Risk and Responsibility in Context

Front matter

EEN ENANA i /

Risk and Responsibility in Context

This volume bridges contemporary philosophical conceptions of risk and responsibility and offers an extensive examination of the topic. It shows that risk and responsibility combine in ways that give rise to new philosophical questions and problems.

PART II Legal Context 4 Risk, Responsibility, and Pre-Trial Detention RA DUFE

5 Risks of Incorrect Use of Probabilities in Court and What to Do about Them ANNE RUTH MACKOR vii

T3 vi

Contents

PART M Bioethical Context

The Failure of Luck Anti-Egalitarianism SVEN OVE HANSSON

Moral Responsibility and Public Health Risks: Examples from the Coronavirus Pandemic JESSICA NIHLEN FAHLQUIST

Responsible Risking, Forethought, and the Case of Human Gene Editing MADELEINE HAYENHJELM

PART IV Technological Context

Emotions, Risk, and Responsibility: Emotions, Values, and Responsible Innovation of Risky Technologies SABINE ROESER AND STEFFEN STEINERT

Responsibility Gaps, Value Alignment, and Meaningful Human Control over Artificial Intelligence SVEN NYHOLM

Radioactive Waste and Responsibility toward Future Generations CÉLINE KERMISCH AND CHRISTOPHE DEPAUS

PART V Environmental Context

Resilience and Responsibilities: Normative Resilience for Responsibility Arrangements NEELKE DOORN AND SAMANTHA COPELAND

Individual Climate Risks at the Bounds of Rationality AVRAM HILLER

Index

List of Contributors

Stearns Broadhead is a post-doctoral researcher at the University of Graz’s Institute of Philosophy, working on the Austrian Science Fund-financed research project Responsibility for Risks: Theory and Practice.

Samantha Copeland is an assistant professor in the Ethics and Philosophy of Technology section of TU Delft’s Department of Values, Innovation and Technology.

Christophe Depaus is a senior expert in safety strategy for the geological disposal of high-level radioactive waste at ONDRAF/NIRAS (Brussels).

Neelke Doorn is a distinguished Antoni van Leeuwenhoek Professor ‘Ethics of Water Engineering’ in the Ethics and Philosophy of Technology section of TU Delft’s Department of Values, Innovation and Technology and Director of Education of the Faculty of Technology, Policy and Management.

RA Duff is an emeritus professor in Philosophy at the University of Stirling.

Jessica Nihlén Fahlquist is an associate professor of Practical Philosophy and a Senior Lecturer in Biomedical Ethics at Uppsala University’s Centre for Research Ethics and Bioethics.

Benjamin Hale is an associate professor in the Philosophy Department and the Environmental Studies Program at the University of Colorado Boulder.

Sven Ove Hansson is a professor in the Department of Philosophy and History at the Royal Institute of Technology (Stockholm).

Madeleine Hayenhjelm is a senior lecturer in Philosophy at Umea University.

Avram Hiller is an associate professor in the Philosophy Department at Portland State University.

viii List of Contributors

Céline Kermisch is a lecturer at the Ecole polytechnique de Bruxelles (Université Libre de Bruxelles) and a senior expert in ethics of science and technology.

Anne Ruth Mackor is a professor of Professional Ethics, in particular legal professions, at the University of Groningen’s Faculty of Law, and she is chair of the North Sea Group, a research network that regularly organizes on- and offline seminars on the theory and methodology of evidence and legal proof.

Sven Nyholm is a professor of the Ethics of Artificial Intelligence at Ludwig Maximilian University of Munich, Germany.

Adriana Placani is an appointed research fellow at NOVA University of Lisbon, Institute of Philosophy (IFILNOVA).

Ibo van de Poel is Antoni van Leeuwenhoek Professor in Ethics and Technology at the School of Technology, Policy and Management at TU Delft.

Sabine Roeser is Antoni van Leeuwenhoek Professor in Ethics and Head of the Department of Values, Technology and Innovation, Faculty of Technology, Policy and Management at TU Delft.

Martin Sand is an assistant professor of Ethics and Philosophy of Technology at TU Delft.

Kenneth Shockley is a professor and Holmes Rolston Professor of Environmental Ethics and Philosophy at Colorado State University.

Steffen Steinert is an assistant professor at the Ethics and Philosophy of Technology section at TU Delft.

