Ethics: The Proceedings of the Twentieth World Congress of Philosophy [1, 1 ed.] 1889680052, 9781889680057

Citation preview

CONTENTS

VII

S ERIES I NTRODUCTION American Organizing Committee

XI

VOLUME I NTRODUCTION Klaus Brinkmann

1

MORAL P LURALISM WITHOUT MORAL R ELATIVISM Alasdair MacIntyre

9

PLURALISM AND THE MORAL MIND Hung-Yul So

19

WHY E THICS IS PART OF PHILOSOPHY: A PLEA FOR A P HILOSOPHICAL E THICS Stephen Darwall

29

MORAL T HEORY AND THE REFLECTIVE LIFE Stuart Rosenbaum

41

FEMINIST ETHICAL T HEORY Virginia Held

51

MORAL THEORIES AND VIRTUE ETHICS Michael Slote

59

CAN A P ARTICULARIST LEARN THE DIFFERENCE BETWEEN RIGHT AND WRONG? Jonathan Dancy

73

PRACTICAL ETHICS AND MORAL OBJECTIVISM Margarita M. Valdés

83

PRINCIPIOS DE EQUIDAD DISCURSIVA Ricardo Maliandi V

C ONTENTS VI

95

D ISCOURSE -E THICAL G RADUALISM : BEYOND ANTROPOCENTRISM AND BIOCENTRISM? Gunnar Skirbekk

107

MORAL K NOWLEDGE AND LINGUISTICS Gilbert Harman

117

EXPLANATION AND J USTIFICATION IN MORAL E PISTEMOLOGY Walter P. Sinnott-Armstrong

129

THE V ALUE OF MORAL RESPONSIBILITY John Martin Fischer

141

PHILOSOPHY AND ECOLOGY John Passmore

151

NATURE AND CULTURE IN ENVIRONMENTAL ETHICS Holmes Rolston, III

159

DEPTH, TRUSTEESHIP AND REDISTRIBUTION Robin Attfield

169

BIOTECHNOLOGY AND THE ENVIRONMENT: FROM MORAL OBJECTIONS TO ETHICAL ANALYSES Matti Häyry and Tuija Takala

179

BEYOND BIOPHOBIC MEDICAL ETHICS Jorge L. A. Garcia

189

DEATH, DYING, AND DIGNITY Felicia Ackerman

203

MORALITY AND HEALTH CARE POLICY Bernard Gert

215

ETHICAL ISSUES IN THE CONSTRUCTION OF COST-EFFECTIVENESS ANALYSES FOR THE PRIORITIZATION AND RATIONING OF HEALTHCARE Dan W. Brock

231

ETHICS IN BIG SCIENCE Russell Hardin

247

ABOUT THE C ONTRIBUTORS

251

NAME INDEX

VOLUME INTRODUCTION Klaus Brinkmann

B

y conventional standards, the papers collected in this first volume of invited contributions to the Twentieth World Congress of Philosophy fall roughly into the two broad categories of moral theory and applied ethics. However, a closer look will reveal that there is nothing conventional about them. Not only do some of the applied ethics papers challenge some widely acknowledged positions in bioethics (Jorge L. A. Garcia, Felicia Ackerman, Gunnar Skirbekk) or suggest a new framework for the reconciliation of nature and culture in environmental ethics (Holmes Rolston, III), the papers in moral theory broadly conceived show a remarkable preparedness to rethink traditional boundaries, positions, and methodological approaches altogether. Thus the traditional boundaries between normative ethics and metaethics come under pressure once it is realized that, for instance, normative disagreements concerning different conceptions of the good life presuppose an agreement to the effect that the good life is indeed something we ought to strive for. Based on observations like these, it is possible to develop a powerful argument for the reintegration of normative ethics with metaethics (Stephen Darwall). Again, the divisions within moral theory become less important, not to say irrelevant, when the established normative positions seem to lose their universalistic standing in a confrontation with culturally based ethical pluralism. One may then even go so far as to speak of the “demise of moral theory” due to its inability to defend an unambiguous notion of morality (Stuart Rosenbaum). XI

V OLUME INTRODUCTION XII

These moves suggest that a new distinction may be opening up between moral theory on the one hand and moral philosophy on the other, where moral philosophy designates the attempt to formulate a new paradigm for what moral theory is about. To take a conspicuous example for this kind of reorientation, it is being claimed that the various efforts in feminist ethics have led to the emergence of a fundamentally new approach to moral theorizing which not only opposes the paradigm of an ‘ethics of justice’ with an ‘ethics of caring’, but also argues that the former needs to be overcome in favor of the latter (Virginia Held). Others are drawing the consequences from the perceived failure of the established normative positions, such as utilitarianism or deontological ethics, to either demonstrate convincingly the existence of moral universals or show their successful application in concrete circumstances by making a case for ethical particularism (Jonathan Dancy) or moral objectivism (Margarita M. Valdés). Meanwhile, attempts are being undertaken in moral epistemology to formulate an empirical procedure for identifying moral universals analogous to the already existing method of establishing the existence of linguistic universals (Gilbert Harman). Not only, then, is there an effort underway to rethink the boundaries, goals, and fundamental orientations in ethics, as the following brief summaries will show more clearly, there is also a simultaneous effort to reconstitute the established positions in moral theorizing and to put them on a more secure or more promising basis. One of the most pressing issues in contemporary ethics is the need for developing a convincing response to moral relativism. The increased awareness of the fact that different cultures live by different value systems has put the universalistic positions in normative ethics on the spot, so to speak. Indeed, both the emergence of non-universalistic positions such as particularism and the call for a moral philosophy beyond traditional normative ethics can be viewed as a reaction to this challenge. One strategy for salvaging universalistic claims has been to limit the unconditionally normative values or moral principles to a minimum. However, in this approach cultural diversity is lost, not preserved. In “Moral Pluralism without Moral Relativism” Alasdair MacIntyre sketches a novel response to the dilemma between moral universalism and moral relativism. The danger of relativism, he points out, is precisely that it encourages us to ignore, rather than begin to appreciate and learn from, the different value systems which form part of the identity of other cultures. If we do not want to lose the possibility for enriching our own understanding of human existence, we must acknowledge that diverse value systems give expression to a particular conception of human nature and

V OLUME INTRODUCTION XIII

the role morality ought to play in our lives. These diverse conceptions of human nature are inherently normative and universalistic, MacIntyre argues. They therefore contain within themselves a commitment to truth which transcends any particular morality. It is this commitment to truth as a universal good which can underwrite what MacIntyre calls an “ethics of inquiry” on the basis of which a fruitful exchange between different moral orientations may take place, an exchange which may even lead to a critical questioning of our own moral culture and thus prevent us from remaining “imprisoned within our own standpoint.” One may want to compare this proposal with Hyung-Yul So’s suggestion, in his “Pluralism and the Moral Mind,” that the moral systems of different cultures should be viewed as expressions of different types of rationality which nonetheless together constitute the unity of human reason. So in fact undertakes a detailed hermeneutical study of how different styles of rationality (deductive, inductive, abductive, dialectical, analogical, and pragmatic) inform the moral outlook and the fundamental moral attitudes of some of the major world cultures. As an aside, it is interesting to note that So’s approach bypasses the appeal to religion in order to explain the ethical diversity of different cultures. Stephen Darwall’s “Plea for a Philosophical Ethics”—this the subtitle of his paper entitled “Why Ethics is Part of Philosophy”—questions the by now historical dissociation between normative and metaethics. Taking core concepts of some of the traditional positions in normative ethics such as normativity, value, obligation, pleasure, or virtue, he argues that no serious work within deontological, utilitarian, or virtue ethics, to name only these, can be done without achieving at least some metaethical consensus about which particular norms constitute values or which character traits should be regarded as virtues. It would therefore be pointless or even fundamentally erroneous to want to “seal off” a normative position from its metaethical presuppositions. Metaethics might thus very well become the Grundlegung for normative ethics. Stuart Rosenbaum’s statement referred to above concerning the “demise of moral theory” in his contribution “Moral Theory and the Reflective Life” naturally prompts the question: What comes after moral theory? Rosenbaum’s reply is: a moral philosophy which focuses on concrete forms of life—paradigms of the good life, one might say—which can function as models of an ideal life form. Rosenbaum sees his proposal as another response to “a growing tendency to reject the idea of a commensurability of all values” and as a way of coming to terms with the pluralism of ethical orientations. His paradigm of an ideal life form is Dewey’s concept of the reflective life and its emphasis on “reflective autonomy,” but he sees similar intentions at work elsewhere, for instance

V OLUME INTRODUCTION XIV

in Annette Baier’s discussion of the pluralism of values in Hume or in Martha Nussbaum’s interpretation of Aristotle’s ethical vision. Perhaps the most radical rethinking of normative ethics comes from a feminist ethics which opposes traditional positions such as Kantian or utilitarian ethics with a new ethical paradigm. As Virginia Held argues in “Feminist Ethical Theory,” while non-feminist ethics is guided by values such as autonomy, self-determination, rights, independence, and abstract universality, feminist ethics favors values such as interdependence, mutual trust, empathy, emotional ties, and community. This new ethical paradigm should not be reduced to an ethics of care (but one might perhaps call it an ethics of caring for others). It does, however, deliberately “challenge the rationalistic and individualistic assumptions” guiding the dominant moral theories as well as the idea of an “impartial morality” in which moral judgments ought to resemble those of a neutral judge in a court of law. A “feminist moral theory” built on the principle of caring should eventually incorporate and subsume the established “ethics of justice,” or so it seems. Since both virtue ethics and the ethics of care are focused on moral practices rather than abstract principles, and since both give a role to the emotions no less than to rational deliberation, the two positions might be good candidates, Held seems to imply, with which to begin the task of integrating the existing normative orientations into the new caring-based model of moral philosophy. How newly evolving positions in normative ethics may affect our assessment of historical positions and give rise to new combinations can also be seen in Michael Slote’s “Moral Theories and Virtue Ethics.” Slote takes Hutchinson’s “hybrid sentimentalist ethics” as an example and shows how it may be transformed into either a non-hybrid utilitarianism without a consequentialist commitment or a non-hybrid virtue ethics. He then goes on to explain how virtue ethics might complement a “feminine” ethic of caring in such a way as to inject a broader humanitarian concern into it. A more broadly humanitarian ethics of caring might have a chance of becoming “a complete approach to morality.” Due to the centrality of the principle of caring as an inner motivational state it could also be described as, or indeed be part of, a virtue ethics. That the universalistic approach in normative ethics is indeed under pressure, if not on the defensive, is also apparent from Jonathan Dancy’s contribution entitled “Can a Particularist Learn the Difference Between Right and Wrong?” Once one has come to the conclusion that there are no moral universals such that particular reasons or actions could be justified by having recourse to them, ethical particularism offers itself as an alternative. However, for particularism to work it needs

V OLUME INTRODUCTION XV

to counter the objection that there are no descriptive features of actions which uniquely identify one action as right and another as wrong—with the undesirable consequence that distinguishing right from wrong would then become a matter of Moorean intuition. The charge from the anti-particularist camp then is that there exists no “pattern to the way in which descriptive information determines moral conclusions.” Dancy’s paper is intended to rebut this criticism by pointing out that a rigid correlation between the concept of a right action on one hand and one or more specific “right-making” descriptive features of that action on the other is unnecessary. Depending on the circumstances one or another feature may be advanced as being the reason, rather than merely the identifying feature, of a right action. All that particularism denies, therefore, is the existence of “a descriptive specification” of the right-making properties, not the existence of such properties. Because evaluative or normative features can thus be linked to descriptive features of actions, the particularist is indeed capable of knowing—and defending—the difference between right and wrong. According to Dancy, therefore, the semantic argument against particularism fails. Dancy then turns to a presentation of two models of particularist rationality, neither of which, he argues, is vulnerable to the semantic criticism. In “Practical Ethics and Moral Objectivism,” Margarita M. Valdés comments on another perceived demise, in this case that of a cornerstone of metaethics, viz., the fact-value distinction. With its collapse, she argues, moral subjectivism has lost its footing and moral objectivism, which claims that morality has an objective basis in moral facts and that moral judgments can be true or false, has been able to gain ground. Relying on Simon Blackburn’s criticism of John Mackie’s critique of moral objectivism she examines ways of avoiding Mackie’s conclusions. Both Ricardo Maliandi (“Principios de Equidad Discursiva”) and Gunnar Skirbekk (“Discourse-Ethical Gradualism: Beyond Anthropocentrism and Biocentrism?”) develop their arguments from a discourse-ethical position which, however, they also see in need of emendation. Maliandi suggests that the so-called Alexy-Habermas principles of discursive equality lack a provision for maintaining discursive fairness under circumstances of real and ineliminable non-discursive asymmetry among discourse participants and proceeds to propose such balancing principles drawing on Rawls’s principles of justice. Skirbekk believes that a version of discourse ethics can help us avoid the dilemma we face in trying to evaluate ‘borderline’ cases between humans and nonhumans in bioethics. Skirbekk shows that the usual paradigmatic approach does not allow us to accord nonhumans any right to our moral concern while the assumption of a fundamental continuity between non-humans

V OLUME INTRODUCTION XVI

and humans—which he calls “ontological gradualism”—is in danger of losing morally relevant distinctions. The solution offered consists in a revised version of discourse ethics which accords moral responsibility to moral agents only but extends the moral universe to all sentient beings, including nonhumans, as moral subjects on the grounds that discourse ethics must recognize a biological foundation (such as embodiment and vulnerability) as part of the competence requirements for being a moral discussant. Gilbert Harman’s “Moral Philosophy and Linguistics” opens a group of three papers dealing with metaethical questions. Harman proposes a new method for discovering moral universals. On the assumption that there exists a parallel between the acquisition of linguistic competence on one hand and moral knowledge on the other, one might try to uncover something like “a universal grammar of morality.” The actual method described by Harman in the sequel of the paper must meet the challenge that just like linguistic universals their moral counterparts represent universal constraints on moral reasoning that must not have been learned through explicit instruction, but instead must have been picked up “ahead of time.” Harman quite ingeniously gets around this difficulty by assuming the existence of “moral idiolects” whose default rules would indicate the existence of underlying moral universals, if indeed these turned out to be universally valid. In a response to Harman’s The Nature of Morality and the discussion about moral explanations ensuing in its wake, Walter Sinnott-Armstrong presents an argument for a “limited version of moral skepticism” (“Explanation and Justification in Moral Epistemology”). Assuming that there are no moral facts, moral explanations of observations concerning behavior that can be morally adjudicated must be relative to contrast classes. However, if the contrast class is unlimited—and thus includes moral nihilism—no moral explanation can meet the necessary independence requirement. It follows that “moral explanations work for some limited contrast classes but not for an unlimited contrast class.” Ethicists often assume that in order to hold agents morally responsible it must be possible to attribute regulative control over their actions to them. To have regulative control over one’s actions means that one could have done otherwise or that one could have prevented something from happening, even if one actually failed to do so. As John Fischer tries to demonstrate in “The Value of Moral Responsibility,” the “freedomrelevant” (as opposed to the epistemic) component of moral responsibility is satisfied by a less comprehensive kind of control which he calls guidance control. This is control of how one reacts to or bears events or how one influences or shapes actions for which no alternative seems

V OLUME INTRODUCTION XVII

available or accessible. As Fischer shows, moral responsibility requires guidance control, but not regulative control. However, what at first might seem to be merely a technical point, turns out to be quite an important consideration for the moral evaluation of a human life as a whole. Clearly we do not have absolute regulative control over how our lives play out, whether they go, say, from rags to riches or from riches to rags. How a moral and prudential evaluation of such biographical ‘narratives’ is nonetheless possible on the guidance control model is worked out persuasively in the second half of Fischer’s paper. John Passmore’s “Philosophy and Ecology” takes the reader from moral theory in the traditional sense to environmental ethics. Passmore reviews a number of difficulties which this branch of philosophy—he calls it environmental philosophy, not ethics—has had to face since its inception, starting with the question as to its scope and position with regard to other philosophical disciplines and continuing with the as yet unresolved normative issues concerning, among other things, a possible justification of our obligations, if any, towards nature (on this last point, however, one might again want to consult Gunnar Skirbekk’s contribution referred to above). Obviously, a mandate to preserve the environment cannot be based on something like rights on the part of nature. Is such preservation then a matter of purely pragmatic prudence on our part? Passmore tentatively suggests that it might be a plausible strategy to invest nature, including rocks and plants, with noneconomic value, but he is quick to point out that such value would still be in conflict with other kinds of value, be they economic, cultural, or aesthetic. There does not seem to be an independent basis for adjudicating between the relative importance of these values. A similar problematic is at issue in Holmes Rolston III’s “Nature and Culture in Environmental Ethics.” However, Rolston broadens its scope by viewing the relationship between humans and nature as one of nature and culture. Moreover, he understands this relationship as an irreversible mutual interdependence, a global symbiosis that is here to stay (unless, of course, we inflict lethal damage on our life-sustaining biosphere). Consequently, he rejects the idea of ‘assimilating’ culture to nature—culture is, after all, not an “emanation of Nature.” The two form the foci of an ellipse which should not be collapsed into a circle. The relationship is dialectical and complementary. A restitution of nature to its pristine state is impossible, nor can culture remain opposed or indifferent to nature. Environmental ethics will have to view nature as a managed environment in which human culture may continue to feel at home. However, whether the managerial approach to nature is indeed the right approach has itself been a matter of some debate. In “Depth,

V OLUME INTRODUCTION XVIII

Trusteeship, and Redistribution” Robin Attfield develops a defense of the trusteeship model against its recent critics. Among other things, this model has the advantage of not being anthropocentric. But the suggestion that man should be the steward of nature naturally prompts the question as to who in particular should be a trustee. Humanity as such does not function as an agent. Responsibility as a trustee must therefore fall to individual societies. Ultimately, however, the burden of stewardship must be born by those societies which are economically healthy and in a position of political leadership, Attfield argues. It will be their task to empower the less developed societies and thus to broaden the basis of trusteeship by minimizing inequalities of power and resources. It is at this point that environmental ethics begins to reach the level of practical consequences and to show its potential for affecting environmental policy decisions. While the scientific debate about how much of the change in the environment is anthropogenic continues, biotechnology is a case where such change is being deliberately engineered. In “Biotechnology and the Environment: From Moral Objections to Ethical Analyses” Matti Häyry and Tuija Takala comment critically on the fact that in their view philosophical ethics has so far failed to be very helpful in giving concrete answers to the question of how to respond to the potential dangers of releasing genetically altered organisms into the environment. Even consequentialism, they argue, is ill equipped to address such issues, since the consequences in this case are typically unpredictable. How then are we to overcome the stage of moral objections based primarily on emotional responses? The authors suggest a risk based assessment of the possible negative consequences which would differentiate between controllable and uncontrollable risks and take into account irresponsible or accidental release of altered material. Whether or not risks are acceptable should be determined by those who would potentially be affected by the consequences. Jorge L.A. Garcia’s “Beyond Biophobic Medical Ethics: What’s the Mercy in Mercy-Killing?” opens a group of four contributions to bioand medical ethics. As the title of his paper indicates, Garcia doubts that we can make sense of the concept of mercy-killing. Indeed, the idea, he maintains, represents an ethical perversion instead of a principle of beneficence as which defenders of assisted suicide and euthanasia usually regard it. The article points out the self-contradictory nature of some of the arguments offered in support of mercy-killing. The deeper roots for what Garcia calls a “degraded intellectual culture” which would allow such perversions are in his eyes the lack of a genuine bioethics, i.e., one “fiercely devoted to human life,” and the tendency to see life in

V OLUME INTRODUCTION XIX

purely instrumental terms. The issue of assisted suicide is further discussed in Felicia Ackerman’s “Death, Dying, and Dignity.” Ackerman subjects the idea that assisted suicide may be a means to maintaining a person’s dignity to critical scrutiny. It would be crucial to determine what does and does not affect or impair human dignity. What, if anything, about a sick or even a terminally ill person’s life is it that degrades his or her existence? Is it the person’s dependence on others, the suffering of pain, a lowering of self-esteem? While these are some of the answers that have been given, Ackerman is not convinced that they are appropriate answers at all. The question we ought to ask ourselves first would rather be: What does it show about our culture that old age, illness and disability are perceived as reducing a person’s dignity? As we saw earlier, a number of contributions question the effectiveness or even the capacity of moral theory to provide solutions to concrete moral issues, partly because of a perceived lack of universal moral standards, partly due to the abstractness of those theories. This situation prompted the emergence of new types of ethical theories such as particularism, moral objectivism, or the call for a new type of moral philosophy altogether. In his contribution “Morality and Health Care Policy,” Bernard Gert challenges what he refers to as the “standard moral theories” (i.e., Kantian-type duty ethics, consequentialism, and Rawlsian contractarianism) from yet another angle by drawing attention to the difference between ethical theory on one hand and our actual moral practices on the other. If we look at moral decision making in an everyday context, say decisions about the allocation of scarce resources in health care, we will see that those decisions are made within the context of what Gert calls “common morality,” an informal public system of moral evaluation(informal because there is no authority to proclaim binding decisions about what is morally right or wrong). In this system, the unique-right-answer assumption of the moral theorist and his idea of the fully informed, impartial and uniquely rational agent (what one might call the ‘ideal agent’ assumption) are not applicable. Gert suggests that not only are these idealizing assumptions mistaken when it comes to actual moral decision making, they are even dangerous, because they obscure the reality of irresolvable moral disagreements and tend to discourage negotiation and compromise. Consequently, the role of moral theory in ‘real life’ should be much more modest than the moral philosopher might think, that is, it should be limited to the negative role of ruling out morally unacceptable policies rather than prescribing solutions. As Gert notes in his paper, one of the difficulties in reaching fair and just decisions in a distributive system such as health care is the problem of how to quantify personal harm with a view to minimizing inequality in

V OLUME INTRODUCTION XX

the distribution of benefits. Dan W. Brock (“Ethical Issues in the Construction of Population Health Measures for the Prioritizing and Rationing of Health Care”) addresses just this problem. The need to factor quality of life assessments into the calculation of cost effective procedures or policies and the necessity to compare the respective quality of life of disabled versus non-disabled persons make considerations of equity not only complicated but also precarious. Other problems arise from basing equity considerations on either a social or an individual approach. Moreover, if life expectancy is a factor in providing a health care measure, it would matter at what stage of life the measure is being provided (a point reminiscent of John Fischer’s argument referred to above). Brock discusses these and other matters in detail, including the question of how to evaluate the benefits of vaccination programs and of minimizing sick days, examining the mathematical models used in calculations of cost effectiveness and highlighting the unresolved moral issues at every turn. His article is a reminder that measures concerning distributive justice and equity are by no means merely technical or economic, but involve serious moral decisions and value choices, down to the question of how to construct mathematical models for calculating quality adjusted life years (QALYs) and disability adjusted life years (DALYs). The volume concludes with a paper addressing the ethical responsibility of the scientist. In “Ethics in Big Science,” Russell Hardin tries to find a new answer to the question of how scientists might be held accountable for upholding the ethical standards of their profession. Existing models such as self-policing may work better in some scientific communities than in others. To a certain extent, one may rely on the scientists self-interest: career prospects usually hinge on “getting it right,” i.e., remaining committed to a scientific idea of truth and professional verifiability. But how are we to treat the case of scientists with a conflict of interest, either perceived or real? Does working for some industry invalidate certain findings? A strict rule against conflicts of interest might have the unwelcome consequence of a loss of outside funding for the academic researcher. Scientists have mostly failed to create working models for professional self-regulation, Hardin concludes, and now they may have waited too long to preempt direct public (political) intervention. Perhaps a self-regulatory model within science can be developed such that a formal regulatory system outside a scientists’ own research arena would act as monitoring body.

MORAL

When we deny the truth of someone else’s moral beliefs and give our

PLURALISM WITHOUT MORAL RELATIVISM

grounds for so doing, we make or imply judgments about the inadequacy of their reasons for belief and about the causes of their belief. And we presuppose a difference between them and us in both respects. In so doing we provide matter for a shared philosophical inquiry about the rel-

Alasdair MacIntyre

evant types of reason and cause. It is a mark of rational disagreement on matters of serious moral import that we who so disagree should be prepared to engage in this inquiry and to recognize its standards as binding on us unqualifiedly. This recognition commits us to a denial of moral relativism. Some of these best examples of rational disagreement are found in some, although only some, of the exchanges between medieval Islamic, Jewish and Christian philosophers.

T

wo of the challenges presented by this World Congress are: first, to show how far rational dialogue between deeply incompatible and conflicting points of view is possible; and, secondly, to achieve this in not much more than three thousand words. This is a necessarily compressed and inadequate attempt to do both. So where to begin? Each of us has to begin where we find ourselves, typically sharing some particular moral culture that our own society has inherited. That moral culture provides rules that structure our relationships. It presents us with conceptions of goods to be achieved and of virtues to be cultivated. And it is the necessary starting-point for critical reflections on how we ought to live. One stimulus to reflection is the discovery that there are 1

A L ASDAIR MAC I NT YRE 2 other different and rival moral cultures whose practice is informed by conceptions of rules, goods and virtues that differ significantly from our own, that sometimes indeed are such that, if their view of things is in the right, then in important respects ours is mistaken. Consider some examples that support this conclusion. In the presently hegemonic moral culture of the United States concepts of individual rights and of the maximization of the satisfaction of individual preferences are central to public discourse. But the standards articulated by means of those concepts are alien and unacceptable to many cultures. Correspondingly, secularized post- Enlightenment moral cultures are generally dismissive both of appeals to custom and tradition and of moralities in which some notion of divine law has a central place. And even over concepts that are important in all or almost all moral cultures conflict occurs. Notoriously there are different and irreconcilable conceptions of distributive justice, a justice that makes desert its standard, a justice that allocates in accordance with entitlements, a justice that requires equal distributions. Honor by contrast is an important concept in some moral cultures, but not in others. And catalogues of the virtues also vary, so that, for example, Confucian catalogues of the virtues include items—notably the virtue of ritual propriety—which other catalogues exclude or ignore. To attend to this range of differences—and it is astonishing how much of Western moral philosophy, unlike anthropology or sociology, has refused to take other than passing notice of them—is to become aware of two possible responses. One is to be resolutely dismissive of everything that is incompatible with the standpoint of one’s own moral culture. The standards internal to that culture become the standards by which other cultures too are to be judged. So most ancient Greeks were dismissive of non-Greeks as barbarians, just as most Chinese of the Ming and Manchu were similarly dismissive of Europeans. So nineteenth and early twentieth century Europeans drew a line between the civilized and the savage designed to confirm them in their belief in their own moral superiority. What this type of stance precludes is the possibility of learning anything of substance from rival and incompatible moral cultures. But so equally does another at first sight very different stance, that of a certain kind of relativism. This kind of relativist begins by noting correctly that all the attempts of Enlightenment and post-Enlightenment thinkers to identify a universal and neutral standard, one compelling to all rational agents whatever their culture, by appeal to which the competing claims of rival standpoints could be rationally adjudicated, have failed. The relativist then further notes that each particular moral standpoint has its own standards of justification and its own modes of justification internal to it. From these two starting-points the relativist concludes that from the standpoint of each

MOR AL P L UR ALISM W ITHOUT M OR AL R EL ATIVISM ORAL URALISM ORAL ELA 3 particular moral culture its adherents have good reason to believe that they are in the right in their disputes with the adherents of other rival moral cultures and that there is no external rationally justifiable standard by appeal to which they could be shown to be mistaken. But it follows, just as it did for those who dismissed others as barbarians or savages, that one moral culture cannot learn from another. Should we accept this disquieting conclusion? Both stances, that of those who reject out of hand rival types of moral belief and practice and that of the relativist, leave out of account some crucial features of moral cultures and of their disagreements and conflicts. One of these is a matter of what is involved in denying what the adherents of rival moral cultures assert. Every major moral standpoint, both in its practical judgments and in its justificatory arguments, presupposes—how far this presupposition is made explicit varies from culture to culture—some distinctive account of human nature and activity and of how human nature and activity are such that morality is whatever the adherents of that particular standpoint hold that it is. So the claim to authority advanced in respect of particular local moral cultures is never merely local. It is never merely the claim that this is how we Athenians democrats or we Japanese Confucians or we Shi’ite Iranians happen to live; it is always the claim, implicit or explicit, carefully qualified or quite unqualified, that ours is the best way for human beings to live. And a crucial ground for this claim is the further claim that our account of human nature and activity is by and large true, from which it follows that, insofar as rival views are grounded in an incompatible account of human nature and activity, those views fail, because their account of human nature is false. To say of our view of human nature and activity that it is true and that rival and incompatible views advanced by others are false is of course to say more than that their views are inconsistent with our view, that they fail by our standards of truth and of rational justification. For that is something uncontroversial, something to which all parties to the disagreements could agree without difficulty. The further controversial claim is that what is judged true by our standards is unqualifiedly true. And this claim is crucial in two ways. First, and less importantly, it reveals the incompatibility of any coherent relativism with the beliefs and practices of all the major moral cultures and in so doing it suggests that relativism, although it does not entail moral scepticism, does entail a rejection of the standpoints of all those cultures. For the only judgment taken by moral relativism to be unqualifiedly true is the judgment that no judgment advanced from the standpoints of those cultures is unqualifiedly true.

A L ASDAIR MAC I NT YRE 4 Secondly, it reveals the extent of the philosophical commitments presupposed by, although often not always accorded explicit recognition by, the adherents of those cultures. For in judging that their own basic affirmations are true and their own justificatory arguments sound, they are also committed to judging that those arguments, by appeal to which the adherents of other rival standpoints purport to show that some of their affirmations are false and some of their arguments unsound, fail. And these are of course philosophical and not only moral claims. So philosophy itself comes on the scene in the form of an invitation to provide sufficiently good reasons for advancing these claims. Philosophy in so doing is no longer merely an external commentator upon, but becomes involved in the constitution of any dialogue between moral standpoints that is able to move beyond the bare rhetoric of assertion and counter assertion. But why should this philosophical invitation be accepted? Why should the adherents of this or that particular moral standpoint choose to move beyond this rhetoric by acknowledging the need to provide reasons for their rejection of incompatible judgments made from rival standpoints? A necessary first step is a recognition by such adherents that importance for others attaches to their own moral judgments, only if and insofar as their presupposed account of human nature and activity is true and that, when they present their own judgments to others as deserving assent by those others, this can only be on the basis of their claim that the account of human nature and activity presupposed by those judgments is true and that all rival and incompatible accounts are false. What are they claiming in claiming this? They are asserting that their understanding both of themselves and of others is not subject to distortion and misrepresentation in the way that the understanding that others have is. They thereby invite those others to radical self-criticism, so that they may identify the difference between how things in fact are and how they have hitherto taken them to be. But this invitation presupposes that truth is a good, independently of one’s own particular moral standpoint, not a good the acknowledgment of which can be independent of all or any standpoints, but a good that is already implicitly acknowledged within the moral practice of any standpoint which in virtue of its claim to truth claims the allegiance of rational individuals. To fail to make explicit this recognition of truth as a good is to deprive claims made on behalf of one’s moral standpoint of the only authority that can successfully legitimate them. But no one’s recognition of truth as a good is adequate, until and unless they have evaluated the strongest arguments that have so far been advanced for conclusions incompatible with their own, that is, until they too have undertaken the tasks of radical self-criticism to which they have invited others. To assert of some

MOR AL P L UR ALISM W ITHOUT M OR AL R EL ATIVISM ORAL URALISM ORAL ELA 5 judgment that it is true is to assert more than that it is able to withstand all attempts at refutation from any standpoint whatsoever. But to assert of any judgment that it is true commits those who assert it to holding that it is so able. And it is a necessary condition of any judgment’s being so able that it has in fact been able to withstand the strongest attempts to refute it so far. It follows that to claim truth for one’s own judgments requires that one should deliberately and systematically expose those judgments to every significant possibility of refutation. Where fundamental disagreement between rival moral points of view is concerned, this involves an act of the philosophical imagination, an imagined shift in one’s vantage-point, so that one understands how one’s own positions appear from the point of view of the other at every level from that of particular episodes of practical judgment and reasoning to that at which basic theoretical claims about practice are made. For without such understanding it will be impossible to appreciate fully the force of those arguments which from the point of view of the other constitute a sufficient refutation of one’s own claims. How is such an imaginative understanding to be achieved? It involves treating the other as a partner in one’s own enquiries, as one from whom much has to be learned, and this type of learning ideally requires that one learn to inhabit the other’s culture, to share, so far as possible, in the other’s practice and so to reason as the other reasons. Much depends here on how hospitable the other is to such enquiry and that in turn depends in part on how far and how effectively the other is invited to participate in the enquiry. What matters most is an acknowledgment that to embark on this type of enquiry requires that one accord authority to the ethics of enquiry and that one therefore find some way of simultaneously giving allegiance both to one’s own moral standpoint and to the ethics of enquiry. What do I mean by the ethics of enquiry? Enquiry has as its goal the achievement of truth through a dialectical development of critical objections to our initial shared beliefs, the discovery of how those beliefs must in consequence be abandoned or revised, and the subsequent development of further objections to our newly revised beliefs until we reach a point at which we have an adequate answer to the strongest objections that anyone has been able to advance from whatever point of view. The beliefs from which we start cannot but be our own; the objections to which we must be open may be those of anyone whatsoever. And this entails that we extend to anyone who is an actual or potential participant in that activity of enquiry the kind of consideration necessary for such participation. What is necessary is that such a one should be free to speak her or his mind in developing whatever arguments are relevant. This freedom is possible only when each participant knows that the others will abide by certain rules and exhibit certain virtues.

A L ASDAIR MAC I NT YRE 6 Those rules must provide security from any threat of bodily or other harm to oneself or to family and friends or to legitimate property. They must warrant the expectation that one will be addressed truthfully and undeviously and trusted to speak truthfully and undeviously. They must provide for justice in conversation, for each contributor to be able to speak in turn and at appropriate length. And they must be directed to the end of enabling and requiring participants to attend to the substance of the arguments and not to who utters them. These rules will have point and purpose to anyone committed to the goods of enquiry. Conformity to them will be part, although only part of what is required by the virtues of enquiry, those qualities of mind and character that are necessary for achieving the good of truth. This set of rules and virtues is not unfamiliar. Versions of them are already ascribed authority in a number of moral cultures. But, if we understand them as constituting the ethics of enquiry, we accord them an additional authority that is independent of moral standpoint, an authority that derives from the fact that without conformity to them we cannot achieve the common good of truth through our enquiries. Here then is the beginning of an answer to the relativist. For it is not just that the adherents of each particular moral standpoint, if they are to give good reason for taking seriously their claims that the beliefs presupposed by their particular standpoint are true, must recognize truth as a good, whatever one’s standpoint, but that the relativists, if they are to give good reason for taking the claim that relativism is true seriously, must also recognize truth as a good, whatever one’s standpoint, and to the extent that they do so they abandon relativism. And since the recognition of truth as a good involves according authority to the virtues and rules that constitute the ethics of enquiry, those virtues and rules too escape the relativistic critique. But this is not the only way in which the ethics of enquiry undermines relativism. I have so far not questioned the relativist’s portrayal of the standpoint of different moral cultures as each having its own standards in such a way that there is no common and neutral standard by appeal to which their rival claims may be adjudicated. But now we do need to put it in question. For when the adherents of some particular moral standpoint find that their claims concerning the truth of their own beliefs have committed them to finding a place for the goods, virtues and rules of an ethics of enquiry within their moral scheme, as they have hitherto understood it, the question is inescapably posed: how well or how badly can that particular set of beliefs and practices accommodate what an ethics of enquiry requires it to accommodate? For it is always possible and often the case that there will be at various points tensions and incoherences between the conceptions of goods, virtues and rules that have hitherto informed the practice of this or

MOR AL P L UR ALISM W ITHOUT M OR AL R EL ATIVISM ORAL URALISM ORAL ELA 7 that particular moral culture and the conceptions of goods, virtues and rules that inform the ethics of enquiry. So problems arise about how to resolve them. Moreover the progress of enquiry may have revealed tensions and incoherences that were already present, but hitherto unnoticed or treated as of no account, within the moral scheme of that particular culture. For either or both reasons that particular moral culture may be unable to avoid becoming to some significant degree an argumentative culture of questions and problems as well as of judgments and affirmations. It is when a moral culture becomes to some significant degree selfquestioning in this way that those who inhabit it may become able to put their own most fundamental standards and principles to the test and in so doing discover what resources are available to them to resolve the problems that are thereby posed. In so doing they will also and incidentally have acquired a new set of standards for evaluating not only their own theory and practice, but also the theory and practice of other rival standpoints. For we are able to compare the relative success or failure of the adherents of each different standpoint in their attempts to resolve through practical and theoretical enquiry the problems internal to their own standpoint and the resources which each possesses for explaining such success and failure. And so the possibility opens up that we may discover not just that our own enquiries in morals and politics have become frustrated and sterile, but that there is some other standpoint from which it can be explained why this failure, given our overall perspective, was inescapable. That is, the possibility has opened up that we may find grounds for judging some other standpoint superior to our own. We are no longer imprisoned within our own standpoint. It may seem to be, but it is not paradoxical, to conclude that the discovery that it is possible for our own particular moral standpoint to be rationally defeated by some rival standpoint is a necessary condition for arriving at a rational vindication of our own point of view. For the strongest vindication that any point of view can receive is that it has so far survived encounters with as wide as possible a variety of other and rival standpoints without suffering such defeat, although it will of course have generally emerged from such encounters modified and revised in various ways. Rational moral enquiry then is always conducted from within some one particular moral standpoint. But insofar as that standpoint has integrated into itself the ethics of enquiry, an ethics that has itself appeared in a number of cultural guises—Jewish, Greek, Confucian, Buddhist, Islamic, and Christian among others—its adherents will become able to enter into a constructive dialogue with the adherents of some other point of view, having discovered that there are standards that they share with those who also engage in enquiry, whatever their point of view. And with

A L ASDAIR MAC I NT YRE 8 this discovery the threat that relativism presented, that we could not learn from each other through such dialogue, is dissipated. We can recognize and give respect to a variety of points of view, so remaining moral pluralists, without becoming moral relativists. So I conclude; but am I in fact entitled to this conclusion? It is important to note that in at least three respects my argument is incomplete. First, I have relied upon, but never spelled out, a particular understanding of the nature of truth, one that is very much at odds with some currently influential theories of truth. Secondly, my account of what I have called the ethics of enquiry is far too brief to be adequate. And thirdly, I have not considered what reply to my argument an insightful relativist might make. So that what I have presented is perhaps a gesture towards an argument, rather than argument, not a conclusion to which I am as yet entitled, but a conclusion to which I might become entitled. But that at least is a beginning. Alasdair MacIntyre, Department of Philosophy, Duke University, Durham, NC 27708

PLURALISM AND THE MORAL MIND

Cultural pluralism has caused disturbing problems for philosophers in applied ethics. If moral sanctions, theories, and applications are culturally bound, then moral conflicts ensuing from cultural differences would seem

So Hung-yul

to be irresolvable. Even human nature, good or evil, is not free from cultural determination. One way out of this pluralistic impasse is the expansion of the moral mind. It is the outlet taken by religion, the arts, and philosophy from the earliest time in human culture. In philosophy we find an authentic example of this in Socrates. Following the practice of Socrates, we can try to expand the moral mind philosophically, that is, by working on various forms of reasoning, both deductive and non-deductive, including induction, abduction, dialectics, analogy, and pragmatics.

I

n this age of pluralism, applied ethics must deal with two problems. It must argue against ethical relativism as well as provide solutions to moral problems in the real world. The first task is theoretical; nevertheless, it concerns philosophers in applied ethics, for ethical relativism taken seriously would mean that philosophy has no concrete response to moral problems. If philosophy has nothing concrete to say regarding moral problems, what is the point of going around the world in order to make this point? It will only confuse the people. They will not know what to make of a philosopher publically proclaiming that philosophy has nothing to say to them. Perhaps a better service would be to keep silent. He certainly cannot do any philosophical service by telling them not to expect anything from

9

S O HUNG- YUL 10 philosophy. However, he could cause severe damage and disservice to fellow philosophers who are trying hard to make philosophy meaningful with general truths and universal values even in this age of pluralism. Ethical relativism seems harmless, even beneficial, when limited to philosophy classes or academic discussions. Even sophistry is useful when deployed as the devil’s advocate among philosophers themselves. If we leave ethical relativism behind as a theoretical problem, we face the real problem for moral philosophy in practice: cultural pluralism. We have always lived in a world of pluralistic cultures. We have been aware of the differences in cultures. But we have not been able to learn to respect other cultures as equal to our own. Cultures have been destroyed and annihilated by invaders, without the slightest concern from the rest of the world. But now slowly we are becoming aware of other cultures and beginning to respect and cherish them. Cultural pluralism has sounded in a new key. Accepting cultural pluralism presents a challenge to moral philosophy: To provide a universal solution for the problems arising from cultural diversity. However, this appears to be an insurmountable problem because values are culturally dependent and, further, there are no moral sanctions recognized by all cultures alike. ‘Being universalizable’ is an essential meaning of ‘moral’. A moral solution must be universalizable in application. But what is the ground for such universalizability? Going back to human nature does not help because we know that disputes over optimistic versus pessimistic views of human nature can never be resolved. We can assume men are all evilnatured just as surely as that men are all good-natured. Besides, philosophers no longer seem to talk about the nature of something because that subject belongs to science. Science with its reductionist methodology will continue to supply evidence for how man’s moral nature, good or evil, is determined genetically. However, if morality is reduceable to genetics, universal moral judgements are out of the question. But scientific interest does not stop here. Science can go beyond the descriptive and explanatory function to that of control and manipulation. Scientific technology in genetics might be able to make all men good-natured if it is so desired. But is it desirable? What would it do to the diversity of our cultures? Is it reasonable to give up pluralistic cultures in order to eliminate moral problems of pluralism? We are proud because we are more than our genes. We would not have genetic engineering meddle with our destiny. But this is not to say that our moral nature has no basis whatsoever in genetic determination. The differences between man and other animals are primarily genetic. No other animals could grow and develop to become moral agents in

P LURALISM

AND THE

MORAL MIND 11

a human society because of their genetic shortcomings. These gentic differences have developed an astonishing human culture with its unique achievements in art, religion, philosophy, and science. Morality belongs to the unique human culture that is both continuous and discontinuous with the genetic nature of man. So, if we are to find a human capacity to make universal moral judgements, we will have to find it in the same human mind that has brought all the wonders of human culture. That is, we must be able to find—in the very same mind that gives rise to the vast diversity of cultures—the ground for universal moral judgements. Needless to say, it is difficult for the moral mind to deal effectively with moral problems arising from rapidly changing situations. Yet, if a solution must be found, it will be when we keep our minds ever expanding in order to meet these modern demands. How do we expand our minds? What does it mean to have our minds expanded? Having universal concepts or beliefs may help the mind to expand, but that alone does not seem to be sufficient to make the mind function universally. A conceptual belief or a propositional attitude on its own could be without any functional meaning. Concepts or propositions could be held in belief without any experiential meaning or functional effect. Let us take an analogy from religion. Understanding the concepts and propositions of a religious faith is quite different from having the experience of enlightenment or of God’s saving grace. Enlightenment may require understanding, but it is more than understanding. One can understand the conceptual meaning of God’s grace without the personal experience of it. Also, the experience of enlightenment or being saved, not the conceptual or propositional understanding alone, instills into our minds the religious commitment. Religion can help only those who have the commitment of faith, because this commitment is an act of acceptance of the religious way of life, a willingness to grow in faith and to expand the mind accordingly. Even so, the process of growing up in faith and expanding the mind commits one to the incessant striving towards the goal and the standard set by the religious institution. It is a guided effort, an institutional program in which one must submit oneself to be helped and judged as to one’s approximation to the universal mind of God or the Enlightenment. If this is the religious way for the expansion of our minds, what can be the way of philosophy? What can philosophy offer to the expanding mind? What guidance can philosophy give us in our efforts to reach the universal mind? In order to answer these questions, we turn to Socrates. His dialogues were all directed toward the expansion of mind. It was not a mere academic exercise. He was seriously engaged in philosophy, to the point

S O HUNG- YUL 12 of jeopardizing his own life. Socrates’s dialogues demonstrated a practical way of doing philosophy in which the result could be an experience of mind expansion for the participants. This is clear by the reaction of those participants in his dialogue. For them, it was philosophical enlightenment. They were disturbed and embarrassed for their narrow mindedness to be exposed. Also telling is the reaction of those in the establishment. They accused Socrates of corrupting the minds of young people. Corruption of young minds concerned those in hegemony because open minds tend to become disobedient and rebellious with respect to the prevailing moral or legal order. An expanded mind finds it difficult to compromise with the establishment. How did Socrates achieve this? His approach was not a religious one, nor an artistic technique. It was a philosophical way, a rational, cognitive way, a way that worked with the power of human reasoning and logic. We have here a practical proposal for applied ethics: we must assign ourselves to the kind of role that Socrates played and to become mindful of how Socrates was free to apply all forms of reasoning. Socrates was free because philosophy was not solely theory building. Philosophy was still soft enough to run free in the outskirts of the rigid, deductive domain. His method was called ‘dialectic’, but what it really meant was a free and open method to use whatever forms of reasoning appropriate for the purpose at hand. We suggest that we expand our tools of reasoning by turning to some non-deductive forms of logic. Accordingly, we will introduce five such nondeductive forms and see how their applications might help expand the mind. They are induction, abduction, dialectics, analogy and pragmatics. They are introduced because there are events, processes and relations in the real world that need to be explained and understood with non-deductive as well as deductive forms of reasoning. Deductive reasoning does not require any explanation. However, the application of inductive reasoning is really uninteresting when it is meant to be a statistically significant generalization. Inductive generalization in this sense is statistical and it serves to supply the general premise in deductive reasoning. Yet there is another application of induction which is more interesting and uniquely inductive. Induction in this sense is applied when we deal with events distributed in time, in space, or in space-time in a wide range. These events appear to be unrelated and random to an inattentive mind but when observed with an experienced mind, they reveal a hidden connection or a general meaning beyond and above the particularities of scattered events. The ability to see such a hidden connection in the distributed events through an extended, patient observation constitutes inductive wisdom. The wise man who

P LURALISM

AND THE

MORAL MIND 13

sees such inductive connections can predict what is unpredictable and unexpected to the ordinary mind. Abduction is another form of non-deductive reasoning for which we must give a stipulative meaning. We want to limit the meaning of abduction to the kind of reasoning applied to the explanation or understanding of phenomena in hierarchical structures. We make abductive assumptions at the higher level in order to explain events at the lower. Scientific explanation of events by subsumption under a covering law would be a special case of abduction. But to limit and distinguish abduction from deduction we will prescribe its use for non-deductive, hierarchical explanation and understanding. Explanations in terms of abstraction or generalization are abductive. So are the explanations in transcendentalism such as we find in Plato, Kant, and in Christianity. An abductive assumption could help us open our minds to new explanations, different interpretations, better understanding or alternative world views. Dialectic reasoning does not need any explanation. But here we want to share a beautiful application of dialectical reasoning in science. Lynn Margolis’s explanation of the evolution for the eukaryotic cell reads as follows: How did the eukaryotic cell appear? Probably it was an invasion of predators, at the outset. It may have started when one sort of squirming bacterium invaded another—seeking food, of course. But certain invasions evolved into truces; associations once ferocious became benign. When swimming bacterial would-be invaders took up residence inside their sluggish hosts, this joining of forces created a new whole that was, in effect, far greater than the sum of its parts: faster swimmers capable of moving large numbers of genes evolved. Some of these newcomers were uniquely competent in the evolutionary struggle. Further bacterial associations were added on, as the modem cell evolved.1

Analogical reasoning, too, does not need any explanation. Nature has given us an abundance of examples of analogical ways of doing things. Nature is analogically repetitive. Nature does not repeat mechanically. But nature uses the same kind of form or structure to accomplish the same kind of function. When the function is similar, so is the form. The dictum “form follows function” seems to apply to the work of nature, as well as to the things of human ingenuity and creativity. Pragmatics or pragmatic reasoning is another form that we would like to invoke. It is a form of reasoning applied to the relation of ends and means. It is a reasoning for selecting a means, from among varied alternatives, for a given end. Or, it could be a reasoning for choosing the right end from different alternatives with the given means. Pragmatic

S O HUNG- YUL 14 reasoning as such could involve the relation of multiple means with multiple ends. In any case, excellence in pragmatic reasoning is what we call ingenuity. Now, in working with these six forms of reasoning we want to see how they could be applied to characterize the rationality of any particular culture. An example of a culture with deductive rationality as its characteristic is the culture of the scientific world. The world of science is made possible by deductive reasoning, exemplified in the language of mathematics. The same scientific world has brought us a new world of cyberspace with all its unpredictable, awesome cultural implications. This brave new world in cyberspace has been realized by digitalized computational logic, which is an application of deductive rationality. We may consider Chinese patience as an example of inductive rationality. The patience of the Chinese people is a cultural trait to be noticed and respected. It can be explained in terms of their Taoist tradition, especially its emphasis on inductive reasoning. Using inductive reasoning to discover real connections between widely distributed events in time and space takes time and patience. Excellence in inductive reasoning is a type of wisdom and the accompanying patience is a virtue in itself according to Taoist teaching. Chinese patience—which might be characterized as an inductive rationality—is evident, for instance, in their business dealings. They go through a long and tedious process of personal interactions in order to judge the personality of a potential business partner. They want to know his personality thoroughly so that they may be able to predict even his unpredictable behavior. As for abductive rationality, we find an example in the tolerance characteristic of the Indian people. How the Indian people tolerate vast inequalities both in economic life and social status caused by the long outdated caste system requires an explanation. We are suggesting that one possible explanation might be found in the transcendental thought of the Indian people. By means of transcendental thinking, the Indian people seem to elevate themselves in their value perceptions. They would, then, look down from the spiritual realm upon all the myriad differences and discriminations in their socioeconomic life and find them truly insignificant. This transcendental way of experiencing the values of life is a form of abductive reasoning; such a cultural characteristic with its power of tolerance is an example of abductive rationality. A case of dialectic rationality may be seen in the Buddhist culture of renunciation or unattachedness. The reasoning that leads to this state of mind is dialectic. It is a form of reasoning that directs us to view everything in the relationship of both continuity and discontinuity at the same time. For instance, every being may be seen as dependent and at the same

P LURALISM

AND THE

MORAL MIND 15

time independent of every other being. Or, every relation may be viewed as having an aspect of agreement as well as an aspect of disagreement. Even the relation of the creator and the created or the agent and its action must be both continuous and discontinuous. Realizing the dialectic nature of relationship in continuity and discontinuity would free us to renounce or be unattached because everything, no matter how dear it may seem to us, could be seen as discontinuous in some respects and defiant of any permanent tie. Of course this is only what our dialectic reasoning would lead us to do, but our emotion, unyielding to reason, might have its own reasons to cling to the object of our attachment. Buddhism as a religious practice works for both the cognitive aspect and the emotive aspect of the mind leading towards the state of unattachedness or renunciation. Dialectical reasoning will carry us only to a certain point, however, where dialectic reasoning must relinquish its place to the religious way. Examples of analogic rationality are abundant in almost every culture. But, in general, a culture that has remained less civilized is more naturalistic and shows a more pervasive tendency toward analogical reasoning. Being less civilized the primitive culture of a tribal society has never been able to withstand the highly civilized culture of an industrial society. Cultures of the Native American tribes and nations are good examples because they had the most unfortunate destiny of being destroyed and annihilated by the European civilization that had the power of highly advanced technology on their side. Today, with a global sensitivity and awareness of ecological issues and problems, we begin to see the importance of analogic rationality in naturalist cultures. The best example of pragmatic rationality is, of course, American culture and ‘know-how’. Pragmatic reasoning, being instrumental, values efficiency and expediency more than any other value. The measure of excellence in pragmatic reasoning is taken from the consequences, for which the best means have been free and fair competition. The cultural power of American know-how or ingenuity is the fruit of pragmatic rationality, which has been surpassed by no other culture on earth or in heaven—including the Stoic rationality characteristic of the Roman Empire. A prevailing form of reasoning in a culture could be misapplied or overindulged to the point that its own rationality might turn to irrationality, causing adverse effects to its own culture as well as to others. Let us see how irrationality may ensue with each form of reasoning in application or misapplication. Pragmatic reasoning fails when the consequences of a pragmatic measure does not really serve its purpose or when the means and the ends are reversed by the consequences. Also, when an exclusive adherence to pragmatic values of efficiency and expedience disregards some other important

S O HUNG- YUL 16 values, the overall effect could be negative. For example, accepting Black American culture for pragmatic reasons alone on the part of some White Americans would not convince Black Americans. Likewise, when any culture accompanied by superior economic power tries to engage other cultures with inferior economic conditions the reasoning behind such an intercultural encounter is crucial for the outcome. When the dominant reasoning is pragmatic with its obsession for efficiency and expediency the encounter could become an instance of intrusion, or even cultural imperialism. The pragmatic rationality of the intruder will degenerate into pragmatic irrationality in the eyes of those who have been subjected to the power of pragmatic efficiency. Analogic rationality could change to analogic irrationality when the analogy is pushed too far, or some disanalogies inherent in the context are overlooked. For example, there are disanalogies inevitable between the natural kinds and those that are products of human civilization and culture. But naturalism of the primitive tribes failed to see the importance of such disanalogies when their culture encountered the culture of advanced civilizations. The native’s analogical reasoning was not sufficient to deal with the threats to their cultural identity and integrity. They became defensive and turned to isolationism in order to guard the narrow perspective of their analogic rationality. Isolationism as such could easily turn anti-ecological, betraying nature’s analogy in a greater perspective. Russia, under the socialist regime, was building a culture of dialectic rationality. But to have history develop in accordance with dialectical reasoning is a time-consuming process. Russia, under turbulent circumstances following the Revolution, was not given the needed historical time, nor was it able to reason dialectically to achieve synthesis out of antithetical discontinuities and contradictions. The urgent situation demanded that the regime rely on totalitarian measures in order to maintain stability and order. The unity for their pluralist cultures and societies had to be sustained by coercion. The socialist regime also abused dialectical reasoning by playing with false contradictions in order to mitigate the atmosphere of tension at home and abroad. Perhaps the failure of their dialectical reasoning proved most apparent when it was unable to come up with a working synthesis between pragmatic capitalism and dialectic socialism. Socialism in Russia failed for one reason: dialectic rationality faltered. Nevertheless, reaching a true synthesis from the contradictions of pragmatic capitalism and dialectic socialism still remains an idealistic goal for today’s world. Abductive rationality has pitfalls in empty concepts and irrelevant abstractions. Transcendentalism in abductive reasoning may weaken the sense of reality so as to become less mindful of concrete and specific

P LURALISM

AND THE

MORAL MIND 17

matters of practical concern. Transcendental elevation without a firm footing in reality would be escapism. The consequences of transcendental escapism would be disheartening. We often witness such consequences in the behavior of transcendentally-minded people who become utterly intolerant and impatient with the practical affairs of the world. Not being accustomed to pragmatic reasoning, their behavior becomes militant and destructive. Abductive rationality turning to escapist transcendentalism degenerates to abductive irrationality. Inductive reasoning as a means for finding the real connection in the seemingly random distributions of events runs the risk of missing the connection or being mistaken regarding a false connection. True inductive wisdom may be confused with false inductive beliefs. Such confusion, then, causes much disorder in society as a whole. We see a similar phenomenon in a culture intoxicated with magic, mystery, pseudo-science, parapsychology and what not. A culture with undependable inductive rationality or one that induces its people to confuse inductive irrationality with inductive rationality, or superstition with wisdom would falter on its own chaotic consequences; it would eventually be replaced by another culture with rigid, formalistic orders. Confucianism in China is an example of an orderly culture being established to cope with a disorderly, chaotic state. Deductive irrationality sounds almost like a contradiction in terms. But to equate deduction with rationality itself is an unfounded belief which disregards the conditions that make deductive reasoning a narrow or conditional rationality. Deductive reasoning with its rigorous, formalistic and extensional characteristics must rely on certain presuppositions. For its rigorousness, deductive reasoning must assume the universality of those approximate generalities. For its formalistic consistency, deductive reasoning must begin with an appropriate set of axioms. For its extensional digitalization, deductive reasoning must limit its domain excluding all intensive measurements. The emergence of cyberspace made possible by digitalized computational logic now enables us to visualize almost any logically possible world, that is to say, any deductively possible world. But this powerful world of cyberspace realized by deductive reasoning would show where our deductive rationality could potentially turn into deductive irrationality. Confusing the virtual reality of cyberspace with the actual reality of physical space (or confusing the abstract mathematical world with the concrete physical world) would be considered deductive irrationality. When we run into intercultural conflicts we tend to characterize the opponent’s culture as irrational, while claiming that our own culture is rational. How can we broaden our mind to see the opponent’s rationality more readily? What can we do to open our minds to the opponent’s ratio-

S O HUNG- YUL 18 nality in various non-deductive reasonings? We could begin, on each and every occasion, with questions to ourselves: Is it an inductive rationality? An abductive rationality? A dialectic rationality? An analogic rationality? Or, a pragmatic rationality? By doing so, we might be able to find the hidden rationality in the opponent’s culture, which, in turn, would help expand our own minds. Seeing rationality rather than irrationality in another culture is difficult enough, but an even more difficult feat is seeing irrationality in our own culture. Here again we should not be content with ourselves being free of any deductive irrationality. We must again go down the list and ask ourselves: Are we not being inductively irrational? Abductively irrational? Dialectically irrational? Analogically irrational? Or, pragmatically irrational? By raising these questions and finding a suppressed belief or attitude of irrationality on our part we will find our own minds being expanded just as much with every trial and success. There is no royal road to the universal moral mind. Philosophy, religion, art, and science must all endeavor and work separately and in cooperation towards the goal. Philosophy’s part is the logical or the rational way. It was the way Socrates led us a long time ago. What we have tried here is to follow in his footsteps with the tools we have at hand. So Hung-yul, Department of Philosophy, Ewha Women's University, Seoul, 120750, Korea; [email protected]

NOTE 1. John Brockman, The Third Culture (New York: Simon & Schuster, 1995), 137.

