Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The ethical implications of a subjective model of risk asssessment in product safety cases MacDonald, Christopher John 1994

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

Item Metadata

Download

Media
831-ubc_1994-0505.pdf [ 4.13MB ]
Metadata
JSON: 831-1.0087682.json
JSON-LD: 831-1.0087682-ld.json
RDF/XML (Pretty): 831-1.0087682-rdf.xml
RDF/JSON: 831-1.0087682-rdf.json
Turtle: 831-1.0087682-turtle.txt
N-Triples: 831-1.0087682-rdf-ntriples.txt
Original Record: 831-1.0087682-source.json
Full Text
831-1.0087682-fulltext.txt
Citation
831-1.0087682.ris

Full Text

THE ETHICAL IMPLICATIONS OF A SUBJECTIVE MODEL OF RISK ASSESSMENT IN PRODUCT SAFETY CASES by CHRISTOPHER JOHN MACDONALD B.A.(Hons.) Trent University, 1992 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n THE FACULTY OF GRADUATE STUDIES (Department of Philosophy) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1994 ©Christopher John MacDonald, 1994 In presenting this thesis in partial fulfilment of the requirennents for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Philosophy The University of British Columbia Vancouver, Canada October 11, 1994 Date DE-6 (2/88) ABSTRACT A number of authors from diverse f i e l d s have c r i t i c i z e d , i n recent years, the epistemic assumption that r i s k can be objectively determined. The i m p o s s i b i l i t y of objec t i v e l y i d e n t i f y i n g and quantifying r i s k s poses obvious d i f f i c u l t i e s for those seeking to make reasonable decisions regarding r i s k . More i n t e r e s t i n g l y from the point of view of moral philosophy, however, i t becomes d i f f i c u l t to determine whether our a c t i v i t i e s subject others to unreasonable l e v e l s of r i s k . Product safety provides an i n t e r e s t i n g context within which to examine the implications of a subjective theory of r i s k assessment. I t i s a f i e l d which involves multiple stakeholder groups with a wide var i e t y of in t e r e s t s and obligations. In t h i s t h e s i s , I f i r s t discuss c r i t i c i s m s which authors from several f i e l d s have l e v e l l e d against the claim that r i s k assessment can be done obj e c t i v e l y . I then examine the implications which a subjective model of r i s k assessment might have for interactions among stakeholders i n product safety cases. I suggest that accepting such a model has a wide var i e t y of implications, including among others d i s c r e d i t i n g models of hypothetical consent, l i m i t i n g the value of c e r t a i n c r i t e r i a suggested by economists for decision making under uncertainty, and suggesting new standards for honesty i n product l a b e l l i n g . I l l TABLE OP CONTENTS Abstract i i Table of Contents i i i Acknowledgment i v Chapter 1: Introduction 1 Chapter 2: The Sub j e c t i v i t y of Risk Assessment 4 Chapter 3: Consumers, Choice, and the Su b j e c t i v i t y 22 of Risk Chapter 4: Risks to Others: Sub j e c t i v i t y and Corporate 45 Resp o n s i b i l i t y Chapter 5: Engineers and 'Safe' Products 57 Chapter 6: Regulators and the Reconciliation of 67 Diverse Objectives Chapter 7: Conclusions 82 Bibliography 85 ACKNOWLEDGEMENT The completion of t h i s thesis would not have been possible without the help and encouragement of several people. F i r s t and foremost, I wish to thank my supervisor, Professor Michael McDonald, for h i s encouragement and enthusiasm for t h i s project from i t s e a r l i e s t stages. His comments on d r a f t versions have been i n s i g h t f u l and h e l p f u l , and h i s guidance i s much appreciated. Thanks also to my reader. Professor Peter Danielson, for suggesting important l i t e r a t u r e and for excellent comments on the f i n a l d r a f t . Thanks are also due to Professors James Gaa and Co l i n Boyd. Professors Gaa and Boyd, both v i s i t i n g scholars at the Centre for Applied Ethics during the 1993-4 academic year, provided important insi g h t s into product safety and business ethics, as well as suggesting excellent b i b l i o g r a p h i c a l information. Thanks also to Shannon Shea f o r help i n proof-reading. Her comments regarding both grammatical and conceptual c l a r i t y were invaluable. During the completion of t h i s t h esis, I received essential funding from the University of B r i t i s h Columbia i n the form of a University Graduate Fellowship, a Teaching Assistantship at U.B.C.'s Department of Philosophy during the Spring of 1994, and a Research Assistantship at U.B.C.'s Centre f o r Applied Ethics during the F a l l of 1993 and the Summer of 1994. I thank a l l three bodies fo r t h e i r generous support. F i n a l l y , thanks to Kimberly Proulx, without whose moral support and encouragement t h i s project might never have been completed. This t h e s i s i s dedicated to my parents, Alex and Linda MacDonald, whose continuing support for my pursuit of an academic career reveals very c l e a r l y the f a i t h they have i n me. Chapter 1: Introduction A number of authors from diverse f i e l d s have c r i t i c i z e d , i n recent years, the epistemic assumption that r i s k can be objectively determined. Conrad Brunk and h i s colleagues have argued that the supposedly s c i e n t i f i c process of r i s k assessment i s i n fact subjective and value-laden, and that i t depends c r u c i a l l y on the evaluator's 'value framework.'^ Mary Douglas and Aaron Wildavsky have argued that organizational culture i s the main factor i n determining which r i s k s we s e l e c t f o r attention.^ And Robyn Dawes has summarized the work of numerous psychological researchers i n d i c a t i n g that r i s k evaluation i s heavily prejudiced by r e a d i l y i d e n t i f i a b l e human cognitive limitations.^ The i m p o s s i b i l i t y of o b j e c t i v e l y i d e n t i f y i n g and quantifying r i s k s poses obvious d i f f i c u l t i e s f o r those seeking to make reasonable decisions regarding r i s k . Indeed, i t poses problems for those persons — and I take t h i s to include nearly a l l of us — who wish to be aware of important r i s k s at a l l . Most obviously, i t becomes d i f f i c u l t , at least t h e o r e t i c a l l y , f o r i n d i v i d u a l s to make good decisions regarding which a c t i v i t i e s to p a r t i c i p a t e i n . More ^ Conrad G. Brunk, Lawrence Haworth, and Brenda Lee, Value Assumptions i n Risk Assessment: A Case Study of the Alachlor Controversv. (Waterloo: Wilfred Laurier UP, 1991). ^ Mary Douglas and Aaron Wildavsky, Risk and Culture: An Essay on the Selection of Technolocfical and Environmental Dangers. (Berkeley: U of C a l i f o r n i a P, 1982). ^ Robyn M. Dawes, Rational Choice i n an Uncertain World. (San Diego: Harcourt Brace Jovanovich, 1988). i n t e r e s t i n g l y from the point of view of moral philosophy, however, i t becomes d i f f i c u l t to determine whether our a c t i v i t i e s subject others to unreasonable l e v e l s of r i s k . Product safety provides an in t e r e s t i n g context within which to examine the implications of a subjective theory of r i s k assessment. I t i s a f i e l d which involves multiple stakeholder groups with a wide v a r i e t y of interests and obligations. Product safety i s of sp e c i a l i n t e r e s t of course within the f i e l d of business and professional ethics, but i t also involves many issues of in t e r e s t within p o l i t i c a l and s o c i a l philosophy i n general. In t h i s t h e s i s , I propose to proceed as follows. In chapter 2, I s h a l l discuss some of the l i t e r a t u r e on the s u b j e c t i v i t y of r i s k evaluation. I w i l l present what I r e f e r to as the Subjectivity of Risk Assessment Hypothesis. In Chapter 3, I s h a l l discuss b r i e f l y the implications which the s u b j e c t i v i t y of r i s k assessment has fo r consumer decision making with regard to product safety. Next, I s h a l l evaluate the impact which t h i s s u b j e c t i v i t y has, or ought to have, on the task of producing and marketing 'safe' products. In Chapters 4 and 5, I w i l l examine t h i s problem as i t applies to corporations and the engineers they employ, respectively. In Chapter 6, I w i l l examine the d i f f i c u l t i e s posed by the s u b j e c t i v i t y of r i s k assessment for the task which product-safety regulatory bureaucracies face i n attempting to balance a wide range of i n t e r e s t s . In Chapter 7, I w i l l summarize the arguments presented i n the previous chapters, and o f f e r some concluding thoughts on general implications of a subjective model of r i s k assessment. I turn now to a discussion of three d i s t i n c t arguments favour of a subjective theory of r i s k assessment. Chapter 2; The Su b j e c t i v i t y of Risk Assessment T r a d i t i o n a l r i s k evaluation assumed that there were two components to r i s k : the p r o b a b i l i t y of a given outcome occurring, and the u t i l i t y of that outcome. This basic model has been expanded to acknowledge the importance of values to the decision-making process. According to one common conception of the r i s k assessment process, for example, evaluation of r i s k has three elements : 1) What i s the l i k e l i h o o d of the negative outcome? 2) How severe i s the negative outcome? 3) What evaluation (importance) i s placed on t h i s outcome?** Step 1) involves determining the p r o b a b i l i t y that a given negative outcome w i l l occur, either as a percentage or as a decimal. Step 2) involves determining the severity or magnitude of that negative outcome. For example, i s what i s being risked s l i g h t discomfort, permanent d i s a b i l i t y , or death? Step 3) involves bringing values into the picture, and asking what importance the decision maker places on the negative outcome. This step i s a recognition of the fa c t that i t i s not enough to have a bare description of the '* This model i s based primarily on the discussion of r i s k assessment found i n Nicholas Rescher's Risk: A Philosophical Introduction to the Theory of Risk Evaluation and Management. (Washington, D.C.: University Press of America, 1983). outcome: we must know i t s value, as well. Under t h i s model, then, the l i k e l i h o o d of danger i s objecti v e l y determined (by whatever s c i e n t i f i c or s t a t i s t i c a l means), the magnitude of that danger i s obj e c t i v e l y determined, and then a subjective evaluation of the r i s k i s made based on the decision-maker's values. A decision for action i s based on t h i s subjective evaluation. I t i s important to note that steps one and two have t r a d i t i o n a l l y been seen as objective, and as appropriate subjects of discussion f o r science. William Lowrance, for example, writes that "[m]easuring r i s k — measuring the p r o b a b i l i t y and severity of harm — i s an empirical, s c i e n t i f i c a c t i v i t y . . . . Recent c r i t i c i s m s of t h i s model c a l l into question the presumption that steps one and two can be completed i n some objective, value-neutral way. Conrad G. Brunk, Lawrence Haworth and Brenda Lee, i n t h e i r study of the controversy surrounding de-r e g i s t r a t i o n of the chemical herbicide alachlor for use i n Canada, found that the degree of r i s k a ttributed to alachlor by d i f f e r e n t groups and in d i v i d u a l s depended c r u c i a l l y on the 'value framework' of the person or persons involved. Brunk, Haworth and Lee suggest that, where s p e c i f i c p r o b a b i l i t i e s are uncertain, the values held by individual estimators of r i s k can be the deciding factor i n the r i s k estimate ^ William Lowrance, Of Acceptable Risk: Science and the Determination of Safety. (Los Altos, CA: William Kaufman Inc, 1976) 75. I t should be noted that Lowrance, among others, deserves c r e d i t f o r going to great pains to emphasize the f a c t that, contrary to what some believe, at le a s t some stage of r i s k evaluation i s necessarily subjective. a r r i v e d at. They note, for example, that [o]ne who has a bias i n favour of taking r i s k s , a bias reinforced by the pro-technology and l i b e r a l assumptions, needs a compelling reason for not taking r i s k s , and the compelling reason can only consist i n objec t i v e l y r e l i a b l e data demonstrating that the r i s k i s so serious as to be unacceptable.* One of the main conclusions reached by these authors i s that the i d e a l of objective r i s k assessment may be one which cannot be met: I f the promise of a purely s c i e n t i f i c r i s k assessment i s that i t can replace completely the imputed emotionalism and ' s u b j e c t i v i t y ' that underlies public perception of r i s k by something more substantial and r e l i a b l e , i t simply i s unable to d e l i v e r on that promise.^ Brunk et al. suggest that there i s "a need to go beyond the evidence."* The need which these authors mention — that of going 'beyond the evidence' — r e f e r s to the fa c t that those involved i n estimating product safety l i t e r a l l y must a r r i v e at such an estimate, even i n the absence of what seems l i k e s u f f i c i e n t evidence. The alternatives presented by uncertainty are either to give up (an option often unavailable i n r e a l - l i f e s i t u a t i o n s ) , or to make a judgement which may turn out to under- or over-estimate ri s k . ^ These authors note, however, that decision-makers doing regulatory science, (and, I would add, s c i e n t i s t s and engineers employed by producers i n product safety research), do not have "the * Brunk et a l . , 144. ^ Brunk et al., 105. * Brunk et al., 75. ^ Brunk et al., 75. luxury sometimes enjoyed by laboratory s c i e n t i s t s of responding to the uncertainties by refusing to reach a decision...."^" The former must reach a decision regarding product safety, and thus are "charged with r e s p o n s i b i l i t y for givi n g an answer."" Nor do consumers, f o r that matter, have luxury of refusing to make decisions u n t i l perfect information i s ava i l a b l e . Those involved i n estimating product safety must therefore go beyond the evidence, and base decisions at least p a r t l y on values. The choice of a strategy for dealing with uncertainty i s a normative one. These authors suggest that "[w]hat i s at stake...in a l l r i s k debates, are c o n f l i c t s among fundamental s o c i a l and moral values, which cannot be resolved by s c i e n t i f i c inquiry alone. "^ ^ Values, they suggest, ought to be faced, since they cannot be avoided: "[W]e ought not be misled into thinking that our in t e r p r e t a t i o n of the data w i l l not r e f l e c t a value framework, or, even worse, that i t can substitute f o r such a framework. "^ ^ They suggest that to make t r a d i t i o n a l r i s k assessment, with a l l i t s t r a d i t i o n a l assumptions and i t s p a r t i c u l a r view of r a t i o n a l i t y , the sole a r b i t e r i n r i s k debates, i s " . . . i n e f f e c t to allow one value framework. . .to s e t t l e the issue i n i t s favour."^'' A second c r i t i q u e of the idea of objective r i s k - a n a l y s i s — Brunk et a l . , 89. " Brunk et a l . , 89. 12 Brunk et al., 110. 1^  Brunk et a l . , 152. " Brunk et a l . , 7. one with a very d i f f e r e n t approach, but with conclusions quite s i m i l a r to those reached by Brunk et al., — i s the c u l t u r a l analysis approach taken by Mary Douglas and Aaron Wildavsky i n t h e i r book. Risk and Culture. Douglas and Wildavsky suggest that the culture we f i n d ourselves i n i s a c r i t i c a l determinant of the r i s k s we see as important. They suggest that t h i s i s p a r t i c u l a r l y true of organizations and the cultures engendered by organizational structure. Douglas and Wildavsky present an admittedly s i m p l i f i e d trichotomy of organizational structures (market-individualistic, hierarchical/bureaucratic, and sectarian) , and i l l u s t r a t e the sorts of r i s k s which each type tends to see as important, and those to which each tends to be b l i n d . They suggest that the group choice regarding which r i s k s to see as most worrisome "... i s never made d i r e c t l y but i s s e t t l e d by a preference among kinds of favored s o c i a l i n s t i t u t i o n s . "1^ Attitudes toward c e r t a i n r i s k s w i l l be defining features of any organization. For i n d i v i d u a l s within an organization or group, "...the most relevant of improbable bad outcomes have been anticipated.,. i n a p r i o r s o c i a l commitment. "^ ^ That i s , c e r t a i n s o c i a l commitments r e s u l t i n g from group a f f i l i a t i o n predispose indivi d u a l s to see s p e c i f i c sorts r i s k s as p a r t i c u l a r l y threatening. For Douglas and Wildavsky, "public perception of r i s k and i t s acceptable l e v e l s are c o l l e c t i v e 1^  Douglas and Wildavsky, 187. Douglas and Wildavsky, 187. constructs... The example which serves as the focus f o r Douglas and Wildavsky's investigation i s the b a t t l e fought among corporations, regulators, and environmental groups over environmental p o l l u t i o n . According to these authors, each of these three groups tends on average to have a s p e c i f i c culture. The culture of each group embodies s o c i a l commitments and core b e l i e f s which determine the group's views on environmental r i s k s . According to Douglas and Wildavsky, corporations are e s s e n t i a l l y 'market-individualist' organizations (though of course parts of the corporation w i l l be hierarchies) . Such being the case, a corporation's deepest fear i s of "any threat to the exchange system."^* On questions of environmental regulation then, corporations are l i k e l y to see f a r greater danger i n what they see as excessive government interference than they do i n r e l a t i v e l y small quantities of t o x i c pollutants. The atmosphere i n which such organizations operate, these authors suggest, does not encourage them to plan f a r a h e a d . A s a r e s u l t , corporations are bound to pay more attention to short-term economic threats than to uncertain r i s k s lodged i n the distant future. The r i s k s to which decision-makers within corporations pay heed, then, w i l l often be those which involve threats to the values of market individualism. Next, l e t us look at regulators. Regulators are t y p i c a l l y " Douglas and Wildavsky, 186. Douglas and Wildavsky, 96. Douglas and Wildavsky, 96. bureaucrats, members of regulatory hierarchies. According to Douglas and Wildavsky, organizations which are h i e r a r c h i c a l are that way as a solu t i o n to the problems of voluntary organization.