1st ed.). San Francisco: Jossey-Bass.

References Allen, E. C., Manuel, ]. C., Legault, C., Naughton, M. ]., Pivor, C., & O'Shea,
References Allen, E. C., Manuel, ]. C., Legault, C., Naughton, M. ]., Pivor, C., & O'Shea,

Appadurai, A. (1984). How moral is South Asia's economy? A review article. Journal of Asian Studies, 43, 481-497.

Appadurai, A. (2004). The capacity to aspire: Culture and the terms of recognition. In V. Rao & M. Walton (Eds.), Culture and public action (pp. 59-84). Stanford, CA:

Stanford University Press.

Appendini, K. (2001). Land and livehood: What do we know and what are the issues? In A. Zoomers (Ed.), Land and sustainable livelihood in Latin America. Amster­ dam: Royal Tropical Institute/Vervuert Verlag.

Arcuri, A. S. (2005). Governing the risks of ultra-hazardous activities. Challenges (or con­ temporary legal systems. Rotterdam: Academic Publishers.






(1978). Organizational learning. Reading, MA:



Arnold, D. (1984). Famine in peasant consciousness and peasant action: Madras 1 876-1 878. In R. Guha (Ed.), Subaltern studies: Writings on South Asian history (pp. 63-1 15). Delhi: Oxford University Press.


References Bebbington, A. (1999). Capitals and capabilities: A framework for analysing peasant viability, rural livelihoods, and


Bebbington, A. (1999). Capitals and capabilities: A framework for analysing peasant viability, rural livelihoods, and poverty. World Development, 27(12), 2021-2044.

References Bebbington, A. (1999). Capitals and capabilities: A framework for analysing peasant viability, rural livelihoods, and

Beck, U. (1986). Risikogese/lschaft: Auf dem Weg in eine andere Moderne. Frankfurt am Main: Suhrkamp.

Beck, U. (1992). Risk society: Towards a new modernity. Thousand Oaks, CA: Sage. Beck, U. (1999). World risk society. Cambridge: Polity Press. Becker, M. H. (1993). A medical sociologist looks at health promotion. Journal of Health and Social Behavior, 34(1), 1-6.

Beguin, P., Owen, C., & Wackers, G. (2009). Shifting the focus to human work within complex socio-technical systems. In C. Owen, G. Wackers, & P. Beguin (Eds.), Risky work environment: Reappraising human work within fallible systems (pp. 1- 10). Hampshire, UK: Ashgate.

Bekkers, V., & Thaens, M. (2005). Interconnected networks and the governance of risk and trust. Information Polity, 10(1-2), 3 7-48.

Bellaby, P. (2001). Evidence and risk: The sociology of health care grappling with knowledge and uncertainty. In A. Edwards & G. Elwyn (Eds.), Evidence-based patient choice: Inevitable or impossible? (pp. 78-94). Oxford: Oxford University Press.

Benner, P., Tanner, C., & Chesla, C. (1996). Expertise in nursing practice. Caring, clini­ cal judgement, and ethics. New York: Springer Publishing Company.

Bennett, R., Kambhampati, U., Morse, S., & Imael, Y. (2006). Farm-level economic performance of genetically modified cotton in Maharashtra, India. Review of Agricul­

tural Economics, 28(1), 59-7 1.

Berg, M., & Timmermans, S. (2000). Orders and their others: On the constitution of universalities in medical work. Configurations, 8(1), 3 1-61.

Berkes, F., Colding, ]., & Falke, C. (2003). Navigating social-ecological systems: Building resilience for complexity and change. Cambridge: Cambridge University Press.

References Bebbington, A. (1999). Capitals and capabilities: A framework for analysing peasant viability, rural livelihoods, and

Bhatt, M. R. (1998). Can vulnerability be understood? In ]. Twigg & M. R. Bhatt. (Eds.), Understanding vulnerability: South Asian perspectives (pp. 68-77). London:

References Bebbington, A. (1999). Capitals and capabilities: A framework for analysing peasant viability, rural livelihoods, and

Intermediate Technology Publications.


industrial genocide? A unique compilation of dowments from Indian publications.

References Bebbington, A. (1999). Capitals and capabilities: A framework for analysing peasant viability, rural livelihoods, and

(1985). Hong Kong: ARENA Press.