WHY ETHICS IS PART oF PHILOSOPHY: A PLEA FOR A PHILOSOPHICAL ETHICS

Ethics is frequently divided into three parts: metaethics, normative ethical theory, and the more specific normative et hics. However, only metaethics is explicitly philosophical insofar as it is concerned with fundamental questions about the content, objects, and status of ethical thought and discourse. During the heyday of conceptual analysis, philosophers were admonished to restrict themselves entirely to metaethics. Since, it was said, they lacked any special expertise as phi-

Stephen Darwall

losophers on normative questions, their pronouncements could be no more than hortatory. I will argue that there is no satisfactory alternative to conducting normative ethical inquiry in conjunction with metaethics. Only by investigating normative and metaethical questions together in an integrated philosophical ethics can we hope to make real progress with either. Ethics should remain part of philosophy.

E

thics is frequently divided into three parts: metaethics, normative ethical theory, and more specific normative ethics of the sort frequently, if misleadingly, called “applied ethics.” Of these, only metaethics is explicitly philosophical, concerned, as it is, with fundamental questions about the content, objects, and status of ethical thought and discourse. During the heyday of conceptual analysis, philosophers were admonished to restrict themselves entirely to metaethics. Since, it was said,

19

S TEPHEN DARWAL L ARWALL 20 they lacked any special expertise as philosophers on normative questions, their pronouncements could be no more than hortatory.1 Better to keep their philosophy and their ethics separate. Even during normative theory’s “great expansion” in the late sixties and early seventies, it wasn’t always clear what normative ethics had to do with philosophy (or rather, with the rest of philosophy). The method of “reflective equilibrium” was frequently practiced narrowly, with “intuition pumping” providing a fuel for normative ethical debate that was largely free of other philosophical additives. 2 Rawls himself proclaimed the “independence of moral theory,” arguing that, at least in the short run, progress in ethical theory depended upon pursuing normative ethics independently of fundamental philosophy. 3 Thus the removal by the “great expansion” of earlier philosophical barriers to normative theory did not necessarily reconnect normative ethics with the rest of philosophy. To the contrary, as philosophers’ attention turned during that period toward normative issues, it turned decidedly away from metaethics. From either the perspective of “pure” philosophy or from that of “pure” normative theory or applied ethics, it must seem a mystery why ethical teaching or research should be located in a philosophy department. Metaethics, the philosophy of ethics, would seem to have roughly the same relation to normative ethics as the philosophy of physics does to physics. And although there was no way to sharply demarcate physics from metaphysics before the rise of the experimental method, no one these days would argue for a merger between the departments of physics and philosophy. So why should things be any different with ethics? I believe that the case of ethics is different and that the practice of ethical theory should reflect this fact. Unlike the natural and social sciences, ethics did not spin off from philosophy, nor should it. In what follows, I argue that the reason why ethics is properly located within philosophy is that normative questions are not entirely independent of metaethical issues concerning the nature of value and obligation. A properly philosophical ethics must integrate normative ethics with metaethics. The various sciences became autonomous disciplines when the experimental method gained wide acceptance as the only reliable way of confirming theory about their respective subject matters.4 Disagreements in the philosophy of science or in metaphysics or epistemology leave this consensus largely unaffected. Whether theory is interpreted realistically, as the best explanation of observed experimental results, or instrumentally, as the best device for predicting them, is mostly irrelevant to

W HY ETHICS IS PART OF PHILOSOPHY 21 scientific practice. Scientific method can proceed largely independently of these philosophical disputes. Nothing like this consensus has emerged in the case of ethics, nor is it likely to. It could do so only if ethical method could satisfactorily be modeled either as a narrow reflective equilibrium, that is, one that takes as data only considered normative judgments, or, as some empiricist naturalists argue, as simply a specific instance of general scientific method. But neither of these can gain consensus, partly because each is unacceptable to the proponents of the other. Those who favor narrow reflective equilibrium are often militantly anti-naturalist, and naturalists are frequently skeptical of ungrounded ethical intuition. The main reason why normative ethics remains squarely within philosophy, however, is that, unlike the sciences, its subject matter is itself the focus of a lively philosophical debate to which we can envision no end. Radical skepticism notwithstanding, there is sufficient consensus concerning the phenomena that the natural and social sciences are to explain that they can proceed more or less independently of fundamental metaphysics and epistemology. But this is not the case with ethics. Most importantly, there is no consensus regarding the normativity of (various) ethical propositions and judgments nor even about what normativity (or normative judgment) itself is. The term ‘normative ethics’ signals consensus that ethics is a normative subject, that it concerns how people ought to act, or desire, or feel, or be. But after this, discord reigns. Exactly what normative force do various ethical claims have? And what, after all, is normativity (or normative judgment) anyway? As an example, take claims about moral right and wrong. Many philosophers believe that moral obligations are, or purport to be, categorical imperatives—that moral requirements have an authority or normativity that consists in providing conclusive reasons for anyone to act as they require. But not everyone accepts this. Some writers agree that moral obligations are normative, but hold that they are authoritative for (distinctively moral) attitudes or feelings—for example, for impartial disapproval or impersonal anger—rather than for action.5 Others agree that moral claims concern what a person ought to do, but only what she ought to do morally, without any implication for what she ought to, or has reason to, do simpliciter.6 And yet others agree that they concern what the person should do morally, that is, from a perspective internal to the concerns, rules, principles, and so on, that define morality, but hold that there is no such thing as what a person has reason to or ought to do simpliciter, all reasons for acting being relative to some assumed perspective or system.7

S TEPHEN DARWAL L ARWALL 22 It seems incredible that there could be anything approaching consensus about what phenomena a moral theory is to explain—which actions are right or wrong in specific circumstances—with this degree of disagreement about what it would take to be a phenomenon of the relevant kind in the first place. It is no surprise, therefore, that debates in normative moral theory, such as those between act-consequentialists and their critics, for example, cannot be fully contained at the level of normative moral judgments and frequently turn, ultimately, on fundamental disagreements concerning morality’s normativity and its source. Consider, for instance, the objection that act-consequentialism might be self-effacing, that it might recommend that we (even, indeed, that perfect moral agents) guide deliberation by some nonconsequentialist theory, since the consequences of doing so might be better. This is the objection that led to Bernard Williams’ waggish remark that he would leave it “for discussion” whether, if true, it “shows that utilitarianism is unacceptable, or merely that no one ought to accept it.” 8 To feel the force of the objection though, one must accept that morality is normative for action or feeling, that morality purports to be action-or feeling-guiding. To someone, like Williams, who accepts this, it is simply incoherent to think both that what one is doing is wrong on certain grounds and that these wrong-making grounds might provide no reason to decide against so acting nor any reason to disapprove of one’s failing to do so (lacking adequate excuse). Because consequentialists are frequently externalists, however, they are often unmoved by this objection. They hold that whether an action is right is one thing and that whether anyone has any reason to do it, or even to approve of doing it, is another. It would seem, therefore, that there is simply no way to seal off the normative moral disagreement between act-consequentialists and their critics and bracket such further philosophical issues as the nature and source of morality’s authority. Ultimately, this normative debate must be situated within philosophical ethics. Or consider, to take another example, the question of how “virtue theory” should be positioned within moral or ethical theory. Suppose a theory were to systematize phenomena related to the virtues in the sense of organizing considered judgments about specific virtues or about virtue generally. What would we learn from such a theory about how people should act and be? That all depends on virtue’s normativity. Sometimes virtue is understood as a distinctively moral goodness of the person or of character. The eighteenth-century Scottish philosopher, Francis Hutcheson, proffers a virtue theory of this sort. 9 So understood, virtue’s normativity is related to morality’s, although for virtue theorists like Hutcheson, morality’s normativity is not fundamentally a matter of what we should do,

W HY ETHICS IS PART OF PHILOSOPHY 23 but of how we should be motivated, that being understood in terms of what motives appropriately elicit moral approval and disapproval. More usually, however, virtue theories are put forward in a different spirit, either as encompassing good-making features of persons that extend beyond or perhaps intersect with the moral, or as in some way an alternative to morality. 10 All virtue theories agree that virtue has to do with goodness of the person or character. A person’s virtues are what make her good, just as an automobile’s virtues are what make it good. But how are we to understand goodness of this sort? And what is its normativity? It is impossible to know how to position virtue theory in relation to other ethical theories in order to see where they agree or what issues there might be between them until we can answer questions like these. And to do that we have to move beyond specific normative virtue claims and consider their grounds, content, normativity, and their truth or assertability conditions. Obviously, the virtues concern how we ought to be, in some sense. But in what sense? The virtues of a car are what make it a good thing of its kind, where the relevant kind is defined in terms of the standard purposes and expectations that explain why we have cars, or cars of a certain kind. 11 Should we think of ethical virtues in this way? To do so we would have to think of human beings as a functional kind and reckon the virtues in relation to that function. Even if we could do that, however, it is not obvious what normativity this would have for us as agents, since we could always sensibly ask why we should uphold the standards for a good thing of that kind, and it is not obvious what could provide a satisfactory answer. Alternatively, we might think of human virtue more organically, as we think of what is necessary for a plant or animal to thrive and flourish. More generally, the virtues are sometimes thought of as features of the person or character that are necessary for wellbeing, qualities without which, in Philippa Foot’s words, a person does not “get on well.” 12 If we think of the virtues in this way, their normativity would seem to derive from that of a person’s good or well-being. As I shall argue presently, however, exactly what normative force a person’s good or well-being has is itself controversial. Or perhaps we should understand the virtues in relation to a kind of value that differs from goodness of a kind, from well-being, and even from moral goodness narrowly conceived, namely, in terms of excellence or merit of the sort that consists in being worthy of admiration or esteem. 13 Virtues would then be admirable features of persons, features that are worthy of esteem. Obviously, exactly how we position virtue claims in relation to other ethical and moral claims will depend on how we interpret them,

S TEPHEN DARWAL L ARWALL 24 whether in one of these or in some other way. It seems impossible, therefore, to have a consensus about what traits are virtues without at least some implicit metaethical agreement about what a virtue or a virtue-claim is. So here again, it seems impossible to seal off normative virtue theory from philosophical issues concerning the nature of the phenomenon such a theory would attempt to explain. Ultimately, a satisfactory theory of the virtues must be placed within a philosophical ethics that integrates these normative and philosophical issues. As a final example, consider the normative question of what role a person’s subjective states play in determining his well-being. What I have in mind here is not the question of metaethical subjectivism, that is, of whether being good for someone is the same thing as being thought good by that person. I mean, rather, the normative question of what role a person’s subjective states play in making his life good for him. To what extent are a life’s good-making (more properly, well-beingmaking) features qualities of that person’s subjective states? Hedonists (about well-being) believe that a person’s good is determined entirely by the qualities of his mental states. “Objective list” theorists who maintain that, for example, objective accomplishments are intrinsically relevant to a person’s well-being, deny this. 14 And there are hybrid views that maintain that both objective and subjective elements are necessary, for example, that human well-being centrally involves the appreciation of one’s connection to various objective values. 15 It seems impossible to resolve these normative disagreements without addressing philosophical questions about the nature of well-being and its normativity. For instance, it is frequently argued against hedonism that people rationally desire all sorts of things other than their own subjective states, for example, the flourishing of their children long after they (the parents) are dead. However, for this to be an objection to hedonism as a theory of welfare, the obtaining of something a person rationally desires must make him better off, it must make his life better for him. The objection assumes, that is, that well-being is normative for that person’s desires, or rather, that well-being consists of whatever is normative for his desires. But this is questionable. Indeed, I think it is false. A person’s good is normative, but not, I believe, for that person as agent. Rather a person’s welfare or good is normative for the desires of anyone (perhaps the person himself) who cares, or has reason to care, for that person for his own sake. 16 A hedonist about welfare might grant that there are many things that are appropriately important to us other than our own pleasure, but maintain that that is because their importance to us is not a measure of their contribution to our welfare. The flourishing of my children after I am dead, much as

W HY ETHICS IS PART OF PHILOSOPHY 25 I sensibly want that now, will make no contribution to my well-being now or then. To resolve disagreement about the role of subjectivity in well-being, we would have to resolve this debate about welfare’s normativity. Here again, then, what seems superficially to be simply a disagreement in the normative theory of well-being depends ultimately on philosophical issues about well-being’s nature and normativity. To advance, the debate must be placed in the framework of philosophical ethics. It is only in this century that we have come to distinguish sharply between normative ethics and metaethics. We are right to make this distinction. If I ask what value or value judgment is, I am asking a different question than if I ask the normative question of what is valuable. And, notoriously, a wide variety of metaethical views can go with a large variety of views in normative ethics. One can be a consequentialist in normative ethics, for example, and be prescriptivist, a norm-expressivist, a theological voluntarist, a rational intuitionist, or a naturalist in metaethics. Nonetheless, there do seem to be affinities between various metaethics and roughly corresponding normative ethics.17 For example, metaethical naturalists have almost always been consequentialists, although, as we just noted, a consequentialist need not be a naturalist. Deontologists, on the other hand, require a metaethical account of moral obligation that can explain how the right can diverge from the beneficial, and the wrong from the harmful. Some deontologists have been theological voluntarists, basing their normative deontology on the nonconsequentialist character of God’s commands. Some have been rational intuitionists, holding that we can perceive the truth of some deontological normative claims by direct reflection. As Clarke argued we can see that gratitude is a fitting response to benevolence. And some have followed Kant in maintaining that deontological normative principles are grounded in the structure of deliberation of a free rational agent. These associations are far from accidental. Similarly, when we look within the thought of the great ethical philosophers, such as Aristotle, Kant, Mill, or Nietzsche, we find a dynamic interaction between their normative and metaethical theories. Each has a characteristic philosophical ethics—a distinctive approach that integrates their metaethical and normative thought. They did not treat metaethics and normative ethics as altogether separate and independent. On the contrary, they grounded their normative views about what has value in some metaethical understanding of what value itself is. Mill’s utilitarianism is grounded in his metaethical naturalism, Kant’s deontology, in his metaethics of practical reason, and so on.

S TEPHEN DARWAL L ARWALL 26 It might seem that there is an obvious counterexample to this phenomenon in noncognitivist or projectivist philosophers for whom there is, strictly speaking, no such thing as value or normativity, but only evaluative or normative judgment. How could the position that normative judgments express or project feelings or attitudes tend to support or undermine one normative view rather than another? Even here, however, a philosopher’s metaethics may be relevant in advancing his normative ethics, since he may believe that people are attracted to alternative normative ethics because they mistakenly see them as grounded in metaphysics. Thus, by arguing that value is simply what we project onto things when we desire them, Hobbes hopes to dislodge any attachment that Leviathan’s readers might have to the (normative) doctrine that the sovereign’s political and religious authority should be restricted because they believe in a natural law that is grounded in scholastic metaphysics. On Hobbes’s metaethics, these normative views are nothing but the expressions of desire, but Hobbes believes that the relevant desires nonetheless result from religious rituals that people engage in because of mistaken metaphysics. Convincing his readers of projectivist (or some other noncognitivist or subjectivist) metaethics can thereby loosen their attachment to these rituals and, consequently undermine the desires they express in the normative doctrines that Hobbes opposes. In the end, then, there is no satisfactory alternative to conducting normative ethical inquiry in conjunction with metaethics. Only by investigating normative and metaethical questions together in an integrated philosophical ethics can we hope to make real progress with either. Ethics should remain part of philosophy. Stephen Darwall, Department of Philosophy, University of Michigan, Ann Arbor, MI 48109; [email protected]

NOTES 1. As late as 1970, W. D. Hudson evidently did not expect to raise any eyebrows by writing: “This book is not about what people ought to do. It is about what they are doing when they talk about what they ought to do. Moral philosophy, as I understand it, must not be confused with moralizing. A moralist . . . engages in reflection, argument, or discussion about what is morally right or wrong, good or evil . . . [A] moral philosopher . . . thinks and speaks about the ways in which moral terms, like ‘right’ and ‘good’ are used by moralists when they are delivering their moral judgements” (Modern Moral Philosophy [Garden City, NY: Doubleday, 1970]).

W HY ETHICS IS PART OF PHILOSOPHY 27 2. It is worth noting that Rawls’s own practice or reflective equilibrium was quite wide, as evidenced, for example, by the “Kantian interpretation” of justice as fairness. See A Theory of Justice, (Cambridge, MA: The Belknap Press of Harvard University Press, 1971), 251-257; see also “Kantian Construction in Moral Theory,” The Journal of Philosophy 77 (1980): 515-572. On the distinction between narrow and wide reflective equilibrium, see Norman Daniels, “Wide Reflective Equilibrium and Theory Acceptance in Ethics,” The Journal of Philosophy 76 (1979): 256-282. 3. John Rawls, “The Independence of Moral Theory,” Proceedings of the American Philosophical Association (1975), 5-22. 4. These are matters of degree, of course. Scientific questions are not entirely independent of issues in the philosophy of science. My claim is that theoretical physics is independent of the philosophy of physics to a larger degree than ethical theory is independent of metaethics. 5. See, e.g., Allan Gibbard, Wise Choices, Apt Feelings (Cambridge, MA: Harvard University Press, 1990). 6. This was Philippa Foot’s position in her famous “Morality as a System of Hypothetical Imperatives,” Philosophical Review 81 (1972): 305-316. However, see her “Recantation 1994,” in Moral Discourse and Practice, ed. Stephen Darwall, Allan Gibbard, and Peter Railton (New York: Oxford University Press, 1997), 322. 7. See, e.g., Dan W. Brock, “The Justification of Morality,” American Philosophical Quarterly 14 (1977): 71-78. 8. J. J. C. Smart and Bernard Williams, Utilitarianism: For and Against, (Cambridge: Cambridge University Press, 1973), 135. 9. Francis Hutcheson, An Inquiry into the Original of Our Ideas of Beauty and Virtue, (London, 1725). For selections, see Moral Philosophy from Montaigne to Kant, 2 vols., ed. J. B. Schneewind, (Cambridge: Cambridge University Press, 1990); and British Moralists: 1650-1800, 2 vols., ed. D. D. Raphael, (Indianapolis: Hackett Publishers, Inc., 1991). 10. Aristotle and some neo-Aristotelian accounts can be thought of as instances of the latter. However, I am somewhat unsure of whether Michael Slote’s can. See Slote’s From Morality to Virtue, (New York: Oxford University Press, 1992). 11. On goodness of a kind, see, e.g., G. H. von Wright, The Varieties of Goodness, (London: Routledge & Kegan Paul, 1963), 19f. 12. Philippa Foot, “Virtues and Vices,” in her Virtues and Vices (Berkeley and Los Angeles: University of California Press, 1978), 2. 13. This is the way that Slote seems to think of the virtues. See, e.g., Slote op. cit., 94f.

S TEPHEN DARWAL L ARWALL 28 14. For “objective list” theories, see Derek Parfit, Reasons and Persons, (Oxford: Clarendon Press, 1984), 499-502. For the claim that accomplishment is important to well-being, see James Griffin, Well-Being, (Oxford: Clarendon Press, 1986), e.g., 27. 15. I defend such a view and argue that Aristotle can be interpreted in this way in “Valuing Activity,” in Social Philosophy and Policy, forthcoming. See also Parfit, op. cit., 501f. 16. I argue for these claims in “Self-Interest and Self-Concern,” in Self-Interest, edited by Ellen F. Paul (Cambridge: Cambridge University Press, 1997), 158178, and in “Empathy, Sympathy, Care,” Philosophical Studies 89 (1998): 261-282. 17. The next two paragraphs draw on my Philosophical Ethics (Boulder, CO: Westview Press, 1998).

MORAL THEORY AND

In what follows I want to endorse and to reinforce what seems to me a pragmatic, and more specifically a Deweyan, account of the dim pros-

THE

pects for traditional moral theory. I want further to describe a role for

REFLECTIVE LIFE

moral philosophy that accepts the demise of moral theory, a role exemplified by Dewey himself in his insistence on the place of intelligence and reflection in a satisfactory life. Dewey’s insistence on intelligence

Stuart Rosenbaum

and reflection in the good life gives rise to a large–scale moral ideal, the ideal of the reflective life; I describe that ideal and contrast it with some others.

THE DIM PROSPECTS OF TRADITIONAL MORAL THEORY

T

he dim prospects of moral theory have been remarked by many recent thinkers, some not in any obvious way intellectual descendants of Dewey. Marcus Singer, for example, in his Presidential Address to the Central Division of the American Philosophical Association, May 2, 1986, remarks that “the concept of morality is systematically ambiguous. It may mean either positive morality, or personal morality, or true or rational morality.”1 Singer seems serious about this claim, but he seems not to notice that it undermines the idea that what he calls “rational morality” supplies a perspective that determines the adequacy of what he calls “positive morality” or “personal morality.” How does it undermine this idea? If the concepts of personal, positive, and rational morality are as different as is suggested by the claim that they are “systematically ambiguous,” then none of them may stand in critique of the others any more than you might check the degree of bank of an airplane by measuring the

29

S TUART ROSENBAUM 30 level of water at the bank of a river. “Bank” in this example is, to use Singer’s words, “systematically ambiguous.” If “wrong” in the assertions that, say, abortion is wrong or homosexuality is wrong means such different things as is meant by “bank” in the above example—an implication of Singer’s claim that “the concept of morality is systematically ambiguous”—then no conclusion whatever about “the real” rightness or wrongness of abortion or homosexuality is available. Once one takes seriously the idea that “the concept of morality is systematically ambiguous” then there is no real or ultimate moral wrongness of abortion or homosexuality or of anything else for that matter. Singer’s insight that the concept of morality is systematically ambiguous is one Dewey was attuned to in all of his thought about morality. Dewey saw that the tradition seeking a rationally justified morality, represented by such moral philosophers as Aristotle, Kant, and Mill, had a very different idea of morality from the idea of conventional or customary morality—what Singer calls “positive morality”—of “the man in the street.” In addition, he saw that personal goals of individuals, Singer’s “personal morality,” constituted a kind of morality different from either of those other kinds.2 In addition to Singer, other recent thinkers have come to the same conclusion. Thomas Nagel in “The Fragmentation of Value” mentions five different kinds of value that legitimately enter into moral deliberation. In addition, Nagel argues that these different kinds of value cannot be brought into harmony by “a single scale on which all these apparently disparate considerations can be measured, added, and balanced.”3 Nagel continues this theme as follows: I do not believe that the source of value is unitary—displaying apparent multiplicity only in its application to the world. I believe that value has fundamentally different kinds of sources, and that they are reflected in the classification of values into types. Not all values represent the pursuit of some single good in a variety of settings.4

Having rejected the primary aspiration of moral theory to find a unitary account of moral value and of moral duty, moral right and wrong, Nagel espouses the indispensability of “judgment—essentially the faculty Aristotle described as practical wisdom, which reveals itself over time in individual decisions rather than in the enunciation of general principles.” He finds that . . . ethics is more like understanding or knowledge in general than it is like physics. Just as our understanding of the world involves various points of view . . . so values come from a number of viewpoints, some more personal than others, which cannot be reduced to a common denominator any more than history, psychology, philology, and economics can be reduced to physics. Just as the types of understanding available to us are

M ORAL THEORY AND THE REFLECTIVE LIFE 31 distinct, even though they must all coexist and cooperate in our minds, so the types of value that move us are disparate, even though they must cooperate as well as they can in determining what we do.5

Annette Baier and Thomas Scanlon endorse this same insight. Baier especially insists on it in her “Theory and Reflective Practices.” In addition to agreeing with these ideas of Singer and Nagel, Baier endorses a turn toward the tradition she finds in Aristotle, Butler, Hume, Hegel, Nietzsche, and Dewey. This tradition, in her account of it, reflects on the actual phenomenon of morality, “to see what it is, how it is transmitted, what difference it makes.”6 This tradition not only sees morality as a natural phenomenon but it also respects what Nagel calls judgment; morality in this sense Baier finds to be the heart of Hume’s perspective about morality. She puts this point as follows: “Hume believed that all of morality was a matter of what he called reflection, turning natural responses, not just on their natural target, but on responses, turning self-interest on the workings of self-interest, turning sentiments on sentiments.”7 Thomas Scanlon too finds diversity of value or “fragmentation of the moral” to be a definitive feature of morality.8 He agrees with Alasdair MacIntyre that the most optimistic prospect for moral theory is that it might produce a more accurate and informed understanding of disagreement.9 Martha Nussbaum’s account of Aristotle, on a natural reading, agrees in seeing Aristotle to fall within the broad tradition Baier identifies. In “Non-Scientific Deliberation,” for example, she endorses “Aristotle’s attack on the notion that the major human values are commensurable by a single standard.”10 I hope by these references to Nagel, Scanlon, Baier, Nussbaum, and even to Singer (though his philosophical sympathies differ from theirs) to be identifying representatives of a growing tendency in contemporary philosophy to reject what has been until very recently a more dominant tendency to seek commensuration of all values in a single standard, a unitary value or principle, that can bring rational or moral pressure to bear toward a single ideal outcome in diverse lives or in diverse moral situations. Baier finds this tradition in Hume, Nussbaum finds it in Aristotle, and I want to focus on Dewey’s effort to express this same tradition. A central issue for this tradition, while it endorses careful reflection and deliberation as part of a moral life, is how to understand the idea of moral philosophy. What role is there for philosophy in this moral thought that shuns what moral theory has always sought, viz., the commensuration of apparently diverse values and principles into a single ideal outcome? This question may be expressed in an indefinite number of particular ways. For example: How may we think of Hume—the

S TUART ROSENBAUM 32 skeptical Hume who tried to turn us away from theory with every issue he addressed—as a moral philosopher? How is a thinker who puts sentiment at the heart of morality properly regarded a moral theorist or a moral philosopher? Similar questions arise for Aristotle and for Dewey. Should we regard such thinkers as moral philosophers because they try in intellectually sophisticated ways to turn us away from moral theory? What is moral philosophy without moral theory?11

MORAL PHILOSOPHY WITHOUT MORAL THEORY One answer to this question appears in works of Richard Rorty, Hilary Putnam, and John McDermott, as well as others. These thinkers address specific morally problematic situations in enlightening and useful ways. Rorty’s essay, for example, “Human Rights, Rationality, and Sentimentality,” seeks to bring into focus why humans behave in xenophobic ways, why they treat fellow humans as less than human, as animals even. He believes our dominant moral tradition may reinforce those xenophobic tendencies. Here’s a sample: Plato set things up so that moral philosophers think they have failed unless they convince the rational egoist that he should not be an egoist. But the rational egoist is not the problem. The problem is the gallant and honorable Serb who sees Muslims as circumcised dogs. It is the brave soldier and good comrade who loves and is loved by his mates, but who thinks of women as dangerous, malevolent whores and bitches.12

Rorty sees these xenophobic tendencies as a more proper focus of intellectual effort than are the epistemological problems that usually drive moral theory’s effort to answer “the rational egoist.” His own proposal for constructively addressing the problem is to focus on helping people feel more secure and helping them toward greater sympathy for others. “By security I mean conditions of life sufficiently risk-free as to make one’s difference from others inessential to one’s self-respect, one’s sense of worth.” 13 His suggestion is that insecurity fuels xenophobia because it undermines self-respect. He recommends a redirection of intellectual effort from traditional concerns of moral theory toward a wider realization of conditions that contribute to personal security and to sympathy with others. In his focus on conceptual dimensions of a specific morally problematic situation, Rorty embodies the spirit of the large tradition Baier endorses, and also the dominant spirit of the pragmatic tradition.

M ORAL THEORY AND THE REFLECTIVE LIFE 33 The other thinkers—Putnam and McDermott—have essays that also do well the sort of thing Rorty does in the one I have just discussed, but many others do as well.14 The example from Rorty serves to exemplify one sort of response about what philosophy properly does apart from moral theory and within the context of the large tradition Baier endorses. But there is another role for philosophy that accords with that tradition’s turn away from moral theory. This additional role is that of expressing a general ideal of life that might become, perhaps partly by virtue of expressing it systematically, a more widely and favorably regarded ideal of life. An example of such a general ideal of life might be Aristotle’s intelligent aristocrat, or “great-souled man,” an ideal of life embedded in a particular Greek culture but having nonetheless great appeal beyond Greek culture. Another such example is evident in Hume and more generally in thinkers who share his skeptical commitments, among them Montaigne and Erasmus; this ideal is that of the rational thinker, or skeptic, who sees the futility of efforts to see reality clearly or to wrest knowledge from the ignorance inseparable from our “human condition” and is content instead to seek a modest life guided primarily by prevalent social and cultural tradition. Hume not only argued for this ideal—at least implicitly—in most of his philosophical work but he lived it as well, as is evident in his conversation with James Boswell shortly before his death. These general ideals of life—Aristotle’s and Hume’s—are each specific to a particular cultural situation but they nonetheless have great appeal beyond those particular cultural situations. John Dewey has a similar general ideal, and my next task is to give explicit account of it; like Aristotle’s and Hume’s, Dewey’s is a relatively local cultural ideal but it is nevertheless potentially of great significance to societies attracted to the democratic ideal Dewey found so compelling.

THE REFLECTIVE LIFE Dewey’s ideal, the reflective life or the life guided by intelligence, contrasts most explicitly in Dewey’s treatment of it with the life of custom or convention; one might conceive it, in accord with Baier’s construction, as a reflective response to the life of custom or convention. This reflective response, in Dewey, takes a very specific form, but the form it takes Dewey himself does not present as a straightforward thesis about the ideal of living that one might inspect and evaluate as a thesis about the good life; Dewey presents his ideal in much the way their own ideals are presented by Aristotle and Hume. One sees it emerging in his treatment of a wide variety of issues. His commitment to it is present, for example, in his

S TUART ROSENBAUM 34 reaction against many traditional varieties of moral theory, in his reaction against living by custom simply because it is custom, and in his reaction against the idea of a life lived only for its own sake. His presentation, in other words, constructs the reflective life in contrast with other models or ideals of life. As the commitments of Aristotle and Hume arise out of their own relatively “local” cultural situations, Dewey’s idea of the reflective life likewise arises out of a relatively “local” cultural situation. In Dewey’s case his ideal of the reflective life arises out of his commitments to individualism and to democracy and he is aware that these commitments are historically fortuitous. Nevertheless, he finds these commitments to embody not only social and political ideals worthy of widespread commitment but also a moral ideal likewise worthy of widespread commitment. The reflective life as Dewey conceived it has three constituents. First, to live reflectively is to live autonomously in self-conscious independence of environing custom and tradition. Second, to live reflectively is to live both in harmony with and in creative tension with a nurturing community, seeking both ones own integrity and also seeking the integrity of ones community. Finally, to live reflectively is to seek realization of ideal values that define individual and communal aspirations for a better, and a common, future. These three features of the reflective life are implicit in Dewey’s myriad discussions of issues about value. They are not necessary or sufficient conditions for his idea of the good life; they are, rather, basic commitments of his work and basic to his understanding of the good life. A few remarks about each. Autonomy

Dewey respects individuals and he approves social and political structures that nurture individuals to their fullest realizations. He values democracy as a moral ideal precisely because it allows maximal empowerment of individuals to realize their own goals. Individual growth Dewey notoriously holds to be the goal of morality: The only distinction that can be drawn without reducing morals to conventionality, self-righteous complacency, or a hopeless and harsh struggle for the unattainable, is that between the attained static, and the moving, dynamic self.15

His profound respect for individuals and their potential, and his great respect for individual autonomy, are integral to his respect for democracy. He does see these values as united in their opposition to custom and convention; simply following custom and convention is abdication of reflective autonomy and loss of opportunity for growth into more vital

M ORAL THEORY AND THE REFLECTIVE LIFE 35 regions of selfhood. The health, and thus the autonomy, of individuals is at the center of Dewey’s idea of the reflective life. The following passage makes the point effectively: The growing, enlarging, liberated self . . . goes forth to meet new demands and occasions, and readapts and remakes itself in the process. It welcomes untried situations. The necessity for the choice between the interests of the old and of the forming, moving, self is recurrent. It is found at every stage of civilization and every period of life . . . For everywhere there is an opportunity and a need to go beyond what one has been . . . Indeed, we may say that the good person is precisely the one who is most conscious of the alternative, and is the most concerned to find openings for the newly forming or growing self; since no matter how “good” he has been, he becomes “bad” . . . as soon as he fails to respond to the demand for growth. Any other basis for judging the moral status of the self is conventional. 16

In addition to autonomy Dewey is committed to community. Community

Dewey’s individual autonomy is different from the “cartoon” autonomy lampooned by communitarian thinkers like Alasdair MacIntyre. In Dewey’s thought, even the fullest realization of individual autonomy is not incompatible with genuine community; on the contrary, Dewey’s autonomy is inseparable from deep respect for community. In his understanding of them, individual goods—including the good of individual autonomy—are not only inseparable from, but are supportive of, community goods. Individual goods and community goods, in Dewey’s vision, cohere in symbiotic harmony. The coherence of these two kinds of goods in his work is frequently in evidence, as in the following statement: When selfhood is taken for what it is, something existing in relationships to others and not in unreal isolation, independence of judgment, personal insight, integrity and initiative, become indispensable excellences from the social point of view.17

The community that nurtures individuals to maturity needs their independence as much as it needs their faithfulness; Dewey thus sees no incompatibility between their independence and their faithfulness. The relationship between individuals and their communities should be characterized by creative tension, the tension produced by individuals’ pursuits of their own visions of the good and the natural inertia of the community’s entrenched vision of the good. Even in ideally healthy communities, certainly in actual communities, this tension is a daily fact of individual life.

S TUART ROSENBAUM 36 This tension between individuals’s pursuits of their own goods and the inertia of community traditions and customs concerning the good is prominent in literature and probably is so because of its prominence in daily life. In Thomas Hardy’s Far From the Madding Crowd for example, this tension is a defining feature of Gabriel Oak’s life from the time he first meets Bathsheba Everdeen. Gabriel’s desire to court and marry Bathsheba is frustrated by the fact that they live in different social and economic classes. She lives the leisure life of a privileged aristocrat, while Gabriel earns his wages as her common laborer. The inertia of social convention keeps them apart and directly conflicts with Gabriel’s hope of bringing them together. As it is with Gabriel and Bathsheba, so it is with many others; their situation is typical. The various themes wrought by this ubiquitous tension in individual lives are significantly responsible for the color and texture of great literature. The need to choose between a personal good and a community’s entrenched vision of the good is a familiar fact of life. Imagination usually urges individuals toward their own personal goods and prudence, on the other hand, usually urges away from their own personal goods and toward preservation of the community’s good. From the perspective of the personal, one ought to do a thing—as Gabriel ought to declare to Bathsheba his undying ardor; from the standpoint of the conventional, one ought not to do the same thing—as Gabriel not only ought not to declare his love to Bathsheba but ought even to repress it forever and seek a mate among his own class. Dewey is aware of this “split” in the idea of the moral between customary expectation and personal growth, as are the other thinkers mentioned above (Nagel, Scanlon, Baier, Nussbaum, and Singer), and he sees that individuals must regularly make choices in virtue of it. He acknowledges prudence as a way of preserving the good that already is, and he acknowledges imagination as a way of seeking the good of growth beyond what already is. His realization that these goods, in all the various forms in which they present themselves, are vital, inevitable, and frequently incommensurable is what leads him to see the reflective life as incorporating both the autonomy of individuals as well as the symbiotic harmony—a harmony not without conflict—between individual goods and social goods. Reflective individuals respect the healthy tension between their own good and larger social goods. In addition to autonomy and community Dewey is committed to the ideal.

M ORAL THEORY AND THE REFLECTIVE LIFE 37 The Ideal

The reflective life also involves commitment to ideal values that guide decision making. For individuals this commitment, usually a vision of the self they might realize, guides their striving for personal growth. For communities, this commitment expresses an idea of a healthy properly functioning community, a community perhaps like Plato’s Republic in which all parts work toward the mutual good of all citizens. Dewey’s own ideal is expressed in his life-long commitment to democracy. His commitment to democracy is a function of the fact that it not only respects but it nourishes individual integrity, and it does so as well as or better than any other idea of community structure. He explains his commitment in Freedom and Culture as follows: . . . the source of the American democratic tradition is moral—not technical, abstract, narrowly political nor materially utilitarian. It is moral because based on faith in the ability of human nature to achieve freedom for individuals accompanied with respect and regard for other persons and with social stability built on cohesion instead of coercion.18

That his commitment to democracy was the center of Dewey’s moral life should be uncontroversial. That it was the center of his moral life means that his entire life, and perhaps especially his intellectual life, was devoted to evincing its vitality as an ideal and to realizing conditions for its flourishing. From Freedom and Culture again: The struggle for democracy has to be maintained on as many fronts as culture has aspects: political, economic, international, educational, scientific and artistic, religious. The fact that we now have to accomplish of set purpose what in an earlier period was more or less a gift of grace renders the problem a moral one to be worked out on moral grounds . . . An American democracy can serve the world only as it demonstrates in the conduct of its own life the efficacy of plural, partial, and experimental methods in securing and maintaining an ever-increasing release of the powers of human nature, in service of a freedom which is cooperative and a cooperation which is voluntary.19

Much more about the reflective life as an ideal remains to be said. As I have characterized it here, however, it is consistent with commitment to other ideal ends than the democracy that is so important in Dewey’s own life and work. The commitments to autonomy and community are consistent with commitments, for example, to Marxism or to Christianity; one can imagine an instantiation of Dewey’s reflective life in a Marxist or in a Christian. Perhaps Dewey would urge that many ideals—including even Marxism or Christianity—implicitly include,

S TUART ROSENBAUM 38 at some level, a commitment to democracy not unlike his own. However he or a defender of his particular ideal might face these opponents, what remains is Dewey’s understanding of the reflective ideal as involving not just autonomy, not just commitment to community, but also commitment to ideal ends that bring unity to individuals’ activities and strivings. In Dewey’s own case, this particular commitment to ideal ends is a commitment to democracy as a moral ideal. Although the general ideal of the reflective life as described here has some specificity—it has three components as I mentioned above—it remains unfortunately diffuse; it lacks the crispness of a moral theory, of even a controversial and unsatisfying moral theory. Think of Mill, Ross, Moore, Rawls, Gewirth, Donagan, or even of Alasdair MacIntyre, Mark Johnson, or Allan Gibbard. The large scale differences between any of these thinkers and Dewey are impressive. Dewey’s commitments are vague and diffuse, and frustrating in a way the commitments of these other thinkers are not; they do not, for example, admit of counterexample! Nevertheless, if Baier, Nussbaum, Nagel, Scanlon, and Singer are right about moral theory—and I see their mutual agreement simply as confirming Dewey’s own work against moral theory—then moral philosophy has no recourse other than the effort to conceive and make plausible such vague and diffuse ideals. This recourse for philosophy does give it purpose beyond the futile seeming efforts of moral theorists. This recourse for philosophy also brings us together with fellow intellectuals in a way that moral theory probably separates us from them; as Baier puts this point, . . . we philosophers need to work with anthropologists, sociologists, sociobiologists, psychologists, to find out what an actual morality is; we need to read history to find out how it has changed itself, to read novels to see how it might change again.20

Moral philosophy in this sense is more inclusive than moral theory. Furthermore, although it is vague, this idea of the reflective life is one that endlessly challenges those who seek to live it, for one may always question the extent to which one is successful at living it. The following questions are only some that attend efforts to live the reflective life: In what ways does one succeed or fail in the effort to achieve autonomy? In what ways does one succeed in respecting community values to the extent required to be supportive of those values? When one trangresses community values does one do so in the right way, the way that appropriately respects those values while at the same time acting against them? Are one’s ideals better than alternative ideals? Is democracy as a final value preferable to Marxism or Christianity? These questions are such questions as produce not only widespread concern and conversation but are

M ORAL THEORY AND THE REFLECTIVE LIFE 39 such questions as one lives answers to. Thus, although it is vague, the ideal of the reflective life makes real in thought the real difficulties that attend most efforts to constitute morally satisfying lives. Stuart Rosenbaum, Department of Philosophy, Baylor University, Waco, TX 767987273; [email protected]

NOTES 1. Marcus G. Singer, “The Ideal of a Rational Morality,” Proceedings and Addresses of the American Philosophical Association 60, No. 1 (September 1986): 16. 2. This claim about Dewey of course needs substantiation. At this point, all I can reasonably do by way of substantiating it is to appeal to its intrinsic plausibility for those familiar with his work on ethics and value. I do have a lengthy essay, “Dewey and the Reflective Life,” that defends the claim in detail, but since it is yet unpublished I cannot reference it here. 3. Thomas Nagel, Mortal Questions (Cambridge: Cambridge University Press, 1979), 131. 4. Ibid., 132. 5. Ibid., 138. 6. Annette Baier, Postures of the Mind (Minneapolis: University of Minnesota, 1985), 224. 7. Ibid., 225. Baier’s account of Hume exhibits unmistakeable overtones of Dewey’s notion of “intelligently controlled habit,” or more generally of the first few chapters of Human Nature and Conduct, MW 14: 13-62, esp. 23. (Citations to the work of John Dewey refer to the critical edition published by Southern Illinois University Press. The initials of the series are followed by volume and page numbers. Abbreviations for the critical edition are EW: The Early Works [1882-1898]; MW: The Middle Works [1899-1924]; LW: The Later Works [19251953].) 8. T. M. Scanlon, “Moral Theory: Understanding and Disagreement,” Philosophy and Phenomenological Research LV, No. 2 (June 1995): 348-349. 9. Ibid., 356. 10. Martha C. Nussbaum, The Fragility of Goodness (Cambridge: Cambridge University Press, 1986), 291. 11. An interesting reaction to this question comes from Allan Gibbard, “Why Theorize How to Live with Each Other?” Philosophy and Phenomenological Research

S TUART ROSENBAUM 40 LV, No. 2, June, 1995. Gibbard seems defensive but committed to the theoretical result, and to the expertise, that might issue from what he calls “the hard slog” through the maze of complexity that naturally attends moral theory. 12. Richard Rorty, “Human Rights, Rationality, and Sentimentality,” The Yale Review 81 (October 1993): 11. 13. Ibid., 14. 14. Putnam’s essay is “How Not to Solve Ethical Problems,” in Realism with a Human Face (Cambridge: Harvard University Press, 1990), 179-192. McDermott’s is “Pragmatic Sensibility: The Morality of Experience,” in New Directions in Ethics, ed. Joseph P. DeMarco and Richard M. Fox (New York: Routledge and Kegan Paul, 1986). 15. Ethics, LW 7: 307. 16. Ibid. 17. Ibid., 300. 18. Freedom and Culture, LW 13: 178. 19. Ibid., 186-187. 20. Baier, op. cit., 224.

FEMINIST ETHICAL THEORY

I will treat feminist ethical theory as a distinct type of theory. Although some feminists are skeptical about the need for theory as distinct from cultivating practices of being morally perceptive and sensitive, many others argue for the theory they see as needed. Feminist ethical theory usu-

Virginia Held

ally includes, but is not limited to, the concerns that have been developed under the heading of ‘the ethics of care’ or ‘care ethics’. Care ethics are usually contrasted with ethics of justice, such as Kantian and utilitarian moral theories. Instead of being a theory primarily focused on right action, an ethic of care seeks moral evaluations of relations between persons, and reinterprets both personal and political relations in light of the value of care. I will show how feminist ethical theory differs from virtue theory as well as from Kantian and utilitarian theories.

F

eminist moral philosophers are a diverse group holding a variety of views. Some feminists subscribe to a modified Kantian moral theory, some to modified liberal-contractualist moral theory, some are virtue ethicists. Some criticize the very attempt to capture the concerns of morality in a theory; they favor the cultivation of ethical sensitivity, especially through narrative understanding, though to argue for doing so involves theorizing. Despite such variegated thinking, something has emerged, I believe, that can be characterized as feminist moral theory, and that can be seen as a distinct type of ethical theory.1 I will in this short paper

41

V IRGINIA HELD 42 focus on some important features of this theory, showing how it is a deep challenge to, and not a mere extension of or addition to, the other major types of theory in contrast with which it has been and is still being developed. Feminist ethical theory usually includes the concerns that have been developed under the heading of “the ethics of care” or “care ethics.” But feminist ethical theory should not simply be equated with the ethics of care, since feminist debates have led to rather widespread acceptance of the position that care ethics alone may be inadequate. This does not mean that feminist moral theorists advocating an ethics of care find any other type of ethical theory acceptable if only it is supplemented with some concerns about care. But they may recognize that care ethics, while a fundamental challenge to other types of ethical theory, needs to include or be complemented by other moral theory than itself. How these issues are to be sorted out is still very much under discussion, as should be expected for so distinct a way of thinking that is only a few decades into its explorations. The ethics of care emerged as the bias built into the dominant moral theories was uncovered by feminist critiques (Card, Gilligan, Jaggar, Noddings). In Margaret Walker’s interpretation, the dominant “theoretical-juridical” accounts of morality, of which Kantian ethics and utilitarianism are such clear examples, repeatedly invoke the image of “a fraternity of independent peers invoking laws to deliver verdicts with authority.”2 In another account, the dominant moral theories give primacy to values such as autonomy, independence, non-interference, self-determination, fairness, and rights, and involve a “systematic devaluing of notions of interdependence, relatedness, and positive involvement” in the lives of others.3 In such theories, the experience of women involved in childcare is thought irrelevant to morality: it is seen as natural, instinctive behavior, or the morally neutral exercise of a personal preference, or it is overlooked altogether. Not only is the experience of women caring for children and others excluded; so is that of most people disadvantaged economically, educationally and racially, and so is that of children. And yet, Walker writes, such theoretical-juridical accounts are standardly put forward as appropriate for “the” moral agent, or for deciding rationally how “we” ought to act. “The canonical form of moral judgment in moral theories,” Walker and others conclude, is thus the judgment of someone resembling “a judge, manager, bureaucrat, or gamesman. . . .”4 This shows how moral philosophers have represented in abstract and idealized forms the judgments of some people in a dominant social order. It does not establish that this is how morality ought to be represented, since doing so relegates the experience of so many, for instance, of women in caretaking activities or of blacks

F EMINIST ETHICAL THEORY 43 forging communal ties for survival, to the status of the nonmoral or the morally deficient. Feminists insist that the experience of women is as relevant to morality as any other experience, and not only when women enter male or “public” realms, but anywhere. Those who care for children, for instance, continually face moral problems and engage in activities in constant need of moral evaluation(Ruddick). Feminist moral philosophers challenge the rationalistic and individualistic assumptions of dominant moral theories, and the primacy given to the dominant values. The ethics of care gives expression to women’s experience of empathy, of mutual trust, of emotions highly relevant to morality. It sees human beings as interdependent and connected, and makes clear the deficiencies of the dominant moral theories in evaluating the moral aspects of human life in families and among friends. But the ethics of care is not then limited to the context of “private” life: having rethought the assumptions of the dominant moral theories, it offers a challenge to any moral theory for any context. 5 Within dominant ethical theory, a great deal of attention has been devoted to the differences between deontological, especially Kantian, theory, and consequentialist, especially utilitarian, theory. But from the perspective of an ethic of care, the similarities are more pronounced than the differences: both rely on a single ultimate universal principle (the Categorial Imperative or the Principle of Utility); both are rationalistic in their moral epistemologies; both are built on a concept of the person that stresses our individualism and independence; and both are theories of right action that seek to recommend rational decisions. Both can be characterized as ethics of justice, and it is in contrast with both that the ethics of care has been and continues to be developed. 6 The ethics of care, first, appreciates rather than is suspicious of, our ties to particular others with whom we have actual relationships. The universalistic and abstract aspects of the dominant moral theories of impartial rules thus become subject to critique rather than being beyond question (Benhabib). Dominant moral theories interpret moral problems as standardly occurring in a conflict between an individual ego and universal principles by which an individual should be constrained as he pursues his interests. An ethic of care, in contrast, focuses on the area between these extremes, valuing particular others in relationships with us. Those who sincerely care for others act for particular others and for the particular relationships between us, not for all rational beings or the greatest number, and not for their own individual interests. We do not play with our children out of respect for the moral

V IRGINIA HELD 44 law, and in important ways we care for those we take care of for their sakes, not for our own satisfaction. When dominant moral theories attend to relations of family and friendship, if they do so at all, they may see them as permitted. But they do not allow the values of such actual relations ever to take priority over the universal requirements of impartial morality. Those motivated by particular attachments are suspected of favoritism or tribalism or religious intolerance; universal principles, it is claimed, must always be accorded priority. An ethic of care, in contrast, holds that actual relations between actual persons may at times be more morally compelling than abstract rules. Or it may be held that universal rules are inapplicable or inappropriate in certain contexts (Friedman), though we should certainly make moral evaluations of relations between persons. The dominant moral theories claim to offer moral guidance for all moral contexts: if moral rules do not apply in certain kinds of cases, they are not considered moral cases. As Susan Mendus notes, however, applying impartial moral rules to love and friendship is not only inexpedient, it is a “deformed model” to apply to such contexts. 7 And yet we need to consider when relationships are good or bad, trusting, considerate, and caring, or mistrustful, manipulative, and exploitative; we need to make moral evaluations in this domain as much as in any other. Relationships between people cannot be “outside” morality, as the ties between women and children in the family have standardly been construed as being. Of course care ethics is not saying that we can never make universal judgments about what counts as good care, or what makes a relationship exploitative. On the contrary, care ethics distinguishes itself from the relativism of a purely case by case or situational approach. But in deciding when an exception to a rule should be made, justice ethics must posit another rule, and rationally decide which has priority. Care ethics can argue that informed sensitivity and responsiveness to persons and their relationships and to the situations they are in can be a better guide. Second, the ethics of care values rather than rejects the contribution of the emotions to morality. In trying to figure out what we ought to do, care ethics asserts that sensitivity, sympathy, empathy, and solidarity of feeling are often appropriate sources of insight. When rationalist moral theories consider the emotions, they tend to refer to egoistic or aggressive impulses that threaten an agent’s ability to act in accordance with moral principles. An ethic of care, in contrast, appreciates that there are, as well, “moral emotions” (Baier). Since even these can be misguided, as in the kind of caring that degenerates into undue control and domination, or the kind of excessive concern for others that

F EMINIST ETHICAL THEORY 45 undermines a sense of self-worth, we need an ethic of care, not the mere fact of caring behavior. But both in understanding what morality recommends as well as in acting accordingly, an ethic of care shows us how we should value and cultivate appropriate emotions. It shows us how moral theories and moral epistemologies that rely entirely on reason and rational deductions or calculations are deficient. Third, and perhaps most important, the ethics of care employs a conception of persons as inherently interconnected rather than fundamentally individualistic (Meyers). Dominant moral theories see the person in terms of the liberal notion of rational, autonomous moral agent, or of self-interested individual. Society is seen as made up of independent entities who are constrained by agreed upon contracts and who cooperate to further their own interests. Liberals often point out that these interests are not necessarily selfish: a person may enjoy making a child happy. But an ethic of care sees our relationships to others as much more than optional extras we can choose to engage ourselves in, or not. The liberal concept of the person conceptualizes us as individuals first, autonomously deciding how to act towards others. Our identity depends on our individuality. What connects us to others is thought of as less fundamental. In contrast, the ethics of care sees persons as inherently connected and interdependent, though for certain limited purposes we can agree to treat each other as if we were liberal individuals. To the ethics of care, our relationships are part of what we inherently are, and many such relationships are not voluntary. Every person starts out as a child dependent on others for many years; we will never cease to be the child of our parents, and to have been formed in a certain social context. If, as adults, we are able to think and act independently, this is because a network of social relations enables us to. Of course, for argument’s sake, we can assume, as social contract theorists do, that we are rational, self-sufficient individuals. But doing so can seriously distort morality and social and political thinking, because the degree to which such an assumption does not and should not describe reality is so easily forgotten (Kittay). The currently overwhelming pervasiveness of market-oriented contractualist thinking about morality, education, and healthcare, as well as about political institutions, makes clear how easily it is forgotten. We know we can imagine ourselves as abstract individuals, but we should consider the implications and effects of supposing that this conception of “economic man” is the appropriate concept of the person for our moral thinking. The ethics of care elaborates an alternative: the concept of the relational person. As initially developed, the ethics of care was sometimes proposed as an ethic that could take the place of the ethics of justice. But many

V IRGINIA HELD 46 feminists have argued that the concerns of justice, with its framework of rights and obligations and its attention to equality and liberty, should by no means be abandoned. To many, then, ways need to be found to integrate justice and care, to combine their moral concerns. Some advocates of an ethic of care want to incorporate justice into an expanded ethic of care: within social relations of a caring kind we should, for instance, recognize the claims of justice. Others see the claims of justice and care as comparably important for morality. Grace Clement argues that both the ethics of justice and the ethics of care “capture something real and important about morality, and that insisting on ranking them trivializes the contributions of the ethic deemed less basic.” 8 But like other advocates of care ethics, she opposes assimilating the ethic of care to that of justice, since that would preserve “the traditional hierarchy,” and would interpret care “through the perspective of justice, thereby devaluing and marginalizing it.”9 No advocate of the ethics of care seems willing to see it as a less than equally valuable moral outlook, though many wish to mesh it with the ethics of justice. And almost all advocates of care ethics reject the view that although care may be a suitable ethic for the “private” realm of personal relations, it is not an ethic for the “public” and political domain of relations between strangers. Fiona Robinson develops a “critical ethic of care” quite capable of moving beyond personal relations and dealing with issues of national and global injustice. “The ethics of care,” she writes, “undermines the individualistic moral logic that leads us to believe that rights and obligations are somehow disconnected from the networks of social relations in which actors—from individuals to states—are situated.”10 Neil MacCormick has observed that “justice matters to people who are already in community with each other.” 11 Those advocating an ethic of care can argue that caring relations could gradually build a community encompassing the globe. How to distinguish the ethics of care from the ethics of justice and thus from Kantian, utilitarian, and contractualist moral theory has been extensively explored and is by now fairly clear. How to distinguish the ethics of care from virtue ethics is less well delineated, since many of the concerns of the ethics of care are closer to those of Aristotle, Hume, and contemporary virtue theorists than to those of the ethics of justice. Both the ethics of care and virtue ethics seek morality in practices and the values they embody. Both understand that morality must be nurtured and cultivated in the emotions as well as reflected in thinking. But as so far developed, virtue ethics has not paid adequate attention to the practices of caring in which women and disadvantaged members of societies have had so much relevant experience. And virtue theory has

F EMINIST ETHICAL THEORY 47 often seen the virtues as incorporated in various traditions or traditional communities. The ethics of care, in contrast, is wary of existing traditions and traditional communities: virtually all are patriarchal. The ethics of care is not the actual morality of caring as practiced in societies structured by gender domination. It envisions, instead, how caring should be practiced in all social contexts in postpatiarchal society, of which we have no traditions or experience. The virtues as usually discussed in virtue ethics attach to individual persons, as dispositions. The ethics of care, on the other hand, focuses on caring relations between persons, relations such as trust, and mutual responsiveness to need, on both the personal and wider social level. The values of relations cannot be reduced to the values of the dispositions of the persons in the relation, or to the value of the relation to the person as an individual. Care should thus not be seen as just another disposition in the seemingly endless catalogue of virtues. And the ethics of care characteristically requires objective evaluations of the effectiveness of caring practices, evaluations that virtue ethics is often not equipped to supply. For various reasons, then, the ethics of care should be seen as distinct from both the ethics of justice and virtue theory. Care is more than yet another virtuous disposition, as it is more than a series of acts, and more than a value among others. It seeks to develop caring relations within which we will trust each other enough, respond enough to each other, care enough about each other to respect each others’ rights, act for the good of all, and be virtuous. The history of philosophy that has constructed the “Man of Reason” is a biased enterprise (Lloyd). So is the history that has envisioned the “Man of Virtue.” The ramifications of these in moral philosophy and in all that it influences are now being called into question by feminist ethical theory. The possibilities for a type of ethical theory that can transcend the distortions of gender are beginning to be apparent. Virginia Held, City University of New York, New York, NY 10036; [email protected]

NOTES 1. I do not share the views of those writers who try to distinguish the terms ‘ethical’ and ‘moral’; I here use the terms as more or less interchangeable, though of course we should distinguish actual moral beliefs and ethical practices from moral evaluations of them.