Acknowledgments

We thank the contributors to this volume for their efforts and time. We also thank Andrew Weckenmann at Routledge for his guidance and help, as well as three anonymous reviewers for their feedback. Harald Stelzer was crucial to our work, and we greatly appreciate his support.

The editors’ work on this volume was conducted at the University of Graz’s Institute of Philosophy as part of the research project Responsibility for Risks: Theory and Practice and at NOVA University of Lisbon’s Institute of Philosophy (IFILNOVA). The editors gratefully acknowledge that their research was funded by the Austrian Science Fund (FWF): project number P-31527. Adriana Placani’s work was also financed by national funds through FCT - Fundação para a Ciência e a Tecnologia, I.P., under the Scientific Employment Stimulus - Individual Call - CEECIND/02135/2021.

Risk and responsibility are fertile topics of philosophical investigation. Often, but not always, they are considered separately. While responsibility has a long and varied history, risk as a topic of philosophical focus, in ethics at least, is not as longstanding (Erman and Möller 2018, 207; Hayenhjelm and Wolff 2011, E27). This volume examines risk and responsibility as continuous topics. That is, in its broadest formulation, the volume’s contributions consider some of the many ways in which risk and responsibility relate to each other and combine in philosophy. Such combination, as this book shows, is not limited to a single account of risk or responsibility, nor is it applied to just one issue or within a lone context.

The contributions to this volume examine responsibility and risk by addressing issues that arise out of their interplay within various contexts, whether conceptual, legal, bioethical, technological, or environmental. As such, they raise new and challenging issues across a multitude of philosophical areas of investigation and, ultimately, scrutinize the complexities of the modern world through the lens of risk and responsibility. This points to why risk and responsibility merit such special attention: risk and responsibility, often but not always formulated as responsibility for risks, are at the heart of many central problems of the modern age. Moreover, discussions of how risk should be incorporated into moral and political theories, in which responsibility is a central concept but the concept of risk is less often addressed, are of central and growing interest in philosophy. Such academic interest is supplemented by the fact that risk management with its attendant responsibilities has become a topic of increased public concern (e.g., pandemic risks). Thus, a foray into the topic of risk and responsibility, examined in different contexts and applications, has now become crucial for understanding much of our present world and for guiding its future.

DOI: 10.4324/9781003276029-1 This chapter has been made available under a CC BY-NC 4.0 license.

The spectra of topics and themes considered by contributors to this volume represent areas of research that continue to generate intense discussion. Part I problematizes the idea of control within both responsibility and risk conceptualizations. Part II addresses problems related to risk and responsibility that arise within the law in pre-trial detention and in the statistical use of probabilities in courts. Part III tackles considerations related to risk and responsibility in bioethics by examining luck egalitarianism, responsible risking, and public health risks. Part IV considers issues of risk and responsibility across the technological field by examining the role of emotions in the responsible innovation of risky technologies, artificial intelligence (AI), and radioactive waste management. Part V addresses the topic within environmental ethics by examining a host of considerations pertaining to individual climate risks and resilience.

Regardless of the perspective adopted on the topic at hand, exploring the relationships between responsibility and risk requires clear notions of each. This introduction focuses on doing just that, analyzing risk and responsibility separately and allowing their synthesis and application to come out primarily in the volume’s constitutive chapters. This introduction does, however, identify the concepts and topics explored and elaborated in the respective chapters. In this way, the introduction helps to contextualize and explain the concepts of risk and responsibility, and it also helps to make sense of the relationships between them as discussed in the rest of the volume.

The rest of this introduction starts with an examination of the concept of risk, detailing some of its definitions, dimensions, and conceptualizations. This is followed by an exploration of the concept of responsibility, which outlines some of its senses and dimensions. As noted, these sections are not meant to be exhaustive treatments but rather introductions to the topic by way of outlining its constituent parts. Finally, an overview of each of this volume’s contributions highlights the ways that this volume brings together the concepts of risk and responsibility.

1.2 Risk

Philosophical interest in risk has been intensifying. This is understandable, in part, because of the pervasiveness of risks. Consider, for example, that most of our decisions are made under conditions of risk or uncertainty about the possible consequences of our actions or omissions (e.g., what career to choose, whether to cross the street). Moreover, growing interest in the topic seems also attributable, at least partly, to relatively new concerns, such as the risks posed by anthropogenic climate change, new, emerging, and future technologies, as well as the use of preemptive legal measures. This diversity of interests has meant that, similar to responsibility (as we will see), risk has accrued many senses (Bradbury 1989; Hansson 2004; Renn 1992; Shrader-Frechette 1991; Thompson and Dean 1996).