Bijker, W. E. (2006). The vulnerability of technological culture. In H. Nowotny (Ed.), Cultures of technology and the quest for innovation (pp. 52-69). London: Berghahn Books.

Bijker, W. E. (2007). American and Dutch coastal engineering: Differences in risk conception and differences in technological culture. Social Studies of Science, 37(1),



Bijker, W. E. (2010). How is technology made? That is the question! Cambridge Jour­

nal of Economics, 34(1),


Bijker, W. E., & Law, J. (Eds.) (1992). Shaping technology/building society. Cambridge, MA: MIT Press.

Blount, S., Waller, M. L & Leroy, S. (2005). Coping with temporal uncertainty:

When rigid, ambitious deadlines don't make sense. In W. H. Starbuck & M. Farjoun (Eds.), Organization at the limit. Lessons from the Columbia Disaster (pp. 122-1 39). Oxford: Blackwell Publishing.

Bodewitz, H. ]. H. W., Buurma, H., & Vries, G. H. D. (1987). Regulatory science and the social management of trust in medicine. In W. E. Bijker, T. P. Hughes, & T. ]. Pinch (Eds.), The Social Construction of Technological Systems (pp. 243-260). Cam­ bridge, MA: MIT Press.

Boenink, M. (2009). Tensions and opportunities in convergence: Shifting concepts of disease in emerging molecular medicine. NanoEthics, 3(3), 243-255.

Borlaug, N. (2000). The Green Revolution revisited and the road ahead. http://www .nobelprize.org/nobel_prizes/peace/laureates/ 1970/ borlaug-lec ture. pdf.

Borup, M., Brown, N., Konrad, K., & Van Lente, H. (2006). The sociology of expecta­ tions in science and technology. Technology Analysis and Strategic Management, 1 8(3- 4), 285-298.

Bottelberghs, P. H. (2000). Risk analysis and safety policy developments in the Neth­ erlands. Journal of Hazardous Materials, 71 ( 1-3), 59-84.


References Bijker, W. E. (2010). How is technology made? That is the question! Cambridge Jour­ nal

F., Slavin, D., & Li:ifstedt, R. (2007). The tolerability of risk: A

new framework

for risk management. London: Earthscan.

Bovens, M., & Zouridis, S. (2002). From street-level to system-level bureaucracies. How ICT is transforming administrative discretion and constitutional control. Public Administration Review, 62(2), 1 74-184.

Bowonder, B., Kasperson, ]. X., & Kasperson, R. E. (1994). Industrial risk management in India since Bhopal. In S. ]asanoff (Ed.), Learning from disaster: Risk management after Bhopal (pp. 66-90). Philadelphia: University of Pennsylvania Press.

Brinkman, J. W., Bakker, S. ]. L., Gansevoort, R. T., Hillege, H. L., Kema, I. P., Gans, R. 0. B., et al. (2004). Which method for quantifying urinary albumin excretion gives what outcome? A comparison of immunonephelometry with HPLC. Kidney International, 66, S69-S75.

References Bijker, W. E. (2010). How is technology made? That is the question! Cambridge Jour­ nal

Broekhuis, M., & Veldkamp, C. (2007). The usefulness and feasibility of a reflexivity method to improve clinical handover. Journal of Evaluation in Clinical Practice, 13(1), 109-1 15.



Collingridge, D. (1980). The social control of technology. London: Frances Pinter.

Conradson, D. (2003). Doing organisational space: Practices of voluntary welfare in the city. Environment & Planning A, 35(1 1), 1975-1992.

References Collingridge, D. (1980). The social control of technology. London: Frances Pinter. Conradson, D. (2003). Doing

Costello, K. (2004). Increased dependence on natural gas for electric generation:

Meeting the challenge. National Regulatory Research Institute, Ohio State University.

Crabb, C. (2004). Revisiting the Bhopal tragedy. Science, 306(5 702), 1670-1671.

CSA. (2005). The story of Bt cotton in Andhra Pradesh: Erratic processes and results. Centre for Sustainable Agriculture, Hyderabad, India.

Cullen, W. D. (1990). The public inquiry into the Piper Alpha disaster. London: HMSO. Cutter, S. L. (1993). Living with risk: The geography of technological hazards. London:














Cutter, S. L., Mitchell, ]. T., & Scott, M. S. (2000). Revealing the vulnerability of people and places: A case study of Georgetown County, South Carolina. Annals of

the Association of American

Geographers, 90(4), 713-73 7.