^° Douglas and Wildavsky note that hierarchy has been with us throughout h i s t o r y as a r e s u l t of the f a c t that "... i t i s a f e a s i b l e way f o r a large number of people to collaborate."^^ The primary fear that a l l individuals within hierarchies share, therefore, i s that t h e i r hierarchy may c o l l a p s e , a n d that the benefits to be gained by e f f i c i e n t group e f f o r t may thereby be l o s t . Thus t h e i r most immediate concern i s with r i s k s to the status quo. Hierarchies, according to these authors, t y p i c a l l y place "...the maintenance of the whole system above ind i v i d u a l s u r v i v a l . . . . "^ ^ The structure and purpose which hierarchies share determines which s p e c i f i c r i s k s are foreseen by them, according to Douglas and Wildavsky: Unless things are v i s i b l y bad, standard operating procedures may be used f o r most matters. The se l e c t i o n of r i s k s worth taking and avoiding i s made by a process, not by a person. Problems are solved i n sequence. The need that seems most urgent i n these conditions i s the one whose solution i s r e a l i s t i c a l l y f e a s i b l e . ^ Risks, then, w i l l be selected on the basis of convenience, and the Douglas and Wildavsky, 103. Douglas and Wildavsky, 95. Douglas and Wildavsky, 95. Douglas and Wildavsky, 101. Douglas and Wildavsky, 93. 20 21 22 23 24 problems chosen for attention w i l l most often be those ..to which the le a s t complicated protocol applies.... "^ ^ Risks which seem outside the capacities which the hierarchy has f o r action w i l l tend to be ignored. This suggests that i n order to capture the attention of h i e r a r c h i c a l bureaucracy, environmental r i s k s must be both obvious and subject to cl e a r remedies — two q u a l i t i e s which environmental r i s k s seldom possess. C a p i t a l i s t s and bureaucrats may be blinkered to environmental r i s k s , but Douglas and Wildavsky argue that environmental r i s k s are exactly the sort of issue to which sects (such as leaderless d i r e c t - a c t i o n environmental groups) are a t t u n e d . T h e y argue that sects r e j e c t hierarchy and other s t r i c t forms of organization, and that instead they must keep t h e i r membership cohesive through appeal to such broad values as "...man's natural goodness, his corruption by big h i e r a r c h i c a l organizations, and h i s redemption through a return to the natural, undifferentiated order of things. Douglas and Wildavsky argue that sects are characterized most s i g n i f i c a n t l y by the ease with which the i n d i v i d u a l may withdraw from the group. Such being the case, i n order to maintain i t s 2* Douglas and Wildavsky, 97. 2* Douglas and Wildavsky discuss r e l i g i o u s sects — the Old Order Amish and the Hutterites — i n order to i l l u s t r a t e t h e i r use of the term 'sect,' and i n order to draw i n t e r e s t i n g p a r a l l e l s with the d i r e c t - a c t i o n environmental groups such as Friends of the Earth, to which they also apply that term. See Douglas and Wildavsky, 104-114. Douglas and Wildavsky, 169. membership, a sect must o f f e r a c o l l e c t i v e s p i r i t u a l good, which w i l l be enhanced to the degree that the sect disvalues the world outside.^* Douglas and Wildavsky argue that sects "...choose to be panic struck about dangers from technology...." because t h i s serves t h e i r moral purposes by "focusing on the dangers emanating from large organizations...."^' In order to maintain t h e i r cohesiveness, sects focus on the doom which the impure outside world i s sure to bring. For r e l i g i o u s sects, the r i s k i s of the wrath of God. For environmental sects, i t i s of chemical or nuclear armageddon. The c r i t i q u e of objective r i s k assessment, here, i s c l e a r l y quite compatible with that offered by Brunk, Haworth and Lee. Like the l a t t e r authors, Douglas and Wildavsky suggest that the assessment of r i s k offered by a given assessor w i l l depend c r u c i a l l y upon her p r i o r b e l i e f s and commitments. The main difference l i e s i n the fac t that f o r the most part Brunk et al. f i n d c e r t a i n attitudes inhering i n persons playing c e r t a i n roles as a matter of empirical fact, and do not t r y to formulate an overarching theory r e l a t i n g value frameworks to organizational culture. Douglas and Wildavsky, on the other hand, focus upon the t h e o r e t i c a l reasons f o r the s p e c i f i c value commitments which come as a r e s u l t of group membership. Given t h i s difference, the analyses provided by these two groups of authors complement each other n i c e l y . Brunk et a l . Douglas and Wildavsky, 121. Douglas and Wildavsky, 169. provide Douglas and Wildavsky with empirical observations v a l i d a t i n g at lea s t some of the l a t t e r ' s t h e o r e t i c a l conclusions. And Douglas and Wildavsky provide a t h e o r e t i c a l framework which helps p a r t i a l l y to explain the d i f f e r i n g value frameworks which Brunk et a i . observed among d i f f e r e n t p a r t i e s to the alachlor debate. For example, the attitudes which Brunk et a l . observed being displayed by Monsanto, the manufacturer of alachlor, conform n i c e l y to the c h a r a c t e r i s t i c s which Douglas and Wildavsky t e l l us to expect of 'market i n d i v i d u a l i s t ' groups: Monsanto's main emphasis seemed to be on the threat to f a i r competition and national competitiveness,^° or what Douglas and Wildavsky c a l l "threat[s] to the exchange system. "^ ^ Importantly f o r my purposes here, i t should be noted that Brunk et a l . have i n common with Douglas and Wildavsky that they down-play the extent to which perception of r i s k s depends on the existence of objecti v e l y i d e n t i f i a b l e circumstances and p r o b a b i l i t i e s . The t h i r d c r i t i q u e of the b e l i e f i n objective r i s k assessment to be discussed here i s very d i f f e r e n t from the f i r s t two. I t points not to the t h e o r e t i c a l i m p o s s i b i l i t y but to the p r a c t i c a l unlikelihood of objective r i s k assessment, and i s to be found i n the psychology l i t e r a t u r e . In h i s book. Rational Choice i n an Uncertain Worlds Robyn M. Dawes summarizes the work of numerous psychological researchers, suggesting that r i s k evaluation i s Brunk et al., 17. 1^ Douglas and Wildavsky, 96. heavily prejudiced by r e a d i l y i d e n t i f i a b l e human cognitive l i m i t a t i o n s and dis p o s i t i o n s . Dawes defines a r a t i o n a l choice as being one which meets these three c r i t e r i a : 1. I t i s based on the decision maker's current assets ( f i n a n c i a l , p h y s i o l o g i c a l , psychological, s o c i a l , emotional). 2. I t i s based on the possible consequences of the choice. 3. When these consequences are uncertain, t h e i r l i k e l i h o o d i s evaluated without v i o l a t i n g the basic rules of p r o b a b i l i t y theory. C l e a r l y t h i s model of r a t i o n a l choice i s not uncontroversial. Other writers define r a t i o n a l i t y by means of other c r i t e r i a . Neither i s Dawes' characterization peculiar, however, and a de t a i l e d c r i t i q u e i s well outside the scope of t h i s work. My goal here i s not to dissect Dawes' view, but to discuss i t as a respectable one which has strong negative implications for an objective model of r i s k assessment. Dawes d e t a i l s a wide v a r i e t y of habitual f a l l a c i e s and modes of thought which prejudice our judgments and make us run afoul of the above-stated c r i t e r i a f or r a t i o n a l i t y i n a wide var i e t y of cases. Of the various types of i r r a t i o n a l i t y which Dawes discusses, I w i l l o utline here three which seem to me to be perhaps the most c l e a r l y relevant to the task of product safety evaluation. One type of error discussed by Dawes i s the 'anchoring' of p r o b a b i l i t y estimates on i r r e l e v a n t information. Dawes writes: "In ambiguous si t u a t i o n s , a seemingly more t r i v i a l f actor may have a profound e f f e c t : an 'anchor' that serves as an orien t i n g point f o r estimating p r o b a b i l i t i e s or outcomes."^^ That i s , people's estimates of a v a r i e t y of unknown numbers — p r o b a b i l i t i e s , for instance — tend to be influenced by (that i s , they tend to be close to) whatever i r r e l e v a n t numbers happen to be at hand. One example which Dawes c i t e s involves a study i n which subjects were asked to give a rough estimate of the magnitude of "8!" ("8 f a c t o r i a l " , or 8 x 7 x 6 x 5 x 4 x 3 x 2 x 1 ) . The researchers found that the order i n which factor numbers were presented affected dramatically the estimate at which the research subjects arrived. Those who were asked to estimate the value of 8 X 7 x 6 x 5 x 4 x 3 x 2 x l gave a median answer of 2,250, while those asked to estimate the value of 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8 (the same set of numbers, but i n reverse order) gave a median answer of 512.'^ In other words, the f i r s t number i n the series acted as an anchor. The implications f o r estimating r i s k i n sit u a t i o n s of great uncertainty seem cl e a r . Where p r o b a b i l i t i e s are to be estimated i n situations of uncertainty, these estimates may tend to be influenced by the order of presentation of those numbers which are known. Another sort of error c i t e d by Dawes involves what he c a l l s 'framing e f f e c t s . ' That i s , decision makers tend to give quite d i f f e r e n t , and even contradictory, decisions depending on how a question i s worded or 'framed.' In general, f o r example, "...decision makers are risk-averse when questions are framed i n 2^ Dawes, 121. Cited i n Dawes, pp. 121-122. terms of saving l i v e s and risk-seeking when the i d e n t i c a l questions are framed i n terms of lo s i n g l i v e s . D a w e s c i t e s a large body of l i t e r a t u r e i n support both of t h i s claim s p e c i f i c a l l y , and of the strength of framing e f f e c t s i n general. Dawes r e j e c t s the idea that framing i s j u s t 'verbal t r i c k e r y . ' He c i t e s a 1985 study i n which subjects were asked a series of questions known to be susceptible to framing e f f e c t s . Individual subjects d i d indeed tend to give inconsistent answers to i d e n t i c a l questions framed i n two d i f f e r e n t ways. Not only that, but when these inconsistencies were pointed out to the subjects, who then had the opportunity to change t h e i r responses i n order to achieve consistency, "...almost a l l of them recognized that they had i n fa c t been inconsistent....," but they only chose to change t h e i r responses about h a l f of the time.^^ Dawes concludes that "[wjhereas verbal t r i c k s lose t h e i r effectiveness once they are understood, framing e f f e c t s are stable."^* The implications for product safety are clear, given that these r i s k s must be communicated from those who have knowledge to those who do not. Paul S l o v i c and Baruch Fischhoff note that "...people attempting to communicate information about r i s k s have considerable a b i l i t y to manipulate p e r c e p t i o n s without making any overt ^ Dawes, 35. Dawes, 37. Dawes, 37. Dawes does not d e t a i l the type of choices presented to the experimental subjects, except to say that they involved a choice between r i s k s i t u a t i o n s . misrepresentations. "^ ^ A t h i r d common mode of reasoning c r i t i c i s e d by Dawes as sub-optimal i s the making of global i n t u i t i v e judgments (also 'expert' or ' c l i n i c a l ' judgments). Such judgments are e s s e n t i a l l y evaluations of a s i t u a t i o n or of the status of a subject based on an o v e r a l l impression, often based on expertise. Examples of t h i s sort of evaluation include a physician's global r a t i n g of the ' o v e r a l l s e v e r i t y ' of the condition of a patient, and a university admissions o f f i c e r ' s o v e r a l l impression of a candidate based on an unstructured interview. One a l t e r n a t i v e to t h i s sort of evaluation i s to use "... s t a t i s t i c a l l y derived weighted averages of the relevant predictors."^^ This l a t t e r procedure e s s e n t i a l l y b o i l s down to adding up the pluses and the minuses i n order to estimate the o v e r a l l q u a l i t y of an option or i n d i v i d u a l . The c r i t i c a l point i s that, as Dawes points out, there i s a considerable body of research which indicates that where external standards of accuracy are a v a i l a b l e , "...weighting schemes consistently are found to be superior to i n t u i t i v e global judgment, and judges consistently are found to be overconfident i n t h e i r global judgments."'^ (Emphasis o r i g i n a l . ) Dawes suggests that the r e l a t i v e i n e f f i c i e n c y of global judgments i s p a r t i c u l a r l y alarming i n l i g h t of the modern reliance upon expert witnesses to give assessments "... of the form ' I t i s my Paul S l o v i c and Baruch Fischhoff. "How Safe i s Safe Enough?" i n Jack Dowie and Paul Lefrere, eds.. Risk and Chance: Selected Readings (Milton Keynes: The Open UP, 1980) 134. Dawes, 2 05. Dawes, 203. opinion that '"'*° The implications which the endemic nature of these (and other) forms of i r r a t i o n a l i t y d e t ailed by Dawes holds f o r attempts at objective r i s k assessment are cl e a r . We have a tendency not to make decisions based s o l e l y on the three c r i t e r i a of r a t i o n a l i t y described by Dawes. The numerous psychological studies which Dawes c i t e s indicate that a large number of factors which determine our choices are both i r r a t i o n a l and unknown to us at the time the decision i s made. This f a c t , coupled with psychological i n d i v i d u a l i t y , means that two d i f f e r e n t decision-makers are l i a b l e to a r r i v e at r a d i c a l l y d i f f e r e n t 'objective' decisions based on exactly the same data. Cl e a r l y , there may be times at which i t i s r a t i o n a l (or at lea s t desirable) to be less than f u l l y r a t i o n a l . But the evidence c i t e d by Dawes suggests that we are seldom f u l l y r a t i o n a l i n the choices we make, even when we honestly believe that we are and want to be. I t should be pointed out that unlike the analyses presented by the f i r s t two sets of authors discussed, Dawes' analysis may leave open the t h e o r e t i c a l p o s s i b i l i t y of objective r i s k assessment. The l i t e r a t u r e c i t e d by Dawes appears merely to point to the p r a c t i c a l unlikelihood of objective r i s k assessment. The bare fa c t that humans tend to estimate r i s k s poorly does not imply that they cannot i n p r i n c i p l e estimate them c o r r e c t l y . But the f u l l implications which t h i s evidence has f o r the p o s s i b i l i t y of Dawes, 208. objective r i s k assessment may hinge upon the r e s u l t of an ongoing debate over whether the experimental data point to a basic lack of capacity or to mere poor performance. L. Jonathan Cohen, for example, argues that the types of reasoning i l l u s t r a t e d above involve temporary cognitive i l l u s i o n s quite explainable i n terms of the u n f a m i l i a r i t y of the t e s t questions and the lack of special t r a i n i n g on the part of the experimental subjects.^^ Persi Diaconis and David Freedman, however, point out that many such ' i l l u s i o n s ' seem to be "deeply rooted i n the human mind," and that they often "cannot be d i s p e l l e d by a 'few moments prompted r e f l e c t i o n , ' or several months of college teaching. "'^^ I f , as Diaconis and Freedman and many others believe, such cognitive i l l u s i o n s are a fundamental part of our psycho-cognitive make-up, then each of us i s l i k e l y unique i n the extent to which we display each such i l l u s i o n . I f t h i s i s the case, then our judgments of r i s k w i l l be coloured by i n d i v i d u a l i z e d sets of i l l u s i o n s , and may f a i r l y be termed subjective. I w i l l take i t as evident, then, based on the three works discussed above, that there i s considerable evidence suggesting that r i s k assessment i s , i n one way or another, subjective. For ease of future reference, I s h a l l designate the general thesis that r i s k assessment i s inherently subjective as 'SRAH:' The "1 L. Jonathan Cohen, "Can Human I r r a t i o n a l i t y be Experimentally Demonstrated?" The Behavioral and Brain Sciences 4 (1981): 324. P e r s i Diaconis and David Freedman, "The Persistence of Cognitive I l l u s i o n s , " The Behavioral and Brain Sciences 4 (1981): 333. Subjectivity of Risk Assessment Hypothesis. I t should be noted that i n suggesting that r i s k assessment i s a subjective process, the authors discussed above do not r e f e r to the sort of subjective p r o b a b i l i t i e s u t i l i z e d i n Bayesian theory. Under Bayesian theory, "the p r o b a b i l i t y values [for various states] are subjective p r o b a b i l i t i e s i n that they derive t o t a l l y from the decision maker's personal information about the states"**^ (emphasis o r i g i n a l ) . I t may be possible, of course, to see imperfect consumer information as producing subjective p r o b a b i l i t i e s i n t h i s sense. Then, under the Bayesian theory of learning, "...new information, obtained from experience or by other means, w i l l adjust the 'subjective' p r o b a b i l i t y distribution."'