1.2.1 Definitions

There is not one definition of risk; there are many. This is because the concept of risk is used in a variety of disciplines with different specialized meanings but also in everyday life where the meaning of risk tends to be much looser (i.e., non-specialized). Both the specialized and non-specialized understandings of risk are important, and any precedence of one over the other seems to depend on, inter alia, one’s context and aims.! The following section will provide an overview of some definitions of risk as well as detail dimensions of risk that can illuminate the concept.

In everyday life, in non-specialized contexts, risk usually refers to something undesirable that is possible but not certain to occur. For example, a parent might tell their child to wash their hands after playing outside because they might get sick otherwise. The risk in the example is the risk of sickness, which is an undesirable outcome that may or may not occur.

In specialized or technical domains, risk admits of many perspectives depending on the area of investigation (e.g., psychology, economics, engineering, sociology, philosophy).? Categorizing risk conceptions across various disciplines is a challenging task; however, Hansson (2004, 10) provides a list of conceptions of risk that is a useful guide to the more prominent uses of the term:

. Risk as an unwanted event that may or may not occur

. Risk as the cause of an unwanted event that may or may not occur

. Risk as the probability of an unwanted event that may or may not occur

. Risk as the fact that a decision is made under conditions of known probabilities (this is a decision under risk, which is usually contrasted with a decision under uncertainty)

5. Risk as the statistical expectation value of unwanted events that may

or may not occur

BRwWN Pe

Hansson’s example of the risks associated with cigarette smoking can help clarify the differences in the meanings of risk mentioned above. Consider, then, that:

Lung cancer is one of the major risks (1) that affect smokers. Smoking also causes other diseases, and it is by far the most important health risk (2) in industrialized countries. There is evidence that the risk (3) of having one’s life shortened by smoking is as high as 50%.

The total risk (4) from smoking is higher than that from any other cause that has been analyzed by risk analysts. The probabilities of various smoking-related diseases are so well-known that a decision whether or not to smoke can be classified as a decision under risk (5).

Hansson’s list of risk conceptions is not exhaustive. Moreover, there is no consensus over either the definition or conception of risk that is best suited for philosophical investigation. Some philosophers (Buchak 2014; Pritchard 2015; Shrader-Frechette 1991) criticize the specialized definitions as being too narrow, and the non-specialized everyday sense of the term risk is sometimes favored.

Despite differences, all the definitions have in common the fact that they regard risk either to be or involve something undesirable about which there is some lack of knowledge regarding its occurrence. With regard to knowledge, the first two definitions of risk are non-probabilistic, while the last three refer to probabilities. In other words, we can say that the last three, as opposed to the first two definitions, seek to quantify the degree of knowledge an individual has over the possible occurrence of the unwanted event by employing the notion of probability. The question then becomes what probability is.

1.2.2 Probability

Probability interpretations seem to be no less abundant than those of risk. The complexities of understanding probability are so challenging, in fact, as to prompt Bertrand Russell to state: “Probability is the most important concept in modern science, especially as nobody has the slightest notion what it means” (Stevens 1951, 44). Interpretations of probability are varied (e.g., classical, logical/evidential, subjective, frequentist). However, it bears noting that all probability interpretations agree on at least the fact that probabilities are numbers between zero and one that can attach to certain types of propositions and subjected to a probability calculus (e.g., Kolmogorov’s Probability Calculus, Bayes’ Theorem).° If an event cannot occur, then its probability is zero; while if an event is certain to occur, its probability is one.

In line with the current philosophical literature on the topic, the following will focus on the objective and subjective interpretations of probability, which give us the objective and subjective understandings of risk. Briefly stated, the subjective interpretation views risk and probability as, fundamentally, a matter of some kind of belief; the objective interpretation regards them as features of the world that exist, in a relevant sense, independently of human belief.