V IRGINIA HELD 48 2. Margaret Urban Walker, Moral Understandings: A Feminist Study in Ethics (New York: Routledge, 1998), 21. 3. Fiona Robinson, Globalizing Care: Ethics, Feminist Theory, and International Relations (Boulder, CO: Westview Press, forthcoming), 10. 4. Walker, op.cit. 5. See my Feminist Morality: Transforming Culture, Society, and Politics (Chicago: University of Chicago Press, 1993). 6. See Justice and Care: Essential Readings in Feminist Ethics, ed. Virginia Held (Boulder, CO: Westview Press, 1995). 7. Susan Mendus, “Some Mistakes about Impartiality,” Political Studies 44 (1996): 319-327. 8. Grace Clement, Care, Autonomy, and Justice (Boulder, CO: Westview Press, 1996), 117. 9. Ibid., 5. 10. Robinson, op. cit., 47. 11. Neil MacCormick, “Justice as Anti-Contractualist Impartiality: Assenting with Reservations” Political Studies 44 (1996): 309.

BIBLIOGRAPHY Baier, Annette. 1994. Moral Prejudices: Essays on Ethics. Cambridge, MA:: Harvard University Press. Benhabib, Seyla. 1992. Situating The Self: Gender, Community, and Postmodernism in Contemporary Ethics. New York: Routledge. Card, Claudia, ed. 1991. Feminist Ethics. Lawrence: University Press of Kansas. Clement, Grace. 1996. Care, Autonomy, and Justice. Boulder, CO: Westview Press. Friedman, Marilyn. 1993. What Are Friends For? Feminist Perspectives On Personal Relationships and Moral Theory. New York: Cornell University Press. Gilligan, Carol. 1982. In a Different Voice: Psychological Theory and Women’s Development. Cambridge, MA: Harvard University Press. Held, Virginia. 1993. Feminist Morality: Transforming Culture, Society, and Politics. Chicago: University of Chicago Press. Held, Virginia, ed. 1995. Justice and Care: Essential Readings in Feminist Ethics. Boulder, CO: Westview Press.

F EMINIST ETHICAL THEORY 49 Jaggar, Alison M. 1991. “Feminist Ethics: Projects, Problems, Prospects,” in Feminist Ethics, edited by Claudia Card. Kittay, Eva Feder. 1996. “Taking Dependency Seriously,” Hypatia 10.1: 8-29. Lloyd, Genevieve. 1984. The Man of Reason: “Male” and “Female” in Western Philosophy. Minneapolis: University of Minnesota Press. MacCormick, Neil. 1996. “Justice as Anti-Contractualist Impartiality: Assenting with Reservations.” Political Studies 44: 305-310. Mendus, Susan. 1996. “Some Mistakes about Impartiality,” Political Studies 44, 319-327. Meyers, Diana, ed., Feminists Rethink the Self. (Boulder, CO: Westview Press, 1997). Noddings, Nel. 1984. Caring: A Feminine Approach to Ethics and Moral Education. Berkeley: University of California Press Robinson, Fiona. Forthcoming. Globalizing Care: Ethics, Feminist Theory, and International Relations. Boulder, CO: Westview Press. Ruddick, Sara. 1989. Maternal Thinking: Towards a Politics Of Peace. Boston: Beacon Press. Walker, Margaret Urban. 1998. Moral Understandings: A Feminist Study in Ethics. New York: Routledge.

MORAL THEORIES AND VIRTUE ETHICS

The recent revival of virtue ethics may have a salutary effect on normative ethical theory. Over the past few years, an ‘agent-based’ virtue ethics inspired by the moral sentimentalism of Hutcheson, Hume, Martineau, and (more recently) Nel Noddings has taken shape. Because this approach

Michael Slote

allows room for a generalized humanitarianism that is notably absent in Aristotle, it may have more contemporary promise than neo-Aristotelian views. But agent-based virtue ethics also enables us to make some new distinctions within more familiar views and to that extent, therefore, has something to offer advocates of the very approaches with which it disagrees.

N

owadays, when philosophers distinguish among moral theories,they mention not only differences in their implications for particular cases but structural differences. For example, even where utilitarianism comes close in its particular judgments to what Kantian or common-sense ethics has to say, its explanation of why a given action is morally right or wrong will be highly distinctive, and the character of that explanation will depend on the agent-neutral and teleological character of utilitarianism. Those who have helped to revive virtue ethics during recent decades have by and large objected to the theoretical character of most modern moral theories, but lately various defenders of virtue ethics have argued that virtue ethics has or should have a distinctive theoretical approach of its own. On that assumption, the difference from other moral theories would then lie chiefly in the way virtue ethics is organized as a theory, i.e., in certain general theoretical features that are arguably preferable to what is to be found in other approaches to ethics. Most of 51

M ICHAEL SLOTE 52 the virtue ethics we have seen recently has been in some sense neoAristotelian, but those who nowadays want to see virtue ethics developed in a theoretical fashion are not all followers of Aristotle. Although Rosalind Hursthouse has defended neo-Aristotelianism as a kind of theory, others, like Julia Annas and myself, have been exploring the possibilities of virtue theory grounded, respectively, in Stoic ideas and in the moral sentimentalism of Frances Hutcheson, David Hume, and James Martineau.1 And today I want to show you that the latter approach offers new possibilities for selfunderstanding to utilitarianism and also can be developed in its own right in some surprising, but promising directions. Sentimentalist virtue ethics (which is one form of what can be called “agent-based” virtue ethics)2 grows out of the moral sentimentalism of Hutcheson and Hume, but moves decisively toward characterizing morality in terms of the inner life of the moral agent in a way that sentimentalism left ambiguous or only partly realized. Hutcheson, for example, held that universal benevolence is in itself and without considering its consequences, the morally best of motives. At the same time, however, he was historically an early advocate of the principle of utility, so that his overall theory is something of a hybrid: motives are evaluated “in themselves,” but actions are assessed as morally better or worse depending on their (actual) consequences for overall human happiness. It is well known that utilitarianism as a mature moral philosophy has its roots in eighteenth-century moral sentimentalism, and (non-hybrid) utilitarianism certainly results if one begins with the hybrid theory of Hutcheson and alters it so that motives as well as actions end up being evaluated in terms of their consequences. But it is also possible to change the hybrid character of Hutcheson’s theory in just the opposite direction. If, like Hutcheson, one evaluates motives intrinsically and independently of their consequences, one can also start evaluating actions in terms of their inherent or underlying motivation rather than by reference to their consequences, and given Hutcheson’s original view, this yields a (nonhybrid) theory that says that universal benevolence is the morally best or most admirable of motives and that actions are right or good to the extent they reflect or express that motive. Notice that an act reflecting that motive will aim at good consequences—but that is not to say it will actually achieve them, so a method of act-evaluation based on this criterion will not be consequentialistic. In fact, it will be virtue-ethical inasmuch as it makes the inner life, and in particular a certain motive, criterial for act-evaluation. But since that motive inherently aims at overall (or impartially measured) good consequences, this virtue-ethical view, which we can call morality as universal benevolence, will be a kind of interiorization of utilitarianism: it will be

M ORAL T HEORIES AND VIRTUE ETHICS 53 utilitarianism, as it were, turned outside in.3 Hybrid moral sentimentalism in actual history turned toward utilitarianism, rather than toward the virtue-ethical analogue of utilitarianism just mentioned. But the latter turn is now possible, though I believe it took a considerable revival of virtue ethics to allow us to see it as a distinct possibility and, at the same time, to recognize that Hutcheson’s hybrid theory didn’t have to be resolved in the utilitarian direction. Furthermore, once one sees this possibility, one can easily wonder whether or why a utilitarian should prefer a consequentialistic theory to a virtue-ethical morality as universal benevolence. The utilitarian evaluates actions in terms of whether they achieve the goals of universal benevolence and in so doing he treats universal benevolence as a touchstone or fulcrum for moral evaluation. But at the same time the utilitarian says that that very motive would not be morally good if, unbeknowst to us, it had bad consequences on the whole (it would then not achieve its own goals); and the utilitarian also has to say that we act wrongly when our best efforts to help people end up being counterproductive. Morality as universal benevolence embodies the motive of universal benevolence without having either of these (to my mind, counterintuitive) implications, and this may make such a view more attractive than utilitarianism for (some of) those who are (already) attracted to utilitarianism. In any event, it has to be somewhat surprising that utilitarianism has so much in common with a certain form of virtue ethics, and in the light of our discussion too, it turns out that (act) utilitarianism involves a certain kind of theoretical choice that utilitarianism itself has never recognized. Virtue ethics helps us to see that there are (at least) three possible ways to center a moral theory around universal benevolence, and it is therefore now encumbent on the defender of (act) utilitarianism to explain why her consistently consequentialistic way of doing this is superior to the other theoretical possibilities. However, even if a virtue-ethical morality as universal benevolence is superior to utilitarianism and to Hutcheson’s hybrid moral sentimentalism, the advocate of sentimentalism-inspired virtue ethics may still wonder whether universal benevolence is the morally best of motives. For even if one thinks that concern for (other) human beings is the basis for all morality (rather than the less fundamental and somewhat derivative matter it is for Kantian ethics), it is not clear that such concern is best or most admirably expressed in the aggregative and impartialistic terms endemic to morality as universal benevolence. Let me explain. Morality as universal benevolence, utilitarianism, and Hutcheson’s hybrid sentimentalism all reflect impartiality toward, and a preference for what is best on the whole for, those who are the objects of one’s concern.

M ICHAEL SLOTE 54 But I believe that this way of desiring and seeking people’s well-being is at least sometimes less than morally ideal, and in that case there is probably a better way of embodying humane concern for people within a moral theory or view. In particular, I want now to argue that love is (sometimes) nonaggregative in the way it aims at the good of others, and to the extent we think highly of love, we have reason to question any moral view based in a purely aggregative conception of the aims and concerns of morality. If a parent has two children and loves them equally, what he feels for the one and what he feels for the other do not naturally amalgamate into some overall larger concern for their aggregate well-being. Instead, those concerns in some sense remain separate from one another, and by that I (at the very least) mean that a loving parent will not always seek to do what is good or best on the whole for his or her two (or more) children. Thus imagine that a father has two children in their twenties, one independent and successful, the other dependent and handicapped. If he loves them both (equally), he will invariably make efforts on behalf of both. If, for example, there really isn’t much he can do for the handicapped child, he will tend nonetheless to expend both time and money in attempting to promote that child’s welfare—in effect, he will naturally “show the flag” of concern even where his efforts can deliver very little. But this is not (solely or even chiefly) because that is the worse-off child, so that (Rawlsian) considerations of justice impel one to do the most one can to raise the level of the worst-off objects/subjects of one’s proper concern. For that would mean, for one thing, that he should neglect or skimp on time, money, and efforts for the more advantaged child, and love wouldn’t permit such a thing. For example, consider the different possibility that the father might be in a position to do a great deal for the worse-off child and that the betteroff can manage fairly well on her own. Imagine, further, that efforts on behalf of the former will always achieve more good than the father would have been able to accomplish by instead helping or paying attention to the better-off child. In that case, aggregative thinking of a utilitarian (or Rawlsian) cast will favor always helping the worse-off child. But by its very nature love will reject this “solution” and impel the father to devote substantial time to and make substantial efforts on behalf of the better-off child. And that is because love is not aggregative, even when its concern is impartial or equal as between two individuals. Rather, love seeks some sort of balance between the objects of its concern. But if we think a father should keep individual accounts with each of his two children, rather than acting toward them in aggregative fashion, we shall want to criticize utilitarianism for having precisely the opposite opinion. And if we esteem parental and other love and think more highly

M ORAL T HEORIES AND VIRTUE ETHICS 55 of a father who keeps individual accounts with each of his children, then we also have reason to reject both Hutcheson’s sentimentalism and morality as universal benevolence for treating an aggregating attitude of universal benevolence as (in all circumstances) the morally most admirable of motives. Notice, however, that nothing in the father example yet casts doubt on the impartiality of universal benevolence. It is only the aggregating and, as we may say, the unremittingly impersonal character of such benevolence that the example leads us to question. But if a more personal attitude of love is sometimes more admirable than any aggregating attitude, we might still wonder whether it is always more admirable. The latter would mean that we should ideally love everyone, and such a view is or may well be embodied in Christian ideals of universal love. Christian or agapic love toward others is supposed to resemble God’s love for us as his children, and to the extent, therefore, that love for children or other individuals is non-aggregative in the fashion defended above, the Christian ideal of loving everyone (equally) is as impartial as utilitarianism and morality as universal benevolence, but importantly different nonetheless. This is a point that has frequently been missed by those comparing utilitarianism with the morality of agape. 4 But it also leads us to the question whether non-aggregative impartial agapic love is morally to be preferred to some sort of more mixed concern for others. People often say that impartial Christian love is beyond our powers, but it might also be said that a love that is equally concerned with every other person is or would be less ideal than a more selective attitude of loving. Perhaps love means more if one loves some people more than others, and perhaps the ideal moral attitude is not an impartialistic (though non-aggregative and personal) concern for everyone of the sort arguably recommended by Christianity, but rather a more nuanced and nonuniform attitude of concern for others. Perhaps what is ideal is that one have deep and committed feelings—love—toward some people, but that one also have some sort of broad and substantial concern for other people whom one knows more casually or only knows about. This latter, broad concern might actually be aggregative, insofar as it was not based on love for particular individuals, but on some sort of general (feeling of) humanitarianism. So I am suggesting that, once we see the limitations of aggregating moral attitudes, we have a certain amount of choice in what we advocate as morally ideal in the way of concern for other human beings. We can idealize a non-aggregating but impartial agapic attitude toward all others; but we might actually prefer a less uniform and more partial attitude as a moral ideal. Perhaps the

M ICHAEL SLOTE 56 morally ideal individual has deep love for some and operates in a nonaggregating fashion toward those special individuals; but also has lesser though still substantial aggregating concern for those s/he doesn’t love, so that, faced with a simple choice between helping more distant people or fewer, he will prefer what does the most good overall. I believe that the moral ideal just alluded to is one that grows naturally out of the “feminine” ethic of caring. As described by Carol Gilligan and Nel Noddings,5 that ethic mainly involves being concerned about certain people to whom one is or becomes intimately connected, and, as a result, one standard philosophical objection to the ethic of caring has been that it is at best incomplete, a fragment of the morality human beings need, because it doesn’t tell us what our obligations are to poor or suffering people whom we haven’t met, and this latter issue is very important to modern moral philosophy. Recently, however, both Virginia Held and I have argued that a genuine and self-standing ethic of caring can and should accommodate two kinds of caring: the deep caring for intimates that Nel Noddings, for example, exclusively emphasizes and a broad humanitarian caring that corresponds to the demands of larger moral and social problems.6 An ethic of caring with both emphases has a chance to be or become a complete approach to morality, and, given the way such a view emphasizes caring, an inner motivational state, it seems natural to consider it a form of virtue ethics. But it is a form of virtue ethics that is partialistic and (over some of the range of its concerns) non-aggregative in a way that morality as universal benevolence is not. Still, such an approach involves concern for the well-being of all (other) human beings and to that extent and because caring for others is (in Hume’s sense) a natural attitude, the caring ethic I am so briefly sketching can be traced to the moral sentimentalism of Hume and Hutcheson. It can therefore be said to be a form of sentimentalist (and certainly it is a form of agent-based) virtue ethics.7 The morality of caring I wish to defend has to say more than I just have about what it is to balance deep concern or love for others against broad humanitarian concern for people generally and about how this sort of view can be extrapolated to issues of legal and social justice. But even as a promissory note, an ethic of caring along the above lines clearly puts human well-being at the center of morality, and to that extent, it resembles utilitarianism, morality as universal benevolence, and Hutcheson’s original views more than it does the more indirect and less fundamental treatment of concern for other people that one finds in the Kantian approach to morality.8 Michael Slote, Department of Philosophy, University of Maryland, College Park, MD 20742-7615; [email protected]

M ORAL T HEORIES AND VIRTUE ETHICS 57

NOTES 1. Rosalind Hursthouse, “Virtue Theory and Abortion,” Philosophy and Public Affairs 20 (1991): 223-46; Julia Annas, The Morality of Happiness, (New York: Oxford University Press, 1993); Michael Slote, “Agent-Based Virtue Ethics,” inVirtue Ethics, edited by R. Crisp and M. Slote (New York: Oxford University Press, 1997). 2. See Slote, op. cit. 3. Utilitarianism is based on a (reductive) conception of human well-being as consisting (solely) in pleasure, happiness, or desire-satisfaction. But this is not required for morality as universal benevolence, which says that it is right to pursue human or sentient well-being, but doesn’t commit itself or the pursuers to any one (well-worked-out) conception of what well-being is. However, that is not to deny that the defender of morality as universal benevolence has the same excellent reason we all have for (also) wanting to find a good theory of human well-being. 4. For example, in Percy Shelley’s “Essay on Christianity,” (reprinted in The Necessity of Atheism and Other Essays [Buffalo: Prometheus, 1993], 8), it is assumed that Christian love must be aggregative in the way utilitarianism is. 5. See Carol Gilligan, In a Different Voice (Cambridge, MA: Harvard University Press, 1982); and Nel Noddings, Caring: A Feminine Approach to Ethics and Moral Education (Berkeley: University of California Press, 1984). 6. See Virginia Held, Feminist Morality (Chicago: University of Chicago Press, 1993); and my “The Justice of Caring,” Social Philosophy and Policy 16 (1998). 7. For an earlier form of sentimentalist-inspired virtue ethics, see James Martineau’s Types of Ethical Theory (Oxford: Clarendon Press, 1891). 8. I want to thank Sam Kerstein for extremely helpful comments and suggestions.

CAN A PARTICULARIST LEARN THE DIFFERENCE BETWEEN RIGHT AND WRONG?

This paper is an attempt to answer the charge that extreme moral particularism is unable to explain the possibility of moral concepts and our ability to acquire them. This charge is based on the claim that we acquire moral concepts from experience of instances, and that the sorts of similarities that there must be between the instances are ones that only a generalist can countenance. I argue that this inference is unsound.

Jonathan Dancy

E

thical particularism is the doctrine that moral reasons do not behave in the sorts of regular ways that they would need to behave if there were such things as moral principles. Moral principles are understood as universally quantified expressions that take us from, on the left, specifications of the features that make actions right or wrong, or in general of the morally relevant features of actions, to, on the right, specifications of the features so made—rightness, wrongness or whatever. More simply, a principle of the right, we may say, takes us from a right-making feature or a collection of such features to the rightness so made. Extreme particularists deny that there are any such principles. Moderate particularists claim only that, though there may be some, there are not enough to stand as the basis of moral rationality. Moral rationality, in their account, cannot be conceived as the consistent application of principles to cases. I do not want here to get involved in arguing the toss between extreme and moderate particularism. I am just going to pretend that particularists should be extreme particularists. This is because I want to 59

J ONATHAN DANCY 60 consider a general objection to particularism, one which, were it correct, would suffice to undermine both the extreme and the moderate form of the position. This objection is made in a forthcoming paper by Frank Jackson, Philip Pettit, and Michael Smith. 1 In order to be able to refer to this multiple author, I need a general term. Since they all work in Canberra, I will call them the Canberries. The question raised by the Canberries concerns whether there is any pattern to the way in which descriptive information determines moral conclusions. They take it that particularists deny, and must deny, the existence of any such pattern. For, if there were such a pattern, say for rightness, we could hope to specify that pattern on the left hand side of a moral principle, a principle of the right. If we found that the pattern resisted codification, this in itself is not a terribly interesting result. Particularism is the claim that there neither is nor needs to be any such pattern, not the claim that every such pattern is uncodifiable. The Canberries argue, however, that if there is a concept of rightness at all, there must be a pattern that recurs in every case, a pattern in which purely descriptive features are presented. Their argument is semantic. It starts as follows: We use words to mark divisions. Tables are different from chairs . . . wrong acts are different from right ones . . . What, then, marks off the acts we use ‘right’ for from the acts we use ‘wrong’ for? Or, equivalently, what do the right ones have in common that the wrong ones lack?

The idea is going to be that particularism has no answer to this question. To see why, we need a bit of terminology. Evaluation of every form supervenes on description. We know that if this action is right, there is a right action of the same sort done in any world that is descriptively indistinguishable from the world in which this action is done. There is then a set of world-descriptions, call them Di, such that where any one of that set is instantiated, a right action is done. We can now continue with the argument. What do the right actions all have in common? Particularists might say that the only thing they have in common is that they belong to the set of right actions. But then what is it to grasp the concept of rightness?: Grasp of the predicate ‘is right’ simply consists in a grasp of the various Di which constitute that set. But this cannot be all that unites the class of right actions. There must be some commonality in the sense of a pattern that allows projection from some sufficiently large subset of the Di to new members. If there isn’t, we finite creatures could not have grasped . . . the predicate ‘is right’. So, there must be a pattern—in the weak sense . . . of that which enables projection—uniting the set of right acts.

C AN A PARTICUL ARIST LEARN TH E D IFFERENCE B ET WEEN R IGHT AND W RO NG? THE RONG 61 They go on: But if there must be a pattern uniting the right acts, either it is a descriptive one, in which case particularism is false, or it is one which cannot be understood in terms of the presence or absence of the descriptivesomething un-analysable and nonnatural, as G. E. Moore put it when discussing goodness. Particularists would then have to postulate a class of sui generis moral properties, and their doctrine would become . . . a variety of . . . Moorean intuitionism.

The general idea, then, is that rightness is determined by descriptive features, and there must be a pattern to the way that this is done. There must, that is, be a pattern exemplified by the various sets of descriptive features that determine rightness case by case, and that pattern must itself be descriptive. But if there is such a pattern, particularism is false. I start my response to these arguments with some general points about particularism. Particularists do not deny the possibility of finding some exceptionless general truths linking descriptive predicates on the left with moral predicates on the right, of the sort generated by the supervenience of the moral on the descriptive. What they deny, as I see it, is the possibility of finding (a sufficient range of) truths of this sort that specify features that make actions right—right-making features, as we might put it. The particularist idea is that a feature that here contributes to the rightness of an act may there detract, making its owner less right than it would other-wise have been, and in another case have no moral relevance at all. Contributing features, in this sense, are variable in their moral relevance. So a proposition that had on the left hand side a description of the entire world in which an object is placed, and on the right hand side the predicate ‘is right’, would not count as the sort of principle whose possibility the particularists deny. We can easily see this once we notice that the description of an entire world would be much too indiscriminate; it would be bound to specify far more than the right-making features present in the case. It would also specify features that are vainly attempting to make the action wrong, as well as others that are morally irrelevant here. And it will fail to distinguish between the features that make the action right and those features whose presence or absence is required for the right-making features to do their job. These latter features I call ‘enabling features’. The rightness that the Canberries are concerned with is overall rightness, as we can see in the quotations given above. Focusing our attention entirely on the features that give an action its overall rightness, the rightmaking features present in the case, we notice that an action can be made right by a single feature; an action can be right just because it was friendly.

J ONATHAN DANCY 62 (Further features may need to be in place for this one consideration to be able to make the action right, of course. But these further features will be operating in the present case only as enabling features.) Other right actions, again, are made right by far more complex combinations of features. This surely tells us that the idea that, looking at the different packets of right-making features in different cases, we are likely to be able to discern anything worth calling a pattern is very suspect. Suppose that there are two actions, each right, and each made right by a single feature—but it is a different feature in each case. Any pattern we find here must, surely, be one that holds between any single feature and the action that has it; but such a notion of a pattern is useless. Matters are worse when we recognise that other instances of right actions will be made right by complex combinations of features. Is there any sensible notion of a pattern exemplified by these more complex cases as well as the two simple ones? I doubt it. The attempt to insist that we must be responding to a descriptive recurrable, called a pattern, seems to be plainly at odds with the facts. The recurrable to which we are responding is the recurring property of rightness. One could raise worries about the distinction in terms of which the Canberries make their objection, that between the moral or more generally the evaluative, on the one side, and the descriptive on the other. They say that particularists allow this distinction. But actually particularism can be expressed without it. All we need is the distinction between the moral properties of actions and the features that give those actions those properties. As particularists we deny the existence of general truths linking the generating properties and the properties generated. It is not required that the generating properties be ‘descriptive’. We might be told that the generating properties must be descriptive. But this is not obviously true, at least not at the first stage. We all know that the thin moral properties (right, wrong, good, bad, etc.) obtain in virtue of thicker evaluative properties, among others. The resultance base, as I call it, includes both evaluative and nonevaluative properties. Descending now to the level of the thick, are we forced to admit that each thick property results from (i.e., obtains in virtue of) purely descriptive properties? No: it might be that for each thick property, the ‘resultance base’ includes further thick properties. We might never find an instance of a thick property whose resultance base was purely descriptive. All this is perfectly compatible with the admission that the evaluative supervenes upon the purely descriptive. Or if it is not, so much the worse for the doctrine of supervenience, which anyway requires a rich enough notion of the ‘purely descriptive’ to ensure evaluative invariability where there is invariability of the purely descriptive.

C AN A PARTICUL ARIST LEARN TH E D IFFERENCE B ET WEEN R IGHT AND W RO NG? THE RONG 63 There is, however, not only the right-making features to think about. There is also the relation that holds between those features and the rightness that they generate. This is the ‘right-making relation’. Perhaps this relation exemplifies a pattern. But why must it? What has persuaded us that we are incapable of recognising that relation directly when it occurs? Why have we allowed that the only explanation of our ability to recognise a recurring normative feature is that we are responding to a descriptive pattern present in each case? The answer to this worry is that if we are capable of responding to the relation instance by instance, the relation is itself the pattern. For the notion of pattern at issue is merely the notion of ‘that which enables projection’. The sharp edge of the Canberries’ critique is the claim that the pattern to which we are responding in consistent use of the predicate must be discernible at the descriptive level. But even if this were so in the case of the right-making relation, it would not disprove particularism, since particularism does not deny the existence of a descriptive specification of the right-making relation. What it denies is the existence of a descriptive specification of rightness, or of the right-making properties. We might also hold that, while the features that stand in the rightmaking relation to the rightness made are themselves descriptive, the relation in which they stand can still be non-descriptive. Nothing has yet established that a pattern of descriptive features, even if one could be found, would have to be a descriptive pattern. To this, however, the Canberries have a reply: The pattern uniting the right acts might be . . . something like providing a good reason for acting, or, maybe, a pattern capturable in terms of the thick moral concepts. However, [such concepts] supervene on the descriptive in the same way that rightness does. This means that the same line of argument applies to them. [The predicates we use] for talking about them . . . must apply because of the nature of what they apply to, and there must be a pattern that makes it intelligible how we could have mastered them. But, given their supervenience on the descriptive, the choice is then as before: either there is a projectible descriptive pattern, or else the unifier is sui generis.

This takes us back to the only alternative that the Canberries are willing to countenance. Either there is a descriptive pattern to a concept, constituting that to which we are responding in consistent use of the concept from case to case, or the concept is normative, descriptively unanalysable and sui generis, and this last is held to be a variety of ‘Moorean intuitionism’. What is the worry about intuitionism? It seems to be that Moore thought of evaluative properties as sui generis, and that this

J ONATHAN DANCY 64 view of Moore’s is plain silly. But that point is surely too contentious to stand as an independent argument in the present debate. I do myself believe that evaluative properties are different in style from non-evaluative ones, and in that sense sui generis. As far as I can see, no relevant objection is in place on this front vis-à-vis the argument for or against particularism. It is impossible to object to a particularist form of intuitionism that it is intuitionist. One cannot object to a position merely by describing it. More to the point would be the thought that the intuitionists are incapable of providing any connecting link between the descriptive and the evaluative. Actions have properties of both sorts, descriptive and normative, and that is all that can be said about the matter. Now if that were true, it would be relevant to the present debate, for rightness and wrongness would have been entirely separated from the descriptive world, and our ability to cotton on to such peculiar repeating features would indeed be in doubt. So here we would have something that could stand as an objection to the intuitionist variety of particularism. And perhaps Moore may have said some things that lead one to suppose that he saw things in that way.2 But, whatever Moore may have thought, modern intuitionists make no such mistake. For them, evaluative features are held in place by other features. They express this in terms of reasons. If an action is wrong, there must be a reason why it is wrong, that is to say, features that make it wrong; and the same is presumably true for all evaluative properties, moral or otherwise. The notion of ‘being a reason for’ is sufficient to tie the features that make actions right to the rightness so made. There is no danger that the evaluative features of actions are just further, rather peculiar properties, rather like the presence of a label saying ‘Do me’ tied on to the side of the action but not properly connected to its other properties. So the dilemma that the Canberries offer the particularist fails to be at all threatening. They say that if there is a pattern uniting the right acts, it must either be descriptive or unanalysable, nonnatural, sui generis, not able to be understood in terms of the presence or absence of the descriptive. We think that it is not descriptive, and that it is sui generis, being nonnatural. But we do not think that the pattern is able to be understood without reference to the descriptive. Let us admit (pro tem) that if there are to be moral concepts, those competent with them must be able to respond to what they take to be the same feature case by case. And let us allow that this feature is to be understood as a pattern of right-making features (despite worries raised earlier about patterns and their possibility). Does it follow that the pattern itself is descriptive? It might, for all that I can see, be the case that though the features present in that pattern are descriptive enough, the pattern in which they are present is evaluative. This gives us

C AN A PARTICUL ARIST LEARN TH E D IFFERENCE B ET WEEN R IGHT AND W RO NG? THE RONG 65 a pattern for the concept, but not a descriptive pattern. For the sake of an example, the relevant pattern might be that exemplified by the right-making relation, a blatantly evaluative relation that holds between (let us allow) descriptive features on the left and evaluative ones on the right. This is a relation, and it repeats, and where it repeats we can, if evaluatively sensitive, discern its reappearance. It does not follow from this that the right-making features, case by case, appear in what the Canberries would call a descriptive pattern. 3 The Canberry response to this is that if the relation is evaluative, it must, by supervenience, be held in place by the descriptive. Given this, they write ‘the choice is then as before: either there is a projectible descriptive pattern, or else the unifier is sui generis’. But this reveals the real nature of their objection to particularism. It is an objection to non-naturalism in ethics. They don’t write ‘either there is a projectible descriptive pattern, or there is a projectible evaluative one’. The latter disjunct is ruled out just because such a pattern would be sui generis i.e., non-descriptive. At this stage the question whether one can be a particularist has become the question whether one can be a non-naturalist. Perhaps the point only is that if one is a naturalist, one cannot be a particularist, while if one is a non-naturalist, one inherits all the metaphysical difficulties of that position. 4 To repeat the main point: there is nothing wrong with a sui generis relation that supervenes upon the descriptive, so long as there is an intelligible relation between that which is held in place and that which holds it in place. The notion of a reason provides such an intelligible relation. An evaluative relation may supervene upon the descriptive without this fact generating a relevant descriptive pattern in which the features that make that relation obtain present themselves case by case. There is a projectible pattern, but it is not descriptive. It is evaluative, and sui generis therefore. This does not mean that we are floating free of the descriptive, since such patterns cannot occur in the absence of the (supposedly) descriptive features that are so patterned. That the features patterned are descriptive does nothing to show that the pattern is descriptive as well. I conclude that the semantic argument against ethical particularism fails. To be in possession of an evaluative concept is to be capable of recognising the sort of shape in which descriptive features need to present themselves if they are to generate the relevant property. Nothing that follows from that is of any danger to the particularist—at least not for the non-naturalist. But what about the naturalists? I might say—let them look after themselves. But it does seem somehow odd to suppose that particularism, which is a doctrine about the ways in which reasons relate to each

J ONATHAN DANCY 66 other, is only available to those who adopt a non-naturalistic metaphysics. Certainly that is not what one would have expected in advance. So I end by proposing two models of particularist rationality, i.e., two accounts of how it would be possible to acquire moral concepts from experience of particular cases if particularism were true. The models I propose are intended to be available to naturalists as well as to non-naturalists. Since they are to be compatible with particularism, they will be models of competence with a concept, where that concept is the concept of a resultant property, and there is nothing that repeats at the level of the resultance base and that is worth calling a pattern. Pluralists in moral theory often say that there are many different sorts of ways in which an action can get to be right—many different sets of overall right-making properties, that is—and that as a result the similarities between different right actions are all to be found at the level of the properties so made—at the resultant level, that is. Right actions are similar in being right, that is, but there is no lower-level similarity that binds all instances of rightness together. The Canberries raise the question whether this is possible. All that I am trying to do is to generalise the issue so that our attention is distracted for a while from any peculiarities that the moral case may present. I think there are ways of showing that one can build up a concept without keying into anything worth calling a lower-level pattern that repeats from one instantiation of that concept to the next. The general ideas here are not new. The first is an appeal to the nature of connectionist machines. The second is an appeal to Roschian prototypes. Connectionist machines normally consist of two or more levels of interconnected units or nodes. There are input nodes, which ‘fire’ in response to discrete features of the input or environment, or perhaps to discrete sub-features of minimal conceptual complexity. These nodes may be labelled by the experimenter, but the labels given are not themselves something of which the machine is ‘aware’. There are also output nodes, whose levels of activation are determined by the degrees of activation of the nodes below them together with the degree of ‘weighting’ associated with the links between one node and another. Finally, between input and output nodes there are normally ‘hidden nodes’, which lie like the filling of a sandwich between the input and the output layers and facilitate the processing of information in the system. There are normally no direct linkages between input and output, that is; everything travels via the intermediate, hidden level. Such a machine consists entirely of these nodes and of the connections between them. Nodes ‘fire’ in response to features of the machine’s experience or to the firings of other nodes. They are linked

C AN A PARTICUL ARIST LEARN TH E D IFFERENCE B ET WEEN R IGHT AND W RO NG? THE RONG 67 to the extent that when one fires the other does, and these firings can be stronger or weaker. Nodes can be reflexively or non-reflexively connected; the connection is reflexive if either one fires when the other one does; it may be that the second fires with the first, but not vice versa, in which case the connection is non-reflexive. And the link between one node and another may become stronger or weaker, may be dampened or brightened, by the firing of another node. (In many systems all nodes are to some degree connected to all others.) The crucial point, however, is that there is nothing more to the machine than the nodes and the current excitatory and inhibitory links between them. To the extent, then, that there is anything worth calling representation in the machine, it cannot be in the individual nodes, and must lie in the current patterns of excitatory links. But a system of this sort is capable of changing and developing over time, and so, in effect, of learning better ways of responding to input. This is done by changing the weightings associated with the links between pairs of nodes. These can be changed by the designer of the system, from outside, as it were, but also by the machine itself, in accordance with certain set constraints and goals. In changing them, it changes the strengths of the excitatory linkages between existing nodes, and thereby alters the ensemble of nodes that respond to a given structured stimulus. At any one time, therefore, there is a current pattern of excitatory links set up, with their associated weightings, and so a sort of structure to the way in which the machine responds to a complex input. And we might be tempted to think of this internal structure as the internal representation of a pattern that the machine has found in its ‘experience’, i.e., a pattern present in the input. But the machine can alter the former pattern by adapting its own internal economy, as it were, so that different nodes are now doing the work previously done by others, but in a way that leaves untouched such representing as it is capable of. The internal structure or pattern, such as it is, cannot therefore be an internal representation of a pattern found in experience. The machine’s competences, including crucially its ability to respond to repeated features of its experience, remain untouched even though the ‘support system’, as it were, alters structurally. To go into this last matter in more detail: we should distinguish between occurrent representations and standing representations. On the one hand, there is a system’s representing now that this particular act is generous, which is realised as a pattern of activation levels of particular nodes; this is the occurrent representation. On the other there is the system’s having a representation of generosity stored somewhere within it, in virtue of which it is able to recognise generosity when it comes across it; this, the standing representation, is realised as a pattern of connection

J ONATHAN DANCY 68 weights between nodes—typically hidden layer nodes. This distinction is important because it is in the latter kind of pattern that the system’s standing knowledge, or as we might put it, conceptual competence, is stored, while it is in the production on occasions of the former kind of pattern—patterns of activation—in which that standing knowledge is demonstrated. Importantly from the particularist point of view, at the level of standing representations there is what is sometimes called “deep” or “strong” context sensitivity. That is, what pattern of connectivity represents a particular concept and underlies a particular performance (the production of a certain output, e.g., ‘generous’ in response to a certain input, e.g., ‘giving money to UNICEF’) varies depending on a net’s training history and depending on what else it knows—that is, what other tasks it has been trained to perform. This means that there is no unique pattern of connectivity that can—even over the evolution of a single net—be identified as representing a particular concept. Different patterns of connectivity might hence arise for different inputs, even different inputs classified by output as of the same kind. 5 It seems to me that a machine of this general sort should be quite capable of developing a set of excitatory and inhibitory links in response to instances of a resultant concept, even if those instances share few common features, and certainly even if there is no one pattern common to them all. Particularist accounts of the relation between wrongness and the wrong-making properties case by case would cause it no difficulty. It would perhaps be necessary for the machine to be taught the concept of wrongness directly, with initial instances being presented to it as such. But even if this is so, it would merely be the analogue of a child’s learning the concept ‘wrong’. With a sufficiently wide range of initial ‘given’ cases, the machine should become perfectly capable of coping with the fact that there are many different ways of being wrong, and that properties that on one occasion count in favour of acting can on another count against. It can do this because it is not consulting an inner blueprint for wrongness and seeing how well the present case matches up to it. Rather, its internal workings, the interconnected system of interdependent firings, is exactly suited to fit it to respond to the sorts of effects that different features can have on each others’ ability to make a difference to whether the action should be done or not. The machine will not only be able to respond appropriately to a fully presented situation, in which all relevant information is fairly given. Given an initial but incomplete description of a proposed action it would be able to make something of an informed guess as to

C AN A PARTICUL ARIST LEARN TH E D IFFERENCE B ET WEEN R IGHT AND W RO NG? THE RONG 69 whether further information would lead to an overall judgement against it or in its favour. And given defective information it would still be able to cope, in the manner of which connectionist machines are recognised to be capable. In all these ways it seems to me that the machine is a perfectly adequate model for particularist accounts of moral rationality, or of competence with the concepts of right and wrong. Think, for a moment, in terms of a far simpler but much more limited form of connectionist machine, an imaginary one which only has input and output nodes, and whose input nodes are not the sub-conceptual ones of the real machines, but correspond individually to the sorts of moral concepts which we all think of as the ones that together somehow determine whether an action is right or wrong—concepts such as ‘kind’, ‘interfering’, helpful’, and so on. There are also conceptually rich output nodes for the thin moral concepts of ‘right’ and ‘wrong’. Now ask how the machine will work. Like the real machine, this imaginary one will be sensitive to context in just the sort of way that particularists suppose. If we fire the nodes for ‘action’ and ‘kind’ alone, the node for ‘right’ will fire strongly, let us suppose. But if we fire the nodes for ‘action’, kind’, and ‘interfering’, the node for ‘right’ will fire much less strongly, if at all. The firing of the ‘interfering’ node dampens the excitatory link between the ‘kind’ and the ‘right’ nodes. This seems to me to show that this simpler machine is manifestly a particularist machine. Once it has acquired sufficient experience, it can begin to answer questions about new cases on incomplete evidence, mirroring our capacity to answer questions about inadequately described imaginary cases. The machine might be told that the action is kind, and asked whether it is right. The answer it will give will be yes, if its experience has gone that way or, as we might say, supports that answer. Then it is told that the action, though kind, was thoughtless; and it can respond appropriately to this new information. What this means in the terms of the machine is that if the kindness node is stimulated, the rightness node fires, as it were; but if we stimulate the thoughtlessness node as well, the rightness node may cease to fire. This is explained by the pattern of experiences that the machine has had, but not in any way that requires us to suppose the machine to be master of anything like a principle, blueprint or pattern. The machine can carry on from case to case, correctly discerning which actions are right and which are wrong in a way that involves applying its past experience to new cases. It would probably be best to understand our imaginary machine as operating with a concept of pro tanto rightness, so that the rightness node and the wrongness node can fire at the same time. The question whether the action is overall right is determined by which is firing more strongly, since that is what reveals where the balance of reasons lies.

J ONATHAN DANCY 70 Our machine would agree, then, that a kind action is right, in the sense that if we only fire the kindness node, the rightness node is fired as well, and the wrongness node does not fire at all. But it would not agree that a kind action is always therefore pro tanto right, since it might be that when exposed to the kind action of wiping the torturer’s brow, the rightness node does not fire at all. You should not be kind to torturers, we then say. But perhaps, if we add a further feature, the rightness node will begin to fire again. All this is just as the particularists suppose. If our imaginary machine is a particularist machine, the respects in which it differs from the real one only confirm us in our suggestion that the real one is an adequate model for particularist moral epistemology. The real one does not have individual nodes for contributory moral concepts at the input layer, nor indeed for anything of that level of conceptual richness; it has an entire further layer of nodes, the hidden layer; and it is capable of reorganising its internal economy, its patternings or standing representations, if, as it were, it sees the need. None of these differences affect the essential point; they are merely grist to the particularist’s mill. The second model for non-patterned particularist competence with a concept is that of Roschian prototypes.6 This model has the advantage that it operates, as one might put it, at the software rather than at the hardware level. The general idea is that instead of building up a concept by developing lists of necessary and sufficient conditions for ‘being of that sort’, we do it rather by developing a set of prototypes, clearest cases, or best examples, in such a way that we conceive of less clear cases or less good examples in terms of distance from the prototype or prototypes. The less clear cases are still perfectly good as cases, but they may not count as central or obvious cases. The notion of distance in terms of which similarity to a prototype is judged is not itself something for which we either have or could have a prototype, and if indeed we did need one, this would immediately generate an infinite regress of a familiar sort. Rosch does not, so far as I know, consider our competence with moral concepts, and I am not at all sure that she would like the attempt to extend her work in this direction, but we might suppose that our concept of a good life is developed by accepting a group of prototypes, say the lives of the saints. Other lives are to be judged good by reference to their distance from those lives. Of course we will in this case need a sufficient range of prototypes; but that is all part of the story. This account is attractive to particularists because, again, competence with a concept does not seem to be being understood in terms of the ability to respond to ‘the same thing again’. There need be no pattern at the level of the features to which we are responding in judging distance from a prototype. What is more, if there is more than one

C AN A PARTICUL ARIST LEARN TH E D IFFERENCE B ET WEEN R IGHT AND W RO NG? THE RONG 71 prototype, they need not share a pattern of relevant qualities. The whole idea of the need for a repeating pattern as that which grounds and explains our ability to apply a concept correctly to new cases on the basis of old ones begins to lose its force. Some may think that a prototype just is a pattern. But a prototype is quite unlike a pattern. A prototype is a real concrete instance of the type, and contains many features that are not, and are recognised not to be, important to its functioning as such. If the life of St. Francis is a prototype of a moral life, our ability to use it as such, and to judge the worth of other lives in terms of their distance from this one, includes an ability to tell which features of the prototype are not really contributing much at all, and which are pretty central. To take another instance, a prototype for a dog will have four legs and also a certain number of hairs. The former number is much more significant than the latter. A pattern, by contrast, is an abstract object with respect to which all the selection of ‘relevant’ features has already been done. The relation between prototypes and ‘ectypes’ is different from that between patterns and their instances.

CONCLUSION In this short paper I argued first that the Canberries’ attack on particularism was in the form of a dilemma: either be a non-naturalist, which we all know won’t work, or accept the need for a natural pattern for each concept with which we are competent, which will generate moral principles of a traditional sort. For my own part, I refused to be moved by swift dismissals of non-naturalism. But on behalf of others, I suggested two appropriate models of naturalist particularist epistemology, designed to show that the use that the Canberries made of the notion of a pattern is much exaggerated. 7 Jonathan Dancy, Department of Philosophy, University of Reading, Reading, Berkshire, UK, RG6 2AA; [email protected]

NOTES 1. “Ethical Particularism and Patterns,” in Moral Particularism, ed. B. W. Hooker and M. Little (Oxford: Oxford University Press, forthcoming); quotations given in the text are probably not from the final version of this paper.

J ONATHAN DANCY 72 2. Though I doubt it: see his “Reply to my Critics,” in The Philosophy of G. E. Moore, ed. P. A. Schilpp (New York: Tudor Publishing Co., 1952), 588. 3. I do not claim to be at all clear about the sense of the descriptive/evaluative contrast at issue in this debate: I would much prefer to talk in terms of the contrast between the normative and the non-normative, saying that nonnormative features can form a normative pattern, and that this gives the Canberries all the pattern they could want without creating any difficulties for particularism. In their terms, however, my claim is that descriptive features can form a non-descriptive pattern, that exemplified by the right-making relation. 4. Of course Jackson, leading Canberry, has his own arguments against nonnaturalism, laid out in his recent From Metaphysics to Ethics (Oxford: Oxford University Press, 1998). I don’t accept these arguments. But it is worth quoting one sentence from that book. “When I say that the commonalities are there to be found, I do not simply mean that there is some scientific explanation for our classificatory practices—that goes without saying” (64). For me this does not go without saying. The very idea of a scientific explanation of our distinguishing between the noble and the ignoble seems to me to be most peculiar, and only an inordinate attachment to the scientific could lead one to suppose otherwise. 5. I am particularly indebted to Jay Garfield for help in understanding these more subtle aspects of connectionist architecture. My own inadequate understanding of these matters is derived largely from A. Clark, Microcognition: Philosophy, Cognitive Science and Parallel Distributed Processing (Cambridge, MA: MIT Press/Bradford Books, 1989), and W. Bechtel and A. Abrahamsen, Connectionism and the Mind: an Introduction to Parallel Processing in Networks (Oxford: Blackwell, 1991). 6. See E. Rosch, “On the internal structure of perceptual and semantic categories,” in Cognitive Development and the Acquisition of Language, ed. T. Moore (New York: Academic Press, 1973); also “Universals and cultural specifics in human categorization,” in Cross-Cultural Perspectives on Learning, ed. R. Brislin, et al. (New York: Sage/Halsted, 1975). 7. I would like to thank Jay Garfield and Ian Gold for helpful discussions which have led to great improvements to the paper read at the World Congress. All remaining inadequacies are not their fault.

PRACTICAL ETHICS AND MORAL OBJECTIVISM

Moral philosophers working today on concrete moral issues seem to assume certain views that are opposite to those of their predecessors; chief among these is that morality has an objective basis, that it is not just the result of subjective reactions, but comprises a body of beliefs acquired through some kind of perception of certain traits of reality. However, the

Margarita M. Valdés

reasons for thinking that people who discuss substantive moral issues are committed to moral objectivism are either not very clear or not entirely convincing. In what follows I shall examine the reasons given by John Mackie for considering that the use of first-level moral language—the language frequently used in the discussion of concrete moral problems—commits the user to moral objectivism.

I

f nything has characterized moral philosophy at the end of the present century it has been a notorious increase of interest in discussing substantive moral problems. Some of these problems are old and have to do with matters of life, death, and justice; others are quite new and only became apparent in the course of reflecting on the impact of industrial development on the natural environment; still others seem to be a by-product of some scientific and technological advancements, especially in the field of genetics and reproduction. But, despite their diversity, all of them are topics of great concern for contemporary society and, as I shall try to show, the fact that they are being discussed so profusely within the academic domain seems to reveal a deep change in the prevailing views among moral philosophers. After a long period in which moral philosophers concentrated 73

M ARG ARIT A M. V ALDÉS ARGARIT ARITA 74 almost exclusively on metaethical matters and avoided the question of how to act morally in concrete human situations, attitudes towards substantive moral questions began to change in the sixties. From then on, moral philosophers have increasingly examined a wide variety of specific human situations with the aim of offering or defending reasonable moral evaluations of them.

I. A P ROPOSED EXPLANATION If one asks what has motivated such a change, i.e., why so many moral philosophers got involved in the discussion of concrete moral problems—rather than spending their time analyzing different aspects of moral discourse or addressing questions concerning moral psychology, moral ontology, or epistemology—an acceptable answer would have to mention three dominant views against which the new generation of moral philosophers reacted: first, the traditional view that moral discrepancies were not solvable through rational discussion and hence were not an adequate subject of philosophical debate; according to this view the source of moral value was subjective, and moral judgements were like judgements of taste in the sense that, in case of discrepancies, it was unreasonable to try to find the right answer; second, and relatedly, the view that there exists a clear-cut distinction between fact and value and that truth and rational agreement were possible only with respect to facts, and, correspondingly, that no objective moral facts existed but only subjective reactions or culture-relative moral evaluations; third, the view that philosophical training did not make one especially suited for discussing practical problems, that philosophers could make no special contribution towards solving such problems nor be of any help in their elucidation. As I said earlier, these three assumptions have been challenged. During the past thirty years certain moral philosophers have advanced powerful arguments to show that moral diversity and the existence of moral discrepancies—which were taken by moral subjectivists as the main evidence in favor of their views—are compatible with moral objectivism and with realism about moral facts and values;1 other philosophers—those working in the field of applied ethics—have shown in their practice that reasonable discussion of particular moral problems is possible and that progress can be made in the direction of finding solutions or of approaching a point of agreement in cases of moral discrepancy.2 The distinction between fact and value has also been seriously questioned:3 no one takes for granted any more this philosophical tenet. The fact that evaluative matters can enter, for example, into the explanation of non-moral facts4 has been taken as a

P RACTICAL ETHICS AND MORAL OBJECTIVISM 75 sign that facts and values do not belong to two separate ontological realms. And as regards the third prevailing assumption, moral philosophers have become aware that underlying many moral discrepancies there lie hidden certain conceptual obscurities or defective arguments, and that philosophers can therefore contribute more than other interested parties in disentangling substantive moral problems. Moral philosophers working nowadays on concrete moral issues seem to assume views that are opposite to those of their predecessors: firstly, that morality has an objective basis, that it is not just the result of subjective reactions, but comprises a body of beliefs acquired through some kind of perception of certain traits of reality; secondly, the view that there are moral facts and that moral judgements and evaluations can be true or false; and, thirdly, the idea that philosophical training makes people especially suitable to discuss reasonably those substantive moral problems with which contemporary society is concerned. The contents of almost any contemporary book on practical ethics brings out these commitments on the part of philosophers working on specific moral issues.5 If what I have said is correct, then, practical ethics should not be seen merely as the result of applying general theoretical principles (or universal rules, or categorical imperatives, or whatever) to concrete particular cases, nor as an attempt to test ethical theories against empirical reality. It should rather be seen as the result of a change in some very general philosophical views which were dominant in the philosophical world before its advent. This, of course, does not mean that moral theory has little to offer or is not relevant to the discussion of substantive moral problems. That would be simply false. My point is a more modest one, and it is that, contrary to what some moral philosophers seem to think, practical ethics is not just the result of a systematic application of some a priori moral theory, principle or rules to specific human actions, but is rather the consequence of having accepted moral objectivism and of recognizing that morality is a cognitive enterprise. Moral theory certainly plays a part in the philosophical discussion of substantive moral problems—often supplying the tools necessary to analyze practical problems. But in this regard moral theory is on a par with other philosophical theories in other philosophical fields, including ones from logic, others from the philosophy of language, metaphysics, epistemology, and the philosophy of mind. All of these disciplines can help shed light on moral problems and answer questions about what it is morally correct to do in specific situations. And answering such questions is precisely the goal of practical ethics. I said before that philosophers who address concrete moral problems with the aim of finding a solution seem to be committed to moral objectivism. This is a view stated with great conviction in the writings of John

M ARG ARIT A M. V ALDÉS ARGARIT ARITA 76 Mackie6 and accepted by most philosophers working in the field of applied ethics. However, the reasons for thinking that people who discuss substantive moral issues are committed to moral objectivism are either not very clear or not entirely convincing; indeed, some philosophers have questioned the soundness of arguments trying to connect the two.7 In what follows I shall examine the reasons given by John Mackie for considering that the use of first-level moral language—the language frequently used in the discussion of concrete moral problems—commits the user to moral objectivism. But, first I will characterize very briefly some central claims of moral objectivism and contrast them with others of moral subjectivism.

II. MORAL OBJECTIVISM AND MORAL SUBJECTIVISM Moral objectivism is the view that intrinsic moral values exist, are “part of the fabric of the world” in Mackie’s terminology.8 Moral objectivism implies that actions, persons, characters, plans of life, can be morally good in themselves, independently of our desires or of our subjective reactions to them. Mother Theresa’s actions to the poor would be good even if they hadn’t awoken anyone’s sympathy. On this view, moral judgements and evaluations are factual in the same way as other empirical judgements, and, correspondingly, they can be true or false.9 According to moral objectivism, moral judgements not only have the form of factual statements, but have a factual content:10 they ascribe a moral property to a person or action and can be absolutely true or false. Notice that accepting the existence of moral facts is a way of rejecting the distinction between fact and value, and that objectivism is a form of moral “cognitivism,” that is to say, of the more general view according to which there is moral knowledge. How can moral objectivism explain moral diversity and the fact that there can be moral discrepancies? They can be explained on the basis of cultural differences and of epistemic limitations. Cultural differences account for the possibility of realizing the same objective moral values in different ways; epistemic limitations, for the possibility that people may ignore, as a matter of contingent fact, questions relevant to moral evaluation; and ignorance of such facts would precisely account for moral discrepancies. By contrast, according to subjectivism all values have a subjective source, that is to say, moral judgements, far from describing objective facts, are just the expression of our own inner reactions, of our own feelings, interests, or emotions. This implies that no human action, can be good in itself; no human action, character, or person can have in itself more (or less) moral value than any other. Thus, to the subjectivist, Mother Theresa’s actions towards the poor and dying, for example, are not in themselves

P RACTICAL ETHICS AND MORAL OBJECTIVISM 77 more valuable than those of Adolf Eichmann towards the Jews in concentration camps. Of course, we react to these two kinds of actions in different ways, we approve of the former and condemn the latter, encourage others to follow Mother Theresa’s steps and deter them from following Eichmann’s. We can even say that among the good things there are in the world are Mother Theresa’s actions towards the poor and dying, but our saying this would be nothing but the expression of our own approval. Not even the fact that Mother Theresa’s actions were done with the intention of relieving pain, and Eichmann’s with that of destroying innocent people, gives their corresponding actions different intrinsic value, for, according to the subjectivist, there is no such thing. If moral judgements just express our inner reactions, then, evaluating an action or a person from the moral point of view is not a cognitive enterprise; moral judgements and evaluations teach us nothing about what there is in the world independently of our own attitudes and emotions. Moral subjectivism aims to be a philosophical answer to moral diversity and to the fact that there are discrepancies in moral judgements. According to moral subjectivism, discrepancies in moral judgements cannot be eliminated by any reasonable method.11 And moral diversity, which results from the fact that different communities may react to the same actions in different ways, is a possibility which always exists and which can give rise to moral relativism. Any action, such as eating pork meat, having pre-marital sex, applying physical punishment to a child, could be morally acceptable according to one group of persons and morally comtemptible according to another. And if this were the case that would be the end to the matter; it would be wholy inappropriate to ask whether the action in itself is good, independently of those human evaluations.12

III. MACKIE’S ARGUMENTS Now let us look at Mackie’s arguments for the view that making moral judgements or evaluations presupposes belief in the objectivity of moral values.13 According to him, “moral judgements involve a claim to objectivity” (48). Or, as he writes in another passage, “a belief in objective values is built into ordinary moral thought and language” (49). He gives three different reasons for this view: one semantic, one phenomenological, and one pragmatical. Let us examine briefly the semantic reason. According to Mackie, moral objectivism “has a firm basis . . . in the meanings of moral terms” (30), in the sense that moral terms have as part of their meaning the objective conception of moral value; so, anyone who uses moral language is presupposing moral objectivism. Moral terms such as “good,”

M ARG ARIT A M. V ALDÉS ARGARIT ARITA 78 “ought,” “obligation,” etc., really mean “intrinsically good,” “absolutely ought,” “objective obligation.” In this vein, Mackie writes that the term “good,” for example, in its main moral use “does refer to supposed intrinsic requirements” (63), to “requirements which simply are there, in the nature of things, without being the requiremets of any person or body of persons” (59). So, if someone says that Mother Theresa’s actions to the poor were good, what her words mean is that those actions satisfied the intrinsic requirements of that kind of action, not requirements imposed by anyone’s interests or desires. Notice that if this semantic view were correct, then there would be a kind of tension in Mackie’s own views. For his conjecture about the meaning of moral terms would seem to rule out the consistency of any moral subjectivist theory such as the one he himself defends. Anyone holding that an action could be good but not intrinsically good—as Mackie himself does—would seem to hold something straightforwardly contradictory. Simon Blackburn’s criticism of Mackie’s position on this point seems very powerful: it is erroneous—he argues—to hold that part of the meaning of moral terms used in first-level moral discourse is determined by the kind of theory the user of such terms may accept about their referents.14 Using moral language is neutral with respect to any such theory in the same way as using mathematical language is neutral with respect to a theory about the nature of mathematical objects. So, the mere fact that someone uses moral language to evaluate an action cannot commit her or him to moral objectivism. Mackie’s phenomenological argument is not very convincing either. He argues that when a speaker uses a moral term in a judgement of moral approval she feels that—although she may be expressing a subjective desire—her desire is backed by something objective, something independent of her own desire or of anyone’s desires. This argument is not at all conclusive, for, besides the fact that any subjectivist may refuse to accept that she has such a feeling, having it is fully compatible with moral projectivism according to which the source of moral value is subjective, eventhough we “spread” it on the world. Let us go to Mackie’s pragmatic argument. He claims that when a user of moral language evaluates an action, for example, she does not intend to say something about her own subjective states, but something about the action in itself. In other words, her intended referent is the action she is evaluating, independently of its relations with anything else. “The ordinary user of moral language—Mackie writes—means to say something about whatever it is that he characterizes morally, for example, a possible action, as it is in itself, . . . and not about . . . his or anyone else’s attitude or relation to it” (33). This, of course, may be true of many ordinary unreflective users of moral terms, and also of some

P RACTICAL ETHICS AND MORAL OBJECTIVISM 79 reflective users of them. It certainly seems true that when a moral philosopher, after a long and complicated argument, comes to the conclusion that helping patients with a terminal illness having a quick and painless death is good, or that torturing is bad, or that having an abortion is not immoral, he or she usually intends to say something about the action or kind of action he or she is discussing, rather than about his or her own attitudes towards them. But, there are two problems here: according to the projectivist, one can use moral terms with the intention of saying something about whatever one is morally evaluating, without committing oneself to the existence of objective values. Simon Blackburn argues that there is nothing in the practice of moralizing such that the objectivist can do but the projectivist cannot. “When the context of discussion is that of first order commitment [the projectivist] is as solid as the most virtuous of moralists.” 15 In other words, when the projectivist evaluates an action she can have the same intention as the objectivist, i.e., the intention to talk about the action and not about her own or someone else’s mental attitudes towards it. So, for the pragmatic argument to go through, it has to be shown that it is impossible for the projectivist to have the same intention as the moral objectivist. The second problem with this argument is that people who moralize often believe that the source of moral value is external to the action, and so, when they judge that an action is good, what they intend to say is not that it is good in itself irrespective of anyone’s attitudes towards it, but that it is good because it agrees with God’s will or is approved by the community, for example. If this is so, then it is not always the case that when people make a moral judgement or evaluation they mean to say something about the action in itself independently of anyone’s reactions to it, and again if this is so, Mackie’s pragmatic argument seems to start from a false premise. The arguments I have examined to the effect that practical ethics needs to believe in the validity of moral objectivism seem defective in one or more ways. This, of course, doesn’t show that no such argument could succeed, nor that philosophers working on concrete moral problems do not have good reasons to assume moral objectivism. Moreover, the fact I mentioned at the beginning of this paper, in the sense that practical ethics started flourishing only once moral subjectivism and the fact/value distinction lost credibility, makes us suspect strongly that there must be such an argument to connect one with the other. But, to my mind, this argument has yet to be discovered. Margarita M. Valdés, Department of Philosophy, National Autonomous University of Mexico, Mexico City, Mexico

M ARG ARIT A M. V ALDÉS ARGARIT ARITA 80

NOTES 1. See, for example, Hilary Putnam, Reason, Truth and History (Cambridge: Cambridge University Press, 1981), esp. chaps. 5,6, and 7; Essays on Moral Realism, ed. Geoffrey Sayre-McCord (Ithaca, NY: Cornell University Press, 1988); David O. Brink Moral Realism and the Foundations of Ethics (Cambridge: Cambridge University Press, 1989); and Mark Platts, Moral Realities (New York: Routledge and Kegan Paul), 1991. 2. An good example that this kind of progress is possible can be found in Ronald Dworkin, Life’s Dominion (New York: Knopf, 1993). 3. See Putnam, op. cit., esp. chap. 6. 4. For a good discussion of this topic, see the essay by Nicholas L. Sturgeon, reprinted in Sayre-McCord, ed., op. cit. 5. I will mention just a few of them: Peter Singer, Practical Ethics (Cambridge: Cambridge University Press, 1979); David Rosenthal and Fadlou Shedahi Applied Ethics and Ethical Theory, vol. 1 of Ethics in a Changing World, ed. Margaret Battin and Leslie Francis (Salt Lake City: University of Utah Press, 1988). (Other volumes within this series discuss ethical problems connected with the AIDS epidemic, the use of new reproductive technologies and the use of animals in medical research). See also Dilemas Eticos, ed. Mark Platts (Mexico City: UNAM-FCE, 1996); and Decisiones de vida y muerte, ed. Florencia Luna and Arleen Salles, (Buenos Aires: Editorial Sudamericana, 1995). 6. See J. L. Mackie, Ethics: Inventing Right and Wrong (New York: Penguin, 1977), 33, 34. All page references in the text are to this source. 7. See, for example, Simon Blackburn’s essay, “Errors and the Phenomenology of Value” in Morality and Objectivity, ed. Ted Honderich (New York: Routledge and Kegan Paul, 1985). 8. Mackie, op. cit., 15. 9. For the purpose of this paper I shall assume that moral objectivism and moral realism are just two names for the same doctrine. 10. See, David Brinck, Moral Realism and the Foundations of Ethics (Cambridge: Cambridge University Press, 1989), 25. 11. The issue between objectivists and subjectivists can be put in terms of whether or not moral discrepancies can be eliminated once all parties know all the relevant facts and reasons to evaluate an action, person, or whatever. 12. The subjectivist does not reject every possibility of having the same reaction to the same kind of action in different communities, only the possibility that such a uniformity be achieved due to an exchange of relevant reasons. The subjectivist may think that uniformity of reaction can be achieved by means of persuasion or by some other “emotional,” or “brainwashing” sort of method.