Dalrymple, D. G. (1977). Evaluating the impact of international research on wheat and rice production in the developing nations. In T. M. Arndt, D. G. Dal­ rymple, & V. W. Ruttan (Eds.), Resource allocation and productivity in national and international agricultural research (p. xxviii). Minneapolis: University of Minnesota Press.

Dalrymple, D. G. (1985). The development and adoption of high-yielding varieties of wheat and rice in developing countries. American Journal of Agricultural Economics, 67(5), 1067-1073.

Daniel, L. E. (1998). Vulnerability as a key to authenticity. Image: The Journal of Nursing Scholarship, 30(2), 191-192.

Das, V. (2006). Life and words: Violence and the descent into the ordinary. New Delhi:

Oxford University Press.

Dawson, T., Rounsevell, M., Kluvankova-Oravska, T., Chobotova, V., & Stirling, A. (2010). Dynamic properties of complex adaptive ecosystems: Implications for the sustainability of service provision. Biodiversity and Conservation, 1 9(10),


de Haan, L. (2005). Exploring the frontier of livelihood research. Development and Change, 36(1), 27-47.



Dronamraju, K. R. (1985). The life and work of JBS Haldane with special reference to India. Aberdeen, UK: Aberdeen University Press.

Dworkin, R. (1999). Playing god. Prospect Magazine, May.

References 341 Dronamraju, K. R. (1985). The life and work of JBS Haldane with special reference

Dyer, ]. G., & McGuinness, T. M. (1996). Resilience: Analysis of the concept. Archives of Psychiatric Nursing, 1 0(5), 276-282.

References 341 Dronamraju, K. R. (1985). The life and work of JBS Haldane with special reference

van Eijndhoven, ]. (1994). Disaster prevention in Europe. In S. ]asanoff (Ed.), Learn­ ing from disaster: Risk management after Bhopal (pp. 1 13-132). Philadelphia: Univer­ sity of Pennsylvania Press.

Einarsson, S. (1999). Vulnerability and risk analysis of complex industrial system: A new approach and discussion of main problem areas. Trondheim: The Norwegian University of Science and Technology.

Einarsson, S., & Rausand, M. (1998). An approach to vulnerability analysis of com­ plex industrial systems. Risk Analysis, 1 8(5), 535-546.

Eiser, C., Eiser, ]. R., Mayhew, A. G., & Gibson, A. T. (ZOOS). Parenting the premature infant: Balancing vulnerability and quality of life. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 46(1 1), 1169-1 1 7 7.

Ellery, ]. (2007). Integrating salutogenesis into wellness in every stage of life. Prevent­ ing Chronic Disease, 4(3), A79.

Engineer, A. (2003). The Gujarat carnage. New Delhi: Orient Longman.

Epstein, T. S. (1973). South India: Yesterday, today, and tomorrow: Mysore villages revis­ ited. London: Macmillan.

Eriksson, M., & Lindstrom, B. (2005). Validity of Antonovsky's sense of coherence

scale: A systematic review. Journal of Epidemiology and Community Health,



European Environment Agency. (2001). Late lessons from early warnings: The precau­

tionary principle 1896-2000. Copenhagen:

European Environment Agency.

Evans, G. R. (2008). Transformation from "carbon valley" to a "post-carbon society" in a climate change hot spot: The coalfields of the Hunter Valley, New South Wales, Australia. Ecology and Society, 13(1).

Farjoun, M. (2005). Organizational learning and action in the midst of safety drift:

Revisiting the space shuttle program's recent history. In W. H. Starbuck & M. Farjoun (Eds.), Organization at the limit: Lessons from the Columbia Disaster (pp. 60- 80). Oxford: Blackwell.

Farrell, A. E., Zerriffi, H., & Dowlatabadi, H. (2004). Energy infrastructure and secu­ rity. Annual Review ofEnvironment and Resources, 29, 421-469.



Farrington, ]. (2004). Recognizing and addressing risk and vulnerability constraints to pro-poor agricultural growth. Overseas Development Institute.

342 References Farrington, ]. (2004). Recognizing and addressing risk and vulnerability constraints to pro-poor agricultural growth.

Feenberg, A. (2002). Transfonning technology: A critical theory revisited. New York:

Oxford University Press.

Final Declaration of the UN Conference on Environment and Development, Rio de Janeiro (1992).