To the former, the subjective Bayesian theory sees the probability of a proposition as someone’s degree of belief, credence, or confidence in that proposition.* For example, for Jane, the probability that it will rain tomorrow is her degree of belief that it will rain. If her degree of belief is 1/3, then her degree of belief that it will not rain is 2/3; her degree of belief that it will either rain or not rain is 1. We might say, then, that probabilities in the Bayesian theory are numerical measures of particular individuals’ confidence in some proposition(s). These measures can be arrived at in different ways depending on one’s version of the theory (e.g., on the basis of agents’ betting behaviors).°

Some versions of the subjective theory posit that objective probabilities (e.g., frequencies, propensities) do not exist and that the only measure of probability is individualistic. One difficulty with such a view is that probability judgments can vary widely from person to person, especially in the absence of constraints on what ought to count as a rational belief. Consider Gillies’ version of the subjective theory in order to see this:

Probability is [...] defined as the degree of belief of a particular individual, so that we should really not speak of the probability, but rather Ms. A’s probability, Mr. B’s probability or Master C’s probability.

Such an interpretation of probability would give rise to as many estimations of risk as there are beliefs with potentially no way of privileging one over the other (Oberdiek 2017, 22). Such an extreme version of a personalist account can be tempered by constraints of what counts as a rational belief, by coherence demands or by referring to the evidence for one’s beliefs. It should be noted, however, that the latter would still depend on one’s subjective evaluation of the support that the evidence provides to one’s beliefs (Oberdiek 2017, 22). Nevertheless, versions of the Bayesian theory that posit, inter alia, norms requiring that degrees of belief respect the axioms of probability, empirical norms that require an agent’s degrees of belief to be calibrated with her evidence, and logical norms that require degrees of belief underdetermined by evidence to be as equivocal as possible can rein in the pitfalls of subjectivity.

In turn, objective accounts of risks typically rely on frequentist interpretations of probability.® Frequentists view a risk of an event, E, as the frequency with which E occurs in the general population or some other reference class that is selected. The frequency with which the risk manifests in the reference class is taken to be an objective and scientifically verifiable fact. For example, the probability that a man over the age of 60 is diabetic equals the proportion of men over the age of 60 (the reference class) who are diabetic.

In spite of differences between versions of the frequentist theory, probability can be defined in accordance with a frequentist view as relative to a reference class that must be general, infinite, or at least very large (Gillies 2000, 88-112). Thus, probability in this theory is conceived in general terms. Reference is made to general attributes (e.g., diabetes) and general reference classes (e.g., men over 60), which raises the reference class problem and the problem of the single case. The former appears because probability for the frequentist is, basically, the long-run frequency of repeatable experiments. For example, the probability that a fair coin will land tails is 0.5 because were we to flip the coin enough times, we would get tails 50% of the time. However, singular, unique events are not repeatable by definition. To capture the problem of the single case for the frequentist theory, consider Reichenbach’s take on it:

I regard the statement about the probability of the single case, not as having a meaning of its own, but as representing an elliptical mode of speech. In order to acquire meaning, the statement must be translated into a statement about a frequency in a sequence of repeated occurrences. The statement concerning the probability of the single case thus is given a fictitious meaning, constructed by a transfer of meaning from the general to the particular case. (1949, 376-7)

Then, we also have the problem of the reference class. This is associated with the fact that the probability of an event occurring can change depending on how it is classified, and the same event can be classified in a variety of ways on the basis of it belonging in different reference classes.’ The reference class problem appears in Venn (1876, 194), where he writes: “It is obvious that every individual thing or event has an indefinite number of properties or attributes observable in it, and might therefore be considered as belonging to an indefinite number of different classes of things.” There is yet no established solution to the above problems for the frequency theory (Oberdiek 2017, 27).

1.2.3 Risk and Uncertainty

Having outlined some of the main interpretations of probability and some of their problems, we can return to the “lack of knowledge” aspect that was present in all of the definitions of risk stated earlier. To this, it bears to note that, in decision theory, lack of knowledge is categorized into two main types: risk and uncertainty. Thus, a distinction between risk and uncertainty is drawn, and this can be said to follow the probabilistic/non-probabilistic divide mentioned before. Knight draws the distinction in the following way:

To preserve the distinction [...] between the measurable uncertainty and an unmeasurable one we may use the term “risk” to designate the former and the term “uncertainty” for the latter.

In decision-making under risk, possible outcomes and their probabilities are known; while in decision-making under uncertainty, probabilities are either not known at all or known with insufficient precision (Hansson 2004, 11). It is not often that probabilities are known with certainty; however, when data is available, it becomes possible to determine probabilities that can be called objective in the frequentist sense (Möller 2012, 63). Often, frequency data will have to be supplemented or perhaps even supplanted by expert judgment (Ibid, 63). Such expert judgments cannot be construed as objective fact, but it is not merely subjective (in the classical sense) either because, as noted before, subjective probabilities measure a person’s degree of belief that may need to satisfy various norms (e.g., probability axioms, logical norms) but need not correlate with objective frequencies (Ibid). Thus, such expert judgments might be better described as subjective estimates of objective probabilities (Ibid).