P RACTICAL ETHICS AND MORAL OBJECTIVISM 81 13. I choose to discuss Mackie’s arguments because, despite all their complications, they seem to state the case in the most straightforward and powerful way. 14. See Blackburn, op. cit. 15. Ibid., 11.

PRINCIPIOS DE EQUIDAD DISCURSIVA

This paper relates the concept of “discourse” (in the discourse-ethics’ meaning) with the concept of “social justice.” I maintain that the social nature of reason is a condition of possibility of such justice. According to the Alexy-Habermas system, the rules for the practical discourse mean

Ricardo Maliandi

in their third level the conditions of symmetry, that is, the absence of privileges among the participants. My paper introduces the concept of discursive fairness , referred to the requirements for maintaining the discursive symmetry, even under circumstances of non-discursive, but real asymmetry. Two principles of discursive fairness are offered (by analogy to Rawls’ “principles of justice”): 1. equality of rights among the participants in the practical discourse, and 2. possibility of reviewing all arguments used in the discourse, together with guarantees of benefits for the unsuccessful participants.

L

a ética del discurso recibe ese nombre de la circunstancia de que en ella se pone especial énfasis sobre los “discursos prácticos.” El principio (o “metanorma”) de dicha ética exige, precisamente, que los conflictos de intereses sean resueltos—o que, al menos, se intente resolverlos—por medio de tales discursos, es decir, de procedimientos dialógicos argumentativos en los que cada interlocutor tematice y trate de justificar razonablemente sus pretensiones, en vistas a alcanzar el consenso de todos los involucrados en las posibles consecuencias del tipo de acción que haya de implementarse. En otros términos, y para expresarlo con mayor claridad:

83

R ICARDO MALIANDI 84 la ética del discurso exige que, en lugar de pelearnos, discutamos con argumentos aquello que determina nuestras discrepancias prácticas. Lo que se excluye es el uso de la violencia; no sólo de la violencia física y explícita, sino también de todas las formas de violencia solapada, como, por ejemplo, las amenazas más o menos expresas, o incluso el empleo de artilugios retóricos que confundan al interlocutor, etc., pero—repito—se establece una exigencia ética de apertura al discurso, al diálogo argumentativo. ¿De dónde sale esa exigencia? Parece lícito, en efecto, preguntar por qué y en qué sentido tenemos la obligación moral (ya que se habla de “exigencia ética”) de recurrir a los argumentos cada vez que nos encontramos ante un conflicto de intereses. Pues bien, la ética del discurso no sólo ofrece una respuesta a esa pregunta, sino que en esa respuesta, es decir, en esa fundamentación, consiste su principal y distintivo aporte a la ética filosófica: no se trata de una exigencia gratuita, ni, como podría parecer, meramente remitida al “sentido común,” sino de una exigencia que está ya siempre necesariamente presupuesta cada vez que argumentamos, con independencia de cuál sea el tema de nuestra argumentación. En cada una de nuestras actitudes como usuarios de argumentos hemos admitido de modo implícito, según la ética del discurso, que el uso de argumentos es el único camino correcto para la resolución de conflictos. En esta afirmación se tiene en cuenta la dimensión pragmática del lenguaje. Los signos lingüísticos—y toda argumentación se compone de tales signos—no se agotan en sus referencias semánticas, sino que establecen, en cada caso de su uso concreto, una comunicación entre el hablante y su oyente o interlocutor (o entre el que escribe y el que lee lo escrito, si se trata de signos gráficos). Hay necesariamente presupuestos implícitos, acuerdos tácitos que permiten que alguien use los signos y alguien los interprete. Estos presupuestos no aparecen en el nivel semántico, pero, sin ellos, la comunicación como tal no sería posible. Como son presupuestos a priori, el señalamiento de los mismos se hace mediante procedimientos reflexivos de “pragmática trascendental”: hay que preguntarse, por ejemplo, si sería posible negar o cuestionar un determinado presupuesto sin cometer una “autocontradicción performativa” (es decir, una contradicción entre el contenido semántico de lo que se dice y el particular “acto de habla” que se efectúa precisamente al decirlo). Entre tales presupuestos están, ante todo, las “exigencias de validez” de lo que se dice (exigencias de sentido, de verdad, de veracidad y de rectitud), pero también exigencias normativas, y, primordialmente, el principio de la ética del discurso. Para expresarlo de otra manera, podría decirse que cada vez que argumentamos sobre algo nos comprometemos tácitamente a hacerlo también como manera de resolver conflictos. Aunque,

P RINCIPIOS DE EQUIDAD DISCURSIVA 85 personalmente, tengo algunas objeciones muy puntuales a esta fundamentación, creo que ella, y el “giro pragmático” en el que se inserta, constituye un aporte decisivo al pensamiento ético contemporáneo, un aporte frente al cual la mayoría de las críticas que se le dirigieron hasta ahora provienen de malentendidos o de consideraciones deficientes. Esta presunción, desde luego, debería ser demostrada; pero razones de espacio impiden hacerlo en el presente trabajo. El intento de este último se reduce a señalar relaciones de la enfatización ético-discursiva de los “discursos prácticos” con el concepto de “justicia social,” y en particular con el hecho de que dicho concepto parece estar siendo olvidado u ocultado por el predominio de la economía neoliberal. La insistencia de la ética del discurso en la estructura simétrica que debe prevalecer en los discursos prácticos podría, en principio, servir al mismo tiempo de base para una denuncia de claras contravenciones éticas involucradas en el actual aumento de la concentración de poder político y económico y, paralelamente, de la marginación o exclusión (mensurable también como dificultad o imposibilidad de participación en discursos prácticos) de grupos e individuos—o de sus representantes— que tienen derecho a hacer valer sus necesidades y a expresar sus opiniones sobre las cuestiones de la interacción social. Semejante exclusión forma parte de la exclusión del acceso a bienes económicos básicos y configura así un aspecto relevante en uno de los grandes conflictos sociales de nuestro tiempo. En líneas generales, podría decirse que los discursos prácticos se vinculan a la noción de “justicia social” desde dos puntos de vista: en primer lugar, porque todo auténtico discurso práctico supone que se tenga en cuenta el carácter dialógico y, en tal medida, social, de la razón. Los seres racionales no somos islas, sino que nos comportamos como tales precisamente en nuestras mutuas relaciones comunicativas, y cada individuo racional llega a ser lo que es en virtud de un indeclinable proceso de socialización. Los conflictos de intereses—acontecimientos inevitables en toda sociedad, y que aumentan en proporción directa con la complejidad de las estructuras sociales—representan a menudo intereses grupales y, por tanto, requieren ser tratados en discursos prácticos intergrupales. Aquí aparece entonces el segundo punto de vista desde el cual los discursos prácticos se vinculan a la justicia social: aquellos pueden ser considerados como una “condición de posibilidad” de ésta. En otros términos: el ejercicio de lo que sea la justicia social parece inconcebible sin la apelación a un permanente diálogo simétrico entre los grupos o los representantes de grupos que ven amenazados sus intereses en determinados contextos de la interacción social. El concepto de justicia social tiene que incluir, entre sus notas comprehensionales, la posibilidad de institucionalizar canales

R ICARDO MALIANDI 86 de expresión e intercambio comunicativo-argumental acerca de los intereses de cada individuo y de cada grupo y de las respectivas opiniones acerca de cómo ha de concebirse una sociedad justa. Quizá sea oportuno hacer aquí un par de aclaraciones breves sobre lo que se quiere decir cuando se habla de “discurso práctico.” Ya sin el calificativo de “práctico,” la simple noción de “discurso” acarrea cierta ambigüedad. Etimológicamente, el término alude a “correr de aquí para allá,” lo cual sería un recurso metafórico para expresar la diversidad de temas que pueden tocarse mientras se habla. También es ya tradicional usar el término “discurso” en el sencillo sentido de la articulación coherente de proposiciones e inferencias. Un empleo más popularizado, sin embargo, lo hace equivalente a un género de oratoria, una peculiar alocución con la que se espera conmover, persuadir o entusiasmar a los oyentes. En un sentido menos popular, aunque también extendido desde hace siglos, “discurso” alude a un género literario, un tipo de obra escrita en la que se desarrolla sistemáticamente un determinado tema (de esto sería paradigma el Discurso del método de Descartes). Hoy es muy frecuente, sobre todo en ambientes intelectuales, llamar “discurso” a una específica manera de hablar, característica de algún grupo ideológico, profesional, étnico, etc., y así oímos expresiones como “discurso liberal,” “discurso marxista,” o “discurso jurídico,” “discurso psicoanalítico,” o “discurso latinoamericano,” “discurso europeo,” etc. En esta última acepción (el “discurso” entendido como estilo retórico de una comunidad específica) está más o menos presupuesta una postura relativista, una concepción según la cual toda manera de hablar (y, con ella, toda manera de pensar y de valorar, etc.) está siempre y totalmente condicionada por algún contexto particular, y no sólo reviste, en consecuencia, carácter contingente, sino que, en cada caso, lo que está expreso en el “discurso” es secundario y remite a intereses implícitos, o a prejuicios, etc. Ahora bien, cuando hablamos de “ética del discurso” y, en relación con ella, de “discursos prácticos”, no usamos el término en ninguna de las acepciones mencionadas (si bien tampoco puede negarse que, en definitiva, todas las acepciones tienen ciertas interconexiones). “Discurso” se ha convertido en un término técnico de la filosofía contemporánea, sobre todo a partir de su empleo en las obras de Jürgen Habermas. Este autor lo entiende como “examen crítico argumentativo de las pretensiones de validez presupuestas en una afirmación determinada.” Tal examen es necesariamente dialógico y exige ante todo la simetría y la correspondiente igualdad de derechos entre quienes participan en él. 1 En lo esencial, no se trata de una noción nueva, sino—como a menudo lo señalan diversos representantes de la ética del discurso— de algo así como la adaptación de una actitud proveniente, acaso, del

P RINCIPIOS DE EQUIDAD DISCURSIVA 87 elenchos socrático.2 El “discurso” es, en esta acepción, una forma de diálogo (y, podríamos agregar, de diálogo crítico), en el que, como dice también Habermas, se procura “la fundamentación de las previamente problematizadas pretensiones de validez que están presentes en opiniones y normas.”3 Un discípulo de Apel, Dietrich Böhler, propone distinguir , en el dicurso, el plano de los contenidos y el reflexivo-argumentativo, y, en cada uno de esos planos, dos tipos de discurso: el “teórico” y el “práctico.”4 Pero lo aquí relevante es que, mientras el primero de éstos se refiere a la validez teórica (verdad, sentido), el segundo tematiza la validez práctica (rectitud, veracidad) y, sobre todo, aquello a que remiten las proposiciones deontológicas, es decir, la “obligatoriedad.” Mientras en el discurso teórico interesa especialmente el contenido objetivo (por ejemplo, la cuestión de si lo aludido existe o no), en el discurso práctico lo protagónico es el aspecto comunicativo. Desde un comienzo se advirtió que era de particular importancia reconocer o determinar cuáles son las reglas del discurso práctico, sobre todo por la relación que esto guarda con el problema de la fundamentación de normas o de principios morales. El aporte decisivo, en tal sentido, proviene de Robert Alexy, quien presentó un sistema de reglas que se ha convertido en clásico para la ética del discurso.5 Se trata de un sistema basado en la pragmática de Habermas y en la metaética analítica. Habermas mismo adoptó inmediatamente esas reglas y las adaptó a su propia “lógica del discurso práctico,” convirtiendo los tres grupos reconocidos por Alexy en tres “niveles”: el nivel lógico de los “productos” (los “argumentos convincentes”) el nivel dialéctico de los “procedimientos” (por los cuales se llega a acuerdos entre los interlocutores) y el nivel retórico de los “procesos” comunicativos (por medio de argumentaciones). Con ligeras modificaciones respecto de la tabla de Alexy, Habermas presenta reglas para cada uno de esos niveles. Para el primero se exige la no-contradicción, la coherencia en la atribución de predicados y la unidad de significado para cada expresión utilizada. Para el segundo, la veracidad en lo que se afirma, y la obligación de fundamentar cualquier introducción, al discurso, de proposiciones o normas sobre las que no se está discutiendo. Para el tercer nivel se especifican los derechos de todo hablante a la participación en discursos, a la problematización de cada afirmación y a la expresión de las propias opiniones, deseos y necesidades, y se prohíbe toda coacción sobre el ejercicio de tales derechos. Las reglas del primer nivel son “lógicas y semánticas,” las del segundo tienen un contenido ético vinculado, por ejemplo, al reconocimiento recíproco entre los interlocutores, y las del tercero aluden a algo que aquí nos interesa particularmente: las condiciones de simetría que debe cumplir el discurso práctico. 6

R ICARDO MALIANDI 88 La simetría, en los discursos prácticos, consiste en que ninguno de los participantes ocupa en ellos un lugar de privilegio. Todos están sometidos a las mismas reglas y cada uno debe reconocer a los demás los mismos derechos, especificados, como se acaba de ver, en el nivel 3. Precisamente con respecto a este nivel, ha comprobado Apel el presupuesto de una “comunidad ilimitada de comunicación,” es decir, un modelo ideal en el que los participantes excluyen toda coacción sobre los demás y se atienen a lo que determine el “mejor argumento,” neutralizando cualquier motivo que no sea el de la búsqueda cooperativa de la verdad. Una situación semejante nunca se alcanza de hecho; pero ella está de hecho presupuesta como aspiración común de todos los participantes. Si no fuera así, es decir, si se contara de antemano con que lo decisivo será en definitiva el mayor poder de alguno o algunos de los interlocutores, entonces los argumentos serían superfluos: bastaría con implementar medios adecuados para la medición de ese poder. Creo, sin embargo, que el “discurso práctico” no sólo presupone el criterio del “mejor argumento”—presentado por alguno de los participantes—sino también la disponibilidad de los interlocutores a colaborar en el desarrollo de argumentos colectivos. Sobre la base (a mi juicio implícita en la concepción pragmático-trascendental de los “discursos prácticos”) de que muchas cabezas piensan mejor que una sola, el argumento ideal tiene que surgir como producto dialéctico de la confrontación de los diversos argumentos individuales. En relación con esto, propongo llamar “diálogo crítico” al tipo de discurso práctico en el que a los interlocutores interesa más la cooperación tendiente a la solución del problema que la mera refutación del oponente.7 La simetría que se logre en virtud de una situación semejante es la mejor garantía que puede ofrecerse sobre la equidad del discurso práctico. Un funcionamiento del discurso en el que se cumpla efectivamente con todas las reglas puede, pues, calificarse como equitativo. Y en relación con ello denominaré “equidad discursiva” a la situación de discurso práctico en la que puedan mantenerse las relaciones simétricas hasta alcanzar un acuerdo consensuado. Sin embargo, no se puede dejar de considerar el hecho de que tal equidad discursiva mantiene el carácter ideal: se presenta más como una situación de jure que como una de facto. Todo discurso práctico tiene lugar necesariamente en el marco de un contexto en el que existen condicionamientos económico-sociales, y aunque los participantes se muestren como expresamente dispuestos a respetar las reglas, las posiciones que ellos ocupan fuera del discurso repercuten de algún modo dentro de éste. La veracidad requerida, la intención última (Gesinnung)

P RINCIPIOS DE EQUIDAD DISCURSIVA 89 de cada participante, puede quedar oculta a los demás e incluso, en determinados casos, a él mismo. El participante que ocupe posiciones extradiscursivas de privilegio tratará, consciente o inconscientemente, de hacerlas valer también en el discurso, y esto incidirá negativamente sobre el equilibrio simétrico y, por tanto, sobre la equidad discursiva. Es cierto que incluso la simulación del cumplimiento de las reglas, y la simulación de la simetría resultan éticamente significativas, ya que en tales simulaciones se pone de manifiesto el reconocimiento tácito de que las reglas y la simetría “deberían” cumplirse. Paradójicamente, una forma de incumplimiento viene a probar que incluso un discurso configurado con simulaciones (que equivale a un seudodiscurso) constituye una prueba de que algo así como una “comunidad ideal de comunicación” está presupuesta, como dice Apel, en el uso de la argumentación. Pero también es cierto que si esto ocurriese en todos los discursos prácticos, es decir, si, en definitiva éstos estuvieran condenados a ser seudodiscursos; o, en otros términos, si la simetría y la equidad discursiva sólo pudieran ser fingidas, entonces los discursos prácticos, en definitiva, no existirían, serían utópicos, y la ética discursiva misma resultaría ilusoria. La pregunta que se impone es, pues, la de si efectivamente las asimetrías extradiscursivas perturban el funcionamiento de todo discurso práctico, o si son al menos concebibles recursos de neutralización de tales influencias perturbadoras. Mi propuesta, en el presente trabajo, y como un modo alternativo de responder la mencionada pregunta, consiste en sugerir que quizá el agregado de nuevas reglas—con carácter de principios de equidad discursiva—podría contribuir a una minimización de las perturbaciones introducidas cuando el acatamiento a las demás reglas es sólo simulado. Estos principios, en todo caso, tendrían que caracterizarse por la insimulabilidad de su cumplimiento, lo cual hace realmente difícil hallarles una formulación adecuada. De manera provisional quisiera también, sin embargo, proponer una formulación en la que, sin que ello implique una aceptación de los famosos “principios de justicia” de Rawls, se aproveche una posible analogía, al menos formal, con los mismos. Las reglas del tercer nivel podrían conjugarse, a mi juicio, en un principio efectivamente formulable de un modo análogo al primer principio de Rawls, aunque añadiendo una alusión clara a la necesidad (y, por consiguiente, al derecho) de tematizar las asimetrías provenientes del contexto económico social que podrían influir a modo de perturbaciones del discurso. Quedaría entonces como sigue: 1) Todo participante en el discurso práctico tiene igual derecho a exponer y hacer valer sus intereses y opiniones, apoyando los unos y las otras en argumentos, y tiene asimismo igual derecho a tematizar argumentativamente

R ICARDO MALIANDI 90 las asimetrías extradiscursivas y a proponer y discutir recaudos metodológicos tendientes a evitar que las mismas ejerzan influencia en los acuerdos logrados como conclusión del discurso.

Un segundo principio de equidad discursiva, algo más complejo que el anterior, serviría como complementación del mismo y permitiría establecer un criterio para resolver aquellos casos en que, no obstante la observancia del primer principio, se genere algún tipo de asimetría en el discurso práctico. Es fácil advertir, a la vez, también en este segundo principio, una analogía con el segundo “principio de justicia” de Rawls. Su formulación sería la siguiente: 2) Las asimetrías que resulten de o en la discusión, o que representen la proyección, en el discurso, de asimetrías extradiscursivas, y que no hayan podido evitarse con la observancia del primer principio, deben satisfacer dos condiciones: a) deben estar asociadas a argumentos revisables en las condiciones de una equitativa oportunidad de revisión, y b) deben procurar el máximo beneficio de los participantes menos aventajados en la discusión.

El principio 2a establece una aceptación meramente provisional de aquellos argumentos que no puedan reconocerse consensualmente como productos dialécticos del intercambio argumentativo entre todos los participantes. El principio 2b, que guarda cierta semejanza con el llamado “maximin” de Rawls, puede ser llamado “principio de Epicuro,” recordando la sentencia del pensador griego según la cual “en las discusiones entre quienes aman razonar obtiene más provecho el que resulta inferior, por lo que del otro aprende.” 8 Pero es un principio que podría, a su vez, entenderse en dos sentidos distintos: 1) en el de que hay participantes con menos recursos argumentativos lo cual conectaría a 2b con 2a), y ello debe tenerse especialmente en cuenta, tomándose los recaudos (tales como las revisiones) para que ese tipo de asimetría, a veces insalvable, no perjudique los intereses de nadie, o bien 2) en el de que quien “pierde” una discusión (es decir, quien expone intereses u opiniones que no alcanzan consenso al final del discurso) debe obtener un beneficio consistente en un logro evidente de mayores probabilidades de “ganar” discusiones futuras, en el marco del reconocimiento de que los discursos prácticos no sólo resuelven conflictos expresamente discutidos

P RINCIPIOS DE EQUIDAD DISCURSIVA 91 en ellos, sino que también contribuyen a la institucionalización del diálogo crítico, formando y alentando la actitud dialógica.

Ambos sentidos presentan, sin embargo, un importante punto de convergencia: la percatación de que, si las asimetrías subsisten aun cuando se hayan tomado los mejores recaudos metodológicos para neutralizarlas, ello no debería determinar una renuncia al discurso práctico, sino la introducción en éste de adecuados procedimientos para asegurar la compensación de tales asimetrías. Si la simetría plena constituye sólo un ideal, puede sostenerse también que ese ideal se convertirá en idea regulativa en la medida en que los discursos prácticos reales cuenten con recursos para minimizar las influencias perturbadoras derivadas de las asimetrías. Como ya lo anticipé, estos principios propuestos tienen carácter provisional, y con ellos sólo pretendo ilustrar la tesis básica de que las reglas del discurso práctico elaboradas por Alexy y Habermas deberían complementarse con principios de equidad discursiva tendientes a minimizar las influencias perturbadoras de asimetrías que son, en definitiva, ineliminables en los contextos reales. La cuestión de la formulación más adecuada de tales principios queda, pues, abierta, y mi tesis se reduce a indicar su necesidad y sugerir fórmulas posibles. Quisiera agregar que la “simetría” deseable en un discurso práctico tiene dos aspectos que conviene distinguir: en primer lugar, ella significa un equilibrio de poder (ya sea el que proviene de posiciones extradiscursivas o el que deriva de la capacidad o habilidad argumentativa de cada participante, etc.), y, en segundo lugar, ella alude a una peculiar reciprocidad discursiva, consistente en que cada interlocutor puede, en principio “ayudar” al otro o a los otros a “alumbrar” sus propias opiniones o “ideas.” Se trata de una forma de mayéutica socrática, aunque con el agregado de que en el diálogo crítico no es un solo interlocutor el que desempeña el papel de “partero,” sino que todos actúan a la vez como “parturientes” y como “parteros,” aunque lo hagan, claro está, de manera sucesiva y no simultánea.9 También me parece importante añadir, finalmente, que el reconocimiento de principios de equidad discursiva (ya se trate de los aquí formulados o de otros que cumplan el mismo propósito) no implica la adopción de una actitud rigorista. No se trata de imperativos categóricos. Deben entenderse como principios que sólo establecen parámetros para aumentar la probabilidad de que el discurso práctico resulte razonable y equitativo. Si se pusiera excesivo énfasis en su observancia, se vería amenazada la espontaneidad que también es “razonable,” y que no conviene perder en las pertinentes discusiones. Estas, por cierto, cuando se llevan a

R ICARDO MALIANDI 92 cabo en un discurso práctico, no se reducen a lo meramente conversacional; pero tampoco deberían alejarse tanto de las conversaciones que derivasen en un solemne reglamentarismo. Por otra parte, sin embargo, una despreocupación total por las reglas del discurso (y por los principios de equidad discursiva) sería, teniendo en cuenta lo que está realmente en juego en los discursos prácticos, una irresponsabilidad manifiesta. La posición que queda sostener, entonces, evocando quizá la mesótes aristotélica, tiene que quedar a mitad de camino entre el rigorismo y un excesivo latitudinarismo. Puede concederse a la hermenéutica la necesidad de no perder de vista la espontaneidad conversacional, siempre y cuando ello no signifique perder de vista los principios. Recordando una célebre fórmula kantiana, y aplicándola a esta cuestión, podría entonces sostenerse que los discursos sin conversación son vacíos, pero también que las conversaciones sin discurso son ciegas, especialmente en el caso de que lo inminente esté constituido por la necesidad de resolver determinados conflictos de intereses. Pero es menester admitir que, aun conservando la espontaneidad conversacional y evitando—incluso rigurosamente—la unilateralidad rigorista, sólo la simetría entre los participantes, o una adecuada compensación de las asimetrías inevitables, podrá, en definitiva, conducir a soluciones razonables. Ricardo Maliandi, Departamento de Filosofía, Universidad de Buenos Aires, Capital Federal, Argentina

NOTES 1. Cf. Jürgen Habermas, Theorie und Praxis, 4e.Aufl. (Frankfurt a.Main: Suhrkamp Verlag, 1971), 23 ss. 2. Cf. Platon, Apología 39c, Sofista 232b-e, Gorgias 471a–472c, y Aristoteles, Metafísica 1006a, lines 6-18. Sócrates llamaba elenchos al tipo de diálogo que él entablaba con sus interlocutores y que se parecía a la confrontación de pruebas y contrapruebas ante un tribunal. 3. Habermas, op.cit., 25. 4. Cf. D. Böhler, “Philosophischer Diskurs im Spannungsfeld von Theorie und Praxis,” en Praktische Philosophie/Ethik, ed. K. O. Apel, D. Böhler, y K. Rebel, Studientexte 2 Funkkolleg (Weinheim und Basel: Beltz Verlag, 1984), 315 y 331ss. Del mismo autor, Rekonstruktive Pragmatik (Frankfurt a.Main: Suhrkamp Verlag, 1985), passim.

P RINCIPIOS DE EQUIDAD DISCURSIVA 93 5. Cf. R. Alexy, “Eine Theorie des praktischen Diskurses,” en Materialien zur Normendiskussion, Band 2: Normenbegründung-Normendurchsetzung, ed. W. Oelmüller (Paderborn: F.Schöningh, 1978), 22-58. Del mismo autor, Theorie der juristischen Argumentation (Frankfurt a.Main: Suhrkamp Verlag, 1978) y Theorie der Grundrechte (Baden-Baden, 1985). Cf. asimismo los comentarios de Jürgen Habermas en Moralbewusstsein und kommunikatives Handeln (Frankfurt a.Main: Suhrkamp Verlag, 1983), 97ss. y en Theorie des kommunikatives Handelns, 2 tomos (Frankfurt a.Main: 1981), passim. Otras referencias: K. Günther, Der Sinn für Angemessenheit. Anwendungsdiskurse in Moral und Recht (Frankfurt a.Main: Suhrkamp Verlag, 1988), 268ss.; A. Cortina, Etica sin moral (Madrid: Tecnos, 1990), 245ss.; E. Tugendhat, Vorlesungen über Ethik (Frankfurt a.Main: Surkamp Verlag, 1993), 168 (en ibid., 358ss. analiza Tugendhat la teoría de Alexy sobre los “derechos básicos”). 6. Cf. Habermas, Moralbewusstsein, 97ss. 7. Cf. Ricardo Maliandi, “El trilema de Aristófanes y los presupuestos normativos del diálogo crítico,” en Crítica, México, XII, 65 (Agosto 1990): 43-55, y Dejar la posmodernidad. La ética frente al irracionalismo actual (Buenos Aires: Almagesto, 1993), 64-65. 8. Epicuro, Sentencias Vaticanas, 74 9. Cf. Maliandi, “El trilema de Aristófanes.”

DISCOURSEETHICAL GRADUALISM: BEYOND

My question is the following: to what extent is ethical anthropocentrism tenable? In a “discourse ethical” perspective I will consider some caseoriented arguments in favor of a paradigmatically unique ethical standing for humans and some ar-

ANTHROPOCENTRISM

guments in favor of an ethical

AND

other mammals and between hu-

BIOCENTRISM?

conclusion in favor of a fair treat-

gradualism between humans and mans and nature, ending with a ment of all moral subjects, human and non-human.

Gunnar Skirbekk

I

n our ethical thinking, I assume, we tend to have some special cases (or clusters of cases) in mind, sometimes without being properly aware of it. I furthermore assume that we tend to argue either in terms of unique cases or in terms of relationships or similarities between various cases. The former way of arguing leads to paradigmatic thinking, the latter to gradualism. Each way has its virtues. The former makes us see differences, the latter makes us see continuities. In philosophy both are required. My question is the following: to what extent is ethical anthropocentrism tenable? In a “discourse ethical” perspective I will consider some case-oriented arguments in favor of a paradigmatically unique ethical standing for humans and some arguments in favor of an ethical gradualism between humans and other mammals and between humans and nature, ending with a conclusion in favor of a fair treatment of all moral subjects, human and non-human.

95

G UNNAR SKIRBEKK 96

CASE-ORIENTED ARGUMENTS FOR AND AGAINST ETHICAL ANTHROPOCENTRISM I begin by briefly recalling the apparent strength of ethical anthropocentrism. Morality is located in the socio-cultural world of acting and thinking human beings, and so are other norms and values, be they juridical or aesthetic. Nature can be the object of aesthetic attitudes and evaluations. She can also be the object of legal regulations, e.g., in terms of the right to own and to use natural resources. And nature can be the object of moral considerations, at least as far as biological nature is concerned: the extinction of endangered species is currently conceived as a moral concern, and the unnecessary infliction of suffering upon sentient non-human beings is in general seen as morally wrong. Sentient animals, at least, are thus the subjects of moral considerations. But these animals cannot themselves take part in these considerations. Nor can they act freely and rationally like human beings. They cannot act morally, nor immorally, only amorally, without responsibility or obligations of any kind. Animal rights are therefore asymmetrical; the rights of animals entail obligations for humans without any obligations on the animal’s side. Not even a chimpanzee is taken to be morally responsible for its deeds. Ethical anthropocentrism is thus prima facie a reasonable position: morality belongs exclusively to man and the human world. Even when we go through a certain repertoire of cases, from physical nature to plants and sentient animals, this conclusion seems safe. However, there are intensive discussions, especially in medical ethics, on the moral status of borderline cases: we have thorough discussions of the ethical status of a fetus and of people with severe brain damage. To what extent do they have the moral status of persons? To what extent do they have human rights? In these borderline cases we have members of the human species who do not perform like the paradigm case of a human being. The question concerning their moral status and their moral rights is therefore discussed by somebody else. This kind of advocatory representation is the standard case for minors: the parents are generally the first to be responsible for their behavior, for their upbringing, and for defending their interests, but other people or institutions might also be entitled to assume this role. This advocatory responsibility is supposed to be reduced gradually in accordance with the process of maturation; it formally ends at the time when the child reaches full legal age. The cases of advocatory representation in medical ethics are more extreme, as it were; they take place further away from the paradigm case of a mature and morally responsible human being than in the case of rearing

D ISCOURSE-ETHICAL GRADUALISM 97 children: we have fetuses at different stages, and we have those human beings who are still alive, but who at the end of their lives are no longer able to participate in a discussion about their own situation. We have people who are severely senile and people with severe brain damage. We have babies born with anencephaly who will never be able to participate in any such discussion. Along the same lines we have the cases of advocatory concern future generations, and thus for hypothetical people, that is, for those who am not yet individualized, but who are somehow statistically recognized. To the extent that medical ethics and our ethical concern in general tend to include all such cases within the realm of human morality—giving each of these human beings thereby the status of a person with moral rights—we have taken a major step away from the paradigm case of human morality. In this expansion we include everyone who belongs to the human species regardless of his or her capabilities. We include everybody who can become a mature human being. We include everybody who has been a mature human being. And we include everybody who once could have become a mature human being. There are certainly great differences between these different cases. And in quite a few cases there is no unanimity about the moral status of the human being concerned. The debates on abortion and euthanasia illustrate this point: membership in the human species is in one sense a biological question and in another sense a normative question, and there is an open debate concerning potential and full membership in the human species in a normative sense. When is a member of Homo sapiens a person, with full moral standing? A fetus is a potential person, but in what sense is it a person? A terminal patient with severe brain damage has been a person, but in what sense is be or she still a person? When all these kinds of human existence are considered, we am forced to adopt a certain gradualism, an “ontological” gradualism, not necessarily an ethical gradualism: i.e., we may consider the biological and psychological continuity (and variety) between different individuals of the human species, but still insist on the recognition of a moral status for them all. Their characteristics differ, but they are all human beings, with human dignity and human rights. This is an ethical anthropocentrism with an awareness of the borderline cases from medical ethics. So far we have not questioned the distinction between Homo sapiens and other species. However, at this point we ought to look carefully into borderline cases on the other side of the species border. Everybody is aware of the fact that chimpanzees can act and communicate, and that they can also experience pain and pleasure. Genetically,

G UNNAR SKIRBEKK 98 chimpanzees are close relatives to humans, and they have well-developed brains and nervous systems. Nevertheless, a chimpanzee does not possess the higher capabilities delineated above. It is not a moral agent; it is not a morally responsible being. Its use of language is neither reflexive nor creative. But it does act, feel and communicate. It has some self-awareness and sense of identity. And it clearly has higher mental capabilities than some of the members of the human species represented advocatorily for the sake of their moral status as human beings. The latter is a disturbing fact. In self-defense we, the humans, could add that we do take than pain of sentient non- human beings into moral consideration. It is regarded as morally wrong to inflict unnecessary pain on sentient animals; in most countries it is even forbidden by law. However, painless killing of animals is regarded as morally acceptable, an underlying assumption being that none of these animals has any awareness of its own death, except when a higher animal is threatened by death. What started as human self-defense here ends in embarrassment: some of our patients as well as early human fetuses, do not have any awareness of their own death. Even worse, they would probably experience even less of a trauma by being killed than most higher mammals probably experience when we kill them. Here we are confronted with a well entrenched ethics principle: if there is a moral difference between two cases, there must also be some morally relevant difference in properties between the two. (i) However, the decisive difference between the ‘deviant’ cases of Homo sapiens and the higher cases of other mammals can hardly be found on the level of actual properties; arguments from actuality are thus not conclusive for a sharp distinction between humans and other mammals. To begin with, there is a problem as to what actual properties should be considered. What are the morally relevant properties? Some major candidates are: an ability to act rationally and freely, an ability to make interest claims, and self-consciousness. But even if we should come to agreement on a clear and consistent notion of such actual properties, it is unlikely that such a notion would support a moral difference between all mammals that genetically belong to the human species and all other mammals. For whatever actual properties we choose there will most probably be cases of humans who fail to have these properties, and for quite a few such properties there will be some highly developed non-humans who have more of them than some defective or less developed humans do. If, for instance, we choose conscious self-identity as such a morally relevant property, we cannot

D ISCOURSE-ETHICAL GRADUALISM 99 include newborn human infants. If we choose some basic brain functions as a morally decisive actual property, we will have a problem in making a moral distinction between a human fetus and a chimpanzee fetus, and we will have to recognize that normal chimpanzees have this actual property to a higher degree than some humans with brain damage. If we include human infants or people with severe brain damage we will have to include adult chimpanzees as well. It is therefore hard to see how any actual property could do the job of providing a sharp demarcation between humans and non-humans. (ii) For this purpose arguments from potentiality are more promising, though not quite conclusive. If, for instance, we make the potentiality of a future self-conscious life to be the morally decisive property, we will certainly have cases of severely defective human infants with less of this potentiality than we find in normal chimpanzee infants. However, at this point we could introduce the notion of a potentiality of a second order, i.e., a potentiality to have potentialities, and on this basis we could argue in favor of a morally relevant distinction between these two cases: a brain-damaged human infant could have had a potentiality to become a self-conscious being in a sense in which the normal chimpanzee infant could not. This is clearly an interesting point. But the twist of the argument consists in a change from considering the real potentialities of an individual to considering the potentialities of the species, or the essence (in an Aristotelian sense) to which this individual belongs. I would therefore maintain that arguments from potentiality might gave some support to a notion of ethical anthropocentrism, though not unconditionally. (iii) There is also a theologically grounded anthropocentrism claiming (for instance) that all humans and only humans are created in the image of God, this giving them a unique ethical standing. But in this paper I choose to leave this theological version of ethical anthropocentrism aside. (iv) There is, however, a sociological version of the anthropocentric position which I will briefly comment upon. This argument does not defend genetic anthropocentrism, but a social anthropocentrism. The argument is one in favor of personhood and moral rights, not as individually given independently of society, but as a moral status inherently ascribed to members of a community. Any being included in this community has moral rights in accordance with his or her role and function within this community. According to such a societal notion, all that matters is whether a being is included in a community, to a large extent regardless of individual properties. If a being is included, this

G UNNAR SKIRBEKK 100 being enjoys a moral status with certain rights. Beings that are not included have no such rights, regardless of their actual and potential properties. For instance, a society may include a severely brain-damaged human being as a member with a basic moral standing, and the same society may choose to exclude a chimpanzee with higher actual (and potential) properties. If such a society decides to make a sharp distinction between those who are genetically humans and all other mammals, we have a case of a socially based anthropocentrism. In short, by tradition or by decision, a society can posit an ethical anthropocentrism. However, the, price paid is the danger of an ethical contextualism: we are at the mercy of a given tradition or of a given form of decision-making, as to the inclusion and exclusion of community members. The demarcation line between insiders and outsiders might in a given context be one that excludes groups of genetic human beings, regarding them as sub-human. An open discussion of the legitimacy of the given tradition and the given form of decision making is therefore required. We say: only to the extent that these conditions are freely discussed and agreed upon by everybody concerned could we rightly claim that these conditions are legitimate. Through such an attempt at legitimizing or criticizing a given tradition, we transcend ethical contextualism and move toward discursive ethics. (v) Discourse ethics, as a procedural ethics, implies a certain anthropocentrism since all actual participants, in the world known to us, are humans. But this is merely an empirical fact; in principle any speechacting sentient being could be a candidate for participating in a discourse. This claim is in need of a brief comment. In discourse ethics there is a demarcation problem concerning the competence requirements for participation in ethical discourse, and there is a problem of how to represent those who are themselves unable to participate, and of how to decide on who is deserving of advocatory representation: who are qualified to participate, who are qualified to be advocatorily represented in practical discourse? Should speech-acting and intelligent Martians participate? According to the basic assumptions of discourse ethics the answer is affirmative. Should only mentally undeveloped humans be advocatorily represented, and not higher mammals who actually are mentally more developed? According to the assumptions of discourse ethics the answer is negative: all moral subjects should be advocatorily represented, each in accordance with its moral status—which opens for an ethical gradualism, e.g., from more sentient to less sentient moral subjects.

D ISCOURSE-ETHICAL GRADUALISM 101 In this perspective the demarcation between humans and nonhumans gets blurred. The discourse-ethical support for ethical anthropocentrism is therefore conditional.

BASIC VIEWS ON WHAT C OUNT AS MORALLY RELEVANT PROPERTIES There is more to be said about morally relevant continuity and discontinuity between humans and non-humans. There are, for instance, various basic views on what count as morally relevant properties; the various ones may be conceived in different conceptual perspectives: hermeneutically, cases and conceptual perspectives are intertwined. I will therefore briefly recall some major theoretical positions, and I choose those located around utilitarianism, the deontological position, and discourse ethics. A classical passage from Jeremy Bentham gives the core of the utilitarian position: The question is not, Can they reason? nor Can they talk? but, Can they suffer?

At this point I will just summarize my own opinion by saying that the utilitarian focus on suffering (sentience) is conceptually too narrow to grasp the borderline problems between humans and non-humans with sufficient subtlety. This does not mean that the descriptive analyses of the utilitarians are wrong. Indeed, I think that utilitarian arguments about suffering and sentience (in humans and non-humans alike) have a normative strength which cannot easily be ignored. But still I regard their perspective as conceptually insufficient: utilitarians underplay the paradigmatic difference between humans and non-humans since they tend to neglect the morally relevant aspects of the acquisition and maintenance of a social identity. I also assume the quite a few of the proponents of individual rights, based on some deontological position, tend to operate with a conceptual scheme which is too narrow to grasp the realm of socialization and of reflective and discursive justification with sufficient subtlety. In my opinion, the idea of a paradigmatic difference between humans and non-humans is most satisfactorily conceived by the theorists of social identity and of discursive rationality. The gradualist argument, as utilitarianism is thereby not denied, only “sublated,” as it were.

G UNNAR SKIRBEKK 102 In my view there are thus two major advantages to discourse ethics relative to utilitarianism and to classical theories of pre-social individual rights. Bluntly stated: (a) These two latter positions presuppose socialized individuals, without themselves questioning and elaborating the implications of human socialization for the interrelationship between individuals and communities, for the human need for a mutual recognition of one’s vulnerable social identity, and for the discursive interpretation of needs in general. (b) Often the emphasis is given to detailed analysis of actual and potential properties in the various creatures, but then there is a danger of a naturalistic fallacy and thus of ethical skepticism. One way out of this skepticism is that of a reflection on the constitutive conditions for a normative discussion, in short that of a discourse ethics. Discourse ethics focuses on the self-reflective insight of argumentation, using the undeniability of its constitutive preconditions as its foundation, supplemented with arguments from universal pragmatics and developmental theory. As an ethics based on the self-reflective insight of argumentation and on speech-act inherent preconditions, discourse ethics includes universalization in terms of general role taking and presupposes mutual recognition among the discussants, thus underlining socialization as a core element. Both social identity through socialization and normative justification through self-reflection and discourse are thus essential elements in discourse ethics. Even if discourse ethics has its inherent problems, for instance the problem of the exclusion and inclusion of participants, and even if this problem is extended to advocatory representation for humans and non-humans (including future generations and future moral subjects of a non-human kind), a revised version of discourse ethics still remains, in my view, the best proposal for a fundamental ethical theory.

THE RELATIONSHIP BETWEEN MORAL SUBJECTS, MORAL AGENTS AND MORAL DISCUSSANTS I will end these reflections on the ethical borderline between humans and higher mammals by discussing some conceptual distinctions, pertinent for the discourse-ethical approach, concerning the notion of a moral subject in its relation to the notions of a moral agent and of a moral discussant. In discussing the moral status of humans and non-humans it is helpful to make a distinction between moral agents and moral subjects. The former are capable of acting morally, the latter are capable of being harmed in a morally relevant sense. Moral obligations are connected both to the interrelationship between moral agents and to their relationship to moral

D ISCOURSE-ETHICAL GRADUALISM 103 subjects. But moral subjects who are not moral agents are unable to have obligations (either to each other or to moral agents). If we were to work out a concept of an agent capable of acting morally, we would include notions about capabilities to understand a situation and to evaluate the moral importance of what one does or does not do, and to act in accordance with that understanding and evaluation. In trying to make such a concept of a moral agent more precise we move into current philosophical debates on the nature of action, of rationality and of accountability. I assume, however, that the crude distinction between moral agent and moral subject makes enough sense for the time being, At this point I will approach these philosophically controversial issues by introducing the concept of a moral discussant, and consider the interrelationship between moral discussants, moral agents and moral subjects. It could be claimed that the concepts of moral agent and of moral discussant are coextensive: those who are able to act morally are able to discuss moral actions, and those who are able to discuss moral actions are also able to act morally. But even if this were empirically true, it does not imply that the two concepts converge into one. However, it is reasonable to assume that those who are able to act morally should normally be able to talk about their evaluations and actions, and even be able to explain why they think their actions were morally right in a given situation. That would be the first step toward a moral discussion. In this sense it is reasonable to think of moral agents as potential moral discussants. This is an important point, and I assume that the claim of an interrelationship between moral agency and moral discourse can be philosophically elaborated along the lines of a universal pragmatics (as in Apel and Habermas) and of a theory of socialization and of modernization (as in Kohlberg or Weber). But these are tricky problems, and we should at least be aware of the various levels involved: even if moral agents paradigmatically are moral discussants, that does not mean that all are. Even if moral agents are potentially moral discussants, that does not mean that they always actually are. It is possible to think of cases of moral agents who are relatively unable to discuss the moral aspects of their actions. The ability to discuss moral questions requires considerable intellectual skills. It presupposes a certain intelligence and a certain training. The required social and intellectual training comprehends an ability to step back and reflect upon a case from different perspectives, an awareness of the possibility of applying different concepts, and an ability to discuss their strengths and weaknesses in a given case. These kinds of reflection require not only mature (and sane) individuals, but also a certain cultural development, that is, a certain degree of modernization.

G UNNAR SKIRBEKK 104 This means, all in all, that in claiming an interrelationship between the concept of a moral agent and the concept of a moral discussant, we are not talking in empirical terms, but in terms of presuppositions and idealizations, that is, in term of a universal pragmatics concerning competences inherent in speech-acts and in arms of a theory of modernization and socialization concerning conceptual development and the development of social identity. Only if we are willing to argue for some such presuppositions can we talk about moral agents as moral discussants, tout court, If not, we cannot claim that being a moral agent is sufficient far being a moral discussant. Could we claim, the other way around, that being able to participate in a moral discourse is sufficient for being able to art morally? The answer depends on the presuppositions built into the concept of a moral discussant. We could try to spell it out in a thought-experiment (and a pretty weird one at that). Let us imagine a robot and a god, both having adequate intelligence, knowledge of all relevant facts, and an ability to speak and to listen. In short, we assume that both are capable of participating in theoretical discourses. I also assume that they are able to intervene in worldly affairs. But are the requirements for participating in a theoretical and in a practical discourse the same? If the answer is affirmative, then these theoretical discussants are moral discussants, and we could ask whether they are also to be regarded as moral agents. But could discussants without a biological body (like our robot or a bodiless god) count as moral agents? A being without a biological body has no biological needs, no experience of biologically rooted pains or pleasures, no biological birth and growth, no biological aging or mortality, no biological vulnerability. Such a being could, according to our presuppositions, discuss all morally relevant facts concerning moral subjects, be they humans or non-humans, Such a being could also apply legal and formal principles, such as the principle of treating equal cases equally. But how could this being possibly be in a position to understand and evaluate biological life, with its vulnerability and death? It could get information about these facts, but without acquiring through experience the notions necessary to understand these facts. This being could get information about people’s reactions towards these facts of life, but how could this being understand these reactions? This robot or bodiless god has no experience rooted in life processes from the psychological and social world, including socialization and learning based on bodily existence and interaction. This being could only register people’s actions mechanically, without understanding what the passions and interests were all about. The problem here is that of how an observer might acquire act-constitutive notions when he or she

D ISCOURSE-ETHICAL GRADUALISM 105 is excluded from participating in the activities for which the notions are constitutive: as a non-participant in the biological world, and in the social world that is based directly or indirectly on biological existence, this being would not be able to participate in all the role taking which is supposed to take place in a practical discourse, and which again opens the way to universalization and to solidarity. My thought-experiment is focused on the importance of bio-bodily existence for the concept of a moral discussant. The robot, I assume, has no body, that is, no biological body, but merely a mechanical one. The god has no body, or he has a body that is invulnerable and eternal, never born and immortal. My suggestion is that a biological bodily existence is necessary for a competent moral discussant. Those who cannot be morally harmed, since they lack vulnerability, cannot be moral subjects, and therefore they cannot be moral discussants, even if they have the intelligence, the information, and the semantic competence required: one has to be a moral subject in order to be a moral discussant. Furthermore, in order to acquire act-constitutive notions one has to be a participating agent: one has to be a moral agent in order to be a moral discussant. (This latter argument, I assume, holds for the acquisition of language in general and therefore also for theoretical discussants— but for the sake of the thought experiment, I here disregard this point.) If this argument is tenable it means that discourse ethics has a biological foundation as part of the competence requirements for being a participant in practical discussions. It means that biologically rooted learning and vulnerability represent a shared foundation for moral discussants, moral agents and moral subjects.

CONCLUSION Regarding the moral standing of human beings, both paradigmatic and gradual ways of thinking are possible, all depending on context and perspective—but not unconditionally, not without an appropriate conceptual explication. A crude version of ethical anthropocentrism is philosophically not tenable. This conclusion does not imply a reduction of human dignity and hence a diminishment of the realm of ethics. The other way around: it implies a gradualist inclusion of all sentient non-human beings into the realm of moral subjects, and hence, compared to ethical anthropocentrism, it represents an extension of the moral universe. This conclusion has substantial normative implications, not only for our personal lives, but also for our legal and economic institutions.

G UNNAR SKIRBEKK 106 We are forced to answer the question: how should we fairly represent all moral subjects, human and non-human, both in our personal lives and with regard to these social institutions? Gunnar Skirbekk, Center for the Study of the Sciences and the Humanities, and Department of Philosophy, University of Bergen, Bergen, Norway, 5020; [email protected]

MORAL PHILOSOPHY AND LINGUISTICS

Any acceptable account of moral epistemology must accord with the following points. (1) Different people acquire seemingly very different moralities. (2) All normal people acquire a moral sense, whether or not they are given explicit moral instruction. Language resembles morality in

Gilbert Harman

these ways. There is considerable evidence from linguistics for linguistic universals. This suggests that (3) despite the first point, there are moral universals. If so, it might be possible to develop a moral epistemology that is analogous to the theory of universal grammar in linguistics. In what follows, I will try to sketch what might be involved in such a moral epistemology.

INTRODUCTION

M

orality is in many ways like language. Both admit of great apparent variation. Both appear at first to be group phenomena but are probably best studied as instantiated in particular individuals. Both are acquired without explicit instruction through appropriate experience of the world. These similarities between morality and language suggest there may be other similarities. Perhaps morality can be studied in the way that language is studied in linguistics.1 Linguistics seeks to uncover a universal grammar behind the surface diversity between languages. Might moral philosophers discover something like a universal grammar of morality? In what follows I begin with a brief description of the extent of apparent moral diversity. I then sketch how linguists are able to accommodate linguistic diversity within universal grammar. Finally, I consider how there might be an analogous study of universal morality. 107

G ILBERT HARMAN 108

APPARENT M ORAL DIVERSITY I will mention two kinds of apparent moral diversity, societal and individual. Societal Differences

Apparent differences in the moralities associated with different societies range from trivial differences in etiquette (burping after dinner being polite in some places, impolite in others), to differences concerning cannibalism, forms of marriage, and caste systems. It might be thought that there is a universally recognized central core of morality involving at least prohibitions against killing and harming others, against stealing, and against lying to others. But much depends on what counts as acceptance of the same principle in different societies. Consider first theft. Societies that do not recognize private property do not have rules against stealing; without property, there can be no theft and no rules against theft. On the other hand, it seems trivial to say that all societies that recognize private property have rules against stealing, because having rules against stealing is a necessary condition of recognizing private property. Next consider rules against murder and harming others. Typically, a morality does not prohibit all killing or harm. In some societies a master is thought to have an absolute right with respect to slaves, including arbitrarily beating and killing them, and there may be no limitations on what a husband can do to his wife or a father to his young children. Where moral prohibitions on harming and killing exist, they typically apply only with respect to some privileged group and not with respect to outsiders with different moralities privileging different groups. Prohibitions on lying and cheating are similarly often limited to the local group and not to outsiders. Someone who can successfully cheat outsiders may be treated as an admirable person. Similarly, for someone who is able to harm and kill outsiders. Rules against killing, harm, lying, and cheating may occur in all societies, but not exactly the same rules, since the protected group changes from one society to another. Individual Differences

A further aspect of apparent moral diversity is that it occurs not just between societies but also within societies and in a way that often leads to seemingly intractable moral disagreements that rest on irreconcilable differences in basic values and not just on differences in opinion about the nonmoral facts. Moral vegetarians, who believe that it is wrong to raise animals for food, can exist in the same society (even the same family) as non-vegetarians. A disagreement between moral vegetarians

MORAL PHILOSOPHY AND LINGUISTICS 109 and non-vegetarians can seem to survive full discussion and agreement on the facts and certainly appears to depend on a basic difference in the significance assigned to animals as compared with humans. In a similar way, disagreements about the moral acceptability of abortion or euthanasia seem to survive extensive discussion and awareness of all relevant information about abortion. Such disagreements appear to depend on basic differences concerning the degree to which human life has intrinsic value as compared with the degree to which the things that life makes possible are of intrinsic value, like pleasurable experience and fulfilling activity.

LANGUAGE Before comparing the study of morality to the study of language, I need to say something about language and linguistics. Linguistic Variation

At least on the surface, linguistic diversity is even greater than moral diversity. Languages differ in vocabulary and in what sounds are used to form vocabulary. Where the English say “white” the French say “blanc,” which sounds like the English word “black.” Languages also differ in syntax or word order and in whether thematic relations like subject and object are normally revealed through word order, as in English, or through case, as in Latin. Language, Dialect, Idiolect

It might be supposed that a language is always the language of a group and is based on the linguistic conventions of that group, but this seemingly plausible idea is hard to reconcile with dialectical differences within what are normally called languages. Dialects blend into one another in ways that do not fit national or geographical boundaries. Even for speakers of what we normally consider the “same language,” it is unlikely that any two people speak exactly the same language. Your English differs somewhat from mine; her French differs from his. There will be some difference in vocabulary, pronunciation, or grammar. The dialect of a particular person that is in certain respects unique to that person is that person’s “idiolect.” Different people speak at least slightly different idiolects. The serious study of language therefore takes an idiolect to be the unit of language. The more popular, nonscientific notion of a language, like French or German or Chinese, is a political notion. “A language is a dialect with an army and a navy.”