*^ But product safety cases, or at l e a s t the most d i f f i c u l t ones, seldom lend themselves to learning by experience. Indeed, that i s p r e c i s e l y why they are problematic. In many product safety cases, i n d i v i d u a l learning, i n so f a r as learning requires error, i s f a t a l . Any comment beyond t h i s l i e s well beyond the scope of my project: my main point here i s to point out that subjective r i s k assessment does not imply subjective p r o b a b i l i t i e s i n the Bayesian sense. I now turn to examine the implications which a subjective model of r i s k assessment might have for decision-making among Peter Gardenfors and N i l s - E r i c Sahlin, "Introduction: Bayesian Decision Theory — Foundations and ProlDlems," i n Peter Gardenfors and N i l s - E r i c Sahlin, Decision. P r o b a b i l i t y . and U t i l i t y : Selected Readings. (Cambridge: Cambridge UP, 1987) 4. Gardenfors and Sahlin, 4. consumers. Chapter 3; Consumers. Choice, and the Sub j e c t i v i t y of Risk What impact might a subjective theory of r i s k assessment have for the options open to consumers? I t has generally been assumed that consumers who purchase products accept the r i s k s which go along with the possession and use of those products to the extent that they understand — or can be expected to understand — the r i s k s involved. These r i s k s are not accepted for t h e i r own sake: they are among the costs traded for the benefits of consumer goods. The most i n t e r e s t i n g questions to be asked i n l i g h t of SRAH (the Su b j e c t i v i t y of Risk Assessment Hypothesis) involve: 1) the best basis f o r decision making concerning such trade-offs; 2) what consumers can reasonably demand of both producers and regulators i n terms of help i n making such decisions; and 3) the epistemic r e s p o n s i b i l i t i e s which consumers themselves might be considered to have. In the f i r s t sections of t h i s Chapter, I discuss the implications which a subjective model of r i s k assessment has for these three questions. I then discuss the extent to which the role played by consumer associations and public i n t e r e s t groups might be affected by such a model. What impact might a subjective model of r i s k assessment have upon the bases available to consumers for making sound decisions concerning product r i s k s ? As Harvey M. Sapolsky notes, product safety presents unique challenges to consumers: Personal experience i s usually s u f f i c i e n t to allow consumers to cope with the hazards of the marketplace. [...] But product health r i s k s are a d i f f e r e n t sort of market hazard. Few consumers f e e l comfortable with the measure of parts per m i l l i o n , l e t alone parts per b i l l i o n . As Elaine Vaughan notes, "The growth and complexity of technological systems has resulted i n [proportionally] fewer i n d i v i d u a l s having the necessary t r a i n i n g to understand on a tec h n i c a l l e v e l the possible consequences of new technologies."'** Indeed, as Sapolsky reminds us, consumers play a r e l a t i v e l y minor r o l e i n even identifying r i s k s from consumer products: " I t i s not the consumer who i d e n t i f i e s the existence of r i s k , but rather the news media, public i n t e r e s t groups, businesses, and s c i e n t i f i c organizations. ... "•*' Forced to r e l y on others f o r product safety information, the thinking consumer w i l l eventually see cause to question even the experts. As William Lowrance suggests, "[t]he question becomes, how do we know the experts are 'right'T"*** Given that the primary impact of r i s k assessment i n product safety i s on the consuming public, i t i s su r p r i s i n g how l i t t l e attention the epistemic needs of consumers have attracted i n the r i s k assessment l i t e r a t u r e . There i s , of course, a body of l i t e r a t u r e d i s s e c t i n g and attempting to explain the apparently '*^  Harvey M. Sapolsky, "The P o l i t i c s of Product Controversies," i n Harvey M. Sapolsky, ed.. Consuming Fears: the P o l i t i c s of Product Risks. (New York: Basic Books, 1986) 182. "** Elaine Vaughan, Some Factors Influencing the Nonexpert's Perception and Evaluation of Environmental Risks. (New York: Garland, 1990) 3. "*' Sapolsky, 5. •** Lowrance, 106. i r r a t i o n a l r i s k perceptions of non-experts. A va r i e t y of explanations are advanced, primarily with an eye to solving the p o l i t i c a l problems which other groups face as a r e s u l t of such perceptions. For example, a n t i c i p a t i n g p u b lic response to new r i s k s i s a highly valued goal from the point of view of regulatory bodies. As Elaine Vaughan notes, ". . .unanticipated public reaction to regulatory decisions has been known to complicate the implementation of a decision and sometimes necessitate a delay or modification of the intended course of action."'*' Other writers have focused on the problem which organizations face i n attempting to convey c r u c i a l r i s k information i n order to allow members of the p u b l i c to make informed decisions.^" But the confusion facing consumers has seldom been discussed, i t seems, as a problem for consumers themselves. The i n i t i a l question was what trade-offs are reasonable between the r i s k s and benefits of consumer products. The question presented by SRAH: what ought the prudent consumer to do when offered a product which of f e r s known benefits but unknown risks? Per hypothesis, a l l products s t r i c t l y speaking f a l l into t h i s category, since no expert can a u t h o r i t a t i v e l y t e l l consumers what the r i s k s of a given product are. There seem to be three main strategies or groups of strategies, by no means mutually exclusive, which might offe r '^^  Vaughan, 6. °^ See, f o r example, F. Reed Johnson, Ann Fisher, V. Kerry Smith, and William H. Desvousges, "Informed Choice or Regulated Risk?" Environment. 30.4 (1988): 12-15 and 30-35. guidance to consumers seeking to make such d i f f i c u l t choices. The f i r s t such group i s the well-known group of general strategies developed by decision t h e o r i s t s for decision making under uncertainty. The second option involves strategies developed by regulators f o r dealing s p e c i f i c a l l y with unknown r i s k s . The t h i r d strategy involves a recognition of the necessity of what might be c a l l e d an e x i s t e n t i a l choice. That i s , sometimes values must be enough to inform one's decision, even i n the absence of what might seem to be adequate information. F i r s t , l e t us look at general strategies for decision making under uncertainty. Luce and R a i f f a d i s t i n g u i s h decision making under uncertainty from both decision making under r i s k and decision making within the context of g a m e s . B r i e f l y , a decision made under uncertainty i s one i n which the decision-maker does not know the r e l a t i v e p r o b a b i l i t i e s of various s a l i e n t e v e n t u a l i t i e s . Of course, decisions of whether or not to t r u s t a given product might i n f a c t be u s e f u l l y analyzed as a problem for game theory (in order to act, the consumer must guess, for example, what action the producer has or has not taken) , but the main implication of SRAH i s that decisions about whether or not to t r u s t products are to a large degree decisions made under uncertainty. Luce and Ra i f f a characterize decisions under uncertainty thus: 1^ R. Duncan Luce and Howard R a i f f a , "Individual Decision Making Under Uncertainty," i n Decision. P r o b a b i l i t y and U t i l i t y ; Selected Readincfs. Peter Gardenfors and N i l s - E r i c Sahlin eds. Cambridge: Cambridge UP, 1988. Reprinted from Luce and Raiffa's Games and Decisions. John Wiley & Sons, Inc., 1957. 48. A choice must be made among a set of acts A^ , Aj, . . ., A„, but the r e l a t i v e d e s i r a b i l i t y of each act depends upon which 'state of nature' p r e v a i l s , either s^, S2, •••/ s„. [...] As the decision maker, we are aware that one of several possible things i s true; which one i t i s i s relevant to our choice, but we do not even know the r e l a t i v e p r o b a b i l i t i e s of t h e i r t r u t h . . . l e t alone which one obtains. For the consumer deciding whether or not to t r u s t a given product, the relevant set of acts includes buying and not buying the p r o d u c t . T h e relevant possible states of nature include at least one i n which the product produces f o r her a negative balance of u t i l i t i e s and one i n which the product produces a net benefit. If the r e l a t i v e p r o b a b i l i t i e s of these two states of nature were known — that i s , i f the consumer a c t u a l l y knew the precise r i s k involved — she could structure the problem "...as one of decision making under r i s k — as a choice among lotteries"** and choose that act which had the maximum expected value. But per hypothesis, the consumer cannot know the r e l a t i v e p r o b a b i l i t i e s of either of these states coming to be. What c r i t e r i a does decision theory o f f e r to the thoughtful consumer, then? The following p r i n c i p l e s are offered by Luce and Ra i f f a as possible p r i n c i p l e s f o r decisions under uncertainty.^^ Luce and R a i f f a , 48-49. Even i f the product i s purchased, there may be a subsequent question regarding how much to t r u s t i t . This well may influence the way i n which the product i s used. ^ Luce and R a i f f a , 50. A l l of the following discussion of strategies f o r decision-making under uncertainty i s summarized from Luce and R a i f f a , 51-59. Only s p e c i f i c statements of opinion on the part of those authors 1) The maximin c r i t e r i o n : choose that act which maximizes the minimum payoff. For the consumer, t h i s means choosing that action which involves the lea s t d i r e worst-case-scenario. Presumably, t h i s c r i t e r i a would most often d i c t a t e doing without non-essential consumer goods. As Luce and R a i f f a note, t h i s extremely conservative c r i t e r i o n has been c r i t i c i z e d for precluding the p o s s i b i l i t y of r i s k i n g a marginally lower payoff i n the hopes of achieving a much higher one.*^ 2) The minimax r i s k c r i t e r i o n : choose that act which minimizes the maximum r i s k index for each act. That i s , choose that act which allows f o r the smallest quantity of possible regret fo r missed opportunities. In terms of our consumer, t h i s c r i t e r i o n suggests that i f not purchasing a given product involves possibly missing out on r e a l l y t e r r i f i c benefits but only assures the avoidance of f a i r l y minor detriments, then the product ought to be purchased. Luce and R a i f f a note that there are some rather complex objections to t h i s c r i t e r i a , having to do generally with the l i k e l i h o o d that u t i l i t i e s are non-linear and with the fac t that, where more than two choices are possible, some decision problems under uncertainty involve i n t r a n s i t i v i t i e s which hinder optimization. 3) The pessimism-optimism index c r i t e r i o n : use a fixe d index i n d i c a t i v e of one's own degree of pessimism (some number between 0 and 1, say) and the inverse of that index to weight the value one w i l l be footnoted further. Luce and R a i f f a , 52-53. a t t r i b u t e s to maximum and minimum pay-offs respectively. From the point of view of SRAH, t h i s c r i t e r i o n has the advantage of r e l a t i v i z i n g the choice to the consumer's own values. But from the point of view of SRAH, there are d i f f i c u l t i e s with t h i s c r i t e r i o n ' s suggested way of determining one's own pessimism-optimism index (the number between 0 and 1) . The c r i t e r i o n suggests that one look at how one evaluates other simpler decision problems under uncertainty to determine what index number r e s u l t s i n indifference between options. I t suggests that i f one tends to weight maximum pay-offs o p t i m i s t i c a l l y i n one case, then one ought to weight maximum payoffs equally o p t i m i s t i c a l l y i n other cases. But one of the main implications of SRAH i s that r i s k preferences are not uniform across various r i s k s i t u a t i o n s . As a r e s u l t , i t looks as i f one's preferences i n one decision s i t u a t i o n are a poor reference point f o r evaluating other, d i f f e r e n t , choices. 4) C r i t e r i o n based on the ' p r i n c i p l e of i n s u f f i c i e n t reason': i n the absence of any good reason to assume otherwise, assume a l l p o s s i b i l i t i e s to be equally probable, and sel e c t that option with the highest expected value calculated on the basis of that assumption. Luce and R a i f f a note a number of c r i t i c i s m s of t h i s procedure. From the point of view of the consumer, the most s a l i e n t point i s that while she i s indeed operating under uncertainty, she i s l i k e l y not operating under complete uncertainty. She w i l l most l i k e l y know, f o r instance, that harm (that i s , the state of nature which implies that she suffers harm) from a product holds a p r o b a b i l i t y considerably lower than 1/2. Equiprobability w i l l often turn out to be a poor assumption. Let us now look at the solutions offered by the d i s c i p l i n e of r i s k assessment for decision under uncertainty. The following c r i t e r i a are derived from the l i t e r a t u r e on r i s k assessment, and while they were o r i g i n a l l y intended as p r i n c i p l e s for use by regulators, they may also serve i n d i v i d u a l consumers to some extent. 1) The Delaney P r i n c i p l e : accept no r i s k f o r which there i s any s c i e n t i f i c evidence. This p r i n c i p l e , part of an amendment to the U.S. Food and Drug Act and named for congressman James J. Delaney, states that no food additive should be considered safe i f there i s any evidence at a l l of carcinogenicity i n either humans or animals. As a possible generalized p r i n c i p l e f o r consumer decision making, t h i s p r i n c i p l e suggests that any product which o f f e r s any r i s k should be avoided. This p r i n c i p l e might be made more reasonable — or at least more p r a c t i c a l — by modifying i t to require that any product about which there i s ' s u f f i c i e n t ' expert suspicion be avoided. The p r i n c i p l e might also be t a i l o r e d to a given consumer's personal fears by mentioning s p e c i f i c types of i n j u r i e s or s p e c i f i c diseases to be avoided. Some people place p a r t i c u l a r disvalue on neurological damage, for example, and thus might follow a r u l e of s t r i c t l y avoiding any substance implicated i n neurological syndromes. 2) Custom of usage: i n the hands of r i s k assessors, t h i s p r i n c i p l e has generally meant the designation of c e r t a i n substances Lowrance, 82. as 'generally recognized as safe' ('GRAS').^^ I t i s generally regarded as a designation which reduces the number of products about which one need a c t i v e l y worry. For i n d i v i d u a l consumers, the r u l e might be to accept any product which i s a t r a d i t i o n a l part of one's l i f e s t y l e , and which 'common knowledge' says i s safe. Of course, t h i s c r i t e r i o n only makes sense i n the absence of strong s c i e n t i f i c evidence of danger. I t i s c l e a r from the two sets of strategies j u s t reviewed — one from decision theory and one from r i s k assessment — that no a v a i l a b l e strategy can t e l l the consumer i n unequivocal terms what choices to make. Yet decisions must s t i l l be made. As Slovic and Fischhoff note, "[t]he option of dropping out [of the marketplace] gives us some control over the l e v e l of technological r i s k to which we are exposed." However, these authors continue, "reduction of r i s k t y p i c a l l y e n t a i l s reduction of benefit."^' In the modern world, dropping out i s simply not an option — most of us cannot do without the benefits offered by consumer products. So when decisions cannot be made according to the r a t i o n a l c r i t e r i a of decision theory and r i s k assessment, what i s l e f t ? I suggest that when f a c t s are unavailable or unhelpful, decisions must be made based s o l e l y on values. Let us imagine that we wish to make a personal evaluation of the r i s k s involved i n a given product. What sort of information do we desire of producers and regulators? The most obvious sort of Lowrance, 80. S l o v i c and Fischhoff, 122. information with which an expert might provide us would be a s t a t i s t i c a l measurement of the p r o b a b i l i t y of harm. But SRAH implies that such measurements are impossible to make objectively. They depend to too large a degree on c h a r a c t e r i s t i c s of the i n d i v i d u a l or group making the estimate. We do not commonly know the value assumptions which inform a given r i s k estimate — per hypothesis the estimator herself may be unaware of them. Perhaps, then, the experts could provide a sort of analogy, c i t i n g another r i s k to which the current r i s k can be compared? The r i s k currently under examination, the experts might t e l l us, i s approximately as large as the r i s k associated with well-understood product or a c t i v i t y 'x. ' This, i n fa c t , i s not an uncommon strategy among experts t r y i n g to convey a r i s k estimate to the public.*" But i f we mistrust p r o b a b i l i t y estimates to begin with, then comparing two such estimates only increases the margins of error. *^  We have already noted the degree to which consumers must depend on experts f o r information regarding product safety. We *° William Lowrance, i n Of Acceptable Risk^ 73, quotes the U.S. Atomic Energy Commission WASH-1400 (August 1974) as describing the r i s k of accident at nuclear plants thus: "For accidents causing 1000 or more f a t a l i t i e s the [chance of an accident at one of 100 nuclear plants] i s 1 i n 1,000,000 or once i n a m i l l i o n years. Interestingly, t h i s i s just the p r o b a b i l i t y that a meteor would s t r i k e a U.S. population centre and cause 1000 f a t a l i t i e s . " *^  I f each estimate i s attributed a one order of magnitude margin of error (such margins of error are not uncommon i n r i s k estimates), then the comparison of the two estimates c a r r i e s a margin of error of two orders of magnitude. One of the two p r o b a b i l i t i e s being given as roughly equal may ac t u a l l y be 100 times as large as the other. have also noted that the guidance which such experts can give i s often severely l i m i t e d . I submit that when t h i s i s the situation, the agent i s l e f t with making a sort of e x i s t e n t i a l choice. Stuart Hampshire makes a s i m i l a r point with regard to moral choices. Hampshire suggests that the agent must simply choose, and that she w i l l l i k e l y make the choice based on which course of action i s most nearly i n accord with the sort of l i f e she wishes to l i v e . Hampshire acknowledges that, where possible, choices should be based on reason. But he also suggests that there i s a " . . . r a d i c a l lack of s u f f i c i e n t reason i n cases where...a c o n f l i c t between d i f f e r e n t ways of l i f e arises and where there i s no way of achieving a reasonable and coherent compromise between them."