Knightian uncertainty refers to cases where we lack probabilities, but it should be noted that some theories seek to measure uncertainty by reducing it to probability.’ For example, for subjectivist theories, probability represents all aspects of a decision-maker’s lack of knowledge (Ibid, 65). Bayesians conceive all rational decisions as admitting of probabilities because for them rational decision-makers always assign a probability value to each potential outcome be it implicitly or explicitly (Ibid). Faced with new information, agents may also change their probability assessments (in accordance with Bayes’ theorem), but they always assign determinable probabilities to all states of affairs (Ibid).’

1.2.4 Disvalue

The above sections sketched out some of the ways in which we might come to understand the lack of knowledge aspect that is present in all of the definitions of risk provided at the outset. There is yet one other aspect that is common to most interpretations of risk, be it in an implicit or explicit manner; namely, risk refers to something negative, unwanted, which is to be avoided. It is actually this feature of risk that most easily connects to responsibility questions in virtue of its normativity. If something should be avoided, then questions about whose responsibility it is to avoid it (prospective responsibility sense) and questions about whose responsibility it is if it is not avoided (retrospective responsibility sense) can arise.

In order to avoid charges of triviality, the negative dimension of risk is usually construed as some kind of harm. This construal has the added advantage of providing a unique currency that can be measured and compared. Nevertheless, this is easier said than done. The concept of harm might be basic, but it is not devoid of controversy when it comes to its conceptualization. Harm can be defined as a setback to interests, but this popular conception is not without problems. For example, some critics have pointed out that defining harm in terms of interests is defining one unclear concept in terms of another unclear concept (Miller 2010, 119). Furthermore, although some cases of harmfulness might be easily compared (e.g., losing two arms is worse than losing one, a severed spine is more serious than a headache), many harms are exceedingly difficult to measure and/ or compare (e.g., how do we measure psychological harms, how many headaches equal a severed spine, how do we conceive of the harm of a species becoming extinct). Even when you restrict harm to just one kind — the harm of death - questions remain, such as: Is death a harm (not for the Epicurian)? Is the harmfulness of an 80-year-old’s death the same as that of 19-year-old’s? Is the harmfulness of the death of a cancer patient who is in severe chronic pain the same as that of a healthy 21-year-old? Measurements that consider both the quantity and the quality of life, such as HALY (health-adjusted life years) with its types, QALY (quality-adjusted life years), and DALY (disability-adjusted life years), are controversial and measuring and comparing the severity of harms remains a contested area.

1.2.5 Multidimensional Conceptions

The two dimensions of risk considered above, lack of knowledge and adverse consequences, can be found in many risk conceptions and are constitutive of the most common definition in risk analysis, which equates risk to the expected value of unwanted events. However, there are other interpretations of risk, which introduce new dimensions. Typically, conceptions of risk in psychology, social science, and moral theory are sensitive to contextual factors as well, and they include different aspects within their risk conceptualizations besides the two that were mentioned, such as who runs the risk, whether the risk is imposed or voluntarily incurred, or whether the risk is natural or human-made (van de Poel and Fahlquist 2012, 881). The following will highlight some of these conceptions, albeit in brief, in order to exemplify the rich diversity of views on the topic.

The psychological literature on risk, which has been developing since at least the 1960s, has shown that lay people include a variety of elements in how they perceive and understand risks (Slovic 2000). Such elements include fear, perceived benefits, time delays, voluntariness, familiarity, controllability, catastrophic potential, and exposure (van de Poel and Fahlquist 2012, 881). Studies have found large differences between so-called real or objective risk and perceived risk in some cases. To exemplify, Lichtenstein et al. (1978) found that bad outcomes that were easier to recall tended to be thought of as more likely to occur. This phenomenon is described as the availability heuristic (Tversky and Kahneman 1974). Risk assessment is also driven by affective states. Lay people tend to exploit the so-called affect heuristic, which refers to the fact that people make judgments based on representations of objects or events that are marked with valenced affect (Slovic et al. 2002). Sometimes the fact that lay people employ different views and estimates of risk than experts is seen as a sign of their irrationality (van de Poel and Fahlquist 2012, 881). However, this interpretation presumes that the technical conception of risk is the right one and that lay people should be educated to comply with it and that, were they to be so educated, they would come to understand risk in line with its objective understanding (van de Poel and Fahlquist 2012, 881). However, many authors have argued that the elements included by lay people in their risk conceptualizations are relevant (e.g., for risk acceptability and management) and provide richer and sometimes more appropriate conceptions of risk than those of the experts (e.g., Roeser 2006, 2007; Slovic 2000).