G ILBERT HARMAN 110 Competence, Performance, and I-Grammar

An individual’s linguistic performance depends in part on his or her linguistic competence, perhaps as specified by an internal grammar, or I-grammar, but also depends in part on non-linguistic factors, including limited memory, nonlinguistic goals and beliefs, functioning vocal tract and hearing, parsing strategies, etc. Although linguistic intuitions are the primary source of evidence for linguistic theories, such intuitions are influenced by more than linguistic competence. Sentences that seem ill formed may be grammatical but hard to process. Possible examples are: 1. The horse raced past the barn fell. 2. The malt that the rat that the cat that the dog bothered chased ate lay in the house that Jack built.

On the other hand, sentences that seem well formed to the person may actually be ungrammatical in his or her idiolect. One possible example is: 3. The not unhappy student finished his term paper.

The evidence of linguistic intuition bears only indirectly on linguistic theory. Language Acquisition and Universal Grammar

All normal children acquire language. The only exceptions are children with specific brain deficits and children brought up in terribly isolated conditions. How do children first acquire language? It may seem that children are normally taught language by their parents. But, explicit instruction in language is not necessary for language learning and may not be very useful. Children who interact with people speaking a local dialect will acquire (a version of) that dialect, whether or not their parents ever do. Important aspects of language are not learned in any explicit way, as in the following somewhat oversimplified example. In dialects of English, questions can be formed by moving a question word to the beginning of the sentence and following it with the first auxiliary verb, as in “Who is Alfred asking Bob to talk to?” which comes from “Alfred is asking Bob to talk to who?” But this rule cannot apply if the question word occurs in grammatical conjunction with another noun phrase, as in “Alfred is asking Bob to talk to who and Mary?” because the result, “Who is Alfred asking Bob to talk to and Mary?” is not a well-formed question.

MORAL PHILOSOPHY AND LINGUISTICS 111 Now it is very difficult to see how a child might learn such a constraint (the “coordinate structure constraint”) given the available evidence. Indeed, the constraint was unknown to linguists before 1950! But, if children do not learn the constraint, that can only be because it is somehow built into language acquisition ahead of time. So it should apply no matter what language is learned, which is to say it must be a universal constraint, one that holds in every language that is acquired as a first language in the normal way. Reasoning in this manner, linguists can use facts about a single dialect of English to make predictions about all dialects that are acquired in the normal way as first languages. And, indeed, the prediction about the constraint on question word movement has been confirmed through the study of a great variety of languages, as have other similar predictions. Considerations of learnability have led linguists to universal grammar. Notice that such universal principles of language must hold not only for already existing languages but for newly created natural languages as well. For example, deaf children of hearing parents sometimes invent their own new sign language. Such an invented sign language will have all the complexity of spoken natural languages and admit of the same linguistic universals, despite being newly invented. Furthermore, when people who speak different languages come together, they may for business purposes invent a crude “pidgin” mixture of languages that does not satisfy the principles of universal grammar, but their children will acquire a corresponding “creole” that does satisfy the principles of universal grammar. Pidgins are not natural languages in the relevant sense, even though they arise in a natural way when people speaking a variety of languages need to do business together. Pidgins are not “natural languages” because they are not learnable in the natural way in which children learn a first language. I said that my discussion of the constraint on question word movement is oversimplified. A rule that is not learned might be built into universal grammar merely as a default case, so that there might be exceptions to it. If a child had evidence that the rule did not hold in the local dialect, the child would learn a dialect that is not subject to the rule, even though in other circumstances a child would learn a language that is subject to the rule without having to learn the rule. Further refinements of this approach have led to a “principles and parameters” theory of language, which says that languages are basically the same except for vocabulary and a few parameters whose values are picked up during early language acquisition, parameters that determine for example whether the head of a phrase goes at the beginning or end of the

G ILBERT HARMAN 112 phrase. The task of universal grammar is then to specify the relevant principles and parameters.

LINGUISTICS AND MORALITY Moral Idiolects, I-Morality

As with language, it might at first be supposed that a morality is always the morality of a culture or group of people. But, just as different people in the culture or group have different idiolects, they also have at least somewhat different moral understandings and the moral understandings of different people vary continuously in ways that do not respect national, cultural, or group distinctions. Two people in the same family might have slightly different (or even very different) moralities. We must therefore concentrate on the characterization of an individual person’s morality—that person’s moral idiolect, as it were. Also as with language, we must recognize that a person’s actions and conscious judgments about right and wrong depend not only on that person’s moral competence or I-morality but also on many other factors, including various beliefs, goals, limits of information, memory, and processing strategies. Although moral intuitions are the primary source of evidence for moral theories, such intuitions can be influenced by more than just the individual’s moral competence.2 The evidence of moral intuition bears only indirectly on the correct characterization of an individual’s I-morality. Moral Education

Despite rather large apparent differences in moral idiolects, all normal people acquire a moral sense. The only exceptions are people with certain sorts of brain damage and people whose upbringing is unusually deprived. Although many parents explicitly teach morality to their children, such teaching is probably not needed and may not be particularly effective. Just as children acquire a version of a local dialect simply through being exposed to others who already speak it, without any need for explicit instruction, it is likely that this is true for morality also. But, if morality is universal in the sense that all normal people develop it, how much of the content of morality is universal?

MORAL PHILOSOPHY AND LINGUISTICS 113 Universals of Language and Morality

Despite obvious surface differences between languages (English and Japanese, for example), there are universals of language, characteristics which all languages share. There is universal grammar. Individual languages are actually very similar except for vocabulary and a few choices about syntax. Is the same true of morality? Despite large obvious surface differences in moralities are there underlying universal features, so that differences in moralities depend on a small number of parameters? If so, what are the universals and how can we find them? Where should we look for them?

POSSIBLE MORAL UNIVERSALS The I-grammars of all languages are based on categories like phoneme, morpheme, phrase, clause, etc., with nouns, verbs, prepositions, adjectives, etc. Similarly, we might expect that all I-moralities make use of basic notions like right, wrong, permitted, not permitted, good, bad, fair, unfair, etc. Questions might arise about some notions: do all I-moralities contain a notion of individual rights in anything like the modern sense? Is some relation to guilt crucial to I-morality? Whatever principles there are, they do not hold absolutely but are themselves default rules. There are situations in which I-morality permits one to kill or injure or interfere with someone in the relevant protected group. The need for a systematic way of dealing with default rules complicates I-morality.3 (Or maybe some other part of the mind figures out what to do when there are conflicts among the requirements of I-morality.) Unlearned Moral Principles, Moral Pidgins, and Moral Creoles

That part of moral philosophy that looks most like linguistics tries to find principles that would characterize intuitive judgments about trolley cases and related examples that derive from Philippa Foot’s famous article on “Abortion and the Doctrine of Double Effect.” 4 Suggested principles include the Principle of Double-Effect, the principle that Negative Duties are stricter than Positive Duties, and Thomson’s Deflection Principle, which specifies an exception to the general rule that one is not permitted to harm someone to save others, the exception being when one deflects an existing harm from a larger group of people to a smaller group of people.5 It is interesting that the suggested principles

G ILBERT HARMAN 114 are not principles that ordinary people are aware of accepting and philosophers disagree about which principles are appropriate and how they are to be formulated. If any of these principles should turn out to be adequate to an ordinary person’s I-morality, the situation would be like that with respect to the constraint on question-word movement discussed earlier. An ordinary person was never taught the principle of Double-Effect or the deflection principle, and it is unclear how such a principle might have been acquired from the examples available to the ordinary person. This suggests that the relevant principle is built into I-morality acquisition, in which case we should expect it to occur in all moralities (or be a default case, or something of the sort). In other words, the principles should be part of universal moral grammar. 6 A further implication is that we might be able to distinguish between as it were pidgin and creole moralities. A pidgin morality would be an invented morality such as act utilitarianism that did not adhere to such universal principles as Double Effect or the Deflection Principle. The corresponding creole morality would be the I-morality acquired in the normal natural way by children of utilitarians. Principles and Parameters

It may be that all I-moralities contain certain principles which involve parameters that differ from one I-morality to another. Perhaps all I-moralities contain a rule prohibiting killing of members of a protected group G, where G may vary. Similarly, I-moralities may contain a rule prohibiting harm to members of a protected group H, where H may or may not be the same as G. Is it true that, where there is a rule against harming Hs, there is also a rule against killing Hs? Not obviously, because many people seem to accept a rule against cruelty to animals while allowing animals to be killed for food. There may be a universal principle of noninterference with the activities of the members of protected group H. Or perhaps all moralities accept some sort of Golden Rule or variant of Kant’s Categorical Imperative: One thinks about cases by imagining the situation reversed. Or one does this at least for situations involving members of some favored group G. Notice that disputes about the extent of the protected group G yield some of the most compelling examples of intractable moral disagreement. Are animals included? Are fetuses? Such disputes may reflect genuine parametric variation!

MORAL PHILOSOPHY AND LINGUISTICS 115

CONCLUSION An account of moral diversity might be modeled on linguistics. The plan would be to give a relatively precise account of particular moral idiolects and to consider whether the principles involved could be learned in order to see what universal constraints there are on morality. Such an approach promises new ways of looking at morality. Gilbert Harman, Department of Philosophy, Princeton University, Princeton, NJ 08544; [email protected]

NOTES 1. John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971), 47. John Mikhail, “The Moral Faculty,” (workshop presentation, 23 April 1996). 2. For preliminary discussion of possible other factors, see Tamara Horowitz, “Philosophical Intuitions and Psychological Theory,” Ethics 108 (1998): 367-385; F. M. Kamm, “Moral Intuitions, Cognitive Psychology, and the Harming-versusNot-Aiding Distinction,” Ethics 108 (1998): 463-488. 3. Robert Nozick, “Moral Complications and Moral Structures,” Natural Law Forum 13 (1968): 1-50; reprinted in Robert Nozick, Socratic Puzzles (Cambridge, MA: Harvard University Press, 1997). 4. Philippa Foot, “Abortion and the Doctrine of Double-Effect,” in Foot, Virtues and Vices and Other Essays in Moral Philosophy (Oxford: Basil Blackwell, 1978). 5. Judith Jarvis Thomson, “Killing, Letting Die, and the Trolley Problem,” and “The Trolley Problem,” in Thomson, Rights, Restitution, and Risk: Essays in Moral Theory, edited by William Parent (Cambridge, MA: Harvard University Press, 1986). (Thomson no longer accepts this principle.) 6. In this paragraph I am indebted to John Mikhail, “Outline of a Research Program in Moral Psychology,” (manuscript, 11 August 1997).

EXPLANATION AND JUSTIFICATION IN MORAL EPISTEMOLOGY

Recent exchanges among Harman, Thomson, and their critics about moral explanations have done much to clarify this two-decades-old debate. I discuss some points in these exchanges along with five different kinds of moral explanations that have been proposed. I conclude that moral explanations cannot provide evidence within an

Walter Sinnott-Armstrong

unlimited contrast class that includes moral nihilism, but some moral explanations can still provide evidence within limited contrast classes where all competitors accept the necessary presuppositions. This points towards a limited version of moral skepticism.

G

ilbert Harman’s discussion of moral explanations at the beginning of his textbook, The Nature of Morality,1 is rightly famous for raising a profound “problem with ethics.” In his central example, Harman argues that a scientist’s belief that a proton is in a cloud chamber at a time can be justified by an inference to the best explanation of the scientist’s observations of a cloud in that chamber at that time. This justification works, because the actual presence of a proton and hence the truth of the belief is a necessary part of the best complete explanation of why the scientist forms the belief at that time rather than earlier or later. More generally, protons are needed to explain experimental results and thereby to explain why scientists come to believe in protons on the basis of those experiments. The problem for ethics is that moral facts and properties never seem to be needed for explanations of this kind. 2 Harman tries to solve this problem by arguing that moral facts can be reduced to psychological facts that are needed for the best explanation of observations. 117

W AL TER S INN O T T -A RMSTRO NG ALTER INNO RMSTRONG 118 However, without such a reduction of morality to science, morality seems more problematic than science. At first, it was not completely clear which kind of problem was being raised. On one interpretation, with some basis, Harman’s argument raises a metaphysical or ontological problem about whether or not moral facts exist. However, certain background beliefs and interests affect which explanation is best, but these beliefs and interests do not affect which facts exist. This makes Harman’s argument dubious if it is directly about moral ontology. Harman’s argument becomes more plausible if it is interpreted so that it raises a problem in moral epistemology. In recent exchanges, Harman makes it clear that epistemology is his main concern: “[M]y interest in moral explanations has always been an interest in the extent to which uncontroversial data might be better explained by one rather than another competing moral framework, so that the data could provide evidential support for the one framework over the other.” 3 Harman’s conclusion is that “[w]hether or not [Thomson’s examples] are cases in which a moral claim explains some uncontroversial facts, they are not cases that support one moral framework as compared with another.”4 Thus, Harman now seems to allow the possibility of moral explanations, but he still denies that they have any epistemological force. I will argue that Harman is correct to some extent but not completely. Moral explanations can be used to adjudicate between some rival moral frameworks if, but only if, these frameworks share certain assumptions. To show this, the first step must be to clarify the basic principle in Harman’s argument. It seems to be this: An observation O is evidence for hypothesis H1 above hypothesis H2 if the truth of hypothesis H1 is a necessary part of the best complete explanation of observation O, and if neither observation O nor any presupposition of the explanation is in dispute between defenders of hypotheses H1 and H 2.

The truth of H1 need not be the complete explanation, since subsidiary principles will be needed; but the truth of H1 must be necessary and not just an optional add on.5 Also, this kind of evidence requires not just any explanation but the best explanation, where the best explanation is, roughly, the simplest and most conservative that explains the most data. Most importantly, the observation to be explained must be independent, that is, not controversial or in dispute at present. If one theology provides the best explanation of why God told Abraham to kill his son, that is not evidence for that theology against an atheist who denies that God told Abraham anything.

E XPL ANA TI O N AND J USTIFIC ATI O N I N M OR AL E PISTEM O LOGY ANATI TIO USTIFICA TIO ORAL PISTEMO 119 This independence requirement makes such evidence relative to a contrast class. An observation that provides evidence for hypothesis H1 above hypothesis H2 might not provide evidence for hypothesis H1 above hypothesis H3 if the observation is admitted by proponents of H1 and H2 but is denied by proponents of H3. We cannot, then, say simply that the observation is evidence for H1. We need to say that it is evidence for H1 as opposed to a certain contrast class of competitors. This should come as no surprise. Explanations are relative to contrast classes, so this relativity is inherited by evidence based on inferences to the best explanation. Given this relativity, one issue is whether these requirements are met by any moral explanation out of any limited contrast class of moral beliefs or theories. If so, the explained observations provide evidence for one moral belief or theory above others in the contrast class. If not, they do not. A separate issue is whether any moral explanation can provide evidence for any moral belief when the contrast class is unlimited, so it includes all competitors. To address the latter issue, I will focus on one extreme competitor, moral nihilism, which is the view that there are no moral facts, including facts about which acts are morally required or wrong, good or bad, just or unjust, etc. I will argue that no moral explanation can provide evidence against moral nihilism, since every such explanation depends on some moral presupposition that nihilists deny, so it fails the independence requirement. Consequently, moral explanations cannot provide evidence within an unlimited contrast class. Nonetheless, some moral explanations can still provide evidence within limited contrast classes where all of the competitors accept the necessary presuppositions. There will be other limited contrast classes within which the moral explanation cannot provide evidence, but my point here is just that moral explanations work for some limited contrast classes but not for an unlimited contrast class. Many moral explanations have been proposed. Nobody claims that moral facts directly explain purely physical events, but moral facts are claimed to explain various aspects of human life, including acts and beliefs of individuals or groups of humans.6 So there are four main patterns: WHAT IS EXPLAINED?

Individual

Group

Act

Woodworth

Abolitionism

Belief

Amazing Grace

Convergence

W AL TER S INN O T T -A RMSTRO NG ALTER INNO RMSTRONG 120 The first pattern occurs when a moral judgment about a person or a person’s character trait is cited to explain that person’s acts. A classic example, given by Nicholas Sturgeon, is that of Passed Midshipman Selim Woodworth, as described by Bernard DeVoto. Woodworth accepted command of the mission to rescue the Donner party, who were trapped and dying in the High Sierra mountains. Instead of leading rescue parties into the mountains, Woodworth “spent time arranging comforts for himself in camp, preening himself on the importance of his position.”7 As a result, many people died who could have been saved. DeVoto concludes, “Woodworth was just no damned good.” 8 Sturgeon concurs. Their idea seems to be that Woodworth would have led a rescue if he had had any goodness at all in him. More specifically, his moral vices of cowardice and vainglory are supposed to explain his actions. To assess this explanation, we need to specify the relevant contrast classes. If the question is why Woodworth did bad acts instead of good acts, then it seems natural to cite the moral badness of Woodworth’s character. But this question obviously assumes that Woodworth’s acts were bad. So it cannot provide evidence against anyone who denies that assumption, as do moral nihilists. To get evidence that does not beg the question against moral nihilism, we need the contrast class to be specified in morally neutral terms. Presumably, then, what is to be explained is why Woodworth stayed in the base camp instead of going into the mountains to save the Donner party. Of this, there are several possible explanations: Maybe (1) he was coerced (by being held at gun point in the base camp). Maybe (2) he did not know that the Donner party needed him (because he thought someone else had already rescued them). Maybe (3) he thought he had to stay in the base camp to prevent some greater evil (such as the spread of an infectious disease carried by the Donner party). But none of this was true. Woodworth had no justification or excuse for his acts. So why did Woodworth do what he did? In contrast with explanations (1)–(3), it seems better to explain his actions by the fact that (4) Woodworth was no damned good. Opponents of moral explanations will insist on considering other alternatives, such as (5) Woodworth was more concerned about his own comfort and ambition than about the lives of those whom he had promised to rescue. If some non-moral explanation like (5) is available, then (4) does not seem necessary. However, defenders of this moral explanation will respond that (5) is a reason why (4) is true on the assumption that anyone is no damned good if he is more concerned about his own comfort and ambition than about the lives of others whom he promised to rescue. So (5) is not a

E XPL ANA TI O N AND J USTIFIC ATI O N I N M OR AL E PISTEM O LOGY ANATI TIO USTIFICA TIO ORAL PISTEMO 121 competitor to (4). Whereas (1)–(3) exclude (4) (and also (5)) as the best explanation, (5) does not undermine but supports (4), at least on certain assumptions. Defenders of moral explanations might go further and claim that (4) is better than (5), because (4) is more general, so (4) would explain more actions than (5) would explain. Moreover, suppose Woodworth would have stayed in the base camp if he had been no damned good in some way other than that described by (5), such as if he had been more concerned about his physical appearance than about the Donner party. Then (4) rather than (5) would capture the level of generality that best explains Woodworth’s overall pattern of actions. Even if so, however, this same level of generality can be achieved by describing the non-moral supervenience base for this moral property. Let’s stipulate that “bbadness” is the complex non-moral base property that underlies the supervening moral property of being bad (or no damned good). Then we can cite Woodworth’s bbadness to explain everything that could be explained by his badness (or by his being no damned good). One might object that there might be nothing that unifies bbadness.9 But there also might be nothing that unifies badness (or being no damned good). So there is no reason to prefer the moral explanation over the subvening non-moral explanation. Admittedly, there is also no reason to prefer the non-moral explanation. Moreover, if supervenience ruled out all moral explanations and evidence, it would also rule out all biological and economic explanations and evidence; insofar as biological and economic facts also supervene on physical facts, just as moral facts do. So this argument is too strong to be plausible. 10 This makes it seem legitimate to cite the moral fact that Woodworth was no damned good to explain why he stayed in the base camp instead of saving the Donner party, but only if we assume that his kind of character is bad. If this assumption is shared by all of the moral views in a limited contrast class, then DeVoto’s moral explanation can be cited as evidence out of that contrast class for the moral fact that Woodworth was no damned good. Moreover, contrary to Harman’s claim, this moral explanation can “support one moral framework as compared with another.”11 Consider a utilitarian moral theory on which one is morally good if one’s character has good consequences. Compare this with a religious moral theory on which one is morally good if one is connected to God by faith. The religious view still might hold that God would never let anyone with faith be more concerned about his own comfort and ambition than about the lives of others whom he promised to rescue. But there might be strong independent evidence that Woodworth had faith in God. That

W AL TER S INN O T T -A RMSTRO NG ALTER INNO RMSTRONG 122 together with the evidence that Woodworth was no damned good would support the utilitarian moral framework over the religious framework, as long as both continue to share the moral assumption that a person is no damned good if he is more concerned about his own comfort and ambition than about the lives of others whom he promised to rescue. In contrast, this assumption would be denied by moral nihilists. Since moral nihilists deny all moral facts, they deny (4). Instead, they would cite something like (5) (or his bbadness) as the best explanation of Woodworth’s acts. And (5) does compete with (4) on their view, since they believe (4) cannot be true but (5) can, so (5) is not a reason for (4). Consequently, the independence requirement for evidence is not met by this moral explanation when moral nihilism is in the contrast class, so this purported moral explanation cannot provide any evidence against moral nihilism.12 Similar points apply to the second pattern, where a moral fact is supposed to explain some group action or event. In the standard example, we observe that the anti-slavery abolitionist movement grew in the United States during the 1830’s. We wonder why. One possible explanation is that (1) The abolitionists were largely a displaced elite expressing resentment at industrialists who had replaced them.13 Another possible explanation is that (2) The abolitionists reacted to the real evils in slavery (which became worse during the 1820s). The latter explanation seems better than the former if enough abolitionists were industrialists. Opponents of moral explanations will insist on considering other alternatives, such as (3) Abolitionists disapproved of the pains and losses of freedom in slavery (which became worse during the 1820s). But explanations (2) and (3) are not competitors if pain and loss of freedom are the evils in slavery. This assumption would be denied by a nihilist. So the rise of abolitionism cannot be evidence against moral nihilism. But it can still be evidence within some limited contrast classes that share common moral assumptions. Things get trickier with the third pattern, where an individual’s moral belief is explained by its truth.14 If there are moral facts, then it is possible that someone called Amazing Grace is a perfectly reliable detector of all moral facts. Grace believes that an act is morally wrong when and only when the act is morally wrong, and similarly for right, bad, good, and so on. The real moral wrongness of an act would then be the best explanation of Grace’s belief that the act is morally wrong.15 Such an explanation of a moral belief is criticized by Judith Jarvis Thomson on the grounds that Grace believes that an act is wrong because she observes non-moral features of the act, which explain but are not explained by the moral fact.16 This is said to distinguish moral perception

E XPL ANA TI O N AND J USTIFIC ATI O N I N M OR AL E PISTEM O LOGY ANATI TIO USTIFICA TIO ORAL PISTEMO 123 from visual perception, where one’s belief that an apple is red is based on the fact that it appears red, which is explained by the fact that the apple is red. However, symmetry can be reintroduced simply by saying that Grace believes that an act is wrong because it seems wrong to her, and that seeming is explained by the moral fact that it is wrong. So Thomson’s criticism does not cut very deep. A deeper (although more obvious) problem is that this example is fictional. In the actual world, one could cite Grace’s belief as evidence for a moral fact only if one already had adequate independent evidence that Grace is a reliable detector of moral facts. Such evidence would have to include observations that Grace’s past moral beliefs were true. This works fine in some cases. I might have known Grace for a long time and believe that she has always gotten moral matters right in the past. Then, if you and I are having a moral dispute, you can cite Grace’s moral belief as some evidence against my moral view. Our shared moral assumptions about the truth of Grace’s past moral beliefs can make evidence out of our observations of Grace’s current moral beliefs. The best explanation of why she believes what she does might be that her beliefs are true once again. However, this cannot work against a moral nihilist for the simple reason that a moral nihilist would deny both that Grace’s positive moral beliefs were ever true and that she is now a reliable detector of moral facts. Consequently, even if Grace is reliable and moral facts are essential to the best explanation of Grace’s moral beliefs, Grace’s beliefs still cannot be evidence against a moral nihilist because of the independence requirement for evidence. The fourth pattern can be exemplified by the observations that almost everyone agrees that torturing babies just for fun is morally wrong, and modern societies are converging on the belief that slavery is morally wrong. The claim is not that so many people cannot possibly be mistaken. Instead, the argument is that the best explanation of why so many people agree is that they are reacting to a real moral wrong.17 The most common response is that such agreement or convergence can be explained at least as well by sociobiology or some other social science without any moral facts. The fact that such moral beliefs help society survive is enough to explain why we have evolved so that so many people hold those beliefs. Moral explanationists might respond that this sociobiological explanation does not compete with the moral explanation, because the belief that such acts are morally wrong helps society by dissuading people from doing such acts. Preventing these acts helps society because such acts usually cause harm to society or to its members. And that is just what

W AL TER S INN O T T -A RMSTRO NG ALTER INNO RMSTRONG 124 makes those acts morally wrong, or so it is assumed. If this substantive moral assumption is accepted by both parties in a dispute, then observations of moral agreement and convergence might be best explained by moral facts, and this can be cited as evidence for one of the disputing moral frameworks above another. The story is quite different with moral nihilism. As before, moral nihilists will insist that the explanation in terms of moral facts does compete with the sociobiological explanation, and nihilists will reject the moral assumption that acts are morally wrong when they usually reduce the survival value of society. Consequently, observations of moral agreement and convergence cannot be used as evidence against nihilism. These four patterns all cite moral facts to explain non-moral facts. Thomson recently admitted that explanations in this direction never provide evidence for moral beliefs.18 However, she went on to claim that moral evidence can be based on explanations in the other direction, that is, when non-moral facts explain moral truths. In her example, “Alice’s giving Bert a banana was her keeping her word when [she knew] it cost her a lot to do so and she could have got away with not doing so. That, it seems plausible to think, would explain the truth of ‘Alice’s giving Bert a banana was just’.”19 Harman responded, “Clearly, in a context of a dispute as to whether actions like Alice’s constitute justice, [the claim that the cited features of Alice’s act explain why it is just] is controversial and cannot show that there is evidence in favor of one moral framework rather than another.”20 In a different context, however, this kind of explanation might be evidence in favor of one moral framework over another. Suppose Carol and David agree that Alice’s act was just, and the dispute is about whose moral framework can best explain why it is just. Carol is a utilitarian, so David argues forcefully that Alice’s act does not maximize utility, since it will cost Alice a lot to give the banana to Bert, and nobody will be hurt if Alice does not give it, since Bert doesn’t need the banana anyway. If Carol admits that utilitarians cannot explain why Alice’s act is just, but she still believes that Alice’s act is just, then David can offer a better explanation of its justice, namely, Thomson’s. This explanation of a moral fact by non-moral facts can then provide evidence in favor of David’s deontological framework above Carol’s utilitarian framework. Of course, none of this works unless Carol believes that Alice’s act is just, so it cannot work against a utilitarian who sticks to her theory or against a moral nihilist. But that is the point. Moral explanations like this can work in some contexts against some opponents even if they cannot work in all contexts against all opponents.

E XPL ANA TI O N AND J USTIFIC ATI O N I N M OR AL E PISTEM O LOGY ANATI TIO USTIFICA TIO ORAL PISTEMO 125 The final question is: Does this refute moral skepticism? That depends on what moral skepticism claims. An extreme version of moral skepticism claims that there cannot ever be any evidence in favor of any moral belief or theory as opposed to any other. If I am right, moral explanations refute that extreme version of moral skepticism. In contrast, a more limited version of moral skepticism claims only that there cannot ever be any evidence in favor of any moral belief or theory as opposed to all others. One other moral theory is moral nihilism. I argued that moral explanations can never provide evidence against moral nihilism, so they can never provide any evidence in favor of any moral belief or theory as opposed to moral nihilism. If nothing else can provide such evidence either, as I have argued elsewhere, 21 then limited moral skepticism is true. Thus, the “no-explanation argument for moral skepticism” contains part of the truth, but only part. 22 And defenders of moral explanations have part of the truth, but only part. The only way to capture the whole truth is to distinguish different kinds of justification relative to different contrast classes and then to endorse limited moral skepticism. Walter Sinnott-Armstrong, Department of Philosophy, Dartmouth College, Hanover, NH 03755-3592; [email protected]

NOTES 1. Gilbert Harman, The Nature of Morality (New York: Oxford University Press, 1977), chap. 1. 2. In these debates, the term ‘fact’ is and will here be used in a minimalist sense in which it is a fact that p if and only if p. Moral beliefs can and will here be called ‘true’ in a similarly minimalist sense. 3. Gilbert Harman, “Responses to Critics,” Philosophy and Phenomenological Research 58.1 (March 1998): 207f. Compare Harman, op cit., 3; Gilbert Harman, “Moral Explanations of Natural Facts—Can Moral Claims be Tested Against Moral Reality?” The Southern Journal of Philosophy 24, supplement (1986): 60; and Gilbert Harman and Judith Jarvis Thomson, Moral Relativism and Moral Objectivity (Oxford: Blackwell, 1996), 78, 168ff. 4. Harman and Thomson, op. cit., 170. 5. In other words, the moral fact must be a necessary enabler of the explanation. I argue that all reasons are necessary enablers in “An Argument for Consequentialism,” Philosophical Perspectives 6 (1992): 399–421.

W AL TER S INN O T T -A RMSTRO NG ALTER INNO RMSTRONG 126 6. One could also distinguish attitudes from beliefs, as does Thomson in Harman and Thomson, op cit., 87ff, but that distinction would not affect my main points. 7. Nicholas Sturgeon, “Moral Explanations” in Morality, Reason, and Truth, ed. D. Copp and D. Zimmerman (Totowa, NJ: Rowman and Allanheld, 1984), 63. 8. Bernard DeVoto, The Year of Decision: 1846 (Boston: Houghton Mifflin, 1942), 442. 9. This objection might be suggested by Jonathan Dancy’s claims about “natural shapelessness” in Moral Reasons (Oxford: Blackwell, 1993), 76. But see my “Some Varieties of Particularism” (forthcoming in Metaphilosophy). 10. See Harman and Thomson, op. cit., 80, and Peter Railton, “Moral Explanation and Moral Objectivity” Philosophy and Phenomenological Research 58.1 (March 1998): 178–180. 11. Harman and Thomson, op. cit., 170. 12. Another common example of this pattern cites Hitler’s moral depravity as the best explanation of why he ordered the killings of millions of innocent people. However, this explanation assumes that Hitler’s acts were evil (Harman and Thompson, op cit., 86). This assumption would, of course, be accepted by almost everyone, including me, but it would be rejected by moral nihilists. So this example provides no more evidence against moral nihilism than do Woodworth’s acts. 13. This explanation is discussed by Nicholas Sturgeon, “Thomson Against Moral Explanations,” 205, and Judith Jarvis Thomson, “Reply to Critics,” 221, both in Philosophy and Phenomenological Research 58.1 (March 1998). 14. One example of the third pattern is someone who does not believe that slavery is morally wrong but converts to the belief that slavery is morally wrong as a result of reading an historical novel or watching a movie like Roots. By hypothesis, the conversion cannot be explained by any of this person’s former moral beliefs, so the best explanation is supposed to be that his new belief results from the real evil in slavery. See Richard Werner, “Moral Realism,” Ethics 93 (1983): 653–679. As before, this explanation would never be accepted by anyone who denies that slavery is evil, so it cannot be evidence against nihilism. The trickier example in the text comes from William Tolhurst, “Supervenience, Externalism, and Moral Knowledge,” The Southern Journal of Philosophy 24, Supplement (1986), 43–55, esp. 46. 15. Indeed, if the supervenience base of moral wrongness has no shape, we might not be able to formulate any non-moral explanation of Grace’s belief. See note 9. 16. Harman and Thomson, op. cit., 84. 17. This argument does not deduce ‘ought’ from ‘is’, because it is inductive. Still, the premises are non-moral (despite ‘best’ explanation being evaluative),

E XPL ANA TI O N AND J USTIFIC ATI O N I N M OR AL E PISTEM O LOGY ANATI TIO USTIFICA TIO ORAL PISTEMO 127 and the conclusion is moral (if it is not just about what we have evidence to believe), so this argument still might seem like pulling a rabbit out of a hat. Anyway, even if this argument works, it is not much like evidence in science. 18. Harman and Thomson, op. cit., 91. 19. Ibid., 93. 20. Ibid., 170f. 21. See my “Moral Skepticism and Justification” in Moral Knowledge?, ed. Walter Sinnott-Armstrong and Mark Timmons (New York: Oxford University Press, 1996), 3–48; and “Nihilism and Skepticism about Moral Obligations,” Utilitas 7 (1995): 217–236. 22. Harman and Thomson, op. cit., 75.

THE VALUE OF MORAL RESPONSIBILITY

Moral responsibility requires control of one’s behavior. But there are different kinds of control. One sort of control entails the existence of genuinely accessible alternative possibilities. I call this regulative control. I believe that an agent can control his or her behavior without

John Martin Fischer

having control over it. In such a circumstance, the agent enjoys what I call guidance control, but not regulative

control.

He

guides

his

behavior in the way characteristic of agents who act freely, yet he does not have alternative possibilities with respect to his decision or action. I contend that moral responsibility requires guidance control, but not regulative control. In this paper, I wish to provide a measure of intuitive appeal to the claim that guidance control is all the control (or freedom) necessary for moral responsibility by sketching the picture that supports this claim.

M

oral responsibility requires (among other things) control of one’s behavior. But there are different kinds of control.1 One sort of control entails the existence of genuinely accessible alternative possibilities; I call this sort of control, “regulative control.” The presence of regulative control is typically signalled by the use of the preposition, “over.” So, when an individual has control over his behavior, he has more than one path available to him; he (say) performs an action, but he could have done otherwise (in the sense of “could” that expresses the distinctive sort of ability involved in free will).

129

J OHN MARTIN FISCHER 130 I believe that an agent can control his behavior, and be in control of it, without having control over it. In such a circumstance, the agent has what I call “guidance control,” but not regulative control. He guides his behavior in the way characteristic of agents who act freely, and yet he does not have alternative possibilities with respect to his decision or action. Of course, an agent may have both sorts of control—regulative control and guidance control. But the fact that an agent can have guidance control without regulative control shows that they are distinct forms of control. I contend that moral responsibility requires guidance control, but not regulative control. That is, guidance control exhausts the “freedomrelevant” (as opposed to the epistemic) component of moral responsibility. In this paper I wish to provide a measure of intuitive appeal to the claim that guidance control is all the control (or freedom) necessary for moral responsibility. I shall propose what I take to be the intuitive “picture” that drives the view that alternative possibilities are required for moral responsibility. Elsewhere I have argued against the regulative control model. Here I shall develop an alternative picture which I believe both explains, at an intuitive level, what is going on with behavior for which an agent is morally responsible, and also helps to explain exactly why guidance control is all the control required for moral responsibility.

REGULATIVE CONTROL: MAKING A DIFFERENCE In my view there is an intuitive picture that drives all proponents of regulative control, no matter what sort of alternative possibility they identify as grounding ascriptions of moral responsibility. The idea is that moral responsibility requires making a difference. Slightly more carefully, an agent is morally responsible for his behavior only if he makes a difference to the world in so behaving. But of course an agent can in some sense make a difference when performing an action under coercion, duress, or (say) direct electronic stimulation of the brain. Given that the agent acts in such cases, the world is different than it would have been, had he not so acted. Obviously, this is not the sort of difference envisaged by the proponent of the regulative control model of moral responsibility. Rather, the regulative control theorist believes that moral responsibility requires the ability to make a difference in the sense of selecting one from various paths the world could take, where these various paths are all genuinely available to the agent. The basic idea here is selection from

T H E VAL UE OF M OR AL R ESPO NSIBI LIT Y ALUE ORAL ESPONSIBI NSIBILIT LITY 131 among options that are really accessible to the agent. When one selects from a set of feasible options, one makes a difference: the world goes one way rather than another, or takes one path rather than another, among various paths the agent can cause the world to take. This, I believe, is the basic intuitive idea behind the regulative control model. There can be cases in which an agent deliberates, chooses, and acts freely, on whatever your favorite account of such things is, and yet because of the presence of a fail-safe device which does not play any actual role in the agent’s deliberation or behavior, the agent has no alternative possibilities with respect to choice or action. The fail-safe device does not actually intervene, but would intervene under certain counter-factual circumstances to produce exactly the same sort of choice and action as actually take place. Following recent tradition, I shall call such cases “Frankfurt-type” cases.2 I shall now present a very sketchy version of a Frankfurt-type case. Suppose Jones is in a voting booth deliberating about whether to vote for Clinton or Dole. (He has left this decision until the end, much as some restaurant patrons wait until the waiter asks before making a final decision about their meal.) After serious reflection, he chooses to vote for Clinton, and does vote for Clinton by marking his ballot in the normal way. Unbeknownst to him, Black, a liberal neurosurgeon working with the democratic party, has implanted a device in Jones’ brain which monitors Jones’ brain activities. If he is about to choose to vote democratic, the device simply continues monitoring and does not intervene in the process in any way. If, however, Jones is about to choose to vote (say) republican, the device triggers an intervention which involves electronic stimulation of the brain sufficient to produce a choice to vote for the democrat (and a subsequent democratic vote). Given that the device plays no role in Jones’ deliberations and act of voting, it seems to me that Jones acts freely and is morally responsible for voting for Clinton. And given the presence of Black’s device, it is plausible to think that Jones does not have alternative possibilities with regard to his choice and action. So it appears that Jones is morally responsible for his choice and for voting for Clinton, although he lacks regulative control over his choice and action. This Frankfurt-type case seems to show that regulative control is not required for moral responsibility. Although various objections can be made to this sort of case, I believe that the objections can be answered; I believe that versions of the Frankfurt-type case can be constructed which show to a high degree of plausibility that moral responsibility does not require alternative possibilities, although this will not be the project of this paper. Our moral responsibility is not then based on our capacity to make a difference

J OHN MARTIN FISCHER 132 to the world. I want now to seek to sketch (in what will no doubt be a preliminary way) a different intuitive basis of moral responsibility. Begin with an analogy with artistic creativity. Suppose a sculptor creates a sculpture in the “normal” way—the sculptor is not being manipulated, coerced, and so forth, and is driven by his own creativity. But imagine further that, if he hadn’t created this sculpture, some other artist would have created exactly the same sort of sculpture—a different particular sculpture that is nonetheless molecule-for-molecule isomorphic to the sculpture actually produced by the artist. I am not sure why exactly the production of this sort of sculpture is overdetermined in this way, but it really doesn’t matter exactly why this is so—only that it is so. It may be, for example, that a friend of the artist has discussed the sculpture with him, and is bent on producing it, if the artist doesn’t do so himself. There is a pretty clear sense in which the artist does not make a difference to the world in creating the sculpture. He does not make a difference defined in terms of end-states (individuated in a natural, broad way). The very same kind of sculpture would have been produced, had he not created the sculpture himself. And yet there is also a clear sense in which the artist’s creative activity has value. I suggest that we value the artist’s activity not because he makes a certain sort of difference to the world, but because he expresses himself in a certain way. He does not make a difference; but he does make a statement. My idea is that we can understand the intuitive picture behind moral responsibility in a similar way. When an agent exhibits guidance control and is thus morally responsible for his behavior, he need not be understood to be making a difference to the world; or better, it is unattractive to think that the explanation of his moral responsibility—the intuitive reason why we hold him morally responsible—is that he makes a difference to the world. Rather, the suggestion is that the individual is morally responsible, when he exhibits guidance control, insofar as he expresses himself in a certain way. To a first approximation, the “value” of morally responsible action is understood as analogous to the value of artistic self-expression. But if the value of morally responsible action is self-expression, what exactly is expressed? This is a question that deserves an answer, and yet it is perhaps not as easy as one might have supposed to answer it. Consider, for example, the following passage from Sarah Broadie’s book, Ethics With Aristotle: In voluntary action we pursue an objective which is before us and which figures as a good to us so far as we pursue it; but on another level we enact

T H E VAL UE OF M OR AL R ESPO NSIBI LIT Y ALUE ORAL ESPONSIBI NSIBILIT LITY 133 by our action, and thereby propound into public space, a conception of the kind of practical being that it is good (or at least all right) to be: a kind typified by pursuit of this kind of goal in this sort of way under such conditions. 3

The problem with the sort of view suggested by Broadie, in my view, is the very real phenomenon of weakness of will. That is, one sometimes freely does what one does not believe is good, or rational, or even all right. So, in voluntarily and freely performing some act, it would not in general be accurate to take one to be saying that the relevant goal is good or even all right, or that it is good or all right to be the kind of practical being that typically pursues this kind of goal. I suppose it could be urged that one is at least saying that the goal in question is to some degree good, and so one could be taken to be expressing the idea that it is at least to some degree good or defensible to be the sort of practical being who pursues such a goal. But even this seems implausible, as one can presumably freely do something one does not find to any degree good or morally defensible. Now perhaps it will be replied that whenever one acts, one must have some sort of pro-attitude toward the behavior in question (or the goal it is taken to promote). I agree, but it is somewhat disappointing to be told that the message conveyed by the agent in acting is something we know, as a conceptual point, from the mere fact that the agent has performed an action. Whereas I do not wish to deny that one can find here part of the message of action, I believe that it will be more illuminating to seek an alternative account of what is expressed by the agent in acting. To develop such an account, I begin by suggesting that our morally responsible behavior can be understood not merely to be analogous with artistic self-expression; perhaps it can usefully be understood as a kind of artistic self-expression. Various philosophers have developed the suggestion that our lives have “narrative structures”—that our lives are in some sense stories. 4 This is an intriguing and suggestive idea (even if it is difficult to flesh out precisely), and different philosophers have developed it in different ways (and for different purposes). Here I shall rely on David Velleman’s presentation of the idea in his seminal paper, “Well-Being and Time.” 5 Velleman is concerned to argue that “well-being is not additive.” This claim involves various ideas. One is that we cannot simply add up the welfare values of segments of an individual’s life to get a total value that accurately reflects our judgments about the value of the individual’s life as a whole. Another idea is that the welfare values of the segments depend crucially on their “narrative” or “dramatic” relationships with other parts of the life.

J OHN MARTIN FISCHER 134 Velleman says: Consider two different lives that you might live. One life begins in the depths but takes an upward trend: a childhood of deprivation, a troubled youth, struggles and setbacks in early adulthood, followed finally by success and satisfaction in middle age and a peaceful retirement. Another life begins at the heights but slides downhill: a blissful childhood and youth, precocious triumphs and rewards in early adulthood, followed by a mid-life strewn with disasters that lead to misery in old age. Surely, we can imagine two such lives as containing equal sums of momentary well-being . . . Yet even if we were to map each moment in one life onto a moment of equal well-being in the other, we would not have shown these lives to be equally good. For . . . one is a story of improvement while the other is a story of deterioration . . . the former story would seem like a better lifestory—not, of course, in the sense that it makes for a better story in the telling or the hearing, but rather in the sense that it is the story of a better life.6

Now it might be thought that the moral of Velleman’s story is that we have a general tendency to weight welfare that occurs later in life more heavily. But whereas this would issue in a nonadditive conception of welfare, it is not the moral Velleman wishes to draw. Rather, Velleman says: The reason why later benefits are thought to have a greater impact on the value of one’s life is not that greater weight is attached to what comes later. Rather, it is that later events are thought to alter the meaning of earlier events, thereby altering their contribution to the value of one’s life. [Additionally] . . . [t]he meaning of a benefit depends not only on whether it follows or precedes hardships but also on the specific narrative relation between the goods and evils involved. 7

To illustrate this point, Velleman gives the example of the importance of drawing lessons from one’s misfortunes. We typically think it important to learn from life’s tragedies; the fact that we have been improved as a result of going through a tragic experience adds to the total value of our lives in a distinctive way. As Velleman puts it: If a life’s value were a sum of momentary well-being, learning from a misfortune would be no more important than learning from other sources, since every lesson learned would add so much value and no more to the sum of one’s well-being. On being invited to learn from a personal tragedy, one would therefore be entitled to reply, ‘No, I think I’ll read a book instead’. Edification would offset the losses incurred in the tragedy, but its having been derived from the tragedy would not render edification more valuable . . .8

T H E VAL UE OF M OR AL R ESPO NSIBI LIT Y ALUE ORAL ESPONSIBI NSIBILIT LITY 135 Velleman similarly asks us to consider other two lives. In the first life you have ten years of unhappiness and trouble in a marriage followed by divorce, after which you remarry happily. In the second life the ten years of unhappiness in marriage lead to eventual happiness as the relationship matures.9 About this example, Velleman says: Both lives contain ten years of marital strife followed by contentment; but let us suppose that in the former, you regard your first ten years of marriage as a dead loss, whereas in the latter you regard them as the foundation of your happiness. The bad times are just as bad in both lives, but in one they are cast off and in the other they are redeemed. Surely, these two decades can affect the value of your life differently, even if you are equally well off at each moment of their duration.10

I shall follow Velleman in contending that life has a narrative structure in the specific sense that the meanings and values of the parts of our lives are affected by their narrative relationships with other parts of our lives, and the welfare values of our lives as a whole are not simply additive functions of the values of the parts. In this sense, then, our lives are stories.11 And in performing an action at a given time, we can be understood as writing a sentence in the book of our lives. I suggested above that the distinctive value in acting in such a way as to be held morally responsible lies in a certain sort of self-expression. But the question then arose as to what precisely is expressed by ordinary actions. My answer is that it is not most fruitful to look for a “message” of action of the sort suggested by Broadie—that one believes that the action promotes a defensible goal. Rather, what is expressed by an agent in acting is the meaning of the sentence in the book of his life. And this meaning is fixed in part by relationships to other sentences in this book, i.e., by the overall narrative structure of the life. In acting, an individual need not be “propounding into public space” any sort of vision of the good or defensible life. Rather, his action writes part of the book of his life, and gets its meaning from its place in this story. This suggests a more “holistic” picture of what gets expressed by an agent in acting so as to be morally responsible, and one which is more illuminating than the mere fact that the agent had a pro-attitude toward moving his body in a certain way in the context. I have claimed that the relationship between action and artistic self-expression is not mere analogy. But the idea that when one exhibits guidance control one can be understood to be engaging in a kind of artistic self-expression (writing part of the book of one’s life) faces the following pressing worry. The characteristic modes of evaluation of a piece of art are aesthetic, whereas the central modes of evaluation of a

J OHN MARTIN FISCHER 136 human life do not seem to be aesthetic; arguably, moral and prudential modes of evaluation are the distinctive ways of assessing a human life. For example, the life of a professor at a research university may be lived with the normal attention to duties to others and also with a zest for intellectual development, as well as a healthy preoccupation with self-advancement of various sorts. The professor may well show more than the ordinary concern for morality. Let us suppose that in her professional life she develops various connections between her research and public policy questions, always with an emphasis on seeking to promote the general welfare, and, in particular, the interests of those who are less fortunate. And this moral concern is also a prominent feature of the professor’s personal life; for example, once a week she makes a dinner and brings it to the local soup kitchen. She is always honest and sensitive to the needs of others. The professor’s life can of course be judged by various different criteria. Morally speaking, it seems to be exemplary. It may do reasonably well on prudential grounds, despite the notorious lack of remuneration in the academic profession. But it could be relentlessly boring from an aesthetic point of view. It may be that the story of the professor’s life would make a very boring novel, as judged by aesthetic considerations. It seems that the ways in which we typically judge novels are fundamentally different from the ways in which we judge lives. Of course we might well apply moral criteria to the author of a story. It may be that the author has modelled his characters after reallife characters in a way that is morally problematic, or that the content of the story is likely to increase the probability of violent or exploitative behavior, and so forth. We can assess the morality—or the rationality— of the behavior of the characters in the story, perhaps treating it as part of a hypothetical example. And, similarly, we can apply aesthetic categories to a human life, judging it exciting, beautiful, interesting, and so forth. But even so, it seems that there is a fundamental difference between (say) novels and human lives with respect to their “characteristic” or most central modes of evaluation: novels are evaluated aesthetically whereas human lives are evaluated morally and prudentially. If something like this is true, then in exhibiting guidance control an agent is not simply writing a story; he is writing a story whose distinctive modes of evaluation are moral and prudential. Perhaps a bit more can be said about this “central” or “distinctive” mode of evaluation of a human life. When an author depicts (for example) bad or wrong behavior, the author may be doing something wrong in so presenting the characters, but he is not acting wrongly simply by presenting the characters as acting wrongly. Similarly, an author is not acting in

T H E VAL UE OF M OR AL R ESPO NSIBI LIT Y ALUE ORAL ESPONSIBI NSIBILIT LITY 137 the morally correct way simply in virtue of depicting characters who act in morally exemplary ways. Of course, it is a tautology that in acting wrongly a human agent is thereby acting wrongly; but in depicting wrong behavior, a novelist is not eo ipso behaving wrongly. Thus, there is an important asymmetry between the context of (say) writing a novel and the context of morally responsible action which helps to explain my contention that the moral (and prudential) modes of evaluation are somehow central and distinctive ways of evaluating morally responsible action. To summarize, then, the value of morally responsible action, on the guidance control model, is self-expression. More specifically, when an individual exhibits guidance control and is thus morally responsible for his behavior, he is expressing himself in a certain way: he is writing part of a book which is accessible, in an important and distinctive way, to moral and prudential evaluation. Although the morally responsible agent need not make a difference to the world, he is making a certain kind of statement. I have tried to give an account of what might be called the value we place on acting so as to be morally responsible. In so doing, I have been seeking to sketch what I have suggested is the “picture” which grounds the guidance-control model of moral responsibility—the view that guidance control, and not regulative control, is the freedom required for moral responsibility. Whereas I believe I have put some of the elements of this picture in place, I still don’t think I have fully captured the value of acting so as to be morally responsible. To explain. Consider someone who is, intuitively, not morally responsible for what he does because he is to a significant degree subject to coercion, manipulation, and pressures that render his behavior not suitably responsive to reasons. Such an individual may nevertheless express himself in the relevant way: he may write the story of his life, a story to which certain moral and prudential judgments can attach. That is, presumably we can evaluate this individual’s behavior in such a way as to judge it as right or wrong, good, or bad, and so forth. Now of course we need to distinguish these normative judgments from the further normative judgments and attitudes constitutive of moral responsibility: the reactive attitudes (such as indignation, resentment, gratitude, respect, and so forth). The sort of individual in question can live a life that is legitimately judged in terms of the first kind of normative considerations, but not the second. Why not simply specify that the picture that grounds moral responsibility requires that the individual’s life be subject to normative judgments of the second kind, i.e., those involved in the reactive attitudes? Perhaps one could do this, but I feel uncomfortable doing so because it seems to introduce a troubling circularity. I am trying to identify what exactly we value in cases in which we behave so as to be morally responsible. If I were

J OHN MARTIN FISCHER 138 to say, “Well, what we value is acting responsibly,” this would obviously be circular and uninteresting—although perhaps true! I believe that it is similarly circular and unhelpful to say that the value of acting so as to be responsible is cashed out in part in terms of the value of acting so as to be accessible to the reactive attitudes. A more promising approach is to note that when one is subject to coercive pressures, manipulation, and so forth, one’s self-expression is hindered in certain ways. What one wants to say, I believe, is that the value of acting so as to be morally responsible consists in unhindered or unimpaired self-expression of the relevant sort. Perhaps another way of saying the same thing is to note that when one engages in unhindered or unimpaired self-expression of the relevant kind, one is freely expressing oneself. I will then suggest that the value of acting so as to be morally responsible consists in one’s freely expressing oneself artistically in a way that is accessible in a distinctive way to moral and prudential evaluation. We value freely—in the sense of not being hindered or impaired in certain ways— writing the book of our life. Now someone will say that I have introduced a problematic and contested notion here—the notion of “freely” expressing oneself. I have avoided the circularity mentioned above only by introducing an essentially contested notion: out of the frying pan and into the fire! After all, the proponent of the regulative control model will insist that when one freely does anything, one must have genuinely accessible alternative possibilities. I grant this point. I certainly cannot present any decisive arguments that the ordinary notion of “acting freely” does not require alternative possibilities; indeed, this debate will presumably simply re-inscribe the debate about whether moral responsibility requires regulative control. So I will simply stipulate a special notion of “acting freely,” call it “acting freely*.” When one “acts freely*” one need not have any alternative possibilities. The intuitive idea of acting freely* is that in the actual sequence that leads to one’s behavior, no freedom-undermining factors operate or play a role. So, in the Frankfurt-type cases, one uncontroversially is acting freely*, even if it is controversial whether one is acting freely. Put in other words, in the Frankfurt-type cases a proponent of alternative possibilities as a condition for moral responsibility may say that insofar as acting freely is sufficient for moral responsibility, one of course needs alternative possibilities to act freely. But he should be willing to concede that there is some “actual-sequence” notion of freedom which the agent possesses, acting freely*; the agent possesses this freedom insofar as no freedomundermining factor operates in the actual sequence that issues in the behavior. Of course, the proponent of alternative possibilities will go on to insist that such freedom is not sufficient for moral responsibility, and in

T H E VAL UE OF M OR AL R ESPO NSIBI LIT Y ALUE ORAL ESPONSIBI NSIBILIT LITY 139 fact the agents in the Frankfurt-type cases also possess a more robust (from their point of view) kind of freedom—one involving alternative possibilities. For my purposes, I simply want to crystallize out the “actual-sequence” notion of freedom, acting freely*. So I can now state my account of the value of acting so as to be morally responsible a bit more perspicuously. My contention is that the value of acting in such a way as to be morally responsible consists in freely* expressing oneself artistically in a way that is accessible (in a distinctive fashion) to moral and prudential evaluation. Although this account needs to be filled in various ways, I think that it captures something simple and important: the value of moral responsibility, on the guidance control model, consists in a distinctive kind of self-expression.12 John Martin Fischer, Department of Philosophy, University of California-Riverside, Riverside, CA 92501-0201; [email protected]

NOTES 1. For the distinction between the two kinds of control, the claim that guidance control is sufficient for moral responsibility, and an elaboration of the notion of guidance control, see John Martin Fischer, The Metaphysics of Free Will: An Essay on Control, Vol. 14 of the Aristotelian Society Monograph Series (Oxford: Blackwell Publishers, 1994); and John Martin Fischer and Mark Ravizza, Responsibility and Control: A Theory of Moral Responsibility, Cambridge Series in Philosophy and Law (Cambridge: Cambridge University Press, 1998). 2. The classic source is, Harry Frankfurt, “Alternate Possibilities and Moral Responsibility,” Journal of Philosophy 66 (1969): 829-839. 3. Sarah Broadie, Ethics With Aristotle (Oxford: Oxford University Press, 1991), 159. 4. See, for example, Alasdair MacIntyre, After Virtue (Notre Dame: University of Notre Dame Press, 1981); Charles Taylor, Sources of the Self: The Making of Modern Identity (Cambridge, Mass.: Harvard University Press, 1989); and Alexander Nehamas, Nietzsche: Life as Literature (Cambridge, Mass.: Harvard University Press, 1985). 5. J. David Velleman, “Well-Being and Time,” Pacific Philosophical Quarterly 72 (1991): 48-77; reprinted in The Metaphysics of Death, Stanford Series in Philosophy, ed. John Martin Fischer (Stanford: Stanford University Press, 1993), 329-357. (All subsequent pages references will be to the reprinted paper.) 6. Velleman, op. cit., 331.

J OHN MARTIN FISCHER 140 7. Ibid., 334ff. 8. Ibid., 336. 9. Ibid., 337. 10. Ibid. 11. As Velleman points out, this view should not be confused with the view sometimes attributed to Nietzsche that literary or aesthetic considerations determine the value of a life. (Nehamas, op. cit.) 12. A revised and significantly expanded version of this paper, “Responsibility and Self-Expression,” appears in The Journal of Ethics (1999).

PHILOSOPHY AND ECOLOGY

There was a time when ecological problems were of no interest to philosophy. Now, these issues have raised philosophical problems in several areas. In moral philosophy, one question is what moral obligations, if

John Passmore

any, we have to future generations, and another is how far we have moral obligations relating to the treatment and the preservation of plants, animals and atmospheres. In political philosophy, the issue is the range of such concepts as rights and justice, and whether or not they are limited to human relationships. As to the metaphysical question, we have to ask whether there is something about human beings which entitles us to consider them as being supernatural and whether we can think of Nature as an entity of which each human being constitutes a part.