*^ What form w i l l such an e x i s t e n t i a l choice take i n the realm of product safety? I suggest that t h i s i s i n f a c t the proper de s c r i p t i o n f o r a va r i e t y of choices already made every day — indeed, taken for granted — concerning a vast array of products. Consumer A does not know the exact r i s k s i m p l i c i t i n the choice of a convertible over a hard-top sedan, but she knows that a convertible s u i t s her self-image. Consumer B knows that the evidence l i n k i n g animal f a t s to a range of cancers i s inconclusive, but he also knows that the pleasures which he might derive from the consumption of such products do not rank highly on h i s l i s t of personal p r i o r i t i e s . Consumer C has doubts about the r e a l dangers inherent i n various food dyes, but she also knows that her own *2 Stuart Hampshire, "Public and Private Morality." i n Stuart Hampshire, ed. Public and Private Morality. (Cambridge: Cambridge UP, 1978) 45. image of h e r s e l f as a caring mother i s incompatible with allowing her c h i l d r e n unrestricted access to products containing high levels of such dubious substances. J e f f r e y Harris, a widely respected public health expert, suggests that a si m i l a r choice i s made d a i l y by people who ignore the absolute proscriptions that constitute most public warnings about the r i s k of A.I.D.S. Harris writes, . . . a l l over America, night a f t e r night, ordinary, w e l l -meaning heterosexual men and women are v i o l a t i n g the sharp-line prescriptions. ... These men and women aren't misinformed. They're not out of the loop. They hear the message. They understand the message. They ignore the message.*^ What Harris i s suggesting i s that choices are ine v i t a b l e , and that many c r u c i a l choices are made on bases other than r a t i o n a l concern for one's own health. I suggest that i t ought to be more widely acknowledged that product safety choices can reasonably be grounded i n a wide variety of values. Perhaps consumers should a c t i v e l y acknowledge the p o s s i b i l i t y that someone who eats high-fat foods with abandon might be neither ignorant of relevant s c i e n t i f i c data nor i r r a t i o n a l l y reckless, but may instead be expressing a divergent system of values. Likewise, the consumer who, despite hearing expert testimony to the e f f e c t that nuclear power plants only pose threats comparable to those posed by widely accepted chemical plants, opposes construction of new reactors, may not be su f f e r i n g from a lack of understanding of p r o b a b i l i t y theory, but may instead be Je f f r e y E. Harris, Deadly Choices; Coping with Health Risks i n Everyday L i f e . (New York: Basic Books, 1993) 18. making an i m p l i c i t statement about her core b e l i e f s concerning the r i s k s of new technologies. Lest we should think that, rather than make poor choices, we ought to devolve even greater decision-making r e s p o n s i b i l i t y to various groups of experts, Harvey Sapolsky reminds us that the act of making such choices i s i t s e l f greatly valued. "Choice i s what we cherish, the opportunity to se l e c t our companions, our p o l i t i c i a n s , our causes, our entertainment, our products."^ Sapolsky suggests that the qu a l i t y of such choices may not be as important as the making of them. He writes: "That we p e r s i s t i n ignoring safe pleasures or i n reaching f o r dangerous ones i s to be expected, f o r we cherish the i n s t i t u t i o n s that always permit and, at times, even encourage such errors.""* Next, I turn to examine b r i e f l y the question of what consumers can reasonably demand of both producers and regulators i n terms of help i n making product safety decisions. The ideas presented here w i l l receive further attention i n Chapters 4 and 6, regarding corporations and regulators respectively. F i r s t , l e t us examine what consumers ought to expect of producer corporations. As w i l l be discussed i n d e t a i l i n Chapter 4, I suggest that consumers might reasonably come to expect producers to make clea r any value assumptions inherent i n the producer's a t t r i b u t i o n of safety to the product. B r i e f l y stated, I suggest that i t i s not unreasonable to expect that t r u l y adequate ^ Sapolsky, 2 01. Sapolsky, 201. product l a b e l l i n g might include references to broad value systems contributing to the corporation's claim of safety. One way t h i s might be done would be to l i s t on labels the governmental agencies (such as the National Research Council) who have c e r t i f i e d the product as 'reasonably safe,' and the consumer organizations, i f any, who have been consulted as part of the safety-determination process.** To the extent that consumers know or can f i n d out the values of said organizations, they w i l l also know some of the values which one must hold i n order to see a given product as safe enough. Another way value information might be transmitted i s by including on labels c l e a r l y worded statements of the broad assumptions that went into the corporation's own determination that the product's r i s k s are reasonable. As I discuss i n more d e t a i l i n Chapter 4, i t i s by no means cle a r that such a system i s p r a c t i c a b l e . But I suggest that the idea has s u f f i c i e n t t h e o r e t i c a l value to merit further discussion and perhaps empirical t e s t i n g . Next, l e t us examine what consumers ought to expect of government regulatory bodies. I w i l l b r i e f l y mention two things that consumers ought to expect of regulators (these two involve points discussed i n d e t a i l i n Chapter 6) as well as discussing one ^ One example of such approval i s the Good Housekeeping Seal of Approval programme. See Lowrance, 132. Whether more widespread consumer-group p a r t i c i p a t i o n i n the safety-determination process i s p r a c t i c a l or desirable i s a large enough topi c to merit separate consideration elsewhere. Not a l l groups which sound l i k e consumer groups are t r u l y at arms length from industry. Any but the most vague sort of endorsement might also carry implications for l i a b i l i t y . sort of information which consumers ought not to expect from regulators. F i r s t , I suggest that consumers ought to expect regulators not to use 'regulatory science' to d e p o l i t i c i z e product safety decisions. Deciding on appropriate l e v e l s of r i s k i s a matter of choosing among values, and as such i s an inherently p o l i t i c a l task. For regulatory bodies to allow s c i e n t i s t s and engineers to determine whether or not a given r i s k i s reasonable i s to shirk r e s p o n s i b i l i t y f o r normative decision-making. Consumers should, i n l i g h t of SRAH, demand that value assumptions be made openly and i n a way that i s open to scrutiny and debate, and that such decisions be made by those elected or appointed to make value decisions. S l o v i c and Fischhoff write: "Analysts and regulators are paid out of p u b l i c funds. They should make t h e i r analyses comprehensible, s o l i c i t p u b l i c input, and r e f l e c t public desires i n t h e i r conclusions. ""' Second, consumers ought to demand that regulators not place undue emphasis on models of 'hypothetical consent,' and ought to demand that such models only be used with a great deal of caution. Granted, consumers w i l l generally not make demands i n terms of the use of t h e o r e t i c a l models. The point i s , rather, that consumers need to make clea r to regulators that t h e i r choices of r i s k s are defined by values i n a manner which, while occasionally prima facie inconsistent, i s i n d i c a t i v e of r e a l values. That i s , consumers should reasonably expect of regulators that the l a t t e r not assume Slo v i c and Fischhoff, 143. that consumer consent to a given l e v e l of r i s k i n one s i t u a t i o n necessarily implies consent to a s i m i l a r l e v e l of r i s k i n a d i f f e r e n t s i t u a t i o n . F i n a l l y , consumers ought not to expect more detailed descriptions of r i s k than are possible. Public o f f i c i a l s , i n t h e i r attempt to transmit accurate information to the consuming public, are hampered by at l e a s t two c r u c i a l constraints. The f i r s t i s the by now f a m i l i a r sense i n which there are very r e a l l i m i t a t i o n s on what can o b j e c t i v e l y be said about r i s k . I f there i s no o b j e c t i v e l y determinable f a c t of the matter about safety, then the information which government regulators can transmit to consumers i s somewhat li m i t e d . I f consumers demand more than science i s capable of providing, they w i l l be encouraging regulators to make the sort of o v e r l y - s p e c i f i c statement which t y p i c a l l y hides s i g n i f i c a n t value assumptions. The second l i m i t a t i o n on the transmission of good r i s k information to consumers l i e s i n the main means of communication ava i l a b l e . As J e f f r e y E. Harris notes, the l i m i t a t i o n s on the information transmissible i n b r i e f p u b lic announcements mean that short messages are l i k e l y to be the r u l e . Such l i m i t a t i o n s on the quantity of information mean that public safety messages are l i k e l y to cast safety issues i n rather stark, black and white terms. An excellent recent example i s the statement made by a U.S. safety o f f i c i a l regarding a type of s k i r t believed to be dangerously flammable. Ann Brown, chair of the U.S. Consumer Product Safety Commission, stated c a t e g o r i c a l l y that "Consumers must stop wearing these s k i r t s , " despite the f a c t that there had been no reported deaths or i n j u r i e s . No Author, "'Highly Flammable' s k i r t s believed to have been sold i n Vancouver," Harris suggests that i f a public safety message spoke not of absolutes but of degrees of safety, "[t]here would be an i m p l i c i t message that some degree of r i s k i s acceptable, but the message i t s e l f wouldn't give a formula for evaluating the risk."*' Public health o f f i c i a l s must of necessity work i n broad brush-strokes. Consumers ought not to expect public o f f i c i a l s to provide more d e t a i l s about r i s k reduction than i s possible given l i m i t a t i o n s on the t a i l o r i n g of r i s k advice to personal needs v i a public-safety communications. However, as Harris suggests, the lack of t a i l o r i n g does not mean that such messages are without value: We cannot follow a l l the rules and recommendations exactly, because we each have our sp e c i a l circumstances. The medical messages have not been i n d i v i d u a l l y t a i l o r -made for us, but that doesn't make them useless.'° Next, what implications might SRAH have f o r the epistemic responsibilities of i n d i v i d u a l consumers? What should consumers be expected to know? Individual consumers play two rol e s i n consumer safety. The f i r s t i s that of the consumer as such. The second i s the r o l e of the c i t i z e n involved i n public debate. These two roles involve d i f f e r e n t epistemic r e s p o n s i b i l i t i e s . F i r s t , what implications might SRAH have f o r the epistemic r e s p o n s i b i l i t i e s of the i n d i v i d u a l qua consumer? The amount which consumers do not know i s , of course, enormous. Even i f the amount which consumers know i s growing i n absolute terms, the amount which Vancouver Sun. 13 August 1994: A l . *' Harris, 49. ™ Harris, 206. could conceivably be known i s growing at a f a r greater rate. Morris Kaplan suggests that "[c]onsumers become more uninformed d a i l y i n s p i t e of geometric increases i n consumer education, i n coverage of consumer problems by the media, i n consumer l i t e r a t u r e , and i n consumer l e g i s l a t i o n . I n d e e d , as Harvey M. Sapolsky notes, s c i e n t i s t s themselves face epistemic shortcomings: "Despite extensive e f f o r t , s c i e n t i s t s s t i l l do not know how much one can achieve by reducing intake of red meat, going l i g h t on the s a l t , and drinking sugared as opposed to a r t i f i c i a l l y sweetened drinks. "'^  One commonly mentioned epistemic duty of consumers i s that of reading product labels. This information i s made readi l y a v a i l a b l e . I f a consumer misuses a product, and i f that misuse could have been avoided had the consumer simply read the i n s t r u c t i o n s f o r appropriate use on the product l a b e l , then i t seems that the consumer has v o l u n t a r i l y assumed a r i s k that could have been avoided. A duty to warn on the part of producers implies a duty to l i s t e n on the part of consumers. Just how f a r a consumer should be expected to go i n her attempts to inform h e r s e l f about product r i s k s i s a complex issue. My i n t e r e s t here i s s p e c i f i c a l l y i n the implications which SRAH might have i n t h i s regard. SRAH seems to imply l i t t l e regarding the lengths to which a consumer can be expected to go i n her attempt to become an 'informed' consumer. There i s a sense i n 1^ Morris Kaplan, "Does the 'Informed Consumer' Exist? W i l l He Ever?" i n Product Quality. Performance, and Cost. National Academy of Engineering. (Washington, D.C., 1972) 49. Sapolsky, 199. which SRAH implies a reduced duty to know: i f ought implies can, then one cannot have a duty to know the unknowable. There i s also a sense i n which SRAH implies a greater duty to know: i f objective r i s k i s unknowable, the consumer may have an increased duty to compensate f o r t h i s by fi n d i n g out more about e a s i l y quantifiable aspects of a product. SRAH may also imply, however, that the consumer has an enhanced duty to know more about he r s e l f . I f the determination of whether or not a given product poses a reasonable r i s k i s t r u l y subjective, then the consumer might reasonably be expected to be s u f f i c i e n t l y introspective to have a clear understanding of what her own values are. Next, what implications might SRAH have for the epistemic r e s p o n s i b i l i t i e s of the i n d i v i d u a l qua c i t i z e n ? Qua c i t i z e n , the in d i v i d u a l may have opportunities to engage i n public debate over various products and technologies. One r e s p o n s i b i l i t y which members of the public may have i f they are to p a r t i c i p a t e i n acceptable-risk decisions " i s to make the e f f o r t needed to understand the issues that recurrently a r i s e i n acceptable-risk problems."^^ In order to p a r t i c i p a t e i n acceptable-risk debates i n a responsible manner, c i t i z e n s might be expected to have some understanding of "the p r o b a b i l i s t i c nature of r i s k phenomena," along with a basic understanding of the technologies involved. Baruch Fischhoff, Sarah Lichtenstein, Paul S l o v i c , Stephen L. Derby, and Ralph L. Keeney, Acceptable Risk. (Cambridge: Cambridge UP, 1981) 150. Fischhoff, Lichtenstein, Slovic, Derby, and Keeney, Acceptable Risk. 150. To the extent that r i s k assessment i s subjective, however, the most important r e s p o n s i b i l i t y may again be a r e s p o n s i b i l i t y to know one's own values. The most important contribution which most c i t i z e n s can make to public debate i s that of expressing t h e i r own values. This w i l l only be valuable to the extent to which the values expressed are r e f l e c t i v e and stable. Next, l e t us look at what relevance SRAH might have for the r o l e of consumer organizations. S p e c i f i c a l l y , what e f f e c t might a general acceptance of SRAH have on the r o l e played by consumer organizations i n product safety debates? To what extent would t h e i r r o l e be either strengthened or compromised? In order to examine the r o l e of consumer organizations, two key d i s t i n c t i o n s should f i r s t be made. The f i r s t d i s t i n c t i o n i s that between what I s h a l l c a l l l i t e r a l and hypothetical representation. L i t e r a l representation occurs when a consumer organization has a d e f i n i t e body of registered or subscribing members. When t h i s i s the case, the executive of the consumer organization i s l i t e r a l l y acting on behalf of an i d e n t i f i a b l e group, at that group's c o l l e c t i v e request. An example of such an organization i s the Consumer's Union i n the U.S., which has an ongoing membership based on subscriptions to i t s publication. Consumer Reports. Hypothetical representation, on the other hand, occurs when a group or i n d i v i d u a l acts i n the inte r e s t s of a larger group — usually consumers i n general — without a s p e c i f i c Mark V. Nadel, The P o l i t i c s of Consumer Protection. (Indianapolis: Bobbs-Merrill, 1971) 159-60. mandate to do so. An example of t h i s sort of representation i s the public advocacy work of Ralph N a d e r . N a d e r and advocates l i k e him seek to protect the public i n t e r e s t , without having any s p e c i f i c democratic mandate to do so. The second d i s t i n c t i o n which must be made i s between roles which such groups can play. Mark V. Nadel, i n h i s book The P o l i t i c s of Consumer Protection, i d e n t i f i e s two broad r o l e s played by i n t e r e s t groups i n general. One r o l e i s what Nadel c a l l s the agency r o l e . This r o l e consists of "acting as agents on behalf of the perceived b e l i e f s or objective i n t e r e s t s of a group or class of people."