Fischhoff, B., Lichtenstein, S., Slovic, P., Derby, S. L., & Keeney, R. L. (1981). Accept­ able risk. Cambridge: Cambridge University Press.

Fleischer, N., Weber, A., Gruber, S., Arambula, K., Mascarenhas, M., Frasure, ]., et a!. (2006). Pathways to health: A framework for health-focused research and practice. Emerging Themes in Epidemiology, 3(1), 18.

Folke, C. (2006). Resilience: The emergence of a perspective for social-ecological sys­ tems analyses. Global Environmental Change-Human and Policy Dimensions, 1 6(3),


Folke, C., Carpenter, S., Walker, B., Scheffer, M., Elmqvist, T., Gunderson, L. , e t a!. (2004). Regime shifts, resilience, and biodiversity in ecosystem management. Annual Review ofEcology Evolution and Systematics, 35, 557-581.

Folke, C., Hahn, T., Olsson, P., & Norberg, ]. (2005). Adaptive governance of social­ ecological systems. Annual Review of Environment and Resources, 30, 441-473.

Fombrun, C. ]., & Rindova, V. P. (2000). The road to transparency: Reputation man­ agement at Royal Dutch/Shell. In M. Schultz, M. J. Hatch, & M. H. Larsen (Eds.), The expressive organization (pp. 7 7-96). Oxford: Oxford University Press.

Fortun, K. (2001). Advocacy after Bhopal: Environmentalism, disaster, Chicago: University of Chicago Press.

new global orders.

Foster, H. D. (1993). Resilience theory and system evaluation. In J. A. Wise, V. D. Hopkin, & P. Satger (Eds.), Verif1cation and validation ofcomplex systems: Human fac­ tors issues (pp. 35-60). Berlin: Springer-Verlag.













Foxon, T. ]., Reed, M. S., & Stringer, L. C. (2008). Governing long-term social-eco­ logical change: What can the adaptive management and transition management approaches learn from each other? Sustainability Research Institute, University of Leeds, SRI Paper No. 13.

Freeman, C., & Perez, C. (1988). Structural crises of adjustment, business cycles, and investment behaviour. In G. Dosi, C. Freeman, R. Nelson, G. Silverberg, & L. Soete (Eds.), Technical change and economic theory (pp. 38-66). London: Pinter Publishers.

342 References Farrington, ]. (2004). Recognizing and addressing risk and vulnerability constraints to pro-poor agricultural growth.



Friedmann, H. (1990). The origins of Third World food dependence. In H. Bernstein (Ed.), The food question: Profits versus people (pp. 13-3 1). New York: Monthly Review Press.

Friends of the Earth Malaysia. Malaysia: APPEN.


The Bhopal tragedy-One year after. Penang,

Fro hock, F. M. (1986). Special care: Medical decisions at the beginning of life. Chicago:

University of Chicago Press.

Galanter, M. (1994). The transnational traffic in legal remedies. In S. ]asanoff (Ed.), Learning from disaster: Risk management after Bhopal (pp. 1 33-1 5 7). Philadelphia:

University of Pennsylvania Press.

References Friedmann, H. (1990). The origins of Third World food dependence. In H. Bernstein (Ed.), The
References Friedmann, H. (1990). The origins of Third World food dependence. In H. Bernstein (Ed.), The

Galison, P. (1997). Image and logic: A material culture of microphysics. Chicago: Uni­ versity of Chicago Press.




(2006). Linkages between vulnerability,


and adaptive

capacity. Global Environmental Change, 16(3): 292-303 . Geertz, C. (19 73). The interpretation of cultures: Selected essays. New York: Basic Books. Genus, A., & Coles, A.-M. (2008). Rethinking the multi-level perspective of techno­ logical transitions. Research Policy, 37(9), 1436-1445 . Gherardi, S. (2004). Translating knowledge while mending organisational safety cul­ ture. Risk Management: An International Journal, 6(2), 61-80.

References Friedmann, H. (1990). The origins of Third World food dependence. In H. Bernstein (Ed.), The

Gherardi, S., Nicolini, D., & Odella, F. (1998). What do you mean by safety? Con­ flicting perspectives on accident causation and safety management in a constmc­ tion firm. Journal of Contingencies and Crisis Management, 6, 202-213.

References Friedmann, H. (1990). The origins of Third World food dependence. In H. Bernstein (Ed.), The



(1999). Science's new social contract with society. Natur