C

onsulting the 1971 supplement to the Oxford English Dictionary one finds environmentalism defined as the view that “human beings are formed by their environment rather than by heredity.” Consulting the 1979 edition of an American dictionary (Scribners) one finds that ecology is “that branch of science dealing with the interactions of living organisms with one another and their non-living environment.” In the Oxford example ‘environment’ refers to the ‘social’ as well as, perhaps rather than, the ‘natural’ environment. How, we might ask, can philosophers throw any light on such topics? We might think of Montesquieu who, in 1750, developed a thesis about the effect of climate on human beings and human societies. But that was before

141

J OHN PASSMORE 142 there was a sharp distinction between being a philosopher and being a scientist. (Think of Descartes on physics and of Joseph Priestley with his fundamental contributions to chemistry.) Should we not nowadays leave environmental influences to sociologists and ecology to botanists? But since the nineteen seventies the situation has greatly changed. Environmentalism has become the name of a socio-political movement intent on saving from destruction the natural surroundings—plants, trees, animals, rocks—of human beings. Ecology now considers the interplay between human beings and various forms of life. But ‘what,’ we might still ask, ‘has this to do with philosophy?’ Many philosophers would say: ‘nothing at all’. Let us leave environmental issues to politicians, economists, physicists and professional agitators. Why not? Three personal anecdotes illustrate this attitude. In 1972 I was in Paris. A philosopher whom I met there asked me what I was working on. When I replied that I was exploring environmental issues he responded with: “No French philosopher would dream of considering such questions.” The second anecdote relates to a review of my Man’s Responsibility for Nature, published two years later in 1974. The reviews were not in philosophical journals but in high-grade Sunday newspapers and similar journals. The authors were not philosophers but well-known figures in British cultural life. In the New Statesman Donald MacRae wrote thus: “Professor Passmore is a philosopher. For fifty years no one has expected philosophers to be useful . . . But here is a philosopher who gives us the best net yet.” A similar view was taken by C. P. Snow in the Sunday Times. MacRae’s ‘fifty years’ is in England just about right, if we rule out the Austrian Karl Popper. In the United States Dewey continued to write on social issues as did John Anderson in Australia and indeed myself. But no such publications appear under the names of Ryle, Davidson, Dummett, or Quine. John Stuart Mill, in contrast, had considerable influence not only in England but in Australia, where he was cited in parliaments. On the Idealist side T. H. Green had a very considerable socio-political influence. But all this was, by 1974, more than fifty years ago. What about Bertrand Russell? He wrote, in books, about marriage, education, pacifism, books easier to find in general bookshops than even the most popular of what would now be called his philosophical writings, The Problems of Philosophy. Russell, however, did not regard himself as in these volumes doing philosophy. He wrote them, he says, as a father, a husband, a teacher, a pacifist, not as a philosopher. It was Russell, indeed, who defines philosophy proper as including only logic, epistemology, and ontology, with a doubtful side reference to ethics. It was in the light of this, let us say, ‘pure’ concept of philosophy that some regarded me as having abandoned philosophy when I wrote my environmentalist book. (This is

P HILOSOPHY

AND

E COLOGY 143

the third personal anecdote.) I had kept within Russell’s limits when I wrote my A Hundred Years of Philosophy but that was because I wanted to confine myself to an interwoven set of controversies. Should we think of ourselves as reactionaries, those philosophers who write about the environment? There are certainly philosophers who would argue that although philosophers in the past wrote about sociopolitical issues that was before the emergence of the social sciences. That emergence is why environmental issues can be left to economists and political theorists along with scientists and professional agitators. Philosophers, as such, have nothing to add. There is only one way of countering this view, by considering what counts as being philosophy. For many centuries, of course, philosophy was a general name for almost every kind of organized intellectual activity. Call this Concept One. There are still relics of it in those Universities who use the name ‘natural philosophy’ for physics. But few, if any, philosophers would deny that there are ecological issues which fall outside philosophy in the area we call ‘science’. Thus, for example, philosophers are in no position to contribute to the question whether recent climate changes are the product of industrial emissions or of an ice-age cycle. The most that they claim is that they could contribute to a discussion on whether, granted this uncertainty and remembering that we cannot control the cycle but can control the emissions, we should make the economic sacrifices involved in emission control. A second concept of philosophy is what can be called Russellian purity. He never, so far as I know, had anything to say about environmental issues, which, as I said, did not come to the forefront until the nineteenseventies. Russell died in 1970. But it is clear enough that he would have had something to say about these issues had he lived longer, as he did in relation to all the major socio-political issues of his time. But at the same time, although he might have approved of philosophers taking part, as critical citizens, in these disputes he would not have granted that what they were writing formed part of philosophy. There is, however, a third concept of philosophy which sees it, as in the case of ‘science’, as the name for a wide variety of investigations, not as in the case of ‘biology’, the name of a limited set of closely related investigations, all of them related only to living things. In the case of philosophy, it can be argued, there is a set of what we often call ‘philosophies of’, as in the case of the ‘philosophy of history’ or ‘of science’, in other cases, with an adjective, as in ‘political philosophy’. Environmental philosophy, in my view, is simply one member of that class. Some would rather say that environmental philosophy is a form of ‘applied philosophy’. The concept of ‘applied philosophy’ is, so far as I know, a quite modern one. Certainly the journal of that name dates back

J OHN PASSMORE 144 only to 1983. It is now a quite popular description and has the economic virtue of providing jobs for philosophers in a wide variety of occupations— in, for example, management. Its principal concern is with moral issues; the article I wrote for the first number was on the moral responsibilities of university teachers, an article which no one has since taken up. We are used to the phrase ‘applied mathematics’. It applies mathematical principles or methods to the solution of practical problems in, let us say, engineering. But ‘applied philosophy’ is in a very different situation. For, unlike mathematics, there are no universally accepted philosophical principles. Collingwood attacked the idea of applied philosophy before the phrase was introduced. He contributed a great deal to what I have called ‘philosophies of’, in particular to the philosophy of history and the philosophy of art. ‘Philosophy’, he nevertheless warned, ‘cannot descend like a deus ex machina upon the stage of political life and, out of its superior insight into the nature of things, dictate the solution to this and that problem in morals, economic organization or international politics.’ The extreme example of this is Plato’s Republic. Russell is deliberately separating himself from it. From a different standpoint the British moral philosopher Richard Hare complains that when applied philosophers talk about applying ethics to practical problems, what they appeal to is not the ethics of philosophers but what has now come to be called ‘folk ethics’, everyday moral principles. So that they have no right to say that they are engaged in applied philosophy. Altogether it looks more sensible to say that what is called ‘applied philosophy’ is better described, if it has any value, as ‘philosophy of practical affairs’, taking this to mean that it uses the typical philosophical method of free discussion and critical examination to solve a particular group of problems. (Of course, it can also make use, as philosophers often do, of the historical background of the problem.) So what is called ‘business ethics’ is a critical examination of those business practices which raise moral problems. ‘Philosophies of’ sometimes issue in a criticism of what are thought of as leading principles in pure philosophy. So working on the philosophy of history can lead us to re-examine the generally received conceptions of explanation or objectivity, or even of truth. Working on the philosophy of art can lead us to think more closely about the idea of truth or of morality. So far as I know, business ethics does not involve such re-examination but environmental philosophy certainly does. Let us begin with classical Utilitarianism and its formula that we, or more plausibly, our governments, should act in such a way as to achieve the greatest happiness for the greatest number. That ‘greatest number’ is usually taken to include only living human beings. From a

P HILOSOPHY

AND

E COLOGY 145

different ethical standpoint Kant had written that ‘human nature as such cannot be indifferent to the most remote epoch which may eventually affect our species’ to which he, however, added that this is so ‘only when we can describe future generations with great certainty’. The Utilitarian, Sidgwick, a Kantian reader, writing in 1874, converted this into ‘the interests of posterity must concern a utilitarian’, but adds that it is difficult to determine what future generations will need in order to be happy. It was with the rise of environmentalism, however, that concern for future generations, going far beyond local family concerns, came to the fore. By 1980 the editor Ernest Partridge could publish a collection of papers on this topic, from the United States, Canada, Great Britain, and Australia. It was obvious, after all, that future generations would need food and uncontaminated water, recreation, fresh air, heating, and so on. Critics can rightly say that we cannot, by our actions, ensure the happiness of human beings—that the conditions laid down by Kant and Sidgwick can never be fulfilled and that this is true even at the environmental level. Climate changes cannot be managed, new diseases can always arise, crucial social changes can have a big influence. But at least there are permanent human necessities. In its classical form Utilitarianism concerns itself only with the happiness of human beings as did other ethical doctrines. In the doctrines developed by Aquinas, human beings have no responsibility towards animals. We can treat them as we wish, except that: we should not damage animals owned by our neighbours. This was still Kant’s view. In my Man’s Responsibility for Nature I tell the story of the changes which gradually took place in this respect. But I did not then anticipate the extension of moral responsibilities beyond animals to other forms of life, although I made some reference to Indian and Greek views on this matter. Admittedly, we sometimes say of a flower, a tree, an animal, that it is unhappy. In the case of a pet we may feel a responsibility to reduce or get rid of its unhappiness if we have the power to do so. We may also give water or food or shelter to a plant we describe as being unhappy. But this is for our sake, because we do not get any kind of pleasure out of looking at the ‘unhappy’ plant. If the unhappiness results from, let us say, factory emissions we can seek to have them reduced. But that, again, would seem to be for our sake rather than the plant’s. In the case of what we call ‘weeds’, we would certainly not hesitate to pull them out. Equally we do not feel any obligation to a mosquito which obviously needs our blood to make its life easier; indeed we would not hesitate to kill it as what we call a ‘pest’. Both ‘weed’ and ‘pest’ are not, of course, objective descriptions of the plant or the insect. They incorporate human attitudes, which we use in order to justify our actions in destroying them.

J OHN PASSMORE 146 This might be the time to distinguish two sorts of environmentalism. The first sort I shall call human-centred, the second nature-centred. Human-centred environmentalism is concerned with preserving, or improving, certain sorts of environment for the sake of the human beings who inhabit them. If one is asked to give examples of its success one can refer to the cleansing of air in London, getting rid of dangerous smoke-laden fogs by making coal fires illegal. Of course this cleaning up could help the growth of trees and flowers but that is not why these steps were taken and in any case, in London, they would have been planted for the enjoyment of human beings I should make it plain that human-centred environmentalism is not committed to a human-centred metaphysics. It does not have to support the view that human beings uniquely contain souls; it can be materialistic, naturalistic. Such environmentalists, however, see the preservation of animals, trees, plants, landscapes as being in the interest of human beings. This need not be on economic grounds, with an eye simply, for example, on the tourist trade. Indeed environmentalism may quite consistently be opposed to such a trade, as being damaging to the environment. It can come into conflict, too, with farmers or foresters under certain circumstances. It can be romantic, following the nineteenth-century British poets and landscape painters in their admiration for beauty. In England nature was largely soft and humanized but in Australia painters finally managed to make us alert to the beauty of landscapes which we would earlier have seen as wholly hostile, needing to be humanized, if that was possible. National parks were set up to preserve and display animals, trees, and plants which might otherwise not be seen as cities grew. None of this is in conflict with familiar ethical or metaphysical theory. The situation is different for what I have called nature-centred environmentalism. Darwinism was a major step towards the reduction of distinctions between human beings and other animals. But it was used by such philosophers as Spencer to support capitalism in its most virulent form. Life is everywhere, he said, struggle for survival and, beyond that, for domination. Human beings should take a lesson from this. So we had the emergence of what was called social Darwinism. Subsequent zoologists have tried to reduce the difference between human beings and other animals in just about every possible way, particularly since the discovery of a special group of chimpanzees who, it is said, are not only tool-users—once supposed to be a form of action peculiar to human beings—but are capable of understanding sentences. They do not appear, to be sure, to be producing science, art, or philosophy. But only a small percentage of human beings make contributions in these areas.

P HILOSOPHY

AND

E COLOGY 147

Some environmentalists, Richard Routley for example, make use of such facts, and along with this the existence within the human species of idiots, the insane and small children as forbidding us to define human beings as the rational animals. That view often issues in a metaphysics and an ethics which are anything but conventional. If it retains utilitarianism, it greatly modifies it by demanding that our actions take account of the happiness not only of living human beings, not only of unnamed future human beings, but also of the happiness of other living things. But in accordance with many other modern philosophical developments it prefers to employ an ethics in which the central concepts are rights and justice rather than happiness. I suggested earlier that it was not a revolutionary view that plants, trees, animals can be unhappy, but it was a revolutionary suggestion that a utilitarian calculus had to include their happiness in our decisions about what should be done. That calls upon a great change in a philosophical view, under environmental pressure. But it is a much greater innovation to ascribe rights to animals, let alone to trees. Legislation against, say, cruelty to animals limited what had previously been regarded as the rights of human beings to treat animals in any way they chose. But now, as in the case of Peter Singer, one hears talk of the rights of animals, just as one earlier heard about the rights of women or of indigenous peoples. But in these latter cases to have rights was to be in a position to make certain claims. And to claim rights was to demand that, like others, one should be in the position, to be able, let us say, to vote in elections, to be members of governing bodies, to undertake particular kinds of work, to own land and so on. True enough there are special cases, as Routley emphasizes, where the human being is not able to make such claims. That is true of babies, the insane, the immobilized. But these can be regarded as accidental cases, which are subject to change. The baby grows up, the insane or the immobilized could in principle be cured. The case of plants and animals is very different. Under no circumstances can they make demands, lay claims or do any of the other things I mentioned above. Suppose I want to cut down a tree in my garden. Then, if the tree is a native one, I may have to seek permission to do so from the local government. If I lived in Sydney rather than in Canberra I might have to get permission even if the tree is not a native one. My neighbours, too, have the right to demand or to oppose the cutting down. But it seems to me absurd to say that the tree has any rights in the matter, any more than it can choose to grow in a particular way so as to reduce the likelihood of its being cut down by, for example, avoiding electric wires. Equally I do not think that cutting it down could be regarded as just or

J OHN PASSMORE 148 unjust except in a purely judicial sense. I do not know of any philosopher before the last half century or so who would question these remarks about trees although the view that animals have rights goes back to the last century. Here again environmental ethics, far from simply applying established philosophical views, sets new problems for philosophy or demands a rethinking of accepted views. What is the range of expressions like ‘right’ and ‘just’? Am I obsessed, as some would say, by legalism? There are, of course, other approaches by those who are unhappy with the ‘rights’ attempt to explain why we should take account of the welfare of those living things, the existence of which does not immediately seem to be of any importance to human beings, which might seem, indeed, to be something of an obstacle to human endeavours. That is by way of a theory of values which does not identify being valuable with being financially profitable. I cannot here explore this approach in detail but certainly ascribing non-economic value to plants, trees, animals, rocks does not seem as absurd as ascribing rights to them. It might not, of course, be a matter of their being of direct value to human beings. The value may lie in their being members of an ecological system which we regard, for all sorts of reasons, as being valuable. A major problem, however, remains. There is no philosophical thermometer for measuring values. And in very many cases the argument rages between values of different types. Within the field of economics values can be set by supply and demand. But that is not the case outside economics. The great disputes which arise are not about whether tree X is more valuable than tree Y in an ecological sense but between economic and environmental values, in some countries complicated by the values of indigenous peoples, whether Inuits or Australian aborigines. To take a case, foresters want to cut down a group of native trees, flourishing in their ecological settings. The forester says that the trees are valuable only because one can cut them down and export them as chips. That will make work for those employed to cut the trees down and will enrich the inhabitants of a nearby village who will supply them with resources, food and the like. The indigenous people say that the trees are on sacred ground, and that is their value. Once cut down they are valueless. The environmentalist says that they are of value as being central to an ecological system which cannot survive the departure of the trees. This looks like a philosophical dispute. But how are we to set about settling the dispute? I wish I knew. Now for metaphysics. For centuries there was, in the Christian world, a popular—or what would now be called a ‘folk’—metaphysical chain based on relative degrees of power. At the top came a Trinitarian God with unlimited power, surrounded by what we might call religious

P HILOSOPHY

AND

E COLOGY 149

courtiers. Secondly came human beings, subject to the will of God but with an absolute power, transmitted to them by God, over nature. The weakest form of environmentalism substitutes ‘stewardship’, responsible to God, for absolute power over nature. Philosophers had of course challenged this folk chain, in various ways. Spinoza went so far as to identify God with Nature. Some insisted that human beings simply formed part of Nature. Ecologists had then to determine the relation between human beings and other natural phenomena. The general tendency of philosophy in British-speaking countries, with idealist exceptions in the last decades of the nineteenth century and the early decades of the twentieth century, has been pluralistic, sometimes even atomistic. But quite a few environmentalist philosophers are holistic. Not uncommonly, they take as their bible Aldo Leopold’s 1966 A Sand County Almanac. Sometimes their appeal, rather, is to Asiatic religions. Science is sometimes called upon in the form of chaos theory with its story about the fluttering wings of a butterfly which can bring about changes of weather in a remote corner of the world. Spinoza, too, is appealed to, as identifying God with Nature. As a reaction against the view that the effects of our actions are always purely local this view has its points. But I refuse to go to the opposite extreme. I can see that my coming to Boston can have diverse effects—on my garden, upon the many birds which see in it a source of food and water, on a variety of persons whose income will be reduced, along with much else. But not that it will have an effect on the price of eggs. That human actions can have unexpected ecological effects I am prepared to grant and see as important. The position I adopt I shall call—inventing this name as I write this paper—‘pluralistic connectivism’. It sees the world in the following way: nothing is internally completely simple; it is always complex and hence capable of behaving in ways we did not expect. Secondly, everything is always interacting with its immediate surroundings. The least thing we can encounter is a state of affairs, someone or something being something or doing something. And being and doing thus will have effects on what surrounds them, including both those who observe them and those who do not. The historical background to this view is Heraclitus. If we cut down a tree which is planted in a street, that action can affect the behaviour of birds, of nearby plants, of those human beings who used to seek shelter under it and perhaps in other ways I have never noticed or which happen to the tree only after it is changed, becoming a log burnt to ashes or transformed into furniture. All of these changes can go on to have further effects. Such facts are recognized by connectivists. But connectivism is not holistic; it does not see these manifold states of affairs as manifestations of some single entity. One can say that the death of the tree can be

J OHN PASSMORE 150 a state of affairs in a number of networks but each of these networks is formed by accidental facts. These facts enable environmentalists to argue correctly that the cutting down of a tree is not just that. What else it is depends on the context. John Passmore, The Historical Studies Department, Australian National University, Canberra, Australia 0200

NATURE AND CULTURE IN ENVIRONMENTAL ETHICS

The pivotal claim in environmental ethics is that humans in their cultures are out of sustainable relationships to the natural environments comprising the landscapes on which these cultures are superimposed. But bringing such culture into more intelligent relationships with the natural world requires not so much “naturalizing culture” as discriminating recognition

Holmes Rolston, III

of the radical differences between nature and culture, on the basis of which a dialectical ethic of complementarity may be possible. How far nature can and ought be managed and be transformed into humanized nature, resulting in “the end of nature,” is a provocative question. Environmental ethics ought also to seek nature as an end in itself.

I. NATURE I NCLUDING CULTURE

I

n one sense, “nature” is quite a grand word, referring to everything generated or produced. Natura or physis is the source from which all springs forth. So comprehensive a term becomes troublesome. Is there a contrast class? If one is a metaphysical naturalist, then nature is all that there is. Used in this universal sense, claiming that “everything is natural” is about as informative as insisting that “everything that is, exists.” Metaphysical naturalists may need the word in this sense for their cosmological purposes. Humans and all their activities will be included; humans are generated within nature and they break no natural laws. Everything technological will, on this meaning, be completely natural. So will everything industrial, or political, or economic, philosophical, or religious.

151

H OLMES ROLSTON, III 152 Such scope is also problematic, however, because it prevents discriminating analysis of the differences between spontaneous nature and deliberated culture. A predicate, “natural,” that includes all actual and possible properties, excludes nothing; denoting everything is about like denoting nothing, at least nothing in particular. The most forceful objection to this sense of nature, in the context of doing environmental ethics, is that such definition allows no useful contrast with culture, but we need that contrast carefully analyzed if humans are going to relate their cultures, including their technologies, to nature, asking about sustainable development or about nature conservation as goals.

II. C ULTURE DISTINGUISHED FROM NATURE A straightforward contrast class is culture. If I am hiking across wildlands, the rocks and trees, the birds and even their nests, are natural, but if I come upon an abandoned boot, or a candy wrapper, these are artifacts, unnatural. Expanding such examples into a metaphor, the whole of civilization is producing artifacts in contrast to the products of wild spontaneous nature. Wild animals do not form cumulative transmissible cultures. Information in nature travels intergenerationally on genes; information in culture travels neurally as persons are educated into transmissible cultures. The determinants of animal and plant behavior are never anthropological, political, economic, technological, scientific, philosophical, ethical, or religious. Animals imprint on and learn the behaviors of their parents, and in this sense acquired information sometimes travels from one generation to the next. One sometimes encounters the term “culture” used of animals. Opening an anthology on Chimpanzee Cultures, the authors doubt, interestingly, whether there is any such thing: “Cultural transmission among chimpanzees is, at best, inefficient, and possibly absent.” 1 Chimpanzees clearly influence each other’s behavior, and intend to do that; they copy the behavior of others, including their toolusing. But there is no clear evidence that they attempt to change the mind, as opposed to the behavior, of another chimpanzee. They seem “restricted to private conceptual worlds.” 2 Without some concept of teaching, of ideas moving from mind to mind, from parent to child, from teacher to pupil, a cumulative transmissible culture is impossible. The critical factor is the deliberated modification of nature that separates humans in their cultures from wild nature. Any transmissible culture, and especially a high technology culture, does need to be discriminated from nature. The Boeing 777 jet plane is being built in

N ATURE AND CUL TURE ULTURE 153 the largest building in the world, assembling three million parts per plane. At the Everett, Washington, site, Boeing used in its design 2,200 workstations linked to eight of IBM’s largest mainframes, linked with other computers and databases spread across 17 time zones, bringing the total to 7,000 workstations. The information processed was stored on 3.5 terabytes. 3 Were it stored on ordinary 3.5 inch disks, this would require a stack of two and a half million disks nearly five miles high. In addition, on the economic side, Boeing kept an eye on competing with the A-330 Airbus, subsidized by British, French, German, and Spanish companies, with Boeing more on its own raising capital and encouraged by United’s initial $22 billion order. Boeings fly, as wild geese fly, using the laws of aerodynamics. The flight of wild geese is impressive; scientists can hardly yet be said to understand these “bird brains” and how they migrate. The information storage system in the goose genetics could, in its own way, be the equal of the Boeing system. Some of the information in the geese is transmitted nongenetically, as when they learn migration routes by following other geese. But it is only philosophical confusion to remark that both processes are equally natural, and let it go at that. No interesting philosophical analysis is being done until there is insightful distinction into the differences between the ways humans fly in their engineered, financed jets and the ways geese fly with their genetically constructed, metabolically powered wings. The Boeings are being built within a hundred miles of old-growth forest that American environmentalists are concerned about saving. One could argue that saving the forests is more important than building the new Boeings. But one is unlikely to be guided in the rationale for conserving the forest until one recognizes that the processes that govern the forest are radically different from the processes by which the Boeing 777 is produced. Environmental philosophers, concerned for sustainable relationships with nature, often insist that culture too is natural, that humans are a part of and not apart from nature. Let me cite three: J. Baird Callicott desires a new concept of nature that includes culture. “Nature as Other is over . . . . The modern picture of nature is false and its historical tenure has been pernicious. A new dynamic and systemic postmodern concept of nature, which includes rather than excludes human beings, is presently taking shape.” Callicott hopes to cure us from the “sharp dichotomy between man and nature,” which has too long been a feature alike of religion and philosophy, “both wellsprings of the Western intellectual heritage . . . . We are therefore a part of nature, not set apart from it.” 4

H OLMES ROLSTON, III 154 Val Plumwood, analyzing the wilderness idea, worries about my “dualism” and complains that I “cannot recognize the continuum of nature and culture,” since my dualism “blocks recognition of the embeddedness of culture in nature. . . .This nature/culture dualism distorts the way we can think about land, obliging us to view it as either pure nature or as a cultural product, not nature at all . . . . Recovering the lost ground of continuity that dualistic conception has hidden from us allows . . . recognizing nature in what has been seen as pure culture and culture in what has been seen as pure nature.” Plumwood goes on to say, however, that she wants to eliminate not “the distinction” but “the dualism.” We ought to recognize “the presence of the Other,” reaching “a non-oppositional account of the relationship between humans and the wild Other.” Further, it is not true that “everything that humans do is natural”; it is critical not “to obscure the basis for understanding the difference between anthropogenic from non-anthropogenic elements in country.” 5 At this point I begin to wonder if my mind is subtle enough to catch the distinction between making a distinction between humans and “the wilderness Other,” and distinguishing significant differences between nature and culture, an otherness that indeed sets them apart. Freya Mathews puts it this way: “It is no longer controversial to state that a human individual is essentially a cultural being, and that culture is an emanation of Nature.” 6 “Emanation,” is a flexible term and in the root idea of “flowing forth,” I agree that culture has, over evolutionary history, flowed from sources in nature. Still, the Boeing 777 is not some sort of emanation from the old-growth forest. We need a stronger term; culture is an “emergent” from nature. Nature evolved into culture; culture evolved out of nature, but it did evolve out of it. Over the millennia, humans make an increasing “exodus” from nature. I agree that we humans are “essentially cultural beings,” but that means we are not just emanations from nature.

III. NATURE ENVIRONING CULTURE Still, we must be cautious. Nature is the milieu of culture. Using a metaphor, nature is the womb of culture, but a womb that humans never entirely leave. Nature can do much without culture—the several billion years of evolutionary history are proof of that. Culture, appearing late in natural history, can do nothing without nature as its ground. In this sense, culture will always have to be constructed out of, superposed on nature.

N ATURE AND CUL TURE ULTURE 155 No matter what kind of exodus humans make from nature, they are going to remain male or female, with hearts and livers, and blood in their veins, walking on two feet, and eating energies that were originally captured in photosynthesis by chlorophyll. Culture remains tethered to the biosystem and the options within built environments, however expanded, provide no release from nature. Humans depend on air flow, water cycles, sunshine, nitrogen-fixation, decomposition bacteria, fungi, the ozone layer, food chains, insect pollination, soils, earthworms, climates, oceans, and genetic materials. An ecology always lies in the background of culture, natural givens that underlie everything else. Plants and animals modify the landscapes on which they live. Despite the changes they introduce, however, plants and animals are largely adapted to the environment in which they find themselves. These adaptations are genetic, behavioral, morphological, physiological—fur or horns or teeth, or thorns or deciduous leaves or camouflage. Culture makes possible the deliberate and cumulative, and therefore the extensive, rebuilding of nature. Humans reshape their environments, including new ones into which they expand, rather than being themselves morphologically and genetically reshaped to fit their changing environments.

IV. NATURE AT AN END ? A NATURE-CULTURE ELLIPSE Has nature ended? The question is one of degree. Certainly, nature now bears the marks of human influence more widely than ever before. In one survey, using three categories, researchers find the proportions of Earth’s terrestrial surface altered as follows: 1. Little disturbed by humans, 51.9%. 2. Partially disturbed, 24.2%. 3. Human dominated, 23.9%. Factoring out the ice, rock, and barren land, which supports little human or other life, the percentages become: 1. Little disturbed, 27.0%. 2. Partially disturbed, 36.7%. 3. Human dominated, 36.3%. 7 Most terrestrial nature is dominated or partially disturbed (73.0%). Still, nature that is little or only partially disturbed remains 63.7% of the habitable Earth. If nature means absolutely pristine nature, totally unaffected by human activities, past or present, there is relatively little remaining on Earth—if our detection instruments are keen enough. Still, nature on Earth can be relatively pristine. Nature has not been brought to an end, not yet at least. But we do have to face that possibility in the future. Daniel Botkin agrees: “Nature in the twenty-first century will be a nature that we make.” “We have the power to mold nature into what we want it to be.” Of course he, like

H OLMES ROLSTON, III 156 everybody else, urges us “to manage nature wisely and prudently,” and, to that end, ecology can “instrument the cockpit of the biosphere.” 8 That sounds like high-tech engineering that brings wild nature under our control, remolding it into an airplane that we fly where we please. Humans have always had to rest their cultures upon a natural life support system. Their technosphere was constructed inside the biosphere. But in the future that could change; the technosphere could supercede the biosphere. Michael Soulé faces this prospect: In 2100, entire biotas will have been assembled from (1) remnant and reintroduced natives, (2) partly or completely engineered species, and (3) introduced (exotic) species. The term natural will disappear from our working vocabulary. The term is already meaningless in most parts of the world because anthropogenic [activities] have been changing the physical and biological environment for centuries, if not millennia.9

So it does seem possible to end nature by transforming it into something humanized. This has already been taking place, and the future promises more, at an escalating pace. Over great stretches of Earth, wild nature has been already or likely will be diminished in favor of civilization. In some sense, that ought to be so. This ending may be always, in its own way, a sad thing; but it is an inevitable thing, and the culture that replaces nature can have compensating values. It would be sadder still, if culture had never appeared to grace the Earth, or if cultures had remained so modest that they had never substantially modified the landscape. Humans too belong on the planet; and the epoch of evolutionary nature, and even of ecological nature is over. That is what is right about the view that with the arrival of humans, their cultures, and their technologies, pristine nature vanishes. Nature does not vanish equally and everywhere, but there has been loosed on the planet such a power that wild nature will never again be the dominant determinant of what takes place on the inhabited landscapes. But this is not the whole truth. Nature neither is, nor ought to be completely ended. Or everywhere ended. We do not want entirely to transform the natural into the cultural, nor do we want entirely to blend the cultural into the natural. Neither realm ought to be reduced to, or homogenized with, the other. Humanizing it all does not make us a part of it; rather, the dominant species becomes still more dominant by managing all. That, ipso facto, sets us apart, the one species that manages the place. Rather, we humans, dominant though we are, want to be a part of something bigger, and this we can only do by sometimes drawing back to let others be. This we do precisely by setting aside places as wilderness where we will not remain, which we will not trammel.

N ATURE AND CUL TURE ULTURE 157 Environmental ethics seeks a complementarity. Think of an ellipse with its twin foci. Some events are generated under the control of one focus, culture; such events are in the political zone, where “polis” (town) marks those arts and achievements where the contributions of spontaneous nature are no longer evident in the criteria of evaluation, though they remain among the precursor and sustaining events. This is the artifactual, the technological domain. At the other end of the ellipse, a wild region of events is generated under the focus of spontaneous nature. These events take place in the absence of humans; they are what they are in themselves—wildflowers, loons calling, or a storm at sea. Although humans come to understand such events through the mediation of their cultures, they are evaluating events generated under the natural focus of the ellipse. The constraint of nature is maximal, the contribution of culture is minimal. A domain of hybrid or synthetic events is generated under the simultaneous control of both foci, a resultant of integrated influences from nature and culture, under the sway variously of more or less nature and culture. Nature is redirected into cultural channels, pulled into the cultural orbit. This happens when human labor and craft put natural properties to use in culture, mixing the two to good effect in agricultural, industrial, scientific, medical, and technological applications, or to adverse effect by mistake and spillover. But always culture has to answer to what is objectively out there in nature. Each of the foci critiques the other; the realities of nature test the wisdom of any culture; differing cultures take differing perspectives on the natural world within which they are situated and which they rebuild. “Symbiosis” is a parallel biological word. In the symbiosis zone, we have both and neither, but we do not forget there remain eventzones in which the principal determinant is culture, and other zones in which the principal determinant remains spontaneous nature. We do not want the ellipse to collapse into a circle, especially not one that is anthropocentric. Nature as it once was, nature as an end in itself, is no longer the whole story. Nature as contrasted with culture is not the whole story either. An environmental ethic is not just about wildlands, but about humans at home on their landscapes, humans in their culture residing also in nature. This will involve resource use, sustainable development, managed landscapes, the urban and rural environments. But environmental ethicists can and ought sometimes wish nature as an end in itself. We end with a sense in which nature has not ended and never will. Nature bats last. Humans stave off natural forces, but the natural forces will return, if one takes away the humans. Nature is forever lingering around.

H OLMES ROLSTON, III 158 Given a chance, which will come sooner or later, natural forces will flush out human effects. Even if the original wildness does not return, nature having been irreversibly knocked into some alternative condition, wildness will return to take what course it may. If you wonder whether nature has ended, watch what happens on a vacant lot. One might think that there is no nature left, since the lot is filled with the rubble of artifacts—pop cans and broken concrete blocks. But nature comes back, and soon there are weeds sprouting up, a lush growth. We could say, in a more philosophical mood, that nature still knows how to value the place, or knows what values to put in place that can still be sustained there. In that sense, a vacant city lot, if watched long enough, testifies eloquently to how nature, managed and mismanaged by humans though it may be, does not and cannot end. In, with, and under culture, there is always this once and future nature. Holmes Rolston, III, Department of Philosophy, Colorado State University, Fort Collins, CO 80523; [email protected]

NOTES 1. Richard W. Wrangham et al., eds., Chimpanzee Cultures (Cambridge, MA: Harvard University Press, 1994), 2. 2. Ibid. 3. Henry Petroski, “The Boeing 777,” American Scientist 83 (1995): 519-522. 4. J. Baird Callicott, “La Nature est morte, vive la nature!” Hastings Center Report 22 (no. 5, September/October 1992): 16-23, citation on 16-18. 5. Val Plumwood, “Wilderness Skepticism and Wilderness Dualism,” in The Great New Wilderness Debate, ed. J. Baird Callicott and Michael P. Nelson, (Athens: University of Georgia Press, 1998), 652-690, citations on 670, 675f, 682, and 685. 6. Freya Mathews, The Ecological Self (Savage, MD: Barnes and Noble Books, 1991), 138. 7. Lee Hannah, et al., “A Preliminary Inventory of Human Disturbance of World Ecosystems,” Ambio 23(1994): 246-50. 8. Daniel B. Botkin, Discordant Harmonies: A New Ecology for the Twenty-first Century (New York: Oxford University Press, 1990), 190-193, 200f. 9. Michael E. Soulé, “Conservation Biology in the Twenty-first Century: Summary and Outlook,” in Conservation for the Twenty-first Century, ed. David Western and Mary Pearl, (Oxford: Oxford University Press, 1989), 301.

DEPTH, TRUSTEESHIP,

I review some themes of Naess’s “The Shallow and the Deep, Long-Range Ecology Movements” article and Routley’s “Is there a Need for a New,

AND

REDISTRIBUTION

An Environmental Ethic?” presentation at the 1973 World Congress. Naess’s affiliation to the Deep Ecology Movement deserves acclaim, theoretic entanglements notwithstanding. Routley advocated a new

Robin Attfield

ethic because no Judaeo-Christian ethical tradition could cope with widespread environmental intuitions. However, the ethical tradition of stewardship can satisfy such concerns. It is compatible with environmental values, need not be managerial, and can assume a secular form. But the related res- ponsibilities vary with wealth and power, and structural change is necessary to empower people currently unable to uphold it.

T

he year 1998 marks the twenty-fifth anniversary of two formative essays of environmental philosophy, Richard Routley’s “Is there a Need for a New, An Environmental Ethic?”1 presented at the World Congress of Philosophy at Varna (Bulgaria) in 1973, and Arne Naess’s Inquiry article “The Shallow and the Deep, Long-Range Ecology Movements. A Summary.”2 Together with John Passmore’s work, Man’s Responsibility for Nature, published in 1974, 3 William T. Blackstone’s 1974 collection Philosophy and Environmental Crisis,4 and Holmes Rolston’s essay “Is There an Ecological Ethic?” which appeared in Ethics in 1975,5 these publications could be said to have founded environmental philosophy, such that this Congress marks what anniversary-watchers might call its “silver jubilee.” 159

R OBIN A T TFIELD 160 Besides, Routley’s reference to an unpublished work of Passmore, entitled “Ecological Problems and Western Traditions,”6 which seems to have covered much the same ground as the early chapters of Man’s Responsibility, suggests that by 1973 Passmore too had already contributed to this new field, and may well have provided the stimulus for Routley’s essay. True enough, the ideas of earlier figures such as Aldo Leopold have been taken up and developed by many environmental philosophers (including Routley), as well as environmentalists, while environmental concern was shown in one form or another by earlier philosophers such as Plato, Malthus, Emerson, Mill, and Engels. But in face of a new array of ecological problems (pesticides, nuclear fall-out, ecocide in Vietnam), it was in the early 1970s that philosophers began to take environmental issues seriously. A brief mention of Naess’s paper is in place, prior to discussion of Routley’s, even though no short essay such as this can do justice to either of these classic articles. Besides characterising two kinds of ecology movement, Naess maintained that such movements are not so much ecological and thus scientific as ecophilosophical. “Philosophy,” he claimed, “is the most general forum of debate on fundamentals, descriptive as well as prescriptive, and political philosophy is one of its subsections.”7 ‘Ecophilosophy’ was his term for the philosophical study of ecological issues. This he contrasted with an ‘ecosophy’, a reasoned stance on ecological issues drawing on numerous disciplines including ecophilosophy, and at the same time “a kind of sofia” or wisdom, containing “both norms, rules, postulates, value priority announcements and hypotheses concerning the state of affairs in our universe. Wisdom,” he added, “is policy wisdom, prescription, not only scientific description and prediction.”8 Ecosophies are systems which are global, in the sense (the context suggests) of being broad, comprehensive, inter-disciplinary, and concerned with policy-making,9 as well as with hypotheses and explanations. The earlier parts of his paper provide a framework for ecosophical systems,10 or rather for the kind which he calls “Deep” and “Long-Range.” This is not the place to discuss what he meant by these terms. Indeed, while later work of Routley’s11 suggested that all the positions which reject anthropocentrism in axiology, including my own, are to be classified as deeper positions, readers of Naess with any sense of their own identity and worth might be forgiven for distancing themselves from his characterisation of Depth, with its metaphysic of fields of intrinsic relations which seems to make personal identity depend on relations to the entire universe, and its ethic of biospherical egalitarianism, which seems to place the interests of all living creatures on a par, whatever their

D EPTH, TRUSTEESHIP, AND RED ISTRIBUTI ON EDISTRIBUTI ISTRIBUTIO 161 nature.12 They might even be forgiven for a sneaking sympathy for the Shallow Movement, despite its anthropocentric preoccupation with pollution and resources; after all, Naess considers that even this approach holds promise with regard to recognising the importance of local autonomy and decentralisation.13 However, as Naess points out, a preoccupation with combatting pollution and resource-depletion can serve to increase evils of other kinds, such as the exploitation of some groups by others, something which can adversely affect the potentialities for self-realisation of both.14 Besides defending the interests of nonhuman species, Naess rightly builds into his characterisation of the deep ecological movement struggles against the kinds of economic and cultural domination which are frequently aspects of globalisation, and opposition to class differences, an opposition which he regarded as consistent with encouragement of cultural diversity.15 These social aspects of the deep movement have often been neglected by self-styled “Deep Ecologists,” but they may supply good grounds for leaving behind the Shallower positions and sailing boldly into deeper waters, without necessarily putting in at all the enchanted anchorages of Naess’s favoured ecosophies in metaphysics, axiology, and ethics. The conclusion of Richard Routley’s essay is that respect for the environment requires not just an extension of traditional ethics, but a change to an environmental ethic which would be essentially new. To show this he introduces Passmore’s classification of ethical traditions, and argues that not only is the despotic view beyond repair, the view, that is, which authorises people to treat nature as they please, but so too are the so-called “lesser” traditions, the stewardship view, which regards humanity as a trustee or steward, answerable (at least originally) to God for nature’s care, and the co-operation-with-nature view, according to which the role of humanity is to cultivate and perfect nature by bringing out its potentialities. This conclusion he grounds partly in generalisations about the two non-despotic traditions, intended to show that they cannot go far enough in the direction of an environmental ethic,16 and partly through thought-experiments such as the example of the last man, who destroys every living thing he can without cost to himself or other humans. 17 As Routley recognises, most of his readers shared an adverse verdict on the last man, and on the villains of his other examples, such as the great entrepreneur, who manufactures automobiles simply to maximise gross worldproduct; his claim is that none of the Western ethical traditions can uphold these verdicts.18 However, Passmore maintains in Man’s Responsibility that the seeds are to be found of an acceptable environmental ethic within the non-despotic traditions, 19 and I have argued that the stewardship view need be neither anthropocentric nor managerial nor instrumentalist.20

R OBIN A T TFIELD 162 This interpretation was also independently presented by Val Routley (now Val Plumwood) in her review of Man’s Responsibility,21 but subsequently seems to have been implicitly retracted.22 If this is granted, then the stewardship view, which, incidentally, I have argued to be much too pervasive to comprise a mere “lesser” view in Western ethical traditions, can after all uphold the verdicts which most people form about Routley’s examples.23 The greater entrepreneur, for example, both wastes materials in his charge and harms a wide range of creatures by damaging the climate. An implication would be that Routley does not make his case for the need for a brand new environmental ethic, that no such new ethic is required, and that environmentalists can argue by analogy from received values, at least in the West, and say what needs to be said on this basis. This position is sometimes pilloried as ‘extensionism’, but is probably indispensable, at least for practical purposes. But such large changes of attitude and policy are required to cope with current problems that if Routley is construed as meaning by ‘new ethic’ that we need new practices then there is every reason to agree with him. Nevertheless studies of the stewardship approach bring to light halfforgotten aspects of this continuing tradition. Thus Erik Katz has shown that the Old Testament combines belief in human stewardship of nature, belief in the intrinsic value of the various creatures, including the leviathans of the deep ocean and plants such as Jonah’s gourd, and acceptance of human dominion, albeit in a sense which entirely precludes despotic attitudes, since all belief in human ownership of the land is rejected (let alone of the seas or the heavens). God is held to provide for the beasts of the wilderness, and to care more about the people and about the cattle of Nineveh than about the gourd (and probably more about the people than the cattle), but cares for the gourd too nonetheless; and teleological anthropocentrism, the view that everything was made for humanity, is clearly excluded.24 Further, these various beliefs and values are apparently consistently cotenable. Closely similar values are also implicit in the teaching of the Jesus of the synoptic gospels, about the oxen, the ravens, and the lilies of the field.25 But the subsequent history of the stewardship tradition cannot be discussed in detail here; suffice it to say that while many subsequent adherents persisted in the biblical rejection of teleological anthropocentrism, some highly influential ones such as Calvin did not.26 This teleological or metaphysical anthropocentrism, which seems to have entered Christianity with Origen, would in my view need to be rejected before the ethic of stewardship can be applied fruitfully to current problems. This process could be assisted by using a range of metaphors to interpret that of stewardship itself, like those of humankind as

D EPTH, TRUSTEESHIP, AND RED ISTRIBUTI ON EDISTRIBUTI ISTRIBUTIO 163 nature’s trustees or curators; each of these conveys the high value of the objects held in trust, and belies the charge against the stewardship view, implicit in Routley’s text, of amounting to no more than natural resource deployment or global managerialism.27 Routley, however, raises another problem; he asks for whom man is steward and to whom he is responsible “when the religious backdrop is removed.”28 On the other hand, secular adherents of the traditions of stewardship and of co-operation with nature (such as Murray Bookchin, Karl Marx, and Margaret Thatcher) take it for granted that such questions can be answered. Thus Karl Marx wrote: a whole society, a nation, or even all simultaneously existing societies taken together, are not the owners of the globe. They are only its possessors, its usufructuaries, and, like boni patres familias, they must hand it down to succeeding generations in an improved condition.29

Here “possessors” means “occupants,” and the globe is envisaged as entrusted to the present generation of occupants for the sake of future generations to whom responsibilities of trusteeship are owed. Coming generations, then, are the beneficiaries, and the trust is to be understood as owed either to the transgenerational community, or else to the community of moral agents, for which it is the turn of the present generation to navigate the current watch or to run the current lap of the relay of history. The problems with this passage of Marx concern rather the implications of handing down the globe in an improved condition, problems (once again) of managerialism and of anthropocentrism; but secular adherents of stewardship can correct for these problems just as religious adherents have corrected for tendencies such as these in theologians like Calvin. Thus if responsibilities are focused on the good of succeeding generations of all species, as both Naess and Routley held that they should be, and not just of our own, or at least if some of our responsibilities have this focus and they have a high enough priority among people’s other obligations, then stewardship (thus understood) no longer seems capable of requiring or permitting the total use of the globe for human purposes, as the Routleys used to fear.30 More generally, the stewardship tradition and the co-operation with nature tradition survive Richard Routley’s case against them. However, I should not be taken to be maintaining that humanity as an undifferentiated whole carries the responsibilities of a trustee. This kind of stance abstracts unduly from the material circumstances of particular human agents, and would clash with the valuable emphases of Naess on cultural diversity, and also on differences of power. To expect

R OBIN A T TFIELD 164 the poor to act as trustees would in some cases involve demanding selfsacrifice to the point of suicide. In any case, humanity is not sufficiently unitary or co-ordinated to comprise a single agency, and cannot be regarded as a cosmic steward in actual current practice for this reason if for no other. It may yet be desirable that each human group should develop a form of self-understanding vis-à-vis nature, and that in each case this should include a trusteeship role. But the differences of power, wealth and access to resources suggest instead that different models are in place; the situation of those with sufficient power to act for the overall good if they so choose should not be understood in the same way as that of those with so few resources as to be obliged by circumstances to behave in a more hand-to-mouth manner. Economic and political imperatives sometimes place people in a position where they cannot afford to be stewards or trustees; the need to cut firewood or to cultivate areas of forest can serve as examples. Such people could be compared to agents whose responsibility for their actions is diminished through lack of power or through subjection to pressures which it would be unreasonable to expect anyone to resist. If so, then the stewardship or trusteeship model would only be fully applicable to those relatively immune from such constraints; and while such immunity might be considered the normal state of affairs, it may paradoxically apply to no more than a minority of current human beings. If this much is endorsed, then there are other implications. If trusteeship of the tracts of nature over which a society has some control is the model for normal circumstances, but myriads of people cannot attain these circumstances or live by this model, then those who recognise this should be seeking radical change to emancipate people suffering these constraints. Environmental ethics would seem to supply an additional reason for this view. For the well-being of future generations may well depend on most people acting in the way that trustees would, but this condition is liable not to be fulfilled because of large inequalities of power and resources. So there is an additional reason here for action to rectify these inequalities, and thus empower people to carry out their role as trustees. There may be related implications concerning greenhouse gas emissions. At present, global emissions need to be stabilised and reduced for the sake of future generations, of islands, and of coastal communities, but poorer countries need to increase their emissions in order to provide for the basic needs of their peoples, both present and future. This granted, richer countries have a responsibility to reduce their emissions considerably so as to make all this possible. Further, one way in which this might be

D EPTH, TRUSTEESHIP, AND RED ISTRIBUTI ON EDISTRIBUTI ISTRIBUTIO 165 achieved is through dividing and distributing the total of tolerable emissions on a per capita basis to all countries, as has long been suggested by Michael Grubb,31 making part of each country’s quota inalienable to provide for the basic needs of its citizens, and allowing countries to trade the remaining annual quotas; this approach could both curtail the rise of the oceans and secure a massive redistribution of power and resources. While the stewardship model has yet ampler applications, this much must suffice to show that it can be radical in practice. As such it coheres with many but not all of the planks of the Deep Ecology platform of Naess, and possibly with much of what Routley was seeking in his advocacy of a new environmental ethic, without actually being new or non-traditional in itself. Robin Attfield, Philosophy Section, ENCAP, University of Wales Cardiff, Cardiff, UK CF1 3XB; [email protected]

NOTES 1. Richard Routley, “Is There a Need for a New, An Environmental Ethic?” Proceedings of the XV World Congress of Philosophy, (Varna, Bulgaria: Sofia Press, 1973), 205-210. 2. Arne Naess, “The Shallow and the Deep, Long-Range Eco-logy Movements. A Summary,” Inquiry 16 (1973): 95-100. 3. John Passmore, Man’s Responsibility for Nature, (London: Duckworth, 1974). 4. William T. Blackstone, ed., Philosophy and Environmental Crisis, (Athens, Ga.: University of Georgia Press, 1974). 5. Holmes Rolston III, “Is There an Ecological Ethic?” Ethics 85 (1975): 93-109. 6. John Passmore, “Ecological Problems and Western Traditions.” Routley refers to this work in “Is There a Need,” op. cit., 210, n. 2. 7. Naess, “Deep Ecology Movement,” 99. 8. Ibid., 99. 9. Ibid., 100. 10. Ibid., 99. 11. Richard Sylvan, “A Critique of Deep Ecology,” Radical Philosophy 40 (1985): 2-12, and 41 (1985): 10-22. 12. Naess, op. cit., 95f.

R OBIN A T TFIELD 166 13. Ibid., 98. 14. Ibid., 96f. 15. Ibid., 96f. 16. Routley, “Is There a Need,” 205f. 17. Ibid., 207. 18. Ibid., 207f. 19. Passmore, Man’s Responsibility, chap. 1f. 20. Robin Attfield, The Ethics of Environmental Concern (Oxford: Blackwell and New York: Columbia University Press), chap. 2f. 21. Val Routley, “Critical Notice of Man’s Responsibility,” Australasian Journal of Philosophy 53 (1975): 171-185, esp. 174. 22. Richard Routley and Val Routley, “Human Chauvinism and Environmental Ethics,” in Environmental Philosophy, ed. Don Mannison, Michael McRobbie, and Richard Routley (Research School of Social Sciences, Australian National University, 1980), 96-189, esp.121-125. 23. These claims were both discussed in Robin Attfield, “Western Traditions and Environmental Ethics,” in Environmental Philosophy, ed. Robert Elliot and Arran Gare (St. Lucia: University of Queensland Press, 1983), 201-230; reprinted in Robin Attfield, Environmental Philosophy: Principles and Prospects (Aldershot and Brookfield, Vt: Avebury, 1994), 41-68. 24. Erik Katz, “Judaism and the Ecological Crisis,” in Worldviews and Ecology: Religion, Philosophy & the Environment, ed. Mary Evelyn Tucker and John A. Grim (Lewisburg, Pa.: Bucknell University Press, 1993), 55-70. 25. Richard Bauckham, “What was Jesus’ Attitude to Animals?” in Animals on the Agenda: Questions about Animals for Theology and Ethics, ed. A. Linzey (London: SCM Press, forthcoming). 26. Keith Thomas, Man and the Natural World: A History of the Modern Sensibility (New York: Pantheon Books, 1983), 154-166. 27. Routley, “Is There a Need,” 206. 28. Ibid., 206. 29. Karl Marx, Capital, vol. 3, (London: Lawrence & Wishart, 1972), 776; quoted in John O’Neill, Ecology, Policy and Politics (London and New York: Routledge, 1993), 42. 30. Routley and Routley, op cit. 31. Michael Grubb, The Greenhouse Effect: Negotiating Targets, (London: Royal Institute of International Affairs, 1989). See also by the same author, Energy Policies and the Greenhouse Effect, (Aldershot: Gower, 1990).

D EPTH, TRUSTEESHIP, AND RED ISTRIBUTI ON EDISTRIBUTI ISTRIBUTIO 167

BIBLIOGRAPHY Attfield, Robin. 1983. The Ethics of Environmental Concern. Oxford: Blackwell and New York: Columbia University Press. . 1983. “Western Traditions and Environmental Ethics.” In Environmental Philosophy, edited by Robert Elliot and Arran Gare, 201-230. St. Lucia: University of Queensland Press, University Park: Pennsylvania State University Press, and Milton Keynes: Open University Press. (Only Open University Press supplied the necessary list of errata.) . 1994. “Western Traditions and Environmental Ethics.” Reprinted in Robin Attfield, Environmental Philosophy: Principles and Prospects, 41-68. Aldershot and Brookfield, VT: Avebury. Bauckham, Richard. Forthcoming. “What was Jesus’ Attitude to Animals?” In Animals on the Agenda: Questions about Animals for Theology and Ethics, ed. A. Linzey. London: SCM Press. Blackstone, William T., ed. 1974. Philosophy and Environmental Crisis. Athens, GA: University of Georgia Press. Grubb, Michael. 1989. The Greenhouse Effect: Negotiating Targets. London: Royal Institute of International Affairs. . 1990. Energy Policies and the Greenhouse Effect. Aldershot: Gower. Katz, Erik. 1993. “Judaism and the Ecological Crisis.” In Worldviews and Ecology: Religion, Philosophy & the Environment, ed. Mary Evelyn Tucker and John A. Grim, 55-70. Lewisburg, PA: Bucknell University Press. Marx, Karl. 1972. Capital. 3 vols. London: Lawrence & Wishart. Naess, Arne. 1973. “The Shallow and the Deep, Long-Range Ecology Movements. A Summary.” Inquiry 16: 95-100. O’Neill, John. 1993. Ecology, Policy and Politics. London and New York: Routledge. Passmore, John. 1974. Man’s Responsibility for Nature. London: Duckworth. . “Ecological Problems and Western Traditions.” Rolston, Holmes, III. 1975. “Is There an Ecological Ethic?” Ethics 85: 93-109. Routley (later Sylvan), Richard. 1973. “Is There a Need for a New, An Environmental Ethic?” In Proceedings of the XV World Congress of Philosophy, 205-210. Varna, Bulgaria: Sofia Press. Routley (later Sylvan), Richard and Routley (later Plumwood), Val. 1980. “Human Chauvinism and Environmental Ethics.” In Environmental Philosophy, ed. Don Mannison, Michael McRobbie, and Richard Routley, 96-189. Research School of Social Sciences, Australian National University.

R OBIN A T TFIELD 168 Routley (later Plumwood), Val. 1975. “Critical Notice of Man’s Responsibility.” Australasian Journal of Philosophy 53: 171-185. Sylvan (formerly Routley), Richard. 1985. “A Critique of Deep Ecology, I.” Radical Philosophy 40: 2-12. . “A Critique of Deep Ecology, II.” Radical Philosophy 41: 10-22. Thomas, Keith. 1983. Man and the Natural World: A History of the Modern Sensibility. New York: Pantheon Books.

BIOTECHNOLOGY

Rights can be founded in a variety of ethical systems—e.g., on natural law,

AND THE

ENVIRONMENT: FROM MORAL OBJECTIONS TO ETHICAL ANALYSES

on the duties postulated by deontological ethics, and on the consequences of our actions. The concept of risk we will outline supports a theory of rights which provides at least individual human beings with the entitlement not to be harmed by the environmental impacts of biotechnology. The analysis can, we believe, also be extended to the rights of animals as well as ecosystems, both of which can be harmed by human actions. We argue that further examination of these harms and rights would be the

Matti Häyry and Tuija Takala

best way to proceed from emotional moral objections to truly ethical analyses in the context of biotechnology and the environment.

I

n American television series involving lawyers, clashes between their moral beliefs and their ethical commitments frequently occur. The typical situation is that defense advocates know their clients to be guilty of some hideous crimes, but cannot, due to the ethical rules of their profession, reveal this to anybody, although they themselves feel that their clients ought to be punished for their deeds. The conflict arises, because the lawyers know that their ethical standards serve, as a rule, the best interest of the members of their societies, but also that their moral feelings reflect views which are widely shared in their communities. As the solutions to these conflicts are in the series often reached through breaches of confidentiality, or by other acts which violate the professional code, we, the viewers, are left with the impression that moral demands should in difficult situations always override ethical norms. 169

M AT TI HÄYR Y ÄYRY

AND

TUIJ A T AK AL A UIJA AKAL ALA

170 This impression is, however, deceptive, if ‘moral demands’ are defined as our unreflected reactions to thorny issues, and ‘ethical norms’ as rules or principles which are designed to make human life in ideologically fragmented societies tolerable to everybody. The reason why we accept the solutions given in the television courtroom dramas is that we regard the guidelines of the legal profession as subordinate to better accounts of social ethics, not necessarily that we see the immediate moral responses of individuals or groups as normatively binding. If the—admittedly fragile—conceptual distinction between ‘morals’ and ‘ethics’ is applied to the questions of biotechnology and the environment, the claim can be made that the majority of popular and philosophical responses are, at the moment, founded on the same faulty logic as these television series. This is the methodological starting point and hypothesis of our paper. Because of the relative obscurity of the distinction between ‘morals’ and ‘ethics’, however, we do not employ it systematically in the following, but use, instead, the terms more or less interchangeably.

IMPACTS OR ATTITUDES? Many forms of modern biotechnology can have an impact on our natural environment. These impacts range from the possibly beneficial to the potentially harmful, and they can be felt either by human beings, other living beings as individuals, entire species, or by more abstract entities like ecosystems or the biodiversity of certain regions or of our planet as a whole. Due to these impacts, it would seem reasonable that at least critical philosophical responses to questions concerning biotechnology and the environment would be centered on, or guided by, the harms and benefits produced by genetic engineering, and by their moral acceptability or unacceptability. But philosophical responses in this field are primarily generated by emotional reactions and inapplicable ethical theories rather than by any facts regarding the consequences of our actions. This may not be particularly amazing in the frameworks of virtue-based and duty-based ethics, but our contention is that the observation is valid even in the context of traditional utilitarian moral thinking. Our own proposal is that we can render moral analyses sensitive to the impacts of genetic engineering on our natural environment only by employing a lighter conceptual machinery, which takes into account both the consequences of our actions, or what we can know about them, and the attitudes we have regarding their acceptability.