^ The other r o l e Nadel c a l l s the constituency expansion r o l e . This r o l e "...consists of expanding the constituency of an i n t e r e s t by increasing the awareness that people have of t h e i r own s e l f - i n t e r e s t or to increase the number of people who perceive themselves as holding c e r t a i n i n t e r e s t s . "^ * Next, on to the substantive question. To what extent i s the r o l e of consumer organizations either jeopardized or strengthened by SRAH? Let us look at the two r o l e s described above. F i r s t , what impact might SRAH have on the 'agency' r o l e of consumer organizations? I suggest that to the extent that r i s k assessments involve normative assumptions, the agency r o l e of membership-based Accounts of Nader's work abound. An informal account i s given i n Charles McCarry, C i t i z e n Nader. New York: Saturday Review Press, 1972. A more formal analysis i s given i n Robert D. Hoisworth, Public Interest Liberalism and the C r i s i s of Affluence. Boston: G.K. H a l l & Co., 1980. Nadel, 156. '* Nadel, 156. consumer organizations becomes more c r u c i a l . Such consumer organizations are an important way of aggregating and representing consumer values — although they obviously do so with varying degrees of legitimacy. The value aggregation function of consumer organizations i s a c r u c i a l element of t h e i r nature as p o l i t i c a l organizations. When r i s k assessment i s seen to be subjective, i t becomes c r u c i a l f or consumers to have t h e i r values represented d i r e c t l y . SRAH may suggest a l i m i t a t i o n , however, on the agency r o l e which can be played by the sort of consumer organization which r e l i e s on hypothetical representation. As w i l l be seen i n my discussion of hypothetical consent i n Chapter 6, the idea that r i s k assessment i s subjective poses problems fo r those who seek to make assumptions about what consumers r e a l l y want or value. Clearly, some value assumptions are f a i r l y safe to make. There i s no doubt a very legitimate agency r o l e for groups l i k e Ralph Nader's Center for Study of Responsive Law, despite t h e i r lack of a di r e c t democratic mandate. When such organizations f i g h t f or the most basic sorts of corporate r e s p o n s i b i l i t y , or against the most grievous sorts of product dangers, i t i s u n l i k e l y that they misrepresent the values of any consumers at a l l . I merely suggest that when organizations without mandate seek to influence regulatory decisions, i t ought to be remembered that the degree to which they represent the wishes of consumers may vary with the degree of s p e c i f i c i t y of t h e i r demands. Next, what impact might SRAH have for the 'constituency expansion' r o l e of consumer organizations? Clearly, public education i s a major component of the constituency expansion r o l e . Consumer organizations increase t h e i r constituency — either i n terms of enroled membership or merely i n terms of general public support — by convincing others that a given product i s dangerous or that a given issue i s important. This i s done by the transmission of information, either through standard advertising channels or through the news media. But as mentioned above i n t h i s chapter, SRAH implies l i m i t a t i o n s on how much objective information there a c t u a l l y i s to be transmitted. Both the form and the content of such campaigns should be such that consumers are not pressured into thinking of a given r i s k as ob j e c t i v e l y unacceptable.^' Per hypothesis, the determination that a given r i s k i s unacceptable i s a personal, subjective one. I f consumer organizations represent r i s k s as ob j e c t i v e l y unacceptable, they w i l l be serving only t h e i r own i n t e r e s t i n expansion, and doing so at the expense of responsible representation. Next, I turn to an examination of the implications which a subjective model of r i s k assessment has for the r o l e of corporations i n product safety. '' See footnote 63 above for an example of a categorical statement by a safety bureaucrat suggesting that a product i s obj e c t i v e l y unacceptable. Chapter 4 : Risks to Others : S u b j e c t i v i t y and Corporate R e s p o n s i b i l i t y What r e s p o n s i b i l i t i e s can f a i r l y f a l l upon producer corporations/° i n l i g h t of the l i t e r a t u r e on the s u b j e c t i v i t y of r i s k ? I t seems u n l i k e l y that corporations can displace r e s p o n s i b i l i t y onto the professional judgment of the engineers they employ, onto the bureaucrats who regulate them, or onto the consumers who f r e e l y purchase t h e i r products. Yet i f r i s k assessment i s t r u l y subjective, i t becomes unclear just how a producer can be sure i n advance that h i s product meets the r i s k expectations of the consumer. I hope here to outline a theory of ' d i s t r i b u t i o n i n good f a i t h , ' under which corporations can reasonably make decisions on behalf of consumers even i n the absence of t r u l y informed consent. I also propose a system whereby the value assumptions inherent i n a producer's evaluation of a product's safety can be made clear to the consuming public — thus enhancing, rather than r e s t r i c t i n g , the freedom of choice necessary to liberal-democratic capitalism. I wish to argue that a f u l l appreciation of the s u b j e c t i v i t y of r i s k assessment implies not that we should necessarily expect either more or less from corporations, but that we should expect something d i f f e r e n t . In the absence of omniscient government regulation, product For the purposes of t h i s discussion, i t w i l l s u f f i c e to think of the corporation as a single decision-making unit, and to assume that the locus of moral r e s p o n s i b i l i t y within the corporation l i e s with whomever makes the relevant high-level decisions. A more expansive discussion would d i f f e r e n t i a t e among the r e s p o n s i b i l i t i e s of various groups within the corporation. safety amounts to producers deciding on reasonable l e v e l s of r i s k to impose*' upon consumers. The question of the r e s p o n s i b i l i t i e s of corporations with regard to s e l l i n g products the r i s k s of which are subjective breaks l o g i c a l l y into two separate issues. F i r s t , i n l i g h t of t h i s s u b j e c t i v i t y , what r e s p o n s i b i l i t i e s f a l l upon any person or organization whose actions create r i s k s f o r others? Second, what r e s p o n s i b i l i t i e s f a l l upon the corporation as a r e s u l t of i t s s p e c i a l status within society? We normally think that, a l l other things being equal, we have a duty not to impose r i s k s on others. But, as i s usually the case, a l l things are not equal. Many r i s k s are consented to, either i m p l i c i t l y or e x p l i c i t l y , by those who bear them. The most common sort of r i s k which we impose on others consists of those r i s k s which a l l of us impose upon each other simply by going about our d a i l y business. For example, I impose a c e r t a i n degree of r i s k (hopefully t r i v i a l ) on any number of people every time I decide to drive somewhere. I f I d i d not drive, i t would reduce the chance of my k i l l i n g some pedestrian to 0. Each pedestrian out there would have her chance of being k i l l e d on that day decreased by some small but s t a t i s t i c a l l y determinable amount. These are t y p i c a l l y r i s k s imposed on persons to whom we have no r e l a t i o n s h i p beyond that of f e l l o w - c i t i z e n s h i p . Consent i n t h i s *' I use the word "impose" here with some caution. I r e a l i z e that i t sounds pejorative. Corporations do not generally force consumers to do anything, but i t i s nonetheless the case that t h e i r products do pose inherent threats, threats which would not be present were those products not on the market. Given that, per hypothesis, f u l l disclosure i s impossible, "impose" seems f a i r , so long as we remember the degree to which i t i s a loaded term. sort of case seems unnecessary, probably because the r i s k s involved are r e l a t i v e l y small, and because human l i f e would simply become impossible i f we a l l worried about such things. As Robert E. Keeton notes, "Since the unintended consequences of one's conduct go on i n d e f i n i t e l y , some l i m i t of r e s p o n s i b i l i t y i s a p r a c t i c a l necessity. " *2 Another ta c t might be to suggest that we a l l t a c i t l y consent to these r i s k s simply by p a r t i c i p a t i n g i n the normal a c t i v i t i e s of our community. Either way these r i s k s seem morally unproblematic. The well-known-but-minor r i s k s involved i n using most consumer products seem to f a l l into t h i s category. No one suggests that corporations manufacturing razor blades are i n any way responsible for the minor nicks and cuts suffered by consumers i n the act of shaving. Other r i s k s are more serious, and t h e i r consequences are the sort that f a l l under the domain of the law of t o r t s and negligence. Michael Bayles, i n h i s P r i n c i p l e s of Law, notes that "[a]n ancient t o r t law maxim i s volenti non f i t injuria — a person cannot be v o l u n t a r i l y injured. "^ ^ That i s , where consent i s present, there can be no 'injury' i n the l e g a l sense of 'wrongful harm.' Bayles also notes that, " [ i ] n negligence law, consent has usually gone under the name of assumption of risk..."*^, and that " [ i ] n assuming 2^ Robert E. Keeton, "The Basic Rule of Legal Cause i n Negligence Cases", i n Joel Feinberg and Hyman Gross, eds.. Philosophy of Law. (Encino, CA: Dickenson Publishing Co., 1975) 355. Michael D. Bayles, P r i n c i p l e s of Law: A Normative Analysis. (Dodrecht, Holland: D. Reidel Publishing, 1987) 245. ^ Bayles, P r i n c i p l e s of Law. 246. risk...one waives the l e g a l duty to compensate f o r any injury that might ensue. In order to be j u s t i f i e d i n imposing r i s k s on others, then, we need eith e r t h e i r e x p l i c i t consent, or at l e a s t a reasonable expectation that they would consent. Before we proceed with an a c t i v i t y , we need to evaluate the r i s k s which that a c t i v i t y holds for others. Sometimes t h i s w i l l involve common sense, and sometimes i t w i l l need to be an e x p l i c i t process. We may need to ask the three questions suggested by Rescher: 1) What i s the l i k e l i h o o d of the negative outcome? 2) How severe i s the negative outcome? 3) What evaluation (importance) i s placed on t h i s outcome?** The catch, of course, has always seemed to l i e i n step 3, for i t i s the importance which the outcome has for the person affected — i n the present case, the consumer — which matters. The importance of an outcome i s not an objective c h a r a c t e r i s t i c . In order to know what importance someone else w i l l place on various negative outcomes, we must f i r s t have a clear understanding of that person's values. C l e a r l y t h i s i s no t r i v i a l demand. But the l i t e r a t u r e c i t e d above i n Chapter 2 complicates matters further. I t suggests that Steps 1) and 2) are also quite subjective. Perceptions of both the l i k e l i h o o d and the severity of a negative outcome seem to be matters heavily influenced by the p r i o r s o c i a l commitments and psychological tendencies of the person *^  Bayles, P r i n c i p l e s of Law. 246 ** Summarized from Rescher, 1983. doing the evaluating. I t i s a l l too easy to assume that the r i s k s we become aware of and which we believe to be important are the same ones which others are aware of and believe to be important. Care should be taken to remember that even within our immediate c i r c l e of friends, values are seldom i f ever homogenous. A fortiori, we should not expect the values held by corporations to be the same as those held by any given customer. Even i f we accept the dubious assumption the decision-makers within the corporation are, i n terms of t h e i r values, t y p i c a l of society at large, we must remember that the actions taken by corporations are seldom based on individual values. The personal values of even top decision makers are mediated by r o l e - r e s p o n s i b i l i t i e s and corporate decision-making procedures. What s p e c i a l r e s p o n s i b i l i t i e s might corporations have towards consumers beyond those which private i n d i v i d u a l s have with regard to subjecting others to r i s k ? Various c h a r a c t e r i s t i c s of corporations, such as the nearly coercive power of o l i g o p o l i e s and the corporation's unique a b i l i t y to see aggregate r i s k s , can be seen as imposing s p e c i a l duties. But that much would be true even i f r i s k determination were purely objective. What special obligations f a l l to the corporation s p e c i f i c a l l y as a r e s u l t of the s u b j e c t i v i t y of r i s k assessment? I t has generally been assumed that the act of purchasing a product amounts to consenting to at least some r i s k s . Consumers who purchase products accept the r i s k s which go along with the possession and use of those products to the extent that they understand,or can reasonably be expected to understand, the r i s k s involved. This r e f l e c t s the p r i n c i p l e , c i t e d above, of volenti non f i t injuria. The general p r i n c i p l e as i t applies to the law of negligence i s that "...the person who knows the nature and character of the r i s k that he runs cannot succeed i n a claim for damages f o r i n j u r i e s suffered as a r e s u l t of h i s assumption of risk."*' The s t i p u l a t i o n that the consumer know 'the nature and character of the r i s k ' suggests that what i s required i s not just consent, but informed consent. However, many of the r i s k s inherent i n most modern products are seldom c l e a r and obvious. Granted, there w i l l also be many r i s k s which are clear, but others w i l l be hidden from non-expert eyes. The dangers of butcher knives are obvious to a l l ; the dangers of the herbicide alachlor are not. How, then, can a manufacturer d i s t r i b u t e a product i n good faith ? How can manufacturer corporations j u s t i f y d i s t r i b u t i n g merchandise to a public who w i l l be b l i n d to at lea s t some of the r i s k s involved? In p a r t i c u l a r , how can they do so when then cannot even i n t h e i r own minds a r r i v e at an objective evaluation of risk? I would l i k e to suggest that t h i s seeming i m p o s s i b i l i t y of t r u l y informed consent i s mitigated by three factors. The corporation's duty to warn may be seen as being bounded by consumer assumption of normal r i s k , by the i n t e g r i t y involved i n the *' Ontario Law Reform Commission, Report on Products L i a b i l i t y . (Toronto: Ministry of the Attorney General, 1979) 96. professional expertise of i t s engineers, and by the set t i n g of standards by regulatory bodies. Each of these i s discussed i n a separate chapter (Chapters 3, 5, and 6 r e s p e c t i v e l y ) , but they w i l l be discussed b r i e f l y here i n order to i l l u s t r a t e what they leave i n terms of corporate r e s p o n s i b i l i t y . The f i r s t i s that sometimes, rather than t r u l y 'informed' consent, what we have i s conscious consent i n spi t e of an acknowledged lack of information. The consumer i s informed that r i s k s e x i s t , but, due to s c i e n t i f i c uncertainty or a simple lack of s c i e n t i f i c knowledge, cannot be t o l d the exact extent of the r i s k . In such a s i t u a t i o n , the consumer may f e e l e i t h e r that the probable benefits are so great, or the possible r i s k s so small, that he does not need to know the exact d e t a i l s of the r i s k involved. An example of large probable benefits i s found i n the use of experimental drugs for the treatment of AIDS. For terminal diseases f o r which no c e r t a i n cure i s known, i t may be quite r a t i o n a l f o r a patient to desire an experimental drug the r i s k s of which are quite uncertain. Clearly t h i s j u s t i f i c a t i o n applies only i n a small minority of desperate cases quite unlike the average purchase of a consumer good. An example of small, i f unquantified, r i s k i s the use of safety razors. In t h i s case, the consumer i s w i l l i n g to view the r i s k s to which he i s subjected by the producer as part of the normal a c t i v i t i e s of l i f e . He does not know the exact r i s k , but he i s f a i r l y sure that the r i s k s are small enough to ignored. Also, as Bayles notes, "[pjeople do not necessarily want or expect a completely safe product design; they w i l l r a t i o n a l l y trade safety for e f f i c i e n c y and r e l i a b i l i t y . " * * The second possible way 'around' informed consent i s based on a model of professional engineering expertise s i m i l a r to the model of expertise which applies to doctors. I discuss t h i s proposal i n d e t a i l i n Chapter 5. The basic assertion i s that engineers, i n so fa r as they are asked to c e r t i f y the safety of a product, perform a task that i s not merely t e c h n i c a l . Like doctors, they are expected to make value decisions on behalf of those less t e c h n i c a l l y q u a l i f i e d . To some extent, the consumer's consent i s given i n the form of t r u s t i n the expert opinions of the engineers employed by the corporation. We might envision the consumer as r e l y i n g not so much on the i n t e g r i t y of the corporation but on the expertise of the engineer. Under t h i s model, the corporation becomes something of a go-between, providing a product designed by engineers f o r the consuming public. F i n a l l y , I suggest that there i s some legitimate transfer of moral r e s p o n s i b i l i t y , on the part of corporations, to both regulatory bodies and consumer associations. With regard to regulatory bodies, i n so f a r as regulations of a c e r t a i n type e x i s t , the subjective nature of r i s k assessment dictates that more moral emphasis re s t on such regulations. This point i s discussed i n d e t a i l i n Chapter 6. B r i e f l y stated, I suggest t h i s transfer of moral r e s p o n s i b i l i t y rests on the f a c t that regulatory bodies are capable of representing — not ju s t estimating — the aggregate of ** Bayles, P r i n c i p l e s of Law. 244. consumer i n t e r e s t s . That i s , i t i s the job of regulators to make value judgments on behalf of consumers. Corporations i n general have neither the a b i l i t y nor the p o l i t i c a l legitimacy necessary to perform t h i s function. Of course, as discussed i n Chapter 6, t h i s also implies a change i n the grounds upon which regulatory decisions may be j u s t i f i e d . With regard to consumer associations, I argued i n Chapter 3 that they too are a way of aggregating consumer values, with varying degrees of legitimacy. To the extent that such representation and aggregation is legitimate, corporations which follow guidelines set or r a t i f i e d by consumer associations arguably absolve themselves of a c e r t a i n degree of moral r e s p o n s i b i l i t y . I f at l e a s t some moral r e s p o n s i b i l i t y for product safety l i e s , as j u s t suggested, with consumers, engineers, and regulators, i s any moral r e s p o n s i b i l i t y l e f t to the corporation? In fact, a good deal of r e s p o n s i b i l i t y i s l e f t to the corporation. The main change that should r e s u l t from the r e a l i z a t i o n of the s u b j e c t i v i t y of r i s k assessment i s not that we should expect more or less from corporations, but that we should expect d i f f e r e n t . For example, where regulations e x i s t , meeting these should be seen as going a very long way toward f u l f i l l i n g the corporation's moral resp o n s i b i l i t y . * ' This i s not to say that we should not expect corporations to go beyond minimal adherence to regulations, or to I leave open here the p o s s i b i l i t y that l e g a l r e s p o n s i b i l i t y could be assigned f o r reasons other than moral r e s p o n s i b i l i t y , such as the s o c i a l expediency of assigning r e s p o n s i b i l i t y to those with the deepest pockets. be proactive i n t h e i r search for safety. Reliance on public regulations i s however a way of deferring to the regulatory body's legitimate r i g h t to make value assumptions on behalf of consumers. On the other hand, we should expect more than we currently do i n terms of honest information from corporations regarding the value assumptions that go into putting c e r t a i n products on the market. I suggest that consumers have the r i g h t to expect producers to make clea r any value assumptions inherent i n the producer's a t t r i b u t i o n of safety to the product. An example of such a value assumption i s given by Brunk et a l . These authors note that one assumption made by Monsanto Canada Incorporated as part of i t s assertion that alachlor was 'safe enough' was the assumption that the economic benefits to be r e a l i z e d through the a v a i l a b i l i t y of alachlor were relevant to the decision.'" The contrary assumption, made by the Health Protection Branch of Health and Welfare Canada, was that r i s k s ought simply to be minimized.'