Other richer conceptions of risk come from the literature on risk ethics. Traditionally, ethical theories have dealt with assessing moral problems in contexts of certainty where actions were assumed to have determined outcomes. This may have been because of the benefits of keeping at least some things relatively free from complicating factors, as well as the assumption of a division of labor between ethics and decision theory. Contending with problems associated with a lack of knowledge was seen as a task belonging to decision theory (Hansson 2003, 291).

Moreover, the major ethical branches, such as deontology, utilitarianism, and contractualism, suffer from particular weakness when it comes to considering risk within their respective frameworks due in part to their theoretical commitments and to the fact these were adopted with the implicit assumption of certainty vis-a-vis outcomes.’ Still, contemporary philosophers have recognized that risks are pervasive, introduce genuine ethical dilemmas, and so they must be dealt with in spite of the many challenges they bring.

The diversity of ethical views on the topic is great, but with regard to assessing the moral acceptability of risks, many agree that this depends on more factors than those allowed by the standard technical definitions (i.e., the probability of harm combined with the severity of harm). Relevant factors include voluntariness, justice considerations, rights-based considerations, risks/benefits distribution, responsibilities (e.g., role responsibility), justifiability, and the availability of alternatives (Asveld and Roeser 2009; Caney 2009; Hansson 2009; Kermisch 2012; Shrader- Frechette 1991; Thomson 1986; van de Poel et al. 2012). Interestingly, Jonathan Wolff (2006) develops a new model of the anatomy of risk, which integrates several other factors, including responsibility-related ones, in the definition of risk besides probability and magnitude of harm. Wolff argues that to provide an adequate account of the factors that must be included in order to decide how to manage particular risks, attention must be given to cause, hazard, probability, fear, blame, and shame. In this way, Wolff explicitly connects factors from public perception and responsibility within the definition of risk itself. With regard to the latter, different human-made risks may be different in acceptability depending on whether they were caused by culpable or non-culpable behavior and on the type of culpable behavior (e.g., malice, recklessness, negligence, or incompetence) that caused them.

In the social sciences, we find many conceptions of risk, but perhaps the most influential has been that of Ulrich Beck (1992). In Risk Society, Beck writes about the many definitional struggles surrounding risks (e.g., over their scope and scale, degree, and urgency), the multitude of definitions themselves, as well as the agglomeration of misunderstandings and antagonisms given these issues (Placani 2017). Beck advances his own understanding by writing that: “Risk may be defined as a systematic way of dealing with hazards and insecurities induced and introduced by modernization itself” (Beck 1992, 21). The definition he provides ties the concept of risk to that of reflexive modernization, which is Beck’s call to confront and reflect upon the uncertainties of the modern age (Placani 2017). Beck sees risk as a social construct, which is historically a recent phenomenon that is closely tied to the idea that risks depend on decisions (Beck 1992, p. 183). In Beck’s risk society, all hazards are seen as depending on human choice and, hence, are, according to his definition, conceived as risks. As a result of this, in contrast to the industrial society, the principal issue in the risk society concerns the allocation of risk rather than that of wealth (van de Poel and Fahlquist 2012, 881-2).

1.2.6 Concluding Risk

The above section illustrated some of the more prominent conceptions of risk, but there are yet others. Whether richer senses of risk or sparser ones should be preferred remains an open question whose answer will likely depend on things such as context, aims, preferences, and theoretical commitments. Perhaps among the many differences in risk conceptions, commonalities were adumbrated as well. Still, the concept of risk remains contested and its dimensions are not yet fully explored. Nevertheless, the above concepts should serve the reader well because all the contributions to this volume, to varying extents, rely, as well as expand, on the understandings of risk discussed. However, before we proceed to discussing the entries to this volume, we need to explicate one more piece from our theoretical puzzle — the concept of responsibility.