B IOTECHNOLOGY AND THE ENVIRONMENT 171

THREE TRADITIONAL APPROACHES

TO

NORMATIVE ETHICS

There are three traditional approaches to normative ethics in the West, namely the teleological, deontological, and consequentialist models. These views all provide different answers to two basic questions, namely, ‘What is human nature like?’ and ‘How should individuals behave in order to be moral?’ The proponents of the teleological model hold that all beings have a telos or goal towards which they are inclined to move or to develop. The telos can be secular, as in the ethics of Aristotle, who thought that the natural goal of human beings is a good life in a just society, and beyond that, an elevated state of intellectual contemplation.1 The ultimate end can also be defined theologically, in which case the most likely candidates include an afterlife of everlasting joy, and some other states of being that transcend our earthly experiences. 2 Within the secular reading of the teleological model individuals should live their lives according to the rules of a just society, to be virtuous and to pursue the complicated pleasures of social life and intellectual perfection. The theological version can state, in addition, that we should adjust our life-styles to the received wisdom handed down to us by our parents and religious authorities. The basic deontological view of human nature is that our actions are guided by two competing motives—desires and a sense of morality. This view is open to two main interpretations when it comes to defining how people should find the guidance they need for their lives. In the intellectbased version, reason commands us to obey the moral law, usually against our own desires. Immanuel Kant’s theory is the paragon of this doctrine.3 The emotion-based reading states that feelings tell us what to do in each particular situation.4 The way proponents of consequentialist thinking see human nature, people want to obtain pleasure, happiness or well-being, and they want to avoid pain and suffering. Individuals are equally capable of egoism and altruism, that is, of promoting only their own self-interest and of taking others into consideration. According to the normative part of this doctrine, individuals should aim to be universally altruistic, either by trying to maximize the happiness of humankind (‘positive utilitarianism’) or by trying to minimize suffering (‘negative utilitarianism’).5

VIRTUES AND DUTIES What, then, could be the most typical responses of the different ethical doctrines to our present issue?

M AT TI HÄYR Y ÄYRY

AND

TUIJ A T AK AL A UIJA AKAL ALA

172 To begin with the Aristotelian tradition, it is not easy to apply the teleological model in its original secular form to the questions of genetic engineering and the natural environment. The link between recombinant DNA-techniques and the human good, not to mention the good of the planet as a whole, is obscure, and it seems that an accurate view of the consequences of genetics would be required to support an adequate analysis of the connection. Otherwise all judgements will inevitably be based on the attitudes we already have toward biotechnology, and appeals to virtues and the rules of a good society remain unsupported by the theory. Similar remarks can be extended to Kant’s original views. The fundamental duty postulated by him is our obligation to treat humanity in ourselves and in other persons always also as ends, never as a mere means. 6 This obligation, which we owe to our fellow humans but not to the members of other species, is based on our nature as rational agents. The way we ought to treat animals, plants and other life-forms is determined by our duties towards ourselves and other persons, not by their (nonexistent) worth as ends in themselves. The difficulties of applying Kant’s views to the manipulation of nonhuman beings by recombinant DNA-techniques include the fact that he did not clearly specify what our duties as regards animals, plants and other nonhuman organisms are. He believed that violence and cruelty towards animals set a bad example to our treatment of other people, and that not even inanimate objects should be wantonly destroyed because that would prevent others from making use of them. But how these remarks should be interpreted in the context of genetic engineering in a purely non-consequentialist analysis remains an unanswered question. It seems that Aristotle and Kant are in and by themselves unhelpful as regards the issues of biotechnology and the environment, because the application of their ideas leads us either to ideological choices which have nothing to do with their models, or to the consequences of our actions which have no legitimate place in them. But traces of more theologically-inclined readings of the Aristotelian teaching and of Kantian ethics can be found in many popular objections to biotechnology, especially in the claims that what genetic engineers do is unnatural, or an instance of playing God.

UNNATURALNESS The strand of thought in the Aristotelian tradition since Thomas Aquinas which is relevant to the first claim is the reliance on the notion of natural law. Natural law, according to Thomistic thinkers, is based on our

B IOTECHNOLOGY AND THE ENVIRONMENT 173 genuine being and good as persons, and it defines certain restrictions to our actions and our ways of life. 7 When this doctrine is given its secular reading, its main methodological message is that we ought to act upon the precepts of reason and oppose our inborn inclination to submit to our passions and desires. Kant’s formulation of the same idea was that we should always act in ways which we can accept in the light of our practical reason as universal modes of human behaviour. 8 Depending on the concept of ‘reason’ employed in the formula, it is presumably possible to argue within these views that genetic engineering is unnatural, because it is a violation against the untampered order of things in this world, or a practice which we cannot universally condone. But this argument, if it is an argument somebody would seriously wish to put forward, is problematical on two counts. First, its acceptance would imply that all present and future technological advances should be banned because they go against reason, or the natural order. Secondly, its theoretical tenability requires that we subscribe to the particular definition of reason which forbids us to alter the environment in any way. Both demands seem rather excessive.

PLAYING GOD Another line of argument teleological and deontological moralists can employ is that the prohibition of genetic engineering stems from the fact that there are limits to what we can do as moral agents, and applied biologists are overstepping these limits, or ‘playing God’, by trying to create new forms of life against the dictates of the natural or moral law. This argument has been thoroughly examined by Ruth Chadwick. 9 According to Chadwick’s analysis, the crux of the argument in the context of new technologies is that actions describable as ‘playing God’ can lead to disastrous and unpredictable consequences. But where should the lines of these actions be drawn and by whom? Chadwick considers many possibilities, one of which is particularly relevant here. The playing-God objection in the context of new technologies can, namely, be meant to state that the natural environment as a whole sets certain limits to our actions. Humankind has during the last few decades acquired powers which could be used to destroy most of the biosphere. Many people seem to think that genetic engineering is one of these powers, and they fear that, for instance, the release of genetically altered organisms into the environment may have irreversible ecological consequences.

M AT TI HÄYR Y ÄYRY

AND

TUIJ A T AK AL A UIJA AKAL ALA

174 Assuming that we are interested in the preservation of the biosphere, this objection against genetic engineering does indeed have some moral relevance. But the problem is that the appeal to consequences, which gives this argument its weight, also deprives it of its categorical disguise. It would, no doubt, be pragmatically unwise to destroy the only environment where we can live at present, but this does not amount to a teleological or deontological rejection of genetic engineering. The wrongness of the activity remains conditional upon the consequences.

DISGUST An alternative, emotion-based deontological approach to ethics is provided by Patrick Devlin, who in his influential essay “Morals and the Criminal Law” argued that activities should be reproached and banned by law if they provoke strong feelings of disgust even in individuals who are calm and appreciative of the demands of reason and common sense. 10 Devlin recognized the fact that feelings can vary from one location to another, and confined, accordingly, the prescriptive power of any given set of feelings to the community where it is prevalent. He also explained the moral force of disgust by maintaining that it indicates the boundaries within which the public morality must remain in order to keep the society viable. Applied to our present case, the Devlinian argument would be that biotechnological activities which can have an impact on our natural environment ought to be banned, because they would evoke strong negative feelings in Western societies and damage their moral foundation. But the difficulty with the emotion-oriented model is that it is exceedingly relativistic. In most cases there is no consensus concerning feelings, and many questions remain unanswered. Whose feelings should be respected? Should genetic engineering be banned only if everybody feels that it is bad? Or is it sufficient that the majority feel that way? Or perhaps prohibitions ought to be employed if a significant minority nurtures these feelings? Or should we say that if anybody feels this way, biotechnology ought to be rejected? Furthermore, the question also remains as to how the damage to the moral foundation of societies can be verified. There seem to be no good responses to these queries either at the general level or in the context of biotechnology.

B IOTECHNOLOGY AND THE ENVIRONMENT 175

CONSEQUENCES As we noted at the outset of this paper, it is not all that surprising that the responses of teleological and deontological moralists rely either on e t h i c a l p r e s u p p o s i t i o n s w h i c h i g n o r e t h e i m p a c t s o f biotechnology on the environment, or on ill-defined considerations of the consequences of our actions. What is more amazing, however, is that the situation is similar as regards standard utilitarian models of moral thinking, which should be centered on the advantages and disadvantages of the practices we accept or reject. This is because there are two stories which can be told about these, and the choice between them has very little to do with the actual or expected consequences of our actions. 11 The advantages of genetic engineering, as seen by its proponents, include many actual and potential contributions to medicine, pharmacy, agriculture, the food industry, and the preservation of our natural environment. The applications of genetic engineering to agriculture include the development of plants which contain their own pesticides. As for other food products, gene technologies can be applied to manufacture substances like vanilla, cocoa, coconut oil, palm oil, and sugar substitutes. And biotechnology can even provide an answer to the problems of pollution, as genetically engineered bacteria can be employed to neutralize toxic chemicals and other kinds of industrial and urban waste. The disadvantages of biotechnology, as seen by its opponents, are in many cases closely connected with the alleged benefits. One problem is that, despite the undoubtedly good intentions of the scientists, the actual applications of genetic engineering are often positively dangerous. Consider the case of plants which are inherently resistant to diseases, or which contain their own pesticides. Although there are no theoretical obstacles to the production of such highly desirable entities, corporations—who also sell chemical pesticides—might prefer to market another type of genetically manipulated plant, which is unprotected against pests but highly tolerant to toxic chemicals. The result of this policy would be an increase in the use of dangerous chemicals in agriculture, particularly in the Third World—which is to say that the outcome is exactly opposite to the one predicted by the proponents of biotechnology. An oft-used criticism against agricultural biotechnology is that the introduction of altered organisms into the natural environment can lead to ecological catastrophes. Scientists working in the field of applied biology have themselves noticed this danger, and set for themselves ethical guidelines which are designed, among other things, to

M AT TI HÄYR Y ÄYRY

AND

TUIJ A T AK AL A UIJA AKAL ALA

176 minimize this risk. But as the opponents of genetic engineering have repeatedly pointed out, not all research teams follow ethical guidelines if the alternative is considerable financial profit. Thus, although the expected advantages and dreaded disadvantages of genetic engineering are fairly well publicized, it is difficult to assess objectively what the actual consequences of employing the techniques would be. The results of assessments depend more on the optimism and pessimism of those evaluating the situation than on the consequences themselves.

RISK A possible way out of these dilemmas is, we suggest, to concentrate only on the negative consequences of genetic engineering, or risks, as people perceive them in the light of the facts available to us. ‘Risk’ can be defined as the possibility or probability of a loss, an injury, an unwanted outcome or an undesired result. Some of the potential harms involved in genetic engineering are the following. The release of genetically altered organisms in the environment can increase human suffering when medical measures are concerned, decrease animal welfare in experiments or through the use of recombinant DNA-techniques in breeding, and lead to ecological disasters. The containment of biotechnological material in laboratories and industrial plants involves two dangers: first is the possibility of an accidental release and second is the increased probability with which uncontrolled releases can produce undesired results. A risk that lies between these ‘scientifically controllable’ dangers and the more indirect political hazards of biotechnology is the probability of inadequate containment and irresponsible releases, which can be prompted by the economic selfinterest of research groups and industrial corporations. Given that ‘risk’ can be defined as the probability of harm, then, how should we define the concept of ‘acceptable risk’, on which analyses of the morality of risk-taking often center? Our own view is that the assessment should in each case be left to those who can be harmed by the decision in question. Economic risks are acceptable, if they are condoned by the biotechnological corporations and governments who take them. The risks imposed on laboratory personnel by the containment of dangerous materials ought to be evaluated by the laboratory personnel themselves. All other risks involved in genetic engineering are more or less universal, and should therefore be assessed—and eventually accepted or rejected—as democratically as possible.12

B IOTECHNOLOGY AND THE ENVIRONMENT 177 Scientists, industrialists and autocratic political decision-makers can argue against democratic risk assessment by claiming that their expertise enables them to predict with greater accuracy the consequences of different policies. What this objection overlooks, however, is that the acceptability of a risk for a given group is not determined exclusively by the facts of the matter, but also by the way the members of the group perceive the facts, and by the way they evaluate them. People cannot fully commit themselves to decisions which are based on epistemic and moral values they do not share. If anything goes wrong with the predictions of the experts, people feel, and are, entitled to resent the consequences of the authoritarian choices. The risks taken by experts on behalf of others are therefore unacceptable. But if risk-taking is based upon the considered choices of those who can be harmed by the consequences themselves, the situation is different. Even if the undesired outcome is realized, the risk is acceptable, because it is embedded in their own system of ethical and epistemic values.

RIGHTS In conclusion, let us add a few words on rights. Rights can in ethical analyses be founded on any of the systems we have discussed in this paper—on the natural law, on the duties postulated by deontological ethics, and on the consequences of our actions. The concept of risk we have outlined supports a theory of rights which provides at least individual human beings with the entitlement not to be harmed by the environmental impacts of biotechnology. The analysis can, we believe, also be extended to the rights of animals and possibly ecosystems, who can be harmed by our actions as well as human beings. The further examination of these harms and rights would probably be the best way to proceed from emotional moral objections to truly ethical analyses regarding the possible impact of biotechnology or the environment. 13 Matti Häyry and/or Tuija Takala, Department of Philosophy, University of Helsinki, Helsinki, 00014, Finland

NOTES 1. Aristotle, Nicomachean Ethics. 2. There are germs of this line of thinking in the Nicomachean Ethics, bk. X,

M AT TI HÄYR Y ÄYRY

AND

TUIJ A T AK AL A UIJA AKAL ALA

178 chap. 8, and the idea has been further developed in the works of Thomas Aquinas and his followers. 3. Immanuel Kant, Grundlegung zur Metaphysik der Sitten (1785). Page references are to Kant, Ethical Philosophy, 2nd ed., trans. J. W. Ellington (Indianapolis and Cambridge: Hackett Publishing Company, 1994). 4. The best defence of emotion-based deontological ethics is probably Patrick Devlin’s “Morals and the Criminal Law” (1959), reprinted in The Philosophy of Law, ed. R. M. Dworkin, (Oxford: Oxford University Press, 1977), 66-82. 5. On the many faces of utilitarian thinking, see, for instance, M. Häyry, Liberal Utilitarianism and Applied Ethics (London and New York: Routledge, 1994). 6. Kant, op. cit., 36. 7. For a detailed analysis of the natural law tradition, see, e.g., J. Finnis, Natural Law and Natural Rights (Oxford: Clarendon Press, 1980); J. Finnis, Fundamentals of Ethics (Oxford: Clarendon Press, 1984). 8.

Kant, op. cit., 30.

9. R. Chadwick, “Playing God,” Cogito 3 (1989): 186-193; see also M. Häyry, “Categorical Objections to Genetic Engineering—A Critique,” in Ethics and Biotechnology, edited by A. Dyson and J. Harris, (London and New York: Routledge, 1994), 202-215, esp. 205-209. 10. Devlin, op. cit., 80. 11. This point has been made and thoroughly defended by Heta Häyry in her “How to Assess the Consequences of Genetic Engineering,” in Dyson and Harris, eds., op. cit., 144-156. The following paragraphs have been borrowed from M. Häyry and H. Häyry, “Genetic Engineering,” in Encyclopedia of Applied Ethics, vol. 2, ed. R. Chadwick, (San Diego: Academic Press, 1998), 407-417, esp. 408f. 12. A more detailed analysis of ‘risk’, and a more comprehensive account of our views regarding its acceptability, can be found in M. Häyry and T. Takala, “Genetic Engineering and the Risk of Harm,” Medicine, Health Care and Philosophy 1 (1998): 61-64. 13. Our thanks are due to Professor Robin Attfield, University of Wales Cardiff, whose invitation to prepare this paper for the Twentieth World Congress of Philosophy gave us a welcome opportunity to apply the main theories of ethics to the questions of genetic engineering and our natural environment.

BEYOND BIOPHOBIC MEDICAL ETHICS: WHAT’S THE MERCY IN MERCY-KILLING?

A genuine bioethics would be fiercely devoted to human life (bios) and would express that devotion by articulating as well as advocating moral virtues that rigorously protect that value against the temptation to see life in purely instrumental terms. In my view, no genuine bioethics exists today. In what follows, I will question two fundamental assumptions often presumed in discussions of euthanasia and assisted suicide. These are (i) the agent does will her victim (i.e., her putative beneficiary) some significant

Jorge L. A. Garcia

human good, e.g., relief from pain, escape from becoming a burden to loved ones, a dignified death, or simply self-determination; (ii) in purposely helping someone to kill herself or in killing her for her own good, the agent wills her no serious harm. Put differently, I question the assumption of ‘mercy’ in so-called ‘mercy-killing’.

INTRODUCTION

A

genuine bioethics, one worthy of the name ‘life-ethics’, would be fiercely devoted to human life (bios), and would express that devotion in articulating and advocating moral virtues—acquired dispositions of feeling and action—rigorously protective of that value, even in extreme circumstances, against the temptations to see life as merely of instrumental value. Suffice to say that we have no such bioethics today. When philosophers weigh in on the issues of life—as in the notorious “Philosopher’s Brief” signed by reputed moral thinkers in the United 179

J ORG E L. A A.. GARCIA ORGE 180 States in 1997—it is usually to help rationalize the West’s biophobic Zeitgeist, speaking up for euthanasia, suicide (assisted and self-service), abortion, and so on. I wish here to take a small, preliminary step toward a more genuine bioethics by questioning some assumptions too often accepted without challenge in discussions of voluntary (and involuntary) euthanasia and suicide assistance. These are (i) that the agent does will her victim (i.e., her putative beneficiary) some significant human good: relief from pain, escape from becoming a burden to loved ones, a dignified death, or simply self-determination, and (ii) that in purposely helping someone to kill herself, or in killing her for her own good, the agent wills her no serious harm. It is thought that mercy-killing is a mercy. I want to call this assumption into question by raising doubts both about whether death itself can be a good for someone (taking the ‘eu’ out of euthanasia), and about whether the things for whose sake people are euthanized or assisted in suicide are really important goods for them.

I. WHAT GOOD DOES MERCY-KILLING AND S UICIDE ASSISTANCE D O? Pain-relief. It is said that these measures provide patients with relief from their suffering and that this is a great service to them. However, real relief from pain is a benefit to someone only insofar as it improves her life. Ending her pain by ending her life does not really relieve her pain and it certainly does not improve her life. It becomes difficult to see this or how this does the patient good. I have developed this point at greater length elsewhere and won’t rehearse that discussion here. 1 Permit me only to observe that, my critics notwithstanding, I see no commitment in my position to the thesis that life has “intrinsic value,” though I should affirm that its value is more than instrumental. Also, I see little merit to my critics’ suggestion that, even if I am right that someone cannot be better off dead, she may nonetheless be “worse off alive.” Can someone be ‘worse off alive’ even when not ‘better off dead’? Certainly, we need an explanation of why this is not, as it appears to be, merely a misleadingly better-sounding redescription. Why should we think that the statement ‘S is worse off in Circumstance C1 than S is in C2’ implies that ‘S is better off in C2 than S is in Circumstance C1’? Compare the emptiness of this advice. Don’t say, ‘The number 7 is heavier than an orange’ because that implies that the number has weight; just say, ‘An orange is lighter than the number 7’ instead. Is there any reason to think the recommended formulation an improvement? 2

B EYOND BIOPHOBIC MEDICAL ETHICS 181 Not burdening others. Some maintain that resorting to such interventions as suicide-assistance and mercy-killing benefits the patient by enabling or helping her to avoid becoming a burden to others, whether it be loved ones, close relatives, or society more generally. However, human beings are always dependent on others to some extent, and especially so at the beginning and end of life. We are morally obligated to help one another, according to any plausible moral theory, and it is good for us to do so in a variety of ways. Being dependent on others helps a person to overcome certain forms of the vice of pride, traditionally, the demonic vice. It builds community within networks of dependency, helps individuals to overcome the temptation, at first exhilarating but often dispiriting in the end, to see themselves as a burden. It may be thought that our dependence on others for food, provision of minor comforts, and so on, places one in an undesirable and precarious position. This, however, is not necessarily so. Anyone is better off dependent in part on others than dependent solely on oneself. Others have more resources to help me than I have alone, and they may also have not only inclination, but often more inclination to help me than I have myself. This is perhaps most often encountered in cases where— as is oftentimes the case—the medically afflicted are tempted to succumb to despair, shame, or self-disgust. Of course, whether others care enough about me to help me will depend largely on how I have treated them. In this way, the web of mutual dependence and help tends to benefit the agent as well as the beneficiary. It can build bonds of gratitude and feelings of indebtedness, which can motivate others, especially loved ones, to help those who earlier helped them. Being on the receiving end of such help for one’s own children may, it is true, be experienced as humiliating by the now-dependent parent, who may prefer to think of herself always as the one helping, never the one helped; always the one with the surplus of energy, resources, and concern, never the one who is needy. Puncturing this sort of self-importance is a significant opportunity for moral growth, for a deepened appreciation of the human condition of mutual dependency, and for a heightened ability genuinely to sympathize and to feel with the needy. For many, it is their final opportunity, one they—and we—should not be deprived of. Thus, pandering to a patient’s fantasies of independence is an ignoble business, not a real service to her at all. It is important for all of us to learn and to relearn that human beings are valued and worth caring for no matter what their medical condition. Dignified dying. Nowadays, it has become commonplace to defend suicide assistance and mercy-killing for what they do to secure for the patient a death with dignity, a dignified end of life. However, this presupposes an unanalyzed and undefended view of dignity. Genuine

J ORG E L. A A.. GARCIA ORGE 182 indignity is an improper response to dignity: that is, it consists in disrespectful treatment. To live, at the end, a life of dependency and some indelicacy—incontinence and all the rest—can be embarrassing and even humiliating. However, by itself it is not the sort of thing that can constitute a loss of dignity. Here, what we need to do is to remind the one suffering that her dignity as a human being does not depend on these details, but remains intact. On the contrary, we offend another’s dignity by adopting (and thus encouraging her to adopt) the view that a life or a phase of life characterized by such dependence is without dignity. That view must be either dehumanizing or antihumanistic. It is dehumanizing when it recognizes that dignity inheres in every human being in virtue of her nature and irrespective of the vagaries of her condition; it is antihumanistic when it is does not. Felicia Ackerman has been helpful in exposing efforts by the devotees of medicalized killing to introduce a new type of dignity that, allegedly, is not inherent and is lost in unfortunate circumstances. 3 Of course, we are all familiar with socially-recognized forms of dignity accorded to great personages. What is doubtful is that these matters of social status can bear the needed moral weight to be conclusive in such dire situations as those of mercy-killing. Respecting autonomy. Mercy-killing and suicide assistance, it is maintained, help the patient because, in so behaving, the agent honors and preserves the patient’s autonomy by allowing and facilitating her self-determination. I cannot here pursue in any adequate detail the complicated matters of the nature, scope, and moral importance of autonomy. I will have to make do with only a few observations. Plainly, this position assumes a wide scope and high value of autonomy, and that needs to be defended. Here, I suspect, too often those who appeal to autonomy-based arguments in defense of homicidal ‘medical’ procedures are trading on the august associations the term acquired from Kant’s magisterial treatment. This is unfair and illegitimate, because our contemporaries’ conceptions of autonomy differ in some crucial particulars from Kant’s conception. The demands of Kant’s autonomy were the same in all human beings. They were rooted in pure practical reason. They were free from determination in and by human appetites or interests. They inherently observed the moral law grounded in reason. The autonomy that even today’s neo-Kantian theorists defend is nothing like that. Why, then, ought we defer to it so much, even to the extent of offering it our neediest, most desperate, and despairing as human sacrifices? The Philosophers’ Brief approvingly quotes the majority opinion of the United States Supreme Court in its Cohen decision, which grounded some freedoms in the belief that “no other approach

B EYOND BIOPHOBIC MEDICAL ETHICS 183 would comport with . . . individual dignity.” 4 About this, note, first, the irony in appealing to dignity to ground actions premised on the view that some people’s lives are not worth living or saving. Second, it raises the question: why decisions about how long to live, but not about how much money to have, etc., are ones to which we should necessarily accede? Again, The Philosophers’ Brief, quoting the Court this time in its Casey decision, finds decisions about death “central,” “intimate,” and deeply “personal.” It is true, undoubtedly, that deciding whether to live or to die soon has far-reaching effects. Yet, so can many decisions, such as choosing marriage or a career. That cannot be enough to show that all decisions about when to die—let alone, about whether to live—are legitimate. Similarly, The Brief affirms that “we want the last act [of life] to reflect our own convictions.” 5 Again, this is true, but insufficient. For it is not the case that the frustration of every desire is a moral wrong. What requires that this one remain inviolate? Why should we think that autonomy properly extends beyond the uncontroversial range of some questions of how to live to questions of how long and whether to live? What my respecting your dignity seems to require is that I allow you generally to shape your life, including determining some elements of its content. Setting its temporal boundaries is a very different matter and it needs to be argued that this also is properly within your rights. To do that, we need to decide what the point of autonomy is that requires or licenses this extension. I should think autonomy can have no basis other than human dignity. However, as we observed, frequently it is precisely a denial of the agent’s dignity—the idea that her life is not worth living, let alone, saving—that underlies and motivates mercy-killing. In any case, here we confront a variant on the old Lockean problem of whether we are free to do such things as enslave ourselves, therein ending the freedom from which the decision to become a slave originated and from which it derives whatever legitimacy it had. As Locke rightly rejected such a right to permanently destroy one’s own freedom, it is hard to see how we serve someone’s autonomy in acts that destroy that autonomy by destroying one’s self. I am aware that my remarks here have been brief and programmatic. My intent is not to refute anything. On the contrary, I wish only to raise problems and therein to help open discussion of the ways in which mercy-killing and suicide-assistance are alleged to benefit the patient rather than simply benefit her relatives who may wish to get on with their lives, and her insurers who wish to cut costs. I think the claim of mercy-killing to be a mercy has gone underexamined too long. My remarks here are intended to begin the process of correcting that.

J ORG E L. A A.. GARCIA ORGE 184

II. DOES KILLING SOMETIMES BENEFIT THE ONE W HO DIES? I have so far called into question the claim, usually assumed by both sides in the controversies over mercy-killing and suicide-assistance but seldom defended, that these quasi-medical procedures in fact help the patient in various ways. There are also deeper and more systematic reasons to insist that death is never a benefit. A human being, whatever else she may be, is a biological organism. Death could no more benefit an organism than could dissolution benefit a team, disintegration benefit a corporation, or disorganization benefit an organization. Like a team’s dissolution or an organization’s disbanding, death destroys the very conditions under which the entity it befalls could be benefited. This is no advantage. Perhaps a life can be more appealing for being shorter: it is not so much, as it is sometimes said, that the patient is better off dead, but that it is better for her to die sooner rather than later. It seems to me that this can be no more than an aesthetic consideration. Indeed, The Philosophers’ Brief lends this view unintended support when it talks of some ends “disfigur[ing]” lives.6 Still, such an aesthetic improvement in someone’s life, an increase in its aesthetic value, does not show that the decedent benefits from dying earlier, even if her life is prettier for that. I do not deny that we can understand, to some extent, preferring death to life in some circumstances. Most of us have moments like that. Again, the fact and partial comprehensibility of this preference does not show that it is justified. Still less, it does not show that the decedent is objectively better off for having died earlier. That is what I deny here.

III. DOES KILLING ALWAYS HARM THE ONE W HO DIES? To my mind, the best reason for thinking that death is never a benefit to the one who dies is that it is, instead, always a harm to her. Cannot something be both a benefit to someone and also, at the same time, a harm to her? Yes, it can. However, if I am right, death is a harm of so fundamental a sort that it is implausible to maintain, as the partisans of assisted suicide and mercy-killing must, that, whatever harm it does, death is in these circumstances a net benefit, or that it is the better choice. Allow me to offer a few reasons in support of the claim, too often denied outof-hand, that death is inherently a grievous harm to the patient even in cases where some find mercy-killing or suicide-assistance attractive.

B EYOND BIOPHOBIC MEDICAL ETHICS 185 Too often, people assume that death is a harm only sometimes, and then only when it deprives them of what Fischer calls the “goods of life.” In short, they see it as, at worst, only instrumentally bad for someone. However, the chief harm in death is that it is a loss of the good of life. This loss is built into death’s nature, not externally or otherwise causally connected to it. There are, of course, puzzles about who suffers the loss and when. 7 Yet, I confess that I cannot see their force. The person who suffers the loss is the one who dies, though she “suffers” it only in the sense that she undergoes it—better, it happens to her—not in the sense that she experiences it. I suspect that the supposed quandary about when she undergoes this loss rests on a confusion. Death is a loss in that there is some good that the decedent had and would have retained if she were alive but, because she is dead, it is not the case that she continues to have it. Unlike most other losses, it is true, we cannot say that she lacks the good in the (later) actual situation. We cannot say that she has to go on without it. But, however common, those further claims are not necessary for loss. Given that this is that in which her loss consists, it is not clear to me what puzzle is supposed to be contained in the question when the loss of this good occurs, or when the decedent suffers it (has it happened to her). Presumably, she loses it when it ceases to be the case that she has it. That is when she dies. True, when I lose a wallet, in addition to the sense in which I lose the money when I lose the wallet, there is also a sense in which I suffer the loss of the money at a later time when I need it. There is nothing like that in the case of death. However, this dissimilarity simply makes death a loss of a different type, hardly a surprising discovery. I cannot see why it should render problematic the claim that it is a loss. We might also ask what the good is that is always lost when someone dies. My suggestion is that the answer is the obvious one: the good is life itself. To put the point another way, what is lost is the person herself, and she is lost not merely to the world but to her very self. Here, I think, people are misled by some philosophers’ recent emphasis on the supposed importance of “having a life” in contrast to the alleged unimportance of merely “being alive”. This privileging of what is called “biographic” life over “biologic” life seems to me merely elitist. It suggests that only those whose existence is filled with the incident and drama that make for good reading have lives, as in the condescending advice “get a life!” which the stylish smugly throw at the less fashionable nowadays. As we observed above, death is always a loss to the one who dies. Therein it is also a harm, and an especially grievous one. First, death is inherently the greatest diminution in health. This follows immediately

J ORG E L. A A.. GARCIA ORGE 186 from the appealing conception of health as consisting in the comprehensive proper functioning of the organism and its subsystems. Death puts an end to all these. It is the limit case of the loss of health. Moreover, death is inherently a grave harm because it is by its nature contrary to the autopoesis toward which life inherently strives. Even if we do not interpret talk of biological teleology strictly, it is commonly recognized that living beings of whatever type tend to evolve in the direction of greater survivability. This is true to such an extent that talk of biological teleology is frequently parsed as talk of what tends to enhance the survival of the organism (whether or not also in the direction of the gene-line’s or the species’s survival). From that perspective, it is hard to see how death, the organism’s destruction, can be other than a grave harm to the organism: it is that in virtue of whose promotion other harms are deemed to be harmful. For what it is worth, it appears that even that account of good, according to which a good is what no rational person will avoid without reason, leaves open the possibility that death is inherently good. 8 After all, why should anyone think it rational actually to avoid (flee) life (that is, an additional period of being alive), when there is no special reason to do so? The interesting question is whether this flight is justified even when there appears to be such reason. (Though my remarks in the previous sections were designed to distinguish that appearance from reality.) Let me conclude this section with an observation. As we said above, mercy-killing and suicide-assistance are often defended on the grounds that they bring the decedent good by permitting her to escape suffering pain, to escape the unwelcome sense that she has become a burden to her loved ones, and to escape the frustration of having her wishes for her life thwarted. It is ironic that today’s antihumanism ethicists thus accord inherent value to subjective states while withholding it from the human subjects themselves who have those states, and from the human lives in which they occur. That death inherently runs counter to the organism’s health and natural orientation is of special practical importance for medical ethics, because health must be the chief goal of medical practitioners qua medical agents. So, even if death were a benefit to the patient, it would remain egregiously illegitimate for physicians, that is, healers, to pursue it (even as a means). Against this fact it is argued that it proves nothing, because the profession and its goals may change. That is true, but what matters is that there are also limits to intelligible change. It makes no more sense to suppose that it could become part of the professional conduct or responsibility of the healer to try to kill her patient

B EYOND BIOPHOBIC MEDICAL ETHICS 187 than it makes to suppose it can become part of the professional role of a teacher or of a journalist to try to mislead.

CONCLUSION For these reasons, I maintain that euthanasia and assisted suicide will often secure their patients (really, it should be clear, their victims) no significant good. These quasi-medical procedures, in contrast, must always deprive those victims of good; they must harm them. I conclude by raising the problem to which this leads: that of the ethics of life with bioethicists. In the current state of professional ethics discourse, we need much more often to have recourse to moral criticism of our positions. To remind a colleague that this or that proposal would be monstrous, not only for us to adopt, but even for her to suggest is simply ‘not done’ in the current climate. To do so would violate the rules of a kind of politesse. I think that both the content and the rules of discourse in professional ethics merit more reflection and attention than they have heretofore received. Elsewhere in the academy, we hear calls for a new sensitivity to the marginalized and the voiceless. It is a sad and revealing irony that it is in ethics programs in various centers, institutes, and schools of higher education that we still hear some people derided as “vegetables,” others classed as “marginal humans,” of the needs of some dismissed as parts of lives “not worth living” (and, by implication, not worth saving), and of suggestions that we should vivisect those who are brain-damaged and who exist in what is cruelly called a persistent ‘vegetative’ state. There is a need for professional ethicists to be reflective about, rather than merely to reflect, the cultural elements in which they operate. To do nothing in such a context is to do something; it is to abide by, and thus implicitly to endorse, our debased status quo. The professional ethics establishment ill serves scholars for public service in the field when it proceeds with no admonitions about the extent to which their academic speculations have distanced them from what is left of civilized and respectable moral opinion. Certainly, the public is ill served when it entrusts decisions over life, death, health, and justice to a clerisy of academic elites trained to evaluate moral positions mainly for the cleverness, rigor, and originality with which they are formulated, oftentimes with scant attention paid to the indecency of their conclusions and the moral poverty of the thought that informs them. Scholars who talk of ‘veterinary’ aspects of neonatology, of human ‘vegetables’, of ‘marginal’ humans, and so on, need to know

J ORG E L. A A.. GARCIA ORGE 188 that their language and the thinking behind it are despicable and deserve chastisement. Surely, it degrades our institutions when such as these staff our medical institutions, advise our governments, or teach our young. Their positions need to be argued against, of course, but these thinkers also need to be remonstrated with, reprimanded, and ultimately called to repentance. The question is how to do this. So, I ask for reflection on the glum practical questions of how to live civilly—without giving offense and without betraying the offended—among the current crop of medical ethicists, especially the growing number with the most repugnant views. We need also to reflect on how to sensitize the applied ethics industry’s degraded intellectual culture, which allows people to think they can talk, act, and feel in these ways as if that were decent. Jorge L. A. Garcia, Department of Philosophy, Rutgers, The State University of New Jersey, New Brunswick, NJ 08901-2882; [email protected]

NOTES 1. See my “Better Off Dead?” American Philosophical Association Newsletter on Philosophy and Medicine 92:1 (1993): 85-88. 2. See John Messerly, “Worse Off Alive: Reply to Garcia,” APA Newsletter on Philosophy and Medicine 94:1 (1994): 121-122, as well as Gary Seay, “Worthless Life,” APA Newsletter on Philosophy and Medicine 93:1 (1994): 87-89. 3. Felicia Ackerman, “Assisted Suicide, Terminal Illness, Severe Disability, and the Double Standard,” in Physician-Assisted Suicide: Expanding the Debate, ed. Margaret Battin, et al. (London: Routledge, 1998). 4. Ronald Dworkin, et al., “Assisted Suicide: The Philosophers’ Brief,” New York Review of Books (27 March 1997), 43 and note. See also George Fletcher, M. Hornick, and R. Dworkin, “‘The Philosophers’ Brief’: an Exchange,” New York Review of Books (29 May 1997). 5. Ibid., 44. 6. Ibid., 45f. 7. See John Martin Fischer, ed., The Metaphysics of Death (Stanford: Stanford University Press, 1993), passim. 8. Bernard Gert, Morality (Oxford: Oxford University Press, 1988).

DEATH, DYING, AND DIGNITY

The word ‘dignity’ is a staple of contemporary American medical ethics, where it often follows the words ‘death with’. People unfamiliar with this usage might expect it to apply to one’s manner of dying—for example, a stately exit involving ceremonial farewells. Instead, conven-

Felicia Ackerman

tional usage generally holds that “death with dignity” ends or prevents life without dignity, by which is meant life marked not by buffoonery, but by illness and disability. Popular examples of dignitydepleters include dementia, incontinence, and being “dependent on machines”—provided the machines are respirators rather than furnaces, refrigerators, and computers.

T

he word ‘dignity’ is a staple of contemporary American medical ethics, where it often follows the words ‘death with’. People unfamiliar with this usage might expect it to apply to one’s manner of dying—for example, a stately exit involving ceremonial farewells. Instead, conventional usage generally holds that “death with dignity” ends or prevents life without dignity, by which is meant life marked not by buffoonery, but by illness and disability.1 Popular examples of dignitydepleters include dementia, incontinence, and being “dependent on machines”—provided the machines are respirators rather than furnaces, refrigerators, and computers. In Sir Thomas Malory’s Le Morte D’Arthur, published in 1485, Blamor, overcome in battle, asks his adversary, Tristram, to slay him, saying, “I have lever [would rather] die with worship [honor] than live with shame.”2 My students often consider this barbaric. Will future generations

189

F ELICIA ACKERMAN 190 consider “Please help me die because I would rather die than live with the degradation and indignity of incontinence and dependence” barbaric? Should they? Whether they will is obviously a speculative empirical question where evidence is unattainable. So I will focus on issues related to the philosophical question of whether they should. This question is important both for theoretical and for practical reasons. It raises theoretical issues about the concept of dignity, about when it is justified to feel shame and degradation, and about when it is moral for a society to endorse people’s feelings of degradation. It raises practical issues because the concept of dignity is often invoked in connection with two major present-day movements concerning the terminally ill. One of these movements is highly controversial. This is the movement to legalize physician-assisted suicide, a movement holding that “it is intolerable for government to dictate that doctors may never, under any circumstances, help someone to die who believes that further life means only degradation.”3 Margaret Pabst Battin, an advocate for the legalization of physician-assisted suicide, has offered the following example of such degradation, an eighty-year-old widow who is almost completely blind from glaucoma, suffering from colon cancer, living in a nursing home, and whose unfortunately realistic picture of old age suggests that she can expect increasing debility, dependence, financial limitation, loss of communication and affection, increasingly poor self-image, increasing depression, isolation, and, due to her glaucoma and cancer, blindness and pain . . . given her physical condition and the social conditions of [her] society [which is presumably contemporary American society] . . . her options are . . . only two: suicide, or the catalogue of horrors just described. Suicide, then, may be constitutive of human dignity in . . . [that] it leaves one less example of human degradation in the world.4

Of course, suicide itself, let alone this sort of “dignity” rationale for it, is highly controversial. Far less controversial is the other prominent present-day movement concerning the terminally ill. This is the hospice movement, which enjoys tremendous uncritical popular support in the United States. The hospice movement eschews any treatment for the terminally ill, even if the treatment is not painful or otherwise disagreeable, whose sole aim is to prolong life. The hospice movement claims to provide “the most effective and least expensive route to a dignified death.”5 Assessing these claims of hospice spokesmen and of assisted-suicide advocates obviously requires careful attention to the concepts of dignity at issue. Battin, perhaps in anticipation of the objection that it

D EA TH, D YING, AND D IGNIT Y EATH IGNITY 191 would be cruel, bigoted, and superficial to categorize the ill old lady of the foregoing example as deficient in human dignity, distinguishes between the “originally Kantian” sense of ‘dignity’, in which “all human beings have dignity, or what we might describe as intrinsic human worth” 6 and a “second, empirical sense of dignity as well, which corrupt or abused persons may not have,” 7 and stresses that it is only in the “empirical” sense that people like the woman in the foregoing example count as lacking dignity or being degraded. But Battin also says of this “empirical” sense of ‘dignity’ that we can distinguish characteristics of the individual himself or herself which show that he or she achieves dignity as a human being. For instance, we can distinguish between the person who bears pain with dignity and the person who, in fright or panic, does not.8

This suggests that even Battin’s “empirical” sense of ‘dignity’ is partly normative in that to say that someone fails to “achieve dignity as a human being” is, unsurprisingly, to say something pejorative about him.9 So there is a serious moral problem in saying the suicide of the sick old woman in the foregoing example serves the interests of dignity, even if one grants, as Battin does, that choosing to stay alive under such circumstances could also exhibit dignity.10 Battin holds that suicide can serve the interests of dignity “[w]hether the threat to one’s self-ideal is from physical illness and pain . . . or from the destruction of other persons or values upon which one’s life is centrally focused,” 11 such as Cato’s suicide and David Wood’s example of “the suicide (by jumping) of an architect who pioneered high-rise apartment buildings and realizes too late what he has done [provided that his suicide is] a considered, courageous statement of principle, a dignified final act transcending [his] own defeat.”12 This raises the question of whether suicide could serve the interests of dignity when committed by a healthy young adult whose “self-ideal” requires that he get tenure (or that he get tenure at Harvard), or that he be a major-league baseball player, or handsome, or a millionaire.13 Such people could doubtless use suicide assistance in the form of prescriptions for painless lethal drugs. No one forms his ideas about dignity or his “self-ideal” in a vacuum. Legalizing physician-assisted suicide for people who seek to avoid the “indignity” of old age, illness, and disability, but not for people who seek to avoid the indignity of failure to achieve success in the major leagues of academia or of baseball would not only reflect but also promote such discrepancies in what people are apt to find degrading. At a World Congress of Philosophy, it seems particularly appropriate to suggest that we consider the foregoing case of the ill old woman from the standpoint of cultures that respect the

F ELICIA ACKERMAN 192 aged. What does it show about American society that so many of its intellectuals are ready to endorse the ideas that the illnesses and disabilities of old age reduce one’s human dignity and that killing oneself under such circumstances can promote this dignity? Healthy young adults who prefer death to the indignity of professional mediocrity do not figure at all in Ronald Dworkin’s book, Life’s Dominion, although the book discusses both cases of the terminally ill and cases of the severely and permanently disabled and does not differentiate between them in arguing “for individual freedom . . . for a régime of law and attitude that encourages each of us to make mortal decisions for himself.” 14 And the Philosophers’ Brief for the legalization of assisted suicide takes an even narrower view of when “it is intolerable for government to dictate that doctors may . . . [not] help someone to die who believes that further life means only degradation.”15 It is only for the terminally ill that The Brief advocates legalizing physicianassisted suicide. It explicitly leaves open the question of legalizing assisted suicide for the severely and permanently disabled. And the possibility of legalizing physician-assisted suicide for healthy people who have nonmedical reasons for wanting to die is mentioned only in dismissive terms; the introduction to the Brief gives “a sixteen-yearold suffering from a severe case of unrequited love” 16 as a clear instance where suicide assistance should not be permitted. On the face of it, this may seem reasonable, as may the rationale that “[s]tates may be allowed to prevent assisted suicide by people who— it is plausible to think—would later be grateful if they were prevented from dying,”17 or who “would one day be glad [they were] forced to stay alive.”18 As I have argued elsewhere, 19 however, it is unreasonable to suppose that this criterion can be met only by the terminally ill, or even only by people who have medical problems at all. Moreover, although Dworkin is the author of the Philosophers’ Brief’s introduction, where the rationale is offered, this rationale overlooks an important distinction he has made elsewhere and earlier. This is the distinction between experiential interests, i.e., interests in having or avoiding certain sorts of experiences simply because of the pleasure or misery of having them, and critical interests, i.e., interests whose satisfaction makes one’s life as a whole a better (more successful, non-wasted) life.20 For example, someone who seeks death simply in order to end or avoid the experience of physical pain has an experiential interest in dying; someone who says he would rather die than undergo the “indignity” of living in an irreversibly comatose or demented condition is affirming a critical interest in dying rather than being kept alive in such a condition.

D EA TH, D YING, AND D IGNIT Y EATH IGNITY 193 If someone wants to die because he has an experiential interest in avoiding future unhappiness, it obviously matters if he is likely to be wrong in his prediction that he will, if remaining alive, be unhappy in the future. But someone who has a critical interest in not being kept alive in a demented state may consider the possibility of his happiness in such a state irrelevant. He may even envision the dementia as enjoyable, but, in accord with conventional values, degrading. Imagine a terminally ill person with a brain tumor that can be predicted to grow in such a way as to make him demented but happy and glad to have been forced to stay alive. Imagine also that he regards this prospect with horror and wants assistance in committing suicide in order to avoid it, and that he is rational, has thought the situation over carefully, and is not being pressured by anyone. Such a case would pose a dilemma for the view expressed in the Philosophers’ Brief. Outlawing suicide assistance in such a case would make this assistance illegal for someone who “believes that further life means only degradation,” 21 and who, since the empirical basis of his belief is sound, would be impeded in the exercise of what the Brief’s introduction says it advocates, “the right to make momentous personal decisions which invoke fundamental religious or philosophical convictions about life’s value for himself.”22 But allowing such a person suicide assistance would amount to allowing it to someone who, it is plausible to think, if denied such assistance would later be glad he was forced to stay alive.23 Would it also amount to social endorsement of the view that certain kinds of lives have more dignity than others, and that avoidance of “genuine” indignity should be a factor in deciding who should be permitted suicide assistance, rather than what Dworkin says he favors, “a régime of law and attitude that encourages each of us to make mortal decisions for himself”? 24 That depends on whether the society in question would also legalize suicide assistance in other cases—for example, for someone who thinks life without tenure at Harvard is beneath him, and who regards as intolerable degradation the likely prospect of his ever becoming happily reconciled to such a situation. This point applies to physical disabilities as well. For example, Dworkin says some people “think it degrading to be wholly dependent . . . At least part of what people fear about dependence is its impact . . . on their own dignity.” 25 The Philosophers’ Brief cites many people’s attitudes toward being “intubated” and “helpless” in support of its position that “it is intolerable for government to dictate that doctors may never, under any circumstances, help someone to die who believes that further life means only degradation.”26 Dr. Timothy Quill holds that “suicide could be appropriate for patients if they did not want to linger

F ELICIA ACKERMAN 194 comatose, demented, or incontinent.” 27 None of these people has suggested legalizing physician-assisted suicide for those who consider professional mediocrity to be beneath their dignity. So what seems to be operating here is not just advocacy of self-determination, but endorsement of the view that incontinence and “dependence” are degrading. But do assisted-suicide advocates really want to endorse the view that human dignity resides in the bladder or rectum? If being unable to control the discharge of one’s urine or feces deprives one of human dignity, then what about being unable to control the discharge of one’s menstrual blood? Should physician-assisted suicide also be legalized for all premenopausal women who believe that this inability undermines their dignity and that the “remedy” of a hysterectomy would also undermine their dignity? Not long ago in American history, most white Americans failed to recognize the human dignity of black Americans. If there are black Americans nowadays who want to die because they buy into this, should suicide assistance be legalized for them? If not, why legalize physician-assisted suicide for people who believe it is their incontinence, rather than their skin color, that deprives them of human dignity? A similar point applies to people who believe it is their “dependence” that undermines their dignity. The notion of “dependence” here cries out for philosophical examination. As Susan Wendell points out, ‘independence’ is “defined according to a society’s expectations about what people ‘normally’ do for themselves and how they do it.”28 She adds that few people in her city would consider her a “dependent” person because she relies on other people to provide her with water out of the tap and electricity, but most would consider her highly dependent if she needed other people to help her get out of bed or go to the toilet. She says that “[t]he philosophical arbitrariness of our ideas concerning which of us is ‘independent’ seems obvious,”29 but makes the much less plausible suggestion that [p]sychologically [our ideas about who is independent] may not be as arbitrary; many of us seem to be priding ourselves on not needing the same kinds of physical help as infants. Perhaps if we could feel and show more respect for children, it would not be as important to distinguish ourselves from them.30

This seems off the mark. People do not generally speak disparagingly of infants for needing such assistance or suggest that such dependence undermines infants’ dignity. In fact, such “dependence” is socially acceptable in infants but not in adults in American society.31 Note also that Wendell’s explanation cannot apply to the parallel illogicality in our judgments about being “dependent on machines.” Few people in my society call me dependent on machines because I rely on

D EA TH, D YING, AND D IGNIT Y EATH IGNITY 195 a furnace and a refrigerator, and virtually no one in the mainstream considers such reliance degrading. But most would call me dependent on machines if I and relied on a respirator, and many if not most would consider such reliance degrading. Obviously, the analogy with children is inapplicable here. In both imbalances, the psychological explanation seems to be bias against the ill and disabled, rather than any analogy between the ill and disabled on one hand and some other group on the other. This parallels the imbalance I have already discussed—the imbalance between many people’s willingness to legalize physician-assisted suicide for people seeking to escape the “indignity” of illness and disability but not for people seeking to escape the indignity of professional mediocrity. These points are especially important in view of Battin’s admission that the prospect of “increasingly poor self-image” is part of what makes suicide in her case of the ill old lady serve the interests of dignity. Is there anything particularly dignified about a sick old woman’s “internalizing” (to use a popular bit of psychological jargon that would make Malory turn over in his grave) her society’s negative view of her to the point where death seems a desirable option? The foregoing remarks do not entail opposition to legalizing physician-assisted suicide. They simply count against the inegalitarian position that would legalize physician-assisted suicide on grounds of “dignity” only for the terminally ill, or only for the terminally ill and for the permanently disabled and the old, debilitated, etc. Such selective legalization, despite its rhetoric of self-determination, cannot be justified except by endorsing the idea that such people “achieve dignity as human beings” less than do the young, able-bodied and healthy. As I have argued elsewhere, physician-assisted suicide should either be illegal for everyone or legalized for all competent adults. 32 Daniel Callahan has also argued that “[i]f we really believe in selfdetermination, then any competent person should have a right to be killed by a doctor for any reason that suits [this competent person].”33 But Callahan intends this as a reductio ad absurdum. He claims that in order to be what he calls “responsible moral agents,”34 rather than what he sneeringly dismisses as “simply hired hands with lethal injections at the ready,”35 doctors “would have to decide, on [their] own, whether the patient’s life was ‘no longer worth living.’”36 Such a claim is diametrically opposed to Dworkin’s liberal vision of “a régime of law and attitude that encourages each of us to make mortal decisions for himself.”37 To be a responsible moral agent by the standards of this liberal régime, a doctor who provided suicide assistance would no more have to decide which patients’ lives were still worth living than would a skydiving instructor have to decide which prospective students’ lives were too

F ELICIA ACKERMAN 196 high-quality to be risked by such an activity. (Technically, when a doctor gives a lethal injection to a patient who requests it, this is voluntary active euthanasia rather than physician-assisted suicide, but Callahan sees these practices as morally on a par, 38 and in this I agree with him.) Callahan’s attitude toward such self-determination is indicated by his article’s title, “When Self-Determination Runs Amok.” But if you are not supposed to control your own death, who or what is? God? This answer cannot be a basis for law in a society without established religion. Nature? Then why take antibiotics or get yourself vaccinated? Other people? Well, whose life is it, anyway? Callahan claims that assisted suicide and voluntary active euthanasia, “because they entail the assistance of another” 39 are not “acts solely expressive of individual autonomy.” 40 The same could be said of sexual activities between consenting adults, but this does not keep laws against such activities from being an invasion of privacy. Moreover, Callahan is not consistent in his view that “[i]t is not medicine’s place to determine when lives are not worth living.”41 When the issue is “letting die,” 42 rather than “killing,” he is willing, in fact eager, to let medicine make such quality-of-life judgments. Thus, he says “[p]atients must have a strong assurance that they will, if they want, and if it is appropriate, be allowed to die.” 43 He even proposes “a number of relatively straightforward ways to make decisions based on quality-of-life standards,” 43 and disparages people “who end their days incompetent and grossly incapacitated” 44 as being “more dead than alive.” 45 Although I have doubts about Callahan’s view of the significance of the distinction between killing and letting die in cases where patients have requested death, this issue is logically independent of the question of who, if anyone, should be making quality-of-life judgments.46 The latter issue also arises in connection with the hospice movement, a movement which, as I have indicated, is far less controversial than the assisted-suicide movement. As I have mentioned, hospices bill themselves as “the most effective and least expensive route to a dignified death.”47 So it is important to look carefully here. One fundamental principle of what the National Hospice Organization calls its “philosophy of hospice” is that “[w]hen death is inevitable, hospice will neither seek to hasten it nor to postpone it.”48 But the principle fails to indicate what counts as “postponing death.” Bringing food to the bedside of a bedridden patient? Feeding by hand a patient who can swallow but is too weak to lift a fork? Feeding intravenously or via a feeding tube? Elsewhere, the National Hospice Organization defines “prolongation of life” as “[u]sing artificial or other medical means to extend a patient’s life beyond what would otherwise be the time of natural death,”49 but it is still

D EA TH, D YING, AND D IGNIT Y EATH IGNITY 197 not entirely clear what this amounts to. Furthermore, the rationale for the principle is obscure. Hospice care can occur in the home (the most usual setting in the United States), or in a hospital-based unit, nursing home, or freestanding hospice facility. 50 All these settings, at least in the United States, are generally well-equipped with air-conditioning and central heating, electric lights, and window shades. Clearly, hospice philosophy is not opposed to hastening or postponing the onset of indoor light, darkness, warmth, or coolness. Why does the National Hospice Organization regard death differently? Why does its “philosophy of hospice” entail what elsewhere I have called a “Goldilocks Principle” for terminal illness (death by assisted suicide is too soon, death after high-tech life-prolonging treatment is too late, “natural” death is just right), when hospices do not eschew intervention through technology or other forms of human ingenuity in other areas? 51 If hospices can make a terminal patient’s life so comfortable, why are they so unwilling to prolong it? Minimizing pain and maximizing comfort are major hospice concerns, but they do not inherently preclude extending life by means of high-tech life-supports. Since “hospice philosophy” does explicitly preclude this, however, its claim to provide the most effective route to a dignified death shows bias against the ill and disabled by denigrating the dignity of those who live by means of such supports. 52 Felicia Ackerman, Department of Philosophy, Brown University, Providence, RI 02912

NOTES 1. For another contemporary sense of ‘death with dignity,’ see my “Late in the Quest: The Study of Malory’s Morte Darthur as a New Direction in Philosophy,” Midwest Studies in Philosophy XXIII (forthcoming): n. 57. 2. Sir Thomas Malory, Le Morte D’Arthur, vol.1 (London: Penguin, 1969): 343. See also The Works of Sir Thomas Malory, edited by Eugène Vinaver, 3d ed. rev. by P.J.C. Field (Oxford: Clarendon Press, 1990), 409, lines 28f. All philosophers would profit greatly from reading Malory. 3. Ronald Dworkin, et al., “Assisted Suicide: The Philosophers’ Brief,” New York Review of Books (27 March 1997), 44. 4. M. Pabst Battin, “Suicide: A Fundamental Human Right?” In Suicide: The Philosophical Issues, ed. M. Pabst Battin and David J. Mayo, (New York: St. Martin’s Press, 1980), 274.

F ELICIA ACKERMAN 198 5. Lonny Shavelson, “What the Dying Really Need,” The New York Times (8 March 1996). See also Elliot J. Rosen, “Demolishing a ‘Straw Man’, Response to ‘Goldilocks and Mrs. Ilych: A Critical Look at the ‘Philosophy of Hospice’’by Felicia Ackerman.” In Cambridge Quarterly of Healthcare Ethics 7 (1998): 208. 6. Battin, op. cit. 275. 7. Ibid. 8. Ibid. 9. Battin grants that this “concept of dignity is not wholly empirical, but contains ideal features as well” (ibid., 276). Elsewhere, she suggests that “severe, sustained physical illness or extreme old age” may be linked with moral degradation because “poor health, pain, discomfort, financial limitation, lack of employment or satisfying occupations, failing senses, and dependency can all produce a kind of self-centeredness . . . a kind of decrepitude in which one is no longer able to treat others as ends in themselves, but uses them only to service one’s needs” (Battin, Ethical Issues in Suicide, [Englewood Cliffs, NJ: Prentice-Hall, 1995], 126). This disparaging picture overlooks the mutually rewarding relationships that old, ill, and disabled people can have with others. (See Susan Wendell, The Rejected Body: Feminist Philosophical Reflections on Disability, [New York: Routledge, 1996], 150f.) It further overlooks the fact that all conditions of life have moral risks. Is there any evidence that old age, illness, disability, and unemployment are more conducive to self-centeredness than youth, health, and professional success? 10. Battin, “Suicide: A Fundamental Human Right?” n. 15. 11. Ibid., 278. 12. Ibid., 279. The example comes from David Wood, “Suicide as Instrument and Expression,” in Battin and Mayo, eds., op. cit., 158f. Wood and Battin neglect to tell us what is so terrible about high-rise apartment buildings, which, as Sara Ann Ketchum has pointed out in discussion, enable city-dwellers to avoid long commutes to work. (Critics of high-rises often point to housing projects, but is the problem architecture or poverty? Many middle-class people seem to live quite happily in well-maintained high-rise apartment buildings.) 13. Battin also discusses nonmedically motivated suicides in Ethical Issues in Suicide. 14. Ronald Dworkin, Life’s Dominion (New York: Knopf, 1993), 239. 15. Dworkin, et al., op. cit., 44. 16. Dworkin, Introduction to “Assisted Suicide: The Philosophers’ Brief,” 41. 17. Ibid.