^ SRAH makes necessary new means of transmitting not techn i c a l information, but value information. I suggest that i t i s not unreasonable to expect that t r u l y adequate product l a b e l l i n g might include references to broad value systems contributing to the corporation's claim that the product i s safe. One way t h i s might be done would be to l i s t on l abels the governmental agencies and consumer groups who have Brunk et al., 62. Brunk et al., 49. r a t i f i e d the product as 'reasonably safe. ''^  To the extent that consumers know or can f i n d out the values of said organizations, they w i l l also know some of the values which one must need to hold to see a given product as safe enough. L i s t i n g such organizations may be a very concise way of communicating general values. Organizational reputations are often r i c h l y laden with value associations. To say, f o r example, that a product i s endorsed by the Dairy Bureau of Canada, for example, implies a d i f f e r e n t set of c r i t e r i a than would be implied by an endorsement by the World Health Organization. Another way value information might be transmitted i s by including on labels c l e a r l y worded statements of the broad assumptions that went into the corporation's own determination that the product's r i s k s are reasonable. This could i n p r i n c i p l e be done i n a way which represents the corporation's assumptions not as biases, but as values. Consider the values implied by these two hypothetical statements. Statement One: 'Alpha Corporation i s committed to making good q u a l i t y products a v a i l a b l e at a price affordable to a l l . ' Statement Two: 'We at Beta Corporation f e e l strongly that the health and safety of our customers should be our primary objective.' Similar statements already e x i s t i n the codes of ethics published by many major corporations. Take, for example, the opening l i n e s of the Credo of Tylenol manufacturers Johnson & Johnson: ^ Reservations regarding such endorsements by consumer groups are mentioned b r i e f l y i n footnote 66 on page 35 above. We believe our f i r s t r e s p o n s i b i l i t y i s to the doctors, nurses and patients, to mothers and a l l others who use our products and services. In meeting t h e i r needs everything we do must be of high quality.'^ Similar statements, transferred to product la b e l s , could help consumers to better understand the standard of q u a l i t y and safety set by the corporation, above and beyond minimal government regulations. I t i s of course by no means clea r that such a system of l a b e l l i n g i s pr a c t i c a b l e . I merely suggest that the idea has s u f f i c i e n t t h e o r e t i c a l value to merit further investigation and possibly empirical t e s t i n g . Next, I turn to an examination of the implications which a subjective model of r i s k assessment might have f o r the professional r e s p o n s i b i l i t i e s of engineers employed by producer corporations. Quoted i n R. E r i c Reidenbach and Donald P. Robin, "A Conceptual Model of Corporate Development," Journal of Business Ethic s . 10 (1991): 181. Reprinted i n Ethics Readings Handbook. Pre-Publication E d i t i o n , Michael McDonald, ed. (Vancouver: CGA Canada, nd). Chapter 5; Engineers''* and 'Safe' Products Professional engineers employed by producer corporations are r e g u l a r l y asked to make judgments regarding the imposition of reasonable r i s k s on consumers. That i s , they are expected to be able to t e l l whether a given product i s 'reasonably safe.''* The s i t u a t i o n i n which such engineers f i n d themselves i s distinguished from the s i t u a t i o n of the private i n d i v i d u a l i n three key ways. F i r s t , being a professional, the engineer has c e r t a i n duties beyond those of a priva t e i n d i v i d u a l . Second, the r i s k s involved are more c l e a r l y s p e c i f i a b l e , at least i n that, per hypothesis, they stem from the use of a s p e c i f i c product. Third, unlike private i n d i v i d u a l s , professional engineers most often do not have the option of saying 'I don't know.' The problem, of course, i s that i n l i g h t of the l i t e r a t u r e reviewed above, i t seems d i f f i c u l t at best f o r engineers to provide consumers with objective estimations of the l e v e l s of r i s k associated with t h e i r products. As safety experts, they are asked to decide whether the r i s k s inherent i n a product are 'reasonable.' Ideally, the question should be whether the r i s k s would be seen as ^ I assume that most of what i s said i n t h i s section also applies to other sorts of s c i e n t i s t s employed by producers. I focus on engineers due to the f a c t that they are most c l e a r l y recognized as a profession, and because as such there i s a s i g n i f i c a n t body of l i t e r a t u r e regarding t h e i r obligations. '* I take t h i s to include a l l engineers employed by manufacturers — not j u s t those d i r e c t l y involved i n product safety t e s t i n g . Even engineers who are not d i r e c t l y responsible for the safety of a f i n a l product can imply by t h e i r s i l e n c e that the product i s safe. reasonable by the average consumer. But Douglas and Wildavsky's analysis implies that as members of organizations, engineers are subject to s o c i a l forces which may colour t h e i r professional judgment i n ways they may not notice.'* The analysis given by Brunk et a l . suggests that an i n d i v i d u a l engineer may also disregard — or f a i l to see — c e r t a i n r i s k s because she personally holds a view of technology which implies that 1) a l l technology has r i s k s , and 2) these r i s k s are j u s t an acceptable p r i c e to pay for progress. How can engineers employed by manufacturer corporations j u s t i f y d i s t r i b u t i n g merchandise to a public who w i l l be b l i n d to at l e a s t some of the r i s k s involved? In p a r t i c u l a r , how can they do so when then cannot even i n t h e i r own minds a r r i v e at an objective evaluation of r i s k ? In Chapter 4, I suggested two p o s s i b i l i t i e s . The f i r s t i s that sometimes rather than l i t e r a l l y 'informed' consent, what we have i s conscious consent i n spite of an acknowledged lack of information. Sometimes, consumers have reasons to accept r i s k s even i n the face of a deficiency of objective r i s k data. The second possible way 'around' informed consent i s to be seen i n a wide v a r i e t y of medical cases, i n which some of t h i s slack created by l i m i t a t i o n s on informed consent seems to be taken up by the physician's professional judgment based on expertise. I t '* This may be p a r t i c u l a r l y true of engineers who also hold management posit i o n s . See Joseph R. Herkert, "Management's Hat Trick: Misuse of 'Engineering Judgment' i n the Challenger Incident," Journal of Business Ethics 10 (1991): 617-21. has been suggested that, i n much medical research, t r u l y informed consent "...on the part of the subject i s impossible...because he or she cannot ever r e a l l y understand the tec h n i c a l substance and methodology of a research experiment."'' But i t seems that, even i n cases i n which a doctor cannot provide the patient with clear s t a t i s t i c a l information, she can provide professional judgment i n i t s place. In fa c t , that i s part of what she i s paid f o r . The patient does not expect to understand the f u l l complexities of his condition, but he does expect to be able to t r u s t h i s doctor's judgment. This, of course, provides the patient with no increase i n objective information, but at least i t becomes transparent where the judgments are being made. I would l i k e to suggest that the s i t u a t i o n faced by engineers employed by manufacturers i s si m i l a r to that faced by doctors i n cases where p e r f e c t l y informed consent i s impossible. Like the doctor, the employed engineer i s a member of a profession. As such, she i s r e l i e d upon to give expert advice to those who lack t h e i r expertise. The engineer's most obvious r e s p o n s i b i l i t y i s her fi d u c i a r y r e s p o n s i b i l i t y to her employer. However Michael Bayles, i n his book Professional Ethics, notes that on the commonly accepted f i d u c i a r y model of the p r o f e s s i o n a l - c l i e n t relationship, professionals "are responsible for e f f e c t s on t h i r d parties and ^ This p o s i t i o n i s summarized and c r i t i c i z e d by Bernard Barber i n Informed Consent i n Medical Therapy and Research. (New Brunswick, NJ: Rutgers UP, 1980). This quote i s from p. 10. subject to obligations to them."'* One might further say that, as a r e s u l t of the professional engineer's r o l e (and privileges) i n society, there i s some sense i n which a great many people — a l l those who use products the design or t e s t i n g of which she pa r t i c i p a t e s i n — may be considered to be her c l i e n t s . The sense i n which a l l consumers may be considered the c l i e n t s of employed engineers provides grounds for the p a r a l l e l which I suggest between the duties of the engineer and that of the physician. I have i n mind here a model under which the consumer i s seen as r e l y i n g not so much on the i n t e g r i t y of the corporation but on the expertise of the engineer. Under t h i s model, the corporation becomes something of a go-between, providing a product designed by engineers f o r the consuming public. To extend the p a r a l l e l with medical expertise, we might say that i n some respects corporations are to engineers as hospitals are to physicians. The extent to which both corporations and hospitals can claim to serve as a 'mere' go-between reaches i t s l i m i t when actions taken by experts are negligent. In writing about professional engineers i n the Canadian manufacturing industry, Howard Bexon notes that under the p r i n c i p l e of vicarious l i a b i l i t y , "the employer i s l i a b l e for the negligent acts of employees...."'' '* Michael D. Bayles, Professional Ethics, 2nd ed. (Belmont, CA: Wadsworth Publishing, 1989) 112. " Howard Bexon, "Professional Engineers i n the Manufacturing Industry," i n Carson Morrison and P h i l i p Hughes, eds.. Professional Engineering Practice: E t h i c a l Aspects. (Toronto: McGraw-Hill, 1988) 56. The problem of balancing the duties which engineers have to t h e i r employers against those of the consuming public i s one which has received much attention i n engineering ethics textbooks and i n the codes of ethics of various professional engineering associations. One example of a c o n f l i c t of duties i s seen i n cases of whistle-blowing. At what point does the engineer's duty to warn consumers of non-obvious dangers override her duty to obey the i n s t i t u t i o n a l authority of her supervisors? For my purposes here, the s a l i e n t question i s , to what extent does SRAH influence t h i s balance of duties? SRAH implies that no determination as to the safety of a product can be completely objective. In some cases, including perhaps the famous Ford Pinto case, the danger i s severe enough and the corporation's willingness to obscure the facts i s cl e a r enough to warrant whistle-blowing. In borderline cases, however, SRAH w i l l give the engineer reason to pause and consider whose standards ought to be applied to determine whether a product involves 'unreasonable' r i s k s . This implies that some normative authority must be claimed i n order to j u s t i f y overriding i n s t i t u t i o n a l practices i n order to blow the whistle. I suggest that the model of engineering expertise outlined above provides j u s t such normative authority. I f we see the task of the engineer as including making c e r t a i n value decisions on behalf of t h e i r c l i e n t s (who, per hypothesis, include a l l consumers), then engineers can be seen as having a greater claim to moral authority than does the corporation when i t comes to deciding on behalf of consumers whether or not a product i s 'safe enough.' The primary duty which employed engineers have toward the general public i s a duty to warn of r i s k s . This leaves open the question "which r i s k s ? " Certainly not all r i s k s need be brought to the public's attention. Some r i s k s w i l l be i n s i g n i f i c a n t (such as the r i s k of paper cuts from c e r t a i n paper products), and others w i l l be so obvious as to obviate the need f o r warning (such as the dangers of sharp knives). Further, SRAH suggests that the decision as to which r i s k s are s i g n i f i c a n t i s not a purely technical one. Perhaps a l l that can be said i s that the public needs to be able to t r u s t professional engineers to act i n good f a i t h . Such being the case, the most we can expect of engineers i s that they do t h e i r best to deserve that t r u s t . How can they do that? I w i l l conclude t h i s chapter with four concrete suggestions. The f i r s t suggestion i s that engineers be w i l l i n g to admit the l i m i t a t i o n s of t h e i r a b i l i t y to determine r i s k . In writing about regulatory science, A l v i n M. Weinberg writes: I should think that a f a r more honest and straightforward way of dealing with the i n t r i n s i c i n a b i l i t y of science to pr e d i c t the occurrence of rare events i s to concede t h i s l i m i t a t i o n and not to ask of science or s c i e n t i s t s more than they are capable of providing. Mike W. Martin and Roland Schinzinger note that public statements involving numbers merit s p e c i a l care and attention. They write: "A willingness to admit uncertainty and bias and to reveal methodology and sources i s p a r t i c u l a r l y important when numerical A l v i n M. Weinberg, "Science and i t s Limits: The Regulator's Dilemma," Issues i n Science and Technology. 2.1 ( F a l l , 1985): 68. data and s t a t i s t i c s are presented."'"' Joshua Lederberg suggests that, i n order to avoid having t h e i r science pushed beyond i t s c a p a b i l i t i e s , experts must refuse to advance opinions on ce r t a i n questions: Although valuable as a t o o l f o r solving the technological problems, expertise has distortions....Besides the inherent problems of c o n f l i c t s of in t e r e s t and d i f f e r i n g l e v e l s of competence, experts may also malfunction when they are asked, and f a i l to r e j e c t , the wrong questions.(emphasis o r i g i n a l ) As noted e a r l i e r , engineers employed by producer corporations seldom have the option of saying 'I don't know.' They are paid to determine the safety of products. They are expected to come to one of two conclusions: yes, the product i s safe — d i s t r i b u t e i t ; or no, the product i s not safe — go back to the drawing board. But we can expect engineers to recognize the l i m i t s of science by refusing to give a product the go-ahead when i t s safety i s uncertain. Engineers should remind themselves of the uncertainty l e v e l s inherent i n many of t h e i r estimates, and should use appropriate margins of safety. My second suggestion i s that engineers involved i n communicating safety information to non-experts recognize the importance of the s c i e n t i f i c language which they use. S c i e n t i f i c language d i f f e r s s i g n i f i c a n t l y from the language used by consumers. Rayna Rapp makes t h i s point regarding the importance of language with regard to genetic counsellors, but I believe the point i s an Mike W. Martin and Roland Schinzinger, Ethics i n Encfineerinq. 2nd ed. (New York: McGraw-Hill, 1989) 135. Joshua Lederberg, quoted i n Lowrance, 109. important one for r i s k experts of a l l sorts. Rapp points out that experts are "inherently b i l i n g u a l s , r a i s e d i n one language community, but having acquired science as a second (or subsequent) language. "1"^ As Rapp points out, s c i e n t i f i c language commands great authority and respect. Thus engineers ought to be careful not to use arcane modes of discourse i n a way which may verge on coercive. Rapp further suggests that although experts may use tec h n i c a l language c l e a r l y i n communicating with the public, the i n t e r p r e t a t i o n of that information by the public w i l l take place outside of the purely s c i e n t i f i c realm i n which language was developed and to which i t i s suited. With regard to genetic counselling, Rapp writes: "accommodation and resistance to s c i e n t i f i c knowledge i s always interpreted i n s o c i a l frameworks larger than the biomedical. "i°* This implies that, even i f non-experts can be made to understand technical terminology, technical concepts w i l l s t i l l have to compete with f o l k b e l i e f s for the r i g h t to be the paradigm under which r i s k s are interpreted. Wise engineers w i l l keep t h i s i n mind, and w i l l seek to communicate with consumers i n ways that are conducive to mutual understanding. The t h i r d suggestion i s that professional engineers should make a conscious e f f o r t to examine from time to time not only the values and i n t e l l e c t u a l commitments of t h e i r profession, but the Rayna Rapp, "Heredity, or: Revising the Facts of L i f e . " Plenary Lecture for the Canadian Anthropology Society and the Canadian Association f o r Medical Anthropology. Vancouver, B.C., May 1994. 7. I'^Rapp, 3. values and s o c i a l commitments of t h e i r larger community. Mike W. Martin and Roland Schinzinger note that some leaders of industry suggest that those who emphasize the environmental r i s k s of technology are "emotional and i r r a t i o n a l or p o l i t i c a l l y motivated."'"* Martin and Schinzinger suggest, however, that such p u b l i c concerns are f a r from i r r e l e v a n t to the task of the professional engineer: " I t i s important that engineers recognize as part of t h e i r work r e a l i t y such widely held perceptions of r i s k and take them into account i n t h e i r designs."'"* Since the professional engineer i s designing products for use by the public, i t i s the concerns and values of the public which should inform safety standards. The fourth and re l a t e d suggestion i s that engineers as a profession must make an e f f o r t not to l e t t h e i r professional values lag behind those of society i n general. Barber notes that t r u s t "... i s l i k e l y to decline where shared values are breaking down or where there are emergent values i n some members of a s o c i a l system not shared by others."'"' New values of consumerism have been emerging at an accelerating rate over the past three decades. I f engineers allow t h e i r judgment to be dominated by the instrumental r a t i o n a l i t y which Brunk et a i . suggest dominates t r a d i t i o n a l r i s k assessment,'"* they may f i n d themselves out of step with t h e i r '"* Martin and Schinzinger, 115. '"* Martin and Schinzinger, 115. '"^  Barber, 27. '"* Brunk et a l . , 143. society, and i n the midst of a c r i s i s of t r u s t . Next, I turn to a discussion of the implications which SRAH might have fo r the task faced by the regulatory bureaucracies responsible f o r product safety. Chapter 6; Regulators and the Reco n c i l i a t i o n of Diverse Objectives Perhaps the most d i f f i c u l t r o l e , i n l i g h t of the apparent s u b j e c t i v i t y of r i s k assessment, i s that of the governmental product-safety bureaucracies. Such bureaucracies are charged with the d i f f i c u l t twin tasks of protecting the public safety and ensuring that producers are treated f a i r l y . A number of c o n f l i c t i n g and overlapping values must be balanced. As we s h a l l see, t h i s task becomes even more complicated i f r i s k assessment i s taken to be subjective. I w i l l deal here with two key questions pertaining to the relevance of the s u b j e c t i v i t y of r i s k assessment to the ro l e of product safety bureaucracies. F i r s t , to what degree can product safety c o n f l i c t s be d e - p o l i t i c i z e d through 'regulatory science,' and to what extent i s such d e - p o l i t i c i z a t i o n desirable? Second, should we see the s u b j e c t i v i t y of r i s k assessment as undermining to any degree the models of i m p l i c i t or hypothetical consent so necessary to the task of regulatory agencies? I turn to these two questions now. To what degree can product safety c o n f l i c t s be d e - p o l i t i c i z e d through 'regulatory science,' and to what extent i s such de-p o l i t i c i z a t i o n desirable? Brunk et a l . define regulatory science as "the enlistment of s c i e n t i s t s and the trappings of s c i e n t i f i c method i n the aid of government regulation, " i * " As these authors note, the idea that Brunk et al., 2. the balancing act inherent i n the r o l e of the regulator "might be c a r r i e d on by the terms of an objective and s o c i a l l y neutral 'science' has become a t a n t a l i z i n g prospect• "''" Standard r i s k assessment i s taken by i t s advocates to be j u s t such a s t r i c t l y s c i e n t i f i c procedure. Brunk et a l . note that i n general, people engaged i n the business of r i s k analysis "imagine that they do not make [value assumptions] and that the c r e d i b i l i t y of t h e i r manner and proceeding depends on t h e i r not doing so."''' To the extent that i t i s recognized that r i s k assessment requires c e r t a i n value assumptions, i t i s assumed that these ought to be minimized, or i d e a l l y eliminated."^ Brunk et a l . suggest that the strongest motivation behind t h i s attempt to minimize the value component of r i s k assessment l i e s i n the l i b e r a l i d e a l that government regulation ought to be "neutral vis-a-vis the d i f f e r e n t goods sought by i n d i v i d u a l s and groups within society.""^ A c t u a l l y , we may d i s t i n g u i s h between two main motivations which can l i e behind the desire to maintain the appearance of administrative n e u t r a l i t y . The f i r s t i s examined by Conrad Brunk i n h i s paper "Professionalism and R e s p o n s i b i l i t y i n the Technological Society.""'* I t involves the idea that experts "" Brunk et al. , 2. '" Brunk et al., 137. " 2 Brunk et al., 6. " 3 Brunk et al. , 6. "'* Conrad Brunk, "Professionalism and R e s p o n s i b i l i t y i n the Technological Society," i n Deborah C. Poff and Wilfred J . Waluchow, eds. Business Ethics i n Canada. Scarborough, ON: Prentice-Hall, should perform t h e i r tasks i n a t e c h n i c a l l y competent but morally neutral manner. Engineers asked to perform r i s k assessments, then, may believe that since they themselves have expertise only i n tec h n i c a l matters, moral aspects of r i s k decisions ought to be l e f t to others. The second motivation behind the wish to maintain the appearance of administrative n e u t r a l i t y i s discussed by Kenneth Kernaghan and John W. Langford i n The Responsible Public Servant. "* I t rests upon the idea that those i n the public service ought to remain p o l i t i c a l l y neutral. Kernaghan and Langford write, "...the ethic of n e u t r a l i t y suggests that the pub l i c i n t e r e s t i s the business of the democratically elected p o l i t i c a l a u t h o r i t i e s , not of public servants.""* Clearly, there i s much to be said for the idea of a neutral public service tending to the public safety i n an unbiased manner. The problem, of course, i s that proponents of standard r i s k analysis see the utility-maximization inherent i n that process as morally neutral. They f a i l to see that the very choice of u t i l i t y -maximization as the standard i s an important moral decision. As Annette Baier writes, to think that c o n f l i c t among sacred values can be resolved by a process which simply maximizes expected preference-satisfaction i s only possible f o r "...those whose own sacred value i s utility-maximization." Baier continues, "to resort 1987. "* Kenneth Kernaghan and John W. Langford, The Responsible Public Servant. Halifax, NS: The I n s t i t u t e f o r Research on Public Policy, 1990. "* Kernaghan and Langford, 46. to utility-maximization i s not to get beyond clashing sacred values, but merely to reveal one's own sacred value."'" Brunk et a i . concur. They write: "To make [regulatory sciences such as r i s k assessment] the sole a r b i t e r s i s i n e f f e c t to allow one value framework within the s o c i a l c o n f l i c t represented i n the r i s k debate to s e t t l e the issue i n i t s favour.""* Can conscientious r i s k assessment at lea s t serve to put to re s t public misconceptions concerning r i s k ? I t might be thought that c a r e f u l measurement of quantifiable aspects of r i s k might serve at lea s t to d i s p e l the sorts of superstitions regarding r i s k which turn regulatory decisions into p o l i t i c a l firestorms. Can regulatory science d e p o l i t i c i z e safety issues to at least that extent? Kip V i s c u s i writes that as a p r a c t i c a l matter, "policymakers may have to overcome the p o l i t i c a l pressures generated by...." i r r a t i o n a l r i s k perceptions among the public."' Clearly, i f the r i s k assessment process brings to l i g h t some gross inaccuracy i n the public's b e l i e f s concerning the r e l a t i v e a c c e p t a b i l i t y of two r i s k s i t u ations, then regulators may seem to have a prima facie reason to discount those opinions, along with j u s t i f i c a t i o n f o r t r y i n g to educate those labouring under misconceptions. A group once seen as a p o l i t i c a l f a c t i o n might thus come to be seen as merely misguided with regard to the "' Annette Baier, "Poisoning the Wells" i n Douglas MacLean, ed., Values at Risk (Totowa, NJ: Rowman & Allanheld, 1986) 51. "* Brunk et al., 7. "' V i s c u s i , 25. r e l a t i v e dangers of various products. For example, the public may believe, i n the wake of a dramatic a i r disaster, that more should be spent on a i r safety. But i f s t a t i s t i c a l evidence suggests that a i r t r a v e l i s a c t u a l l y r e l a t i v e l y safe, and that greater benefits could be achieved through increased spending on highway safety, then regulators might safel y conclude that the public i s simply m i s i n f o r m e d . S o i f lobby groups can i n some circumstances be defused, then i t seems that a form of ' d e p o l i t i c i z a t i o n ' i s i n p r i n c i p l e possible. But i s i t desirable? There seems to be some merit i n the view that such supposed i r r a t i o n a l i t i e s i n the public's r i s k preferences are revealing in d i c a t o r s of what i s valued by the public, and that such i r r a t i o n a l i t i e s ought not to be seen as mere obstacles to r a t i o n a l policy-making. To interpret opposition to a given technology as mere f a i l u r e to understand true l e v e l s of r i s k may i n fact be to misinterpret what the public i s r e a l l y saying. In fact, such opposition may reveal fears beyond the fear of danger to human health. As Douglas MacLean writes, "[p]erhaps the source of fear i n some cases i s about d i f f e r e n t kinds of r i s k to society, or the opposition may be to the way these decisions are being made."'^ ^ To take a l l instances of inconsistency as reason f o r discounting public opinion places too much emphasis on consistency. Members of the p u b l i c may a c t u a l l y be demonstrating a s e n s i t i v i t y to context The tendency to over-estimate the frequency of events which are of a dramatic nature i s well documented. See f o r example Fischhoff et a l . , 28-30. MacLean, 29. which corporations and regulators lack. Granted, i t i s no doubt true that there are genuine cases of misunderstanding r e l a t i v e r i s k s , cases i n which pointing out inconsistencies i s l i k e l y to r e s u l t i n a r e v i s i o n of r i s k preferences. I t i s both e t h i c a l l y and p r u d e n t i a l l y incumbent upon producer corporations and regulatory decision-makers to make every e f f o r t to educate the public i n such cases. But when a f u l l y informed public remains 'unreasonably' opposed to a given risk,'^^ t h i s ought to be taken to be a signal, not merely an obstacle. Rather than seeking to d e p o l i t i c i z e such r i s k regulation decisions, perhaps what i s needed i s a recognition that certain choices are inherently p o l i t i c a l . A l v i n M. Weinberg suggests a r e d e f i n i t i o n of the term 'regulatory science' to t h i s end.'^^ He suggests that regulatory science ought to use relaxed standards of s c i e n t i f i c proof, on the grounds that to demand certainty i n most safety matters i s to demand the impossible. Weinberg contends that such an understanding would amount to r e f r a i n i n g from "asking of science or s c i e n t i s t s more than they are capable of providing...." and would turn what i s now wrongly seen as a s c i e n t i f i c problem into a policy-making problem.'^'* I t seems that an important part of dealing with the inherent shortcomings of science i n t h i s regard i s the r e a l i z a t i o n that r i s k management i s an inherently normative Continuing public opposition to nuclear energy i s the most commonly c i t e d example of public fear of a technology being apparently out of proportion to the r i s k s involved. Weinberg, 68. '2'* Weinberg, 68. a c t i v i t y , and that p o l i t i c a l decisions are a key element of regulating product safety. The search f o r a value-neutral arbiter, i t seems, must be given up as u n r e a l i s t i c . Next, l e t us turn to the question of the impact of a subjective view of r i s k assessment on various theories of in d i r e c t consent. In a growing v a r i e t y of cases, i t f a l l s to governmental bureaucracies to make decisions concerning appropriate r i s k s on behalf of the consuming public. As Douglas MacLean notes, problems of public r i s k can seldom be solved by means of straightforward market solutions, the sort of solutions which allow "individuals to make t h e i r own choices, trading o f f values, as they see them, for t h e i r own advantage, " i ^ * Since, f o r a v a r i e t y of reasons, i n d i v i d u a l action i s either impossible or i n e f f e c t u a l i n such cases, i t most often f a l l s to government to make decisions on behalf of consumers. For evidence of the degree of government involvement i n making decisions for consumers, we need only look at the number of agencies involved i n making safety decisions on behalf of Canadian consumers j u s t i n the area of chemicals used i n agr i c u l t u r e . These agencies include Agriculture Canada, the p r o v i n c i a l m i n i s t r i e s of agriculture, various d i s t i n c t branches of Health and Welfare Canada, and a wide var i e t y of ad hoc bodies such as the Alachlor Review Board. A l l of these are involved i n making decisions on behalf of consumers. And, as MacLean notes, "[s]ince Douglas MacLean, "Risk and Consent: Philosophical Issues f o r Centralized Decisions" i n Douglas MacLean, ed., Values at Risk (Totowa, NJ: Rowman & Allanheld, 1986) 19. these choices are between competing values, they require j u s t i f i c a t i o n . " ' 2 * As noted previously i n Chapter 3, a key concept i n the imposition of r i s k s upon others i s that of consent. I t may not be immediately obvious i n what sense safety bureaucracies need to be concerned with consent. Afte r a l l , they are there to protect the public, not to impose r i s k s . But even i n cases where a regulatory body's mandate i s to be cautious, to err on the side of safety,'^' some degree of r i s k i s usually permitted. I f r i s k i s to be permitted by government regulations, that r i s k i s thereby i n some sense sanctioned by the government. I take i t as given that sanctioning r i s k imposition requires consent. With regard to ce n t r a l i z e d decision making, obvious problems with universal consultation mean that consent must usually come i n some i n d i r e c t form — usually either i m p l i c i t consent or hypothetical consent. There are two basic theories of i m p l i c i t consent — revealed preference theory and expressed preference theory. Revealed preference theory supposes that " i n d i v i d u a l preferences f o r r i s k and safety trade-offs are revealed i n certain areas, where markets function properly...." and that "...we can use data from these areas to j u s t i f y decisions i n other areas. "'^ * Empirical data regarding consumer choices i n situ a t i o n s i n which MacLean, 19. Brunk et. a l . , f o r example, suggest that t h i s i s the case with Health and Welfare Canada's Health Protection Branch. See t h e i r page 49. MacLean, 22. action i s f u l l y consensual, such as the purchase of smoke detectors, are taken as i n d i c a t i v e of general preferences regarding r i s k . Expressed-preference theory "uses psychometric techniques to uncover current attitudes and pref erences. . . . "^ ^^  regarding new and mysterious r i s k s . Hypothetical consent models, on the other hand, ask what sorts of things i d e a l persons i n i d e a l situations would consent to, and take t h i s "to support a conclusion about what we, i n our current circumstances, ought to do.""° The assumption i s that the r e s u l t of such a decision i s procedurally j u s t . The key question f o r my purposes here i s , can either i m p l i c i t or hypothetical consent withstand the r e a l i z a t i o n that r i s k assessment i s subjective? That i s , i s objectivity i n r i s k assessment a necessary condition for v a l i d i m p l i c i t or hypothetical consent? My goal here i s not to pass f i n a l judgment on theories of i n d i r e c t consent. Clearly, such theories do have t h e i r weaknesses. Just as c l e a r l y , i n many cases appealing to even imperfect consent w i l l be better than simply presuming to know what i s best. My goal at t h i s point i s simply to ask whether SRAH (the Subjective Risk Assessment Hypothesis) makes theories of i n d i r e c t consent any less p l a u s i b l e . Let us look f i r s t at i m p l i c i t consent. Paul Slo v i c and Baruch Fischhoff note some of the problems with the revealed-preference theory of i m p l i c i t consent. They note, f o r example, that the theory makes the rather dubious assumption that "...past behaviour MacLean, 23. "° MacLean, 25. i s a v a l i d predictor of present p r e f e r e n c e s . . . . a s well as making strong assumptions "...about the r a t i o n a l i t y of people's decision making i n the marketplace and about the freedom of choice that the marketplace provides."'^' Douglas MacLean makes another c r u c i a l observation: people do not always compare r i s k s straightforwardly i n terms of expected v a l u e . T h a t i s , some r i s k s have symbolic value, f o r example, which makes them either more acceptable (such as skiing) or less acceptable (such as food additives) than others. William Lowrance, drawing on the work of several other authors, i d e n t i f i e s a number of considerations, other than p r o b a b i l i t y and magnitude of harm, which a f f e c t the aversion which people f e e l to p a r t i c u l a r r i s k s . These include, among others, the degree to which the r i s k i s voluntary, the degree to which the p o t e n t i a l effects are immediate, and the degree to which possible e f f e c t s are r e v e r s i b l e . T o what extent does SRAH make extrapolation from 'revealed' preferences less j u s t i f i a b l e ? I contend that SRAH poses serious problems f o r revealed-preference theory. Revealed-preference theory r e l i e s on the assumption that consumer preferences w i l l be consistent across s i t u a t i o n s involving s i m i l a r r i s k s . But SRAH s p e c i f i c a l l y denies that r i s k s are objec t i v e l y measurable. An obvious upshot of t h i s i s that i t becomes impossible to state with confidence that two s i t u a t i o n s pose s i m i l a r r i s k s . Let us look at an imaginary example "' S l o v i c and Fischhoff, "How Safe i s Safe Enough?" 136. MacLean, 23-4. Lowrance, 86 f f . of using i m p l i c i t consent to j u s t i f y a given l e v e l of r i s k . Let us suppose that we know from s t a t i s t i c s regarding consumer spending on optional safety-enhancing extras for automobiles that consumers commonly accept a r i s k of n deaths per 100,000 driv e r s per year. We might conclude from t h i s that i t i s reasonable, as regulators, to assume that any consumer product which presents a s i m i l a r l e v e l of r i s k , measured again i n deaths per 100,000 users per year, w i l l be acceptable to consumers. The c r i t i c i s m suggested by MacLean above i s b a s i c a l l y that not a l l products are of equal symbolic importance. Cars, f o r example, have a s p e c i a l symbolic importance i n our culture, and people may be w i l l i n g to accept extra r i s k s because of that. As Baruch Fischhoff has noted, "[p]eople accept options, not r i s k s . " ^ ^ Options are characterized by more than just t h e i r l e v e l of r i s k . The r e s u l t of SRAH i s possibly even more f a t a l to the project of revealed preference theory. According to SRAH, one can never know — except perhaps i n rare cases where considerable experience with products r e s u l t s i n well-founded a c t u a r i a l tables describing s i m i l a r rates of injury — that products A and B o f f e r similar l e v e l s of r i s k . The comparison i s i n p r i n c i p l e impossible to make. Granted, even producing a c t u a r i a l tables involves making judgments, such as decisions to focus on i n j u r i e s of a c e r t a i n type or severity. I simply o f f e r such ex post facto studies as an example of the most objective sort of knowledge one could have about a 1^ Baruch Fischhoff, "Psychology and Public P o l i c y : Tool or Toolmaker," American Psychologist 45.5 (May 1990): 651. product's safety. In contrast, any estimate of the r i s k s inherent i n a new product are going to be subject to a l l the objections advanced i n Chapter 1. Next, l e t us look at the other v a r i e t y of i m p l i c i t consent theory, namely expressed-preference theory. Expressed-preference theory uses a va r i e t y of d i r e c t methods, such as surveys and interviews, to uncover attitudes and preferences. Rather than guessing what consumers' preferences are, or extrapolating from t h e i r behaviour, why not ju s t ask them? In order to place d o l l a r -values on items for which no market e x i s t s , researchers have developed a "a family of survey techniques c a l l e d contingent valuation methods."'^* These techniques involve asking subjects counter f a c t u a l l y how much they would be w i l l i n g to pay for an outcome i f i t were offered. According to Fischhoff, the p r i n c i p l e problem which t h i s methodology faces i s determining "whether respondents possess s u f f i c i e n t l y well a r t i c u l a t e d values that they can answer the questions."'^* As Fischhoff notes, " i f people cannot answer these complex, novel questions, then t h e i r own responses w i l l misrepresent t h e i r values."'^' Does SRAH have any implications for such attempts to determine consumer values? In fa c t , a b e l i e f i n the s u b j e c t i v i t y of r i s k assessment may be one of the motivating factors behind the Fischhoff, "Psychology and Public P o l i c y , " 651. Baruch Fischhoff, rev. of Usiner Surveys to Value Public Goods; The Contingent Valuation Method, by Robert Cameron M i t c h e l l and Richard T. Carson, Public Opinion Quarterly 54 (1990); 287. '" Fischhoff, "Psychology and Public P o l i c y , " 651. development of expressed-preference theory as an a l t e r n a t i v e to revealed-pref erence theory. SRAH implies that people's preferences cannot be assumed to be consistent across r i s k s which seem, prima facie, to be s i m i l a r . Such being the case, s p e c i f i c actions seem a poor ind i c a t o r of general preferences. Asking consumers d i r e c t l y what values they place on various gambles or on various outcomes seems more reasonable. SRAH may pose a challenge, however, to assumptions regarding the degree to which the views expressed by a few 'representative' consumers can be taken to be an indicator of consumer values i n general. I f r i s k assessment t r u l y i s subjective, then there i s less reason to believe that the views of the few represent the views of the many. This, however, i s a general problem for any attempt to set p o l i c y based on opinion surveys, and i s not unique to questions of product safety. An improvement on t h i s s i t u a t i o n has been suggested by John Stevenson of the University of Toronto. Stevenson suggests that p o l l s be supplemented with "focus groups, where i n i t i a l reactions are explored i n some depth, there i s some free-form discussion, new information i s provided, and so on."i^* The idea, says Stevenson, i s "to get some r e f l e c t i v e feedback and informed opinion. "^ ^^  Granted, even improved methods of determining the values of i n d i v i d u a l consumers w i l l s t i l l be subject to concerns about just how representative the opinions of the few r e a l l y are. However, i t seems to be i n the i n t e r e s t of progressive democracy that public "* John Stevenson, personal communication, August 11, 1994. Stevenson, August 1994. p o l i c y must on occasion be informed by opinion surveys of some sort. Given t h i s assumption, i t i s better that p o l i c y be based on the considered opinions of informed consumers, rather than on knee-jerk reactions to pop-quiz s t y l e surveys. F i n a l l y , l e t us look at one l a s t type of i n d i r e c t consent, namely hypothetical consent. Douglas MacLean writes: "Hypothetical-consent models attempt to evade the problem of determining an indiv i d u a l ' s actual preferences and generalizing from them."'**" Theories of hypothetical consent seek to determine what r i s k s f u l l y r a t i o n a l i n d i v i d u a l s would consent to under ideal conditions. The primary c r i t i c i s m l e v e l l e d at theories of hypothetical consent i s t h i s : to impute such strong normative force to assumptions about what i d e a l persons i n i d e a l situations would do requires substantial j u s t i f i c a t i o n and exp l i c a t i o n . The l o g i c a l step from hypothetical consent to j u s t i f i c a t i o n of imposing r i s k s i s a long step indeed. One might reasonably ask, 'What should I care about what a p e r f e c t l y r a t i o n a l person i n an ideal world would consent to? I am a r e a l person, with considerable emotional commitments to ce r t a i n core b e l i e f s which may or may not be r a t i o n a l , and I do not l i v e i n an i d e a l world.' Again, I wish i n general neither to defend nor c r i t i c i z e theories of hypothetical consent. For my purposes, i t w i l l be enough to ask whether SRAH has any impact on the normative force of hypothetical consent. I wish to argue that SRAH does indeed present a strong MacLean, 24. challenge to theories of hypothetical consent. I f the counterfactual to be considered i s something l i k e , 'would an i d e a l l y r a t i o n a l person accept t h i s r i s k ? ' , i t seems that said r i s k w i l l f i r s t need to be quantified. SRAH implies that accurate q u a n t i f i c a t i o n of r i s k s i s impossible. This leaves open the p o s s i b i l i t y , of course, of foregoing q u a n t i f i c a t i o n , and simply asking 'would an i d e a l l y r a t i o n a l person accept a r i s k of t h i s general nature?' But the q u a l i t i e s a ttributed to such an i d e a l l y r a t i o n a l person are u n l i k e l y to provide a r i c h enough picture to make preferences predictable. By i d e a l i z i n g we take away exactly the sorts of c h a r a c t e r i s t i c s — s o c i a l context and core b e l i e f s — that make choice possible despite the absence of a firm q u a n t i f i c a t i o n of r i s k . Next, I summarize my e f f o r t s and suggest some conclusions regarding the implications of a subjective model of r i s k assessment for product safety discussions. Chapter 7: Conclusion In t h i s t h e s i s , I have examined the l i t e r a t u r e which argues i n favour of a subjective theory of r i s k assessment. I have also explored the impacts which a subjective theory of risk-assessment (which I term 'SRAH:' the Su b j e c t i v i t y of Risk Assessment Hypothesis) might have fo r the rol e s played by various stakeholders i n product safety discussions. In Chapters 1 and 2, I outlined grounds, presented by authors from three d i s t i n c t f i e l d s , f or accepting a subjective model of risk-assessment. The work of Conrad Brunk and h i s colleagues suggests that the supposedly s c i e n t i f i c process of r i s k assessment i s i n f a c t subjective and value-laden, and that i t s r e s u l t s depend c r u c i a l l y on the evaluator's 'value framework.' The s o c i o l o g i c a l perspective of Mary Douglas and Aaron Wildavsky i l l u s t r a t e s how organizational culture can be c r u c i a l i n determining which r i s k s which individuals se l e c t f o r attention. And the psychological l i t e r a t u r e summarized by Robyn M. Dawes implies that r i s k evaluation i s heavily prejudiced by r e a d i l y i d e n t i f i a b l e human cognitive l i m i t a t i o n s . I suggested that together, these three analyses imply that i t i s at leas t p l a u s i b l e that t r u l y objective r i s k assessment i s impossible. In subsequent chapters, I have shown that such a subjective model of r i s k assessment does indeed have s i g n i f i c a n t implications for product safety discussions. With regard to consumers, I suggested i n Chapter 3 that SRAH c a l l s into question the value of certa i n decision-making procedures, modifies consumers' epistemic r e s p o n s i b i l i t i e s , and makes certa i n r o l e s played by consumer organizations more c r u c i a l while making others seem dubious. With regard to producer corporations, I suggested i n Chapter 4 that SRAH implies changes i n what we expect from corporations i n terms of obtaining informed consent from consumers, and that SRAH makes necessary new means of transmitting not t e c h n i c a l information, but value information. With regard to engineers employed by producer corporations, I suggested i n Chapter 5 that acceptance of SRAH implies changes i n the way professional engineers view t h e i r task as experts, and changes i n the way engineering as a profession r e l a t e s to the larger community. With regard to governmental regulatory bodies, I suggested i n Chapter 6 that SRAH casts doubt upon the p o s s i b i l i t y of d e p o l i t i c i z i n g debates over r i s k , and implies strong caveats f o r those who seek to base public decisions on theories of i n d i r e c t consent. What can be said, then, by way of summary regarding the o v e r a l l impact of seeing risk-assessment as a subjective a c t i v i t y ? One general conclusion which presents i t s e l f upon consideration of the implications suggested i n previous chapters i s that r i s k assessment must increasingly be a public a c t i v i t y . Despite ample room fo r i n d i v i d u a l decision making, most safety decisions continue to be made c e n t r a l l y , either by corporations, by consumer groups, or by government agencies. The i n e v i t a b l e p o l i t i c a l aspect of r i s k assessment was f i r s t mentioned i n Chapter 3 above, but i t i s a recurring theme i n subsequent chapters. Bureaucrats, corporations, and consumer advocates must evaluate product safety i n l i g h t of the r e f l e c t i v e values expressed by r e a l consumers. The s u b j e c t i v i t y of r i s k assessment implies that there are no experts i n deciding what r i s k s are reasonable. Perhaps the very f a c t that there are no such experts i s what makes product safety a debate with which a l l consumers must be concerned. Value determination i s a s o c i a l a c t i v i t y . Certainly, values can be said to be p r i v a t e l y established and held. The values which i n d i v i d u a l s hold often c o n f l i c t with those values espoused by society as a whole. But even p r i v a t e l y held values evolve within a s o c i a l context. Consumers must a c t i v e l y engage i n public debate over product r i s k s as i f t h e i r l i v e s depend on i t , because they do. Product safety touches us a l l . William W. Lowrance writes: "Between the two a c t i v i t i e s — measuring r i s k and judging safety — l i e s a discomforting no-man's-land...or every-man's-land.""' Lowrance, 79. Bibliography Baier, Annette. "Poisoning the Wells," i n Douglas MacLean, ed. Values at Risk. Totowa, NJ: Rowman & Allanheld, 1986, Barber, Bernard. Informed Consent i n Medical Therapy and Research. New Brunswick, NJ: Rutgers UP, 1980. Bayles, Michael D. P r i n c i p l e s of Law: A Normative Analysis. Dodrecht, Holland: D. Reidel Publishing, 1987. Professional Ethics. 2nd ed. , Belmont, CA: Wadsworth Publishing, 1989. Bexon, Howard. "Professional Engineers i n the Manufacturing Industry," i n Carson Morrison and P h i l i p Hughes, eds.. Professional Encfineerinq Practice; E t h i c a l Aspects. Toronto: McGraw-Hi11, 1988. Brunk, Conrad. "Professionalism and Resp o n s i b i l i t y i n the Technological Society," i n Deborah C. Poff and Wilfred J. Waluchow, eds. Business Ethics i n Canada. Scarborough, ON: Prentice-Hall, 1987. Brunk Conrad, Lawrence Haworth, and Brenda Lee. Value Assumptions i n Risk Assessment; A Case Study of the Alachlor Controversy. Waterloo: Wilfred Laurier UP, 1991. Cohen, L. Jonathan. "Can Human I r r a t i o n a l i t y be Experimentally Demonstrated?" The Behavioral and Brain Sciences 4 (1981): 317-70. Dawes, Robyn M. Rational Choice i n an Uncertain World. San Diego: Harcourt Brace Jovanovich, 1988. Diaconis, P e r s i and David Freedman, "The Persistence of Cognitive I l l u s i o n s , " The Behavioral and Brain Sciences 4 (1981): 333-4. Douglas, Mary and Aaron Wildavsky. Risk and Culture: An Essay on the Selection of Technoloqical and Environmental Danqers. Berkeley: U of C a l i f o r n i a P, 1982. Fischhoff, Baruch. "Psychology and Public P o l i c y : Tool or Toolmaker," American Psychologist 45.5 (May 1990): 647-53. Rev. of Using Surveys to Value Public Goods: The Continqent Valuation Method, by Robert Cameron M i t c h e l l and Richard T. Carson. Public Opinion Quarterly 54 (1990): 286-9. Fischhoff, Baruch, Robert Kates. Lefrere, eds. UP, 1980. Fischhoff, Baruch, Sarah Lichtenstein, Paul Slo v i c , Stephen L. Derby, and Ralph L. Keeney, Acceptable Risk. Cambridge: Cambridge UP, 1981. Gardenfors, Peter and N i l s - E r i c Sahlin. "Introduction: Bayesian Decision Theory — Foundations and Problems," i n Peter Gardenfors and N i l s - E r i c Sahlin, Decision, P r o b a b i l i t y , and U t i l i t y : Selected Readings. Cambridge: Cambridge UP, 1987. Harris, J e f f r e y E. Deadly Choices: Coping with Health Risks i n Everyday L i f e . New York: Basic Books, 1993. Herkert, Joseph R. "Management's Hat Tric k : Misuse of 'Engineering Judgment' i n the Challenger Incident." Journal of Business Ethics 10 (1991): 617-21. Holsworth, Robert D. Public Interest Liberalism and the C r i s i s of Affluence. Boston: G.K. H a l l & Co., 1980. Johnson, F. Reed, and Ann Fisher, V. Kerry Smith, William H. Desvousges, "Informed Choice or Regulated Risk?" Environment. 30.4 (1988): 12-15 and 30-35. Kaplan, Morris. "Does the 'Informed Consumer' Exist? W i l l He Ever?" i n Product Quality. Performance, and Cost. National Academy of Engineering. Washington, D.C., 1972. Keeton, Robert E. "The Basic Rule of Legal Cause i n Negligence Cases", i n Joel Feinberg and Hyman Gross, eds.. Philosophy of Law. Encino, CA: Dickenson Publishing Co., 1975. Kernaghan, Kenneth and John W. Langford. The Responsible Public Servant. Halifax, NS: The In s t i t u t e f o r Research on Public Policy, 1990. Lowrance, William. Of Acceptable Risk: Science and the Determination of Safety. Los Altos, CA: William Kaufman Inc, 1976. Luce, R. Duncan and Howard R a i f f a . "Individual Decision Making Under Uncertainty," i n Decision. P r o b a b i l i t y and U t i l i t y ; Selected Readings^ Peter Gardenfors and N i l s - E r i c Sahlin eds. Cambridge: Cambridge UP, 1988. Reprinted from Luce and Rai f f a ' s Games and Decisions. John Wiley & Sons, Inc., 1957. Christopher Hohenemser, Roger Kasperson, and "Handling Hazards." i n Jack Dowie and Paul Risk and Chance. Milton Keynes: The Open MacLean, Douglas. "Risk and Consent; Philosophical Issues f o r Centralized Decisions." i n Douglas MacLean, ed. Values at Risk. Totowa, NJ: Rowman & Allanheld, 1986. Martin, Mike W. and Roland Schinzinger. Ethics i n Engineering. 2nd ed. New York: McGraw-Hill, 1989. McCarry, Charles. C i t i z e n Nader. New York: Saturday Review Press, 1972. Nadel, Mark V. The P o l i t i c s of Consumer Protection. Indianapolis: Bobbs-Merrill, 1971. Ontario Law Reform Commission, Report on Products L i a b i l i t y . 1979: Ministry of the Attorney General. Rapp, Rayna. "Heredity, or: Revising the Facts of L i f e . " Plenary Lecture f o r the Canadian Anthropology Society and the Canadian Association for Medical Anthropology. May 1994. Vancouver, B.C. Rescher, Nicholas. Risk: A Philosophical Introduction to the Theory of Risk Evaluation and Management. 1983. Sapolsky, Harvey M. "The P o l i t i c s of Product Controversies," i n Harvey M. Sapolsky, ed. Consuming Fears: the P o l i t i c s of Product Risks. New York: Basic Books, 1986. Slovi c , Paul and Baruch Fischhoff. "How Safe i s Safe Enough?" i n Jack Dowie and Paul Lefrere, eds.. Risk and Chance: Selected Readings. Milton Keynes: The Open UP, 1980. Stevenson, John. Personal communication, August 11, 1994. Vaughan, Elaine. Some Factors Influencing the Nonexpert's Perception and Evaluation of Environmental Risks. New York: Garland, 1990. Weinberg, A l v i n M. "Science and i t s Limits: The Regulator's Dilemma." Issues i n Science and Technology. 2.1 ( F a l l 1985): 59-72. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0087682/manifest

Comment

Related Items