D EA TH, D YING, AND D IGNIT Y EATH IGNITY 199 18. Ibid. For discussion of a logical problem with these formulations, see my “Assisted Suicide, Terminal Illness, Severe Disability, and the Double Standard,” in Physician-Assisted Suicide: Expanding the Debate, ed. M.P. Battin, et al. (New York: Routledge, 1998), n. 18. 19. See my “Assisted Suicide” and my “Late in the Quest,” section II. 20. Dworkin, Life’s Dominion, 201ff. 21. Dworkin, et. al., “Assisted Suicide: The Philosophers’ Brief,” 44. 22. Dworkin, Introduction to “Assisted Suicide: The Philosophers’ Brief,” 41. 23. In Life’s Dominion, Dworkin says that people who would rather be dead than demented might consider signing living wills stipulating that if they become permanently and seriously demented, and then develop a serious disease, they should not be given medical treatment except to avoid pain. They may consider trying to make . . . arrangements . . . if possible, to be killed. . . . a fiduciary could contradict [such advance directives] only by exercising an unacceptable form of moral paternalism (231). In discussion, Sara Ann Ketchum has raised the question of just how demented someone could be while still being in the cognitive state of being glad he was forced to stay alive. But Dworkin allows that dementia can coexist with other (seemingly complex) cognitive states, such as wanting to go on living (230). He even envisions a demented patient who “insists on remaining at home, rather than living in an institution” (221) and who “pleads” to be kept alive and/or especially comfortable (221). 24. Ibid., 239. 25. Ibid., 210. 26. Dworkin, et al., “Assisted Suicide: The Philosophers’ Brief,” 44. 27. Jane Gross, “Quiet Doctor Finds a Mission in Assisted Suicide Case,” New York Times (2 January 1997), B1 (italics added). 28. Wendell, op. cit., 145. 29. Ibid., 146. 30. Ibid. 31. Wendell teaches at a Canadian university, but the illogicality of what counts as being “dependent” on other people is just as characteristic of the United States as of Canada. 32. See my “Assisted Suicide.”

F ELICIA ACKERMAN 200 33. Daniel Callahan, “When Self-Determination Runs Amok,” Hastings Center Report (March-April 1992), 54. See also the amicus brief of Not Dead Yet and ADAPT (Vacco v. Quill, and State of Washington v. Glucksberg, October 1995), as well as Yale Kamisar, “Against Assisted Suicide—Even a Very Limited Form,” University of Detroit Mercy Law Review 72: 4 (1995): 735-69; “Physician Assisted Suicide: The Last Bridge to Active Voluntary Euthanasia,” in Euthanasia Examined, ed. John Keown, (Cambridge: Cambridge University Press, 1995), 225-60; “The Reasons So Many People Support Physician-Assisted Suicide—And Why These Reasons Are Not Convincing,” Issues in Law and Medicine 12:2 (1996): 113-31, and “The ‘Right to Die’: On Drawing (and Erasing) Lines,” Duquesne Law Review 35:1 (1996): 481-521, as well as Leon Kass and Nelson Lund, “Courting Death: Assisted Suicide, Doctors, and the Law,” Commentary 100 (December, 1996): 23. 34. Callahan, op. cit., 52. 35. Ibid. 36. Ibid., 53. 37. Dworkin, Life’s Dominion, 239. 38. See Daniel Callahan, The Troubled Dream of Life (New York: Simon and Schuster, 1993), 240, n. 11. (I endorse Callahan’s conclusion but not his reasoning here.) 39. Daniel Callahan, What Kind of Life? (Washington, DC: Georgetown University Press, 1990): 230. 40. Ibid. 41. Callahan, “When Self-Determination Runs Amok,” 55. 42. Callahan, What Kind of Life?, 246 (italics added). 43. Daniel Callahan, Setting Limits (Washington, DC: Georgetown University Press, 1995), 181. 44. Ibid., 118. 45. Ibid. Note also that Callahan explicitly says that “[i]n determining morally appropriate care for the elderly, three major types of considerations should come into play [one of which is] the quality of life of the patient” (ibid., 180.) See also Callahan, The Troubled Dream of Life, esp. chap. 6. 46. Discussion of the killing/letting die distinction is obviously beyond the scope of this essay, but see Dworkin, et al., “Assisted Suicide: The Philosophers’ Brief,” 45, for a defense of a view opposed to Callahan’s on this distinction. 47. See the references cited in note 5.

D EA TH, D YING, AND D IGNIT Y EATH IGNITY 201 48. B. Manard and C. Perrone, Hospice Care: An Introduction and Review of the Evidence (Arlington, Va., 1994), 4. 49. J. Brennis, et al., Hospice Code of Ethics (Arlington, VA: 1995), 7. 50. “About Hospice,” (Arlington, VA: 1996), 11. 51. See my “Goldilocks and Mrs. Ilych: A Critical Look at the ‘Philosophy of Hospice,’” Cambridge Quarterly of Healthcare Ethics 6 (1997): 314-24. Note also that although the National Hospice Organization has the traditional hospice antipathy to suicide and holds that “the state has an interest in preserving life [although apparently not by “artificial” means!] even in the face of terminal illness,” the Coalition of Hospice Professionals dissents, holding that suicide can be a “competent and rational choice” under certain extreme circumstances. See Linda Greenhouse, “Before the Court, the Sanctity of Life and Death,” New York Times (5 January 1996), 5. 52. I thank Dan Brock, Donna Harvey, and Sara Ann Ketchum for helpful discussions of this material.

MORALITY AND HEALTH CARE POLICY

Medical ethics should show how an adequate description of morality is helpful in dealing with the problems that arise in the context of medical care. However none of the standard moral theories provide such a des c r i p t i o n . Fur t h e r, a l l o f t h e s e theories assume that there must be

Bernard Gert

a unique correct answer to every moral question, though this answer may be that it is indifferent which of the proposed solutions one picks. The failure to recognize that there are unresolvable moral disagreements leads many philosophers to think that their moral theories will enable them to determine which policies ought to be adopted. Howeve r, t h e c o r r e c t ro l e fo r m o ral theories is more limited: to rule out morally unacceptable policies. Moral theories almost never can settle disputes about which of two well supported health care policies ought to be adopted.

MORAL THEORIES

A

moral theory should make explicit, explain, and, if possible, justify morality or the moral system.1 It must provide an explicit account of morality, including its variations. None of the standard moral theories provide anything close to an adequate account of common morality. Even the best of these theories, including those of Hobbes, Kant, and Mill, provide only a partial outline of the moral system that is commonly used. Unfortunately even they are not completely clear about the

203

B ERNARD GERT 204 distinction between their account of this common morality and their attempt to justify it. Many so-called moral theories do not even attempt to explain or justify common morality but are used to generate guides to conduct intended to replace common morality. These proposed guides, those generated by most of the standard consequentialist, contractarian, and deontological theories, are far simpler than the common moral system and sometimes yield totally unacceptable answers to moral problems. Common morality does not provide unique answers to every moral question, but almost all moral theories generate codes that do provide unique answers to every question. This is one explanation why those philosophers who put forward these theories have usually dismissed common morality as confused. Although the better philosophers, such as Norm Daniels and Dan Brock, involved in medical ethics are aware of the empirical complexity involved in making moral decisions and judgments, they still seem unaware of the theoretical complexity. It is not surprising that many who take morality seriously, including many who do medical ethics, and have tried to apply moral theories to real problems faced by actual people are so critical of moral theory. Although morality is known to all those about whom moral judgments are correctly made (moral agents), it is more complex than most philosophers have recognized. This complexity, like the complexity of the grammar of a language, does not conflict with its being understood by all, but it does make it less vulnerable to the kinds of criticisms that are often brought against it by philosophers. One important difference is that common morality acknowledges some unresolvable moral disagreement, whereas almost all moral theories claim that there is a unique correct answer to every moral question; every action is either morally right, morally wrong, or morally indifferent. According to these theories equally informed impartial rational people never disagree about what morally ought to be done. But the proper role of a moral theory is not only to explain and justify moral agreement, but also to explain and justify moral disagreement. In order to adequately explain and justify both moral agreement and disagreement a moral theory must provide explicit accounts of the concepts of rationality and impartiality and that of morality itself. It must also explicitly recognize those features of human nature that explain why morality has the features it has. It must show what features are morally relevant in describing a situation, and it must describe the procedures involved in moral reasoning. I have already started providing an account of morality by pointing out that it has as an essential feature that only those who are not justifiably ignorant of what morality

M OR ALIT Y ORALIT

AND

HEAL TH C ARE P O LIC Y EALTH LICY 205

requires, prohibits, encourages, and allows, can be judged morally. It follows that everyone about whom a moral judgment can be correctly made must know what morality requires, etc. A complete account of the concept of morality must identify and explain all of its other features, such as its relation to rationality and impartiality. The failure of all previous moral theories to provide adequate descriptions of rationality, impartiality, and morality explains why none of them provides an adequate explanation and justification of morality. The moral theory I shall present shows common morality to be justified, that is, to be an informal public system that all rational persons, using only those beliefs shared by all moral agents, would favor adopting. It is this informal public system that is morality, and it is the moral system, not the moral theory, that is applied to the moral problems that arise in business, law, medicine, science, and all other areas of ordinary life. Moral theory is useful because it supplies an explicit account of morality, so that the moral system can be applied to new and difficult situations. It is extremely important for a moral theory to explain why morality does not provide unique solutions to most controversial cases. Recognition that most controversial issues have no unique right answer might prevent people from regarding their opponents as morally corrupt. This could make it easier for both sides to compromise without fearing that they are giving up their moral integrity. Finally, by showing that common morality is justified, a moral theory can justify using morality to decide what policies to adopt, what actions to take, and what judgments to make. Thus moral theory is not merely a theoretical enterprise; it has some practical consequences. These consequences are, however, primarily indirect; it is morality itself that explains moral decisions about what to do and moral judgments on the actions, intentions, motives, and character of others.

MORALITY IS AN INFORMAL PUBLIC SYSTEM THAT APPLIES TO ALL RATIONAL P ERSONS I use the phrase “public system” to refer to a guide to conduct that has the following two features: (1) All persons to whom it applies, all those whose behavior is to be guided and judged by that system, understand it, and know what behavior the system prohibits, requires, encourages, and allows. (2) It is not irrational for any of these persons to accept being guided and judged by that system. Since morality is a public system, any adequate definition of morality must include these two features, that is, (1) everyone who is subject to moral judgment must know what

B ERNARD GERT 206 morality requires, prohibits, encourages, and allows, and (2) it is not irrational for any of them to use morality as a guide for their own conduct. The clearest example of a public system is a game such as baseball or bridge. A game has an inherent goal and a set of rules that form a system that is understood by all of the players. They all know what kind of behavior is required, prohibited, encouraged, and allowed by the game, and it is not irrational for all players to use the goal and the rules of the game to guide their own behavior and to judge the behavior of other players by them. Although a game is a public system, it applies only to those playing the game. If a person does not want the goal sufficiently to abide by the rules, she can usually quit. Morality is the one public system that a person cannot quit. This is the point that Kant, without completely realizing it, captured by saying that morality is categorical. All people are subject to morality, simply by virtue of being rational persons who can control their actions. Morality is a public system that applies to all rational persons. Public systems can be either formal or informal. Formal public systems are those in which there is a decision procedure, usually involving authorities such as judges, umpires, or referees, that resolves all questions of interpretation of the rules of the system as well as all other disagreements between those to whom the system applies. Informal public systems presuppose overwhelming agreement about their interpretation and cannot function unless disagreements are relatively rare. When it becomes important that all disagreements be settled, public systems tend to become formal. Some games such as professional sports have become formal public systems employing elaborate decision procedures with a hierarchy of referees. Many games, however, remain informal public systems, for example, casual card games or neighborhood games of various sports, and disagreements are settled on an ad hoc basis or not settled at all. Morality is an informal public system that has no authoritative judges and no decision procedure that provides unique answers to all moral questions. When it is important that disagreements be settled, societies use political and legal systems to supplement morality. These systems do not provide a moral answer to every moral question; rather, the question, being regarded as morally unresolvable, is transferred to the political or legal system. There are, however, limits to legitimate moral disagreement, just as there are limits to disagreements in all informal public systems. In the vast majority of situations there is no disagreement at all, but for this very reason these situations are never discussed. The claim that some person—for example, the pope—is a moral authority, is the result of failing to distinguish between morality and religion. There can be religious authorities, but there are no moral authorities. It is generally recognized that as long as people are within

M OR ALIT Y ORALIT

AND

HEAL TH C ARE P O LIC Y EALTH LICY 207

vague but generally recognized limits no one has the authority to settle moral disputes in the way that the United States Supreme Court has the authority to settle legal disputes in the United States or the pope religious disputes among Roman Catholics. Further, no one has some special knowledge of morality not available to others, for everyone who is subject to moral judgment must know what it requires, prohibits, encourages, and allows. Of course, some people are better at dealing with moral problems than others, partly due to their training and experience, and partly due to their intelligence and good judgment. But skill in making moral decisions is not an academic specialty, and no one can legitimately claim to be a moral expert. Unfortunately, some people, including some philosophers, have taken the title of “ethicist” as if they were experts concerning moral decisions in fields such as business or medicine in the same way that chemists are experts in chemistry. The primary task of all philosophers, including moral philosophers, is to clarify. They can also sometimes show whether a commonly held view is justified. Moral philosophers clarify the nature of morality and try to show that it is justified, but those who have a better knowledge of the relevant facts and more experience in the relevant field are more likely to make better moral decisions in that field. Clarifying morality cannot settle all moral disputes, because common morality allows for some disagreement. Even in the unlikely event that everyone was in complete agreement on the facts, including their predictions about the consequences of a decision or policy, there would still be some unresolvable moral disagreement. Regarding morality as an informal public system that applies to all rational persons explains many of the features of morality that almost everyone agrees that it has. That all normal adults are regarded as knowing what morality requires, prohibits, encourages, and allows explains why ignorance of morality is not normally allowed as an excuse. It also explains why it is thought not to be irrational for any person to adopt morality as a guide, even as the overriding guide, for her own conduct. The account of morality as a public system that applies to all rational persons also explains why morality is regarded as categorical or inescapable. No one can simply opt out of it, others will continue to judge a person morally regardless of her claim that she is above it or outside of it. Even if they do not necessarily want to follow it themselves, rational persons not only want others to adopt morality as their guide to conduct, they know that all rational persons want them to act morally. That is why the question “Why should I be moral?” is a genuine question and why it is recognized by almost all that it is rationally allowed not to adopt morality as the overriding guide, not even as an important

B ERNARD GERT 208 guide, for one’s own conduct. Those philosophers who claim that the question “Why should I be moral?” is not a genuine question usually also ignore the close relationship between hypocrisy and morality. These errors stem, at least in part, from continuing to regard morality primarily as a guide that each person adopts for her own conduct.

COMMON MORALITY APPLIED TO GOVERNMENT POLICY: HEALTH CARE I am now only concerned with the actions of government. A government that acts in a morally acceptable way acts justly. But what is required for a government to act in a morally acceptable way? A full answer to this question would require a whole book in political theory; I shall not try to provide even an outline of a full answer. However, knowing that common morality applies to the actions of governments as well as to the actions of individuals is sufficient to clarify some points. The recognition that morality is an informal public system suggests the most important point: it is extremely unlikely that there is a unique right answer to how the government should act with regard to almost any controversial political matter, for example, the allocation of health care. Even if, contrary to fact, there were substantial agreement on factual matters like the result of a given law or policy, there probably would still be no unique right answer. There are too many ideological disagreements about human nature and the nature of human society as well as disagreement about the rankings of the harms and benefits involved. One of the primary responsibilities, if not the primary responsibility, of government is to lessen the amount of harm suffered by its citizens. There are several sources of these harms. The first is immoral behavior of other members of the society; the second is immoral behavior by people outside of the society, including other governments. Everyone agrees that governments have duties to enforce the moral rules with regard to their own citizens and to provide security against attacks by those outside of the society. A third source of harms is the environment, both natural and artificial, for example, floods and toxic wastes. Governments should seek to prevent and relieve the harms caused by these conditions. A fourth source of harms is diseases and injuries, and so on, all of which can be classified as maladies. 2 One of the duties of government is to prevent and relieve the harms involved in maladies, especially if they can be most effectively prevented by governmental action, such as ensuring safe drinking water and safe automobiles, and providing safe roads and universal vaccinations. No

M OR ALIT Y ORALIT

AND

HEAL TH C ARE P O LIC Y EALTH LICY 209

one disagrees that these are justifiable governmental expenses. However, after a certain amount of money is spent on health care, including public health, it is appropriate to question whether more ought to be spent. It is doubtful that these questions have uniquely correct moral answers. The same is true of questions about how much ought to be spent on the criminal justice system, education, public defense, and so on. Assuming that agreement has been reached on the amount of government spending on health care, there is still the question of what would count as a morally acceptable way of spending that money or what would count as a just health care system. A formal, but not very useful answer to that question is whatever system a fully informed, impartial rational person could favor adopting. What kind of system could such a person favor adopting? The answer to this is slightly more informative: any system that such a person could regard as resulting in the least amount of harm being suffered due to maladies. If no fully informed person could regard a particular health care system as resulting in the least amount of harm being suffered, then such a system cannot be regarded as just. The present health care system in the United States is not regarded by anyone as resulting in the least amount of harm being suffered, which explains why no one regards the present health care system as just. But this does not mean that there is agreement on what would be a just health care system or how the government should act with regard to health care. All that is required of a just health care system is that a fully informed, impartial rational person can regard that system as resulting in the least amount of harm being suffered. But there may not be agreement on what counts as the least amount of harm being suffered. There is no requirement about equality, nor about providing the most aid to those who are worst off; there may not even be agreement on who counts as being worst off. This is not because equality and aiding the worst off are irrelevant; rather, it is because, insofar as they are relevant, they are encompassed by the goal of lessening the amount of harm suffered. Unlike a capitalist account of justice, which has a goal of increasing the amount of net benefits and so allows massive inequality, the moral goal of lessening the amount of harm suffered sets strict limits on inequality. It also results in great concern for those who are worst off, for they are suffering greater harm than others, and so relieving their harm will normally be included in the goal of the least harm. Justice does not, however, require that the government spend a given amount of money in order to aid one thousand who are worst off if that same amount of money will prevent more harm for a hundred thousand who are not as badly off. For example, it is not required that the government spend a given amount of money on treating a thousand

B ERNARD GERT 210 children with a serious genetic malady rather than spending that same amount on preventing a hundred thousand children from getting some slightly lesser malady. It is also not required that they do not spend the money on the thousand who are worst off, for impartial rational persons can disagree on which alternative most lessens the amount of harm suffered. But keeping the cost the same, if the number of the worst off gets smaller and the number who can be prevented from suffering some slightly lesser disease gets greater, it is quite likely that a point will be reached where it will be unjust to spend that amount of money on the worst off. Since a government that acts in a morally acceptable way acts justly and there is usually more than one morally acceptable way for a government to act, it is very likely that there will not be a unique right answer to the question of how health care should be allocated. Agreement that the goal of health care allocation is to lessen the amount of harm suffered due to maladies still leaves unresolvable any disagreement about what counts as the lesser amount of harm suffered. Some may claim that what is most important is minimizing the suffering of the worst off, those suffering the greatest harm, whereas others may maintain that it is irrelevant whose suffering is minimized, as long as the total amount of harm suffered is minimized. Since there is no agreed-upon way to weigh and balance different evils, there is no way to resolve any plausible disagreement. But, even with all of this disagreement, there is universal agreement that the present health care system in the United States is not just. Often this is expressed by claiming that the health care system is unfair, but this is simply to use the term “unfair” to mean morally unacceptable.

REFLECTIVE E QUILIBRIUM It is, of course, a test of the adequacy of any account of morality that it not be inconsistent with one’s considered moral judgments. However, given that morality is an informal public system, it is extremely unlikely that any considered moral judgment will be incompatible with common morality. Trying to arrive at a correct moral judgment by going back and forth between one’s considered moral judgments and an account of morality, presupposes the mistaken view that there is a unique correct answer to every genuine moral controversy. Common morality allows wide variation in moral judgments, even among those who agree on all of the facts of a particular case. Not only can there be some differences in the interpretation of the moral rules, there is also considerable latitude in the rankings of the different goods and evils, and

M OR ALIT Y ORALIT

AND

HEAL TH C ARE P O LIC Y EALTH LICY 211

there are different ideological beliefs concerning human nature and the nature of human societies. A moral theory cannot and should not be used to settle controversial moral questions. It has, unfortunately, become common for philosophers to talk about “reflective equilibrium” as a method for testing the correctness of one’s moral theory as well as being a test of one’s moral judgments. The phrase “reflective equilibrium” was introduced into moral theory by John Rawls as such a method. It presupposes that a moral theory provides a decision procedure for making moral judgments. When the moral theory yields a moral judgment that is in conflict with a person’s considered moral judgments, he must decide whether or not to change the theory or his considered moral judgments. He must bring the theory and his considered moral judgments into reflective equilibrium. This strongly suggests that if a person changes his moral judgment on some controversial moral question, for example, abortion or the death penalty, he must change his moral theory. This may be true if one conceives of a moral theory, as Kant and the Utilitarians did, as generating moral judgments, that is, as generating a moral system. Rawls seems to share this misleading way of characterizing a moral theory. However, a moral theory is best regarded as providing an explicit description of common morality and then attempting to justify that system. It is only the descriptive part of a moral theory that it even makes sense to bring into reflective equilibrium with one’s considered moral judgments. And even this is misleading, for the descriptive part of a moral theory is simply an attempt to provide an explicit, precise, and coherent description of the considered moral decisions and judgments of all moral agents. That part of a moral theory which is an attempt to justify common morality cannot be inconsistent with the descriptive part, for it is an attempt to justify the moral system that has been described. Reflective equilibrium not only mistakenly seems to presuppose that there is a unique right answer to all moral questions, it also mistakenly seems to regard a moral theory as generating that answer.

DEMOCRATIC IMPLICATIONS OF M ORALITY AS AN INFORMAL SYSTEM Rawls, together with most other moral and political philosophers, seems to hold that a moral system that provides unique answers to every moral and political question is preferable to one that allows for some questions to have more than one morally acceptable answer. For them a moral system is supposed to provide a way of settling disputes, so that a

B ERNARD GERT 212 moral system that settles all disputes is clearly preferable to one that does not. However, as a practical matter, it would not make much difference if the moral system did provide unique answers, for there would still be disagreements about the facts, including the nature and probability of the consequences of alternative policies. And as I have pointed out before, most moral and political disagreements are disagreements about the facts. Nonetheless, it is still a mistake to think that a moral system can provide unique answers to every moral and political question, for equally informed, impartial rational persons do not always agree on all moral matters. The view that morality always provides unique answers is not only mistaken, it is also dangerous. It may lead one to hold that someone who takes a different view in a moral dispute must be either uninformed, partial, or irrational. Regarding one’s opponents in this way is not conducive to fruitful negotiation and compromise. Accepting a theory that claims to provide unique answers also has a tendency to lead one to accept as the ideal form of government a kind of Platonic Republic, where philosopherkings, who are supposed to most closely approximate impartial rational persons, make all of the decisions. Regarding morality as providing unique answers leads to a tendency to regard democracies, where significant decisions are made by the votes of the mass of the population, as employing a defective decision procedure. If there is only one right answer, then why not have those who are most qualified determine what it is? According to Daniels’ earlier work, “fair equality of opportunity” is the primary goal of health care. A just health care system ensures that no one is deprived of fair equality of opportunity because of his suffering of some curable malady, i.e., disease or injury. Curable disability, because it deprives a person of fair equality of opportunity is more important than relief of pain that is not related to fair equality of opportunity. But, it is clear that some people might rank relief of pain, even to a person who is incurably disabled, as higher than curing some disability. Indeed, an individual may choose to be more disabled in order to suffer less pain. In some cases, this will be a rational choice. Daniels now seems to have concluded that there may be no clear resolution to many health care problems and concentrates on a procedure for settling disputes which involves publicity, relevance, appeals, and enforcement conditions. I applaud this move as a move in the right direction, one that acknowledges that there may be no unique right answer to many health care controversies. It is part of my ethical theory that moral disagreements always occur within limits, and that even within these limits arguments for any given course of action must be supported by reasons that all impartial rational persons could accept. However, I

M OR ALIT Y ORALIT

AND

HEAL TH C ARE P O LIC Y EALTH LICY 213

not only acknowledge, but also explain why there will still be disagreements among impartial rational persons. Although less committed to there being, at least theoretically, a unique right answer than Brock still seems to be, Daniels still seems to think that appropriate ethical and empirical research may yet come up with such an answer. Of course, in some cases that might happen. However, I think that the major point of ethical and empirical research will be the discovery of those factors that lead people to disagree. This is not unimportant, for providing a clear basis for the different answers is essential for an informed democratic choice among the alternative policies.

MANY MORAL ISSUES ARE NOT RESOLVABLE Morality sometimes leaves an individual with that dreadful freedom of choice about which some existentialist thinkers have written so fully and brilliantly. Because there is sometimes no morally right course of action, even a morally good person may be forced to choose between alternative courses of action in a situation where not all equally informed impartial rational agree. This kind of situation is likely to be distressing to any morally sensitive person and may explain why many morally sensitive people have claimed that objective morality is a fraud or useless or both. This reaction, though understandable, is not correct. Just because morality does not always provide a unique answer, it does not follow that it never or even generally does not. Of course, those cases where morality provides a unique answer are usually not morally perplexing, so they have not attracted the attention of those fascinated by moral perplexity. To desire morality always to provide a single clear answer may be a rational desire but, perhaps unfortunately, morality does not satisfy this desire. Bernard Gert, Department of Philosophy, Dartmouth College, Havover, NH 03755; [email protected]

NOTES 1. See my Morality: Its Nature and Justification (Oxford University Press, 1998). 2. For the relationship of maladies to the suffering of harms see Bernard Gert, Charles M. Culver, and K. Danner Clouser, Bioethics: A Return to Fundamentals (New York: Oxford University Press, 1997), chap. 5.

ETHICAL ISSUES IN THE CONSTRUCTION OF COSTEFFECTIVENESS ANALYSES FOR THE PRIORITIZATION AND RATIONING OF HEALTHCARE

The

dominant

methodology

in

health policy for prioritizing and rationing health care resources is cost-effectiveness analysis, typically using quality adjusted life years (QALYs) or disa bilit y adjusted life years (DALYs) to measure health outcomes. The construction of these measures involves a number of moral or value choices, including: How should states of health and disability be evaluated, and whose preferences (e.g., the disabled or non-disabled) should be used? How should these evaluations reflect that prioritization will involve tradeoffs between health benefits for different

Dan W. Brock

persons or groups? Do all QALYs count equally, no matter what age at which they are received? Should discount rates be applied to health benefits? I will show the nature of the moral issues at stake in answering these questions, and briefly argue how they should be answered.

INTRODUCTION

A

round the world health care resources are widely recogized to be scarce, which has led in turn to an increasing number of formal attempts to prioritize and ration health care resources, from efforts like the restructuring of the Oregon Medicaid program to the World Bank/ WHO study of the burdens of disease throughout the world. 1 Health policy makers and economists often assume that limited resources should be used in the way that will maximize health benefits for the 215

D AN W. BROCK 216 population served, with cost-effectiveness analysis (CEA) the standard analytic tool for pursuing this aim. Natural, even self-evident, as this assumption may appear to many persons, I will argue that it assumes a utilitarian or consequentialist moral standard, more specifically standard of distributive justice, and that the utilitarian account of distributive justice is widely and correctly taken to be utilitarianism’s most problematic feature. I am going to draw on critical work on consequentialism and distributive justice to address here some of the main ethical issues in the construction of measures of burdens of disease and of CEA in health care. In the companion paper delivered in this symposium, Norman Daniels addresses the ethical issues in the use of CEAs in health care.

FIRST ISSUE: HOW SHOULD STATES OF HEALTH AND DISABILITY BE EVALUATED? Early summary measures of the health status of populations were often measures of a single variable which stood as a more or less crude surrogate measure for the health of a population, such as life expectancy or infant mortality. Measures like Quality Adjusted Life Years (QALYs) and Disability Adjusted Life Years (DALYs), however, have the important advantage that they address and measure both of the main benefits of health interventions—improvements in length of life and in health-related quality of life—and provide a basis for assigning relative value to length versus quality of life, as well as to different impacts on quality of life. The construction of any measure like the QALY or DALY requires a two step process. First, different states of disability or conditions limiting health related quality of life (HRQL) must be described on which individuals’ health states both before and after health interventions can be mapped. A prominent example, the Health Utilities Index (HUI) is reproduced below in Appendix A. Second, different relative values must be assigned to the quality of life with those different conditions, typically on a scale ranging from 0 for death to 1 for full unimpaired quality of life. The determination of people’s different health related conditions both before and after a particular health intervention is an empirical question, which should be answered by appeal to relevant data regarding the burdens of a particular disease and the reduction in those burdens that a particular health intervention can be expected to make. Needless to say, often the relevant data are highly imperfect, but that is a problem to be addressed largely by studies generating better data, not by ethical analysis. The second step in developing measures like the HUI involves assigning relative values or utilities to the different conditions that reduce

E THIC AL I SSUES AND C OST-E FFECTIVENESS IN HEAL THC ARE THICAL EALTHC THCARE 217 people’s HRQL. In the quality of life and cost-effectiveness literature, the standard assumption is that these values should be preference-based; it is not clear that this assumption is correct, but I shall accept it for the purposes of this paper. One problem concerning whose evaluations or preferences should be used of different states of disability or functional limitation arises from the ability of individuals to adjust, cope, and adapt to their disabilities. 2 This results in disabled persons typically reporting a higher quality of life with their disability than non-disabled persons rate the quality of life with that disability. If the evaluations of disability states by the non disabled are used for ranking different states of health and disability, then disabilities will be ranked as more serious health needs, but these rankings are open to two charges. First, they may be biased by the ignorance of the evaluators of what it is like to live with the conditions in question. Second, less value will be given to saving the lives of disabled persons than non-disabled persons because of the lower evaluation of their quality of life. If the evaluations of disabled persons themselves are used, on the other hand, the rankings are open to a different charge of bias against the disabled, namely, that the burden of disability has been unjustifiably underestimated because of the adjustment process that disabled persons have undergone, resulting in less value being given to prevention or rehabilitation of disabilities. The problem here is to determine the appropriate evaluative standpoint for ranking the relative importance of different disabilities, which avoids the potentials for bias inherent in these differing perspectives. 3 The problem that I call the perspective problem is that non-disabled and disabled persons evaluate the quality of life with the disability from two different evaluative perspectives, or plans of life. They are different because of the adaptive change in plans of life and evaluative perspectives that disabled persons have made to their disability, but neither evaluative perspective can be shown to be mistaken as a result of that adaptation. A related problem with any use of QALYs or DALYs for assessing the benefits of health interventions is that they appear to discriminate against the disabled by placing less value on disabled people’s lives.4 For example, as already suggested, an intervention that extends for 10 years the lives of patients with a disability that reduces their HRQL to 0.75 would produce 7.5 QALYs for each patient receiving the intervention. But the same intervention given to patients with an otherwise unimpaired HRQL of 1.0 that extends their lives for 10 years would produce 10 QALYs for each patient receiving the intervention. The use of QALYs for evaluating and prioritizing life-saving interventions appears to discriminate against the disabled and place less value on their lives by assigning less value to extending their lives simply because of their

D AN W. BROCK 218 disability. For this reason, some have argued that quality weights should not be used in the evaluation of saving lives, so long as the individuals in question consider their lives worth living.5 The problem of whether QALYs unjustly discriminate against the disabled, and if they do how that defect should be remedied, remains unresolved.

SECOND ISSUE: SHOULD THE EVALUATIVE PERSPECTIVE FOR DETERMINING RELATIVE HRQL BE I NDIVIDUAL OR SOCIAL? What does it mean to say that the evaluative perspective for determining relative quality of life of different states of disability is individual or social? Consider again QALYs assessed using the HUI scale in Appendix A. That scale contains seven attributes, or areas of functioning, with from three to five levels of function within each attribute or area. How are the preference-based relative values of quality of life, or health utility, determined for the different levels of function within the seven attributes? Typically, individuals are asked what their relative quality of life would be in different health states, on a scale in which one represents full, unimpaired function and zero represents death, for each level of function within each attribute. There are several different technical measures for obtaining the relative values, such as standard gambles, interval scales, and time tradeoffs, but their details for the most part do not bear on the point of concern now.6 Suppose individuals can tell us that their quality of life would be reduced from 1.00 to 0.86 at level three of attribute one if they were able to see, hear, or speak with limitations, even with equipment; they also give us utility or quality of life levels for each of the other attributes and levels. The perspective here is individual because it asks individuals how much an individual’s (or their own) quality of life would be reduced if they suffered the various impairments of function. This individual perspective is appropriate for a number of uses such as monitoring the health status or overall burden of disease in a population, or for comparing alternative health interventions for a given group of patients with a particular medical condition. More generally, the individual perspective is correct for evaluating alternative interventions, or for monitoring changes over time in health status or the burdens of disease, for the same person or group of persons. In the context of resource prioritization, however, alternative health interventions typically do not benefit the same persons, but different persons; for example, comparisons of the cost-effectiveness of treatment interventions for different diseases. Employed in this context, here is the example that I believe is the most ethically implausible implication

E THIC AL I SSUES AND C OST-E FFECTIVENESS IN HEAL THC ARE THICAL EALTHC THCARE 219 of the individual perspective of the HUI. According to the HUI, the benefit of saving the life of a person who will live at a full quality of life, level 1.0, for a year, with no impairment of function in any of the seven attributes or areas of function, is equivalent to the aggregate benefit produced by keeping 20 different otherwise healthy individuals from having to use eyeglasses or a hearing aid to see or hear for a year (which has a utility level of 0.95); each produce 1 QALY. Here, the HUI is employed in a CEA that weighs the tradeoff between using limited resources to meet the different health needs of different groups of individuals. But people who assign the utility level of 0.95 to requiring equipment to see or hear or speak do not understand or intend their assignments to have those implications, and they reject in their explicit tradeoffs these inferred tradeoffs from their utility assignments in measures like the HUI. This individual perspective when used for aggregating QALYs employing the HUI displays two distinct difficulties. The first is that it does not reflect the relative ethical importance people give to saving life in comparison with health benefits that improve the quality of life of other individuals or groups. More important, calculations of aggregate QALYs from different health interventions fail to reflect the ethical importance people place on the fact that health benefits to different individuals or groups are being traded off or prioritized. When summary measures of population health status like the HUI are used to evaluate alternative interventions that will benefit different individuals, as opposed to alternative interventions that will benefit the same individuals, issues of equity and distributive justice are raised; some of these issues are addressed in Daniels’ companion paper for this symposium. There are several broad approaches for taking account of this difference in perspectives. The first is to restrict the use of summary measures like QALYs calculated with instruments like the HUI to the evaluation of health interventions that serve the same individuals, but then they are useless for most health care resource prioritization questions. The second approach, probably the most common in practice, is to explicitly note that use of the CEA ignores issues of equity and distributive justice, typically accompanied by an admonition to policy makers to attend to those issues, but with little if any guidance about what they are and how policy should take account of them. The third approach seeks to develop a quantitative measurement tool appropriate to evaluation of interventions that serve different groups of individuals. The most prominent example is Erik Nord’s “person trade-off” approach which explicitly asks people how many outcomes of one kind they consider equivalent in social value to X outcomes of

D AN W. BROCK 220 another kind, where the outcomes are for different groups of individuals.7 For example, people can be asked how many patients with a very severe disease whose treatment is only partially effective would be equivalent in social value to treating 100 patients with another less severe disease whose treatment is fully effective and so results in a substantially greater health improvement for each patient treated; in this example Nord found that people were prepared to sacrifice substantial aggregate health benefits in order to give priority to treating the sickest or worst off.8 The person trade-off approach is designed to permit people to incorporate concerns for equity or distributive justice into their judgments about the social value of alternative health programs. There has been relatively little exploration and use of this methodology in comparison with the mass of studies and methodological work on measures of aggregate QALYs, in part because many health policy analysts and health economists assume, often with little or no argument, that the social value of health programs is the sum of the individual utilities produced by the program; that utilitarian assumption is rejected in most philosophical work on distributive justice, as well as in the evaluations of ordinary people of the relative importance of different health outcomes and programs. But I do emphasize that for purposes of resource prioritization and allocation, the social approach is the proper perspective, whether by means of a methodology like the person trade-off that seeks to incorporate people’s concern for equity within the measure of the social value of health programs, or by separate attention to issues of equity outside of the CEA.

THIRD ISSUE: DO ALL QALYS COUNT EQUALLY? Typical use of the QALY measure assumes that an additional year of life has the same value, regardless of the age of the person who receives it, assuming that the different life years are of comparable quality. A year of life extension for an infant, a thirty-year-old, and a seventy-five-yearold all have the same value in QALYs, and in turn in a CEA using QALYs, assuming no difference in the quality of the year of life extension. This is compatible, of course, with using age-based quality adjustments to reflect differences in average quality of life of a population at different ages; for example, if average quality of life in a population of persons at age 85 is less than that of persons at age 45, a year of life extension for average 45 year olds would have greater value in QALYs than would a year of life extension for average 85 year olds.

E THIC AL I SSUES AND C OST-E FFECTIVENESS IN HEAL THC ARE THICAL EALTHC THCARE 221 In the World Bank Study, the alternative DALY measure was developed to measure the burden of disease.9 Probably the most important ethical difference between QALYs and DALYs is that DALYs assign different value to a year of life extension at the same level of quality, depending on the age at which an individual receives it; in particular, life extension for individuals during their adult productive work years is given greater value than a similar period of life extension for infants and young children or the elderly. The principal justifications offered for this feature of DALYs were the different social roles that individuals typically occupy at different ages and the typical emotional, physical, and financial dependence of the very young and the elderly on individuals in their productive work years.10 This justification of age-based differences in the value of life extension adopts an ethically problematic social (in a different sense of “social” than that used in the preceding section) perspective on the value of health care interventions that extend life, or maintain or restore function, that is, an evaluation of the benefits to others of extending an individual’s life, or maintaining or restoring his or her function, in addition to the benefit to that individual of doing so. This social perspective is in conflict with the usual focus in clinical contexts only on the benefits to the individuals who receive the health care interventions in question. Typical practice in health policy and public health contexts is more ambiguous on this point, since benefits to others besides the direct recipient of the intervention are sometimes given substantial weight in the evaluation and justification of health programs; for example, with treatment programs for substance abuse, the benefits of reductions in lost work days and in harmful effects on family members of the substance abuser. Using this social perspective is ethically problematic because it gives weight to differences between individuals in their social and economic value to others; it can be argued that this discriminates against persons with fewer dependencies and social ties and so should not be ethically relevant in health care resource allocation. The social perspective justifying the DALY measure is ethically problematic, in a way the alternative QALY measure is not, if the value of health benefits for individuals should focus on the value to those individuals of the health benefits, not on the social value for others of those health benefits. Placing different weight on life extension at different ages, however, might be justified ethically if done for different reasons. For example, Norman Daniels has argued that because everyone can expect to pass through the different stages of the life span, giving different value to a year of life extension at different stages in the life span need not unjustly discriminate against individuals in the way giving different

D AN W. BROCK 222 weight to life extension for members of different racial, ethnic, or gender groups would unjustly discriminate.11 Each individual can expect to pass through all the life stages in which life extension is given different value, but is a member of only one race, ethnic group, and gender. Thus, use of DALYs does not constitute unjust age discrimination comparable to gender, ethnic or racial discrimination. Moreover, individuals, and in turn their society, might choose to give lesser weight to a year of life extension beyond the normal life span than to a year of life extension before one has reached the normal life span on grounds of equality of opportunity. People’s plans of life and central long term projects will typically be constructed to fit within the normal life span, and so the completion of these central projects will typically require reaching, but not living beyond, the normal life span. Life years up to the normal life span, but not beyond it, are necessary to have a fair opportunity to live a full life. 12

FOURTH ISSUE: WHAT LIFE EXPECTANCIES SHOULD BE USED FOR CALCULATING THE BENEFITS OF HEALTH INTERVENTIONS? There are significant differences in the life expectancies of different groups in American society, for example between genders, racial and ethnic groups, as well as differences correlating with differences in socio-economic status. In a broader international context the differences in life expectancies within and between different countries are much greater. Should these differences affect calculations of the life years gained by life extending or quality-of-life improving health care and public health interventions? An accurate estimation of the additional QALYs actually produced by specific health care or public health interventions should not ignore differences in life expectancies that are not caused by the particular condition that the health care intervention affects. Yet the differences in life expectancy between different racial, ethnic, and socio-economic status groups, as well as the very large differences between life expectancies in economically developed and poor countries, are often principally a result of unjust conditions and deprivations suffered by those with lower life expectancies. It would seem only to compound those injustices to give less value to life saving or life extending interventions for groups with lower life expectancies caused by the unjust conditions and deprivations from which they suffer. The developers of the DALY explicitly chose to use a single uniform measure of life expectancy (except for the biologically based gender

E THIC AL I SSUES AND C OST-E FFECTIVENESS IN HEAL THC ARE THICAL EALTHC THCARE 223 difference) for all countries, specifically that observed in Japan which has the highest national life expectancy, to measure gains from health interventions. They justified their choice in explicitly ethical terms as conforming to a principle of “treating like events as like.”13

FIFTH ISSUE: SHOULD DISCOUNT RATES BE APPLIED TO HEALTH CARE BENEFITS? It is both standard and recommended practice in cost effectiveness analyses, within health care and elsewhere, to assume a time preference and so to apply a discount rate to both the benefits and costs of different programs under evaluation.14 It is important to be clear both about what precisely the ethical issue is in whether health benefits should be discounted, as well as why it is important for policy. It is not controversial that a discount rate should be applied to economic costs and economic benefits; for example, if the payment of the costs of a health benefit achieved at a specific time can be delayed for ten years, fewer current dollars are required to produce the benefit. The ethical issue is whether a discount rate should be applied directly to changes in well being or health, and in particular to benefits in the form of increased wellbeing from health interventions. Is an improvement in well-being, such as a specific period of life extension, a reduction in suffering, or an improvement in function, extending, say, for one year of substantially less value if it occurs twenty years from now than if it occurs next year? Distant benefits are appropriately discounted when they are more uncertain than proximate benefits. Proximate benefits, such as restoration of an individual’s function, also are of more value than distant benefits if they make possible a longer period of benefit by occurring sooner. But neither of these considerations require the use of a discount rate—they will be taken account of in the measurement of expected benefits of alternative interventions. The ethical question is whether an improvement in an individual’s well-being is of lesser value if it occurs in the distant future than if it occurs in the immediate future, simply and only because it occurs later in time. This is a controversial issue in the literature on social discounting and my own view is that no adequate ethical justification has been offered for applying a discount rate directly to health and well being. The avoidance of paradoxes that arise if a discount rate is applied to costs, but the same discount rate is not applied to benefits, has influenced many economists to support use of the same discount rate for costs and benefits.15 However, I believe these paradoxes are properly dealt with not through discounting, but rather

D AN W. BROCK 224 through directly addressing the ethical issues raised of equity between different stages of individuals’ lives and between different generations. The policy importance of this in the prioritization of health care interventions is relatively straightforward. Use of a discount rate for evaluating alternative health care programs that take significantly different lengths of time to produce their benefits leads to an unwarranted priority to programs producing benefits more rapidly. Put differently, a program that produces benefits in health and well-being say twenty years into the future will be given lower priority than an alternative health care program that produces substantially less overall improvement in health and well-being, but produces that improvement much sooner. Many public health and preventive interventions, for example, vaccination programs and changes in unhealthy behavior, reap their health benefits years into the future. If those benefits are inappropriately discounted, they will receive less priority and result in a health policy that produces fewer overall health benefits over time than could have been produced.

SIXTH ISSUE: WHAT C OSTS AND BENEFITS SHOULD COUNT IN COST E FFECTIVENESS ANALYSES OF HEALTH PROGRAMS ? It is widely agreed that CEAs in health should reflect the direct health benefits for individuals of their medical treatment, such as improving renal function or reducing joint swelling, and of public health programs, such as reducing the incidence of infectious diseases through vaccination programs. The direct costs of medical treatment and public health programs, such as the costs of health care professionals’ time and of medical equipment and supplies, should also be reflected. But medical and public health interventions typically also have indirect benefits and costs. For example, some disease and illness principally affects adults during their working years, thereby incurring significant indirect economic costs in lost work days associated with the disease or illness, whereas other disease and illness principally affects either young children or the elderly who in each case are not typically employed and so do not incur lost wages or work time from illness. Should an indirect economic burden of disease of this sort be given weight in prioritizing between different health care interventions? From a broad utilitarian perspective encompassing all effects of disease and of efforts to treat or prevent it, indirect benefits and costs are real benefits and costs, even if not direct health benefits and direct treatment costs; they should be reflected in the overall cost effectiveness accounting of how to use scarce health resources so as to produce

E THIC AL I SSUES AND C OST-E FFECTIVENESS IN HEAL THC ARE THICAL EALTHC THCARE 225 the maximum aggregate benefits. What moral reason might there be to ignore these indirect costs and benefits in health resource prioritization? Giving priority to the treatment of one group of patients over another because treating the first group would produce substantial indirect benefits for others (for example, other family members who were dependent on these patients) or would reduce indirect economic costs to others (for example, the employers of these patients who incur less lost work time) could be held to fail to treat each group of patients with the equal moral concern and respect, and in particular the equal moral concern for their health care needs, that all people deserve. Instead, giving lower priority to the second group of patients simply because they are not a means to the indirect benefits produced or indirect costs saved by treating the first group of patients wrongly treats the second group of patients solely as means instead of as equal moral persons or ends in themselves. This Kantian reason for ignoring indirect benefits and costs could serve as a moral basis of the idea of “separate spheres,” that is, that the purpose of health care and of public health is health and the reduction of disease, and so only these goals and effects should guide health care and public health programs.16 The six ethical issues I have discussed very briefly here are all issues involved in developing or constructing a summary measure of the benefits of health interventions and its employment in CEAs in health care. In each case, my point has been that there are important ethical and value choices to be made in constructing the measures; the choices are not merely technical, empirical, or economic, but moral and value choices as well. These issues have by and large not received adequate attention in bioethics or moral and political philosophy and my hope here is to stimulate more work on them.17 Dan W. Brock, Department of Philosophy and Center for Biomedical Ethics, Brown University, Providence, RI 02912; [email protected]

NOTES 1. Harvey D. Klevit et al., “Prioritization of Health Care Services: A Progress Report by the Oregon Health Services Commission,” Archives of Internal Medicine 151 (1991): 912-16; World Bank, World Development Report 1993: Investing in Health (Oxford: Oxford University Press, 1993). 2. C. J. L. Murray, “Rethinking DALYs,” in The Global Burden of Disease: A Comprehensive Assessment of Mortality and Disability from Disease, Injuries,

D AN W. BROCK 226 and Risk Factors in 1990 and Projected to 2020 (Geneva: World Health Organization, 1996). 3. Dan W. Brock, “Justice and ADA: Does Prioritizing and Rationing Health Care Discriminate Against the Disabled?” Social Theory and Policy 12 (1995): 159-84. 4. Brock, ibid.; David C. Hadorn, “The Problem of Discrimination in Health Care Priority Setting,” JAMA 268 (1992): 1454-59; David Orentlicher, “Deconstructing Disability: Rationing of Health Care and Unfair Discrimination Against the Sick,” Harvard Civil Rights-Civil Liberties Law Review 31 (1996): 49-87. 5. F. M. Kamm, Morality/Mortality, vol. 1, Death and Whom to Save From It (Oxford: Oxford University Press, 1993). 6. Marthe R. Gold et al., Cost-Effectiveness in Health and Medicine (New York: Oxford University Press, 1996). chap. 7. 7. Erik Nord, “The Person Trade-off Approach to Valuing Health Care Programs,” Working Paper 38, National Centre for Health Program Evaluation (Fairfield Hospital, Fairfield Victoria, Australia) 7. 8. Erik Nord, “The Trade-off Between Severity of Illness and Treatment Effect in Cost-Value Analysis in Health Care,” Health Policy 24(1993): 227-38. 9. World Bank, op. cit. 10. C. J. L. Murray, “Quantifying the Burden of Disease: The Technical Basis for Disability-Adjusted Life Years,” in Global Comparative Assessments in the Health Sector: Disease Burden, Expenditures and Intervention Packages, ed. C. J. L. Murray and A. D. Lopez (Geneva: World Health Organization, 1994). 11. Norman Daniels, Am I My Parents Keeper? An Essay on Justice Between the Young and the Old (New York: Oxford University Press, 1988). 12. Daniels, ibid.; Dan W. Brock, “Justice, Health Care, and the Elderly,” Philosophy & Public Affairs 18.3 (1989): 297-312. 13. C. J. L. Murray, loc.cit., 7. 14. Gold et al., op. cit. 15. E. B. Keeler and S. Cretin, “Discounting of Life-Saving and Other Nonmonetary Effects,” Management Science 29 (1983): 300-306. 16. Kamm, op. cit., chap. 8; Michael Walzer, Spheres of Justice (New York: Basic Books, 1983). 17. An earlier version of this paper appeared as “Ethical Issues in the Construction of Summary Measures of Health Status,” in Summarizing Population Health: Directions for the Development and Application of Population Metrics (Washington DC, National Academy Press, 1998). I draw here also on my “Considerations of Equity in Relation to Prioritization and Allocation of Health Care Resources,”

E THIC AL I SSUES AND C OST-E FFECTIVENESS IN HEAL THC ARE THICAL EALTHC THCARE 227 in Ethics, Equity and Health For All, ed. Z. Bankowski, J. H. Bryant and J. Gallagher (Geneva: CIOMS, 1997).

APPENDIX A HUI CALCULATION

Health Utility Index--Mark II (Adult/Children) Vaccine preventable conditions: Disease scenario description: Attribute

Level

Description

utility fn

b

1. Sensory

1

Able to see, hear, and speak normally for age

1.00

0

2

Requires equipment to see or hear

0.95

0

3

Sees, hears, or speaks with limitations, even with equipment

0.86

0

4

Blind, deaf, or mute

0.61

0

1. Sensory Total 2. Mobility

0 1

Able to walk, bend, lift, jump and run normally for age

1.00

0

2

Walks, bends, lifts, jumps, or runs with some limitations, but does not require help

0.97

0

3

Requires mechanical equipment (such as canes, crutches, braces, or wheelchair)to walk or get around independently

0.84

0

4

Requires the help of another person to walk or get around and requires mechanical equipment as well

0.73

0

5

Unable to control or use arms and legs

0.58

0

D AN W. BROCK 228

Attribute

Level

Description

utility fn

2. Mobility Total

0

3. Emotion

1

Generally happy and free from worry

1.00

0

2

Occasionally fretful, angry, irritable, anxious or depressed (or suffering night terrors--for children)

0.93

0

3

Often fretful, angry, irritable, anxious, depressed (or suffering night terrors--for children)

0.81

0

4

Almost always fretful, angry, irritable, anxious or depressed

0.70

0

5

Extremely fretful, angry, irritable, or depressed, usually requiring hospitalization of psychiatric institutional care

0.53

0

3. Emotion Total

0 1

Learns and remembers normally for age (e.g., schoolwork--for children)

1.00

0

2

Learns and remembers more slowly than normal for age--for adults; learns and remembers schoolwork more slowly than classmates, as judged by parents and/or teachers--for children

0.95

0

3

Learns and remembers very slowly and usually requires special assistance in learning situations

0.88

0

4

Unable to learn and remember

0.65

0

4. Cognitive

4. Cognitive Total 5. Self-care

b

0 1

Eats, bathes, dresses and uses the toilet normally for age

1.00

0

2

Eats, bathes, dresses, or uses the toilet independently with difficulty

0.97

0

3

Requires mechanical equipment to eat, bathe, dress, or use the toilet independently

0.91

0

E THIC AL I SSUES AND C OST-E FFECTIVENESS IN HEAL THC ARE THICAL EALTHC THCARE 229

Attribute

Level 4

Description

utility fn

b

Requires the help of another person to eat, bathe, dress, or use the toilet

0.80

0

5. Self-care total 6. Pain

0 1

Free of pain and discomfort

1.00

0

2

Occasional pain; discomfort relieved by nonprescription drugs or self-control activity without disruption of normal activities

0.97

0

3

Frequent pain; discomfort relieved by oral medicines with occasional disruption of normal activities

0.85

0

4

Frequent pain; frequent disruption of normal activities; discomfort requires prescription narcotics for relief

0.64

0

5

Severe pain; pain not relieved by drugs and constantly disrupts normal activities

0.38

0

6. Pain Total

0

7. Fertility

1

Able to have children with a fertile spouse

1.00

0

2

Difficulty in having children with a fertile spouse

0.97

0

3

Unable to have children with fertile spouse

0.88

0

7. Fertility Total Health State

Utility fn

0 1.06*(b1*b2*b3*b4*b5*b6*b7)-.06

-0.06

ETHICS IN BIG SCIENCE

In accounts of the ethics of science, we may treat practicing science as an institution of sorts. It has an imputed purpose, roughly, finding the truth about vast classes of causal relations. Scientists have been able to

Russell Hardin

act reasonably with no more than the natural confluence of individual interest with the truth. But in the age of institutionalized science, with career stakes outside the accumulation of scientific findings and with institutional interests often directly conflicting with truth, this ‘natural confluence’ is no longer adequate. Now, incentives must be reinforced from outside the scientist’s personal research arena or the scientist must be normatively governed by a desire for truth. In big science, there is little doubt that a formal regulatory system outside the scientist’s research arena is institutionally

easier

to

design

and

implement than is the inculcation of an attitude of service.

INTRODUCTION

N

atural scientists and, speaking metaphorically, natural science are often asserted to have a single ethical principle: to find the truth. As Charles Peirce expressed this commonplace view, “the scientific man is above all things desirous of learning the truth and, in order to do so, ardently desires to have his present provisional beliefs (and all his beliefs are merely provisional) swept away, and will work hard to accomplish that object.”1 Before the chemical revolution led to big, coordinated, team 231

R USSELL HARDIN 232 science in corporations, the more common image of the natural scientist was of a loner at work in a laboratory, at a desk, or, somewhat romantically, under an apple tree. Among the most romantic visions is that of Peirce, the view that what drives scientists is the pure beauty of truth. Professional codes of ethics were once conceived as regulating the interactions between individual professionals and their individual clients or patients. The ethics codes of the medical and legal professions have shifted during this century from an individual professional focus to an institutional focus. In David Hume’s terms, the duties of professionals are increasingly seen as artificial. That is to say, they are not natural duties to do particular things which, if done, would have generally good effects. Rather, their good effects turn on the way in which they are related to actions of others, especially others in institutions. 2 In part, this merely mirrors, belatedly, the changing structures of these professions, which are increasingly involved in large institutional settings rather than in private, one-on-one relations. But it also reflects a shift of moral focus. Doctors and lawyers have no justification if not to benefit patients and clients. Hence, it is sensible to treat doctors and lawyers as having agency roles—with patients and clients as principals—and therefore to see their principal ethical norms as derived not merely from medical or legal principles, but also from organizational and institutional constraints. The provisions of the early codes of medicine and law were typically directed at what a doctor or lawyer in general should do, as though the practice of medicine or law were a universal, not a particular or contingent, matter. Revisions of the AMA medical and ABA legal codes—adopted in 1980 and 1983, respectively—have eliminated much of that discourse to focus instead on how doctors and lawyers should respond to institutional settings and policies. What is the form of self-policing by professionals? The issue is not an analog of the problem of finding what scientists determine to be true under the best scientific understandings of the time. The focus of professional oversight is often on the procedures rather than on the substance of professional efforts, and on the general character of the professional. A lawyer may be sanctioned for having a conflict of interest with a client, not directly for failing the client. Failures in representing the client will be used as evidence of the influence of the conflict of interest. But the conflict of interest is taken to be wrong tout court. At the core of professional ethics codes are rules requiring some behaviors and proscribing others. Superficially, therefore, the codes appear to be deontological, as though the actions of the relevant professional agent and not the consequences of those actions were at issue.

E THICS IN BIG SCIENCE 233 But the reason a profession is desirable is that it benefits potential clients. Moreover, modern revisions of the two oldest codes of professional ethics cannot be understood without reference to the consequential tendencies of the sanctioned actions. A lawyer might handle a client’s case very well despite a conflict of interest (for example, the lawyer has substantial stock holdings in a firm the client is suing). 3 But it would commonly be prohibitively difficult for the ABA to assess whether the lawyer did do an honorable job. In the face of ordinary human nature, especially the tendency to take own interests into account, it makes generally good consequential sense to stop behaviors that tend to have bad consequences for clients. Of course, it is generally in the collective interest of lawyers to have a code that sanctions certain kinds of behavior, because lawyers will generally be more utilized if they can be confidently expected to act on behalf of their clients without serving other interests instead.4 This is a collective good for the class of lawyers, and an individual lawyer may be better off having such a code in place while violating it, or freeriding on it. And the fact of the collective good gives reason to create an enforcement mechanism to secure it against freeriding that would undercut it. Professional self-regulation by scientists, if it is strictly analogous to that by lawyers or doctors, would similarly focus on procedures and character. This might be the correct focus for such matters as fraud in science. The problems of David Baltimore and Robert Gallo could come under procedural rules on record-keeping and on how to handle competing work by other researchers. Unlike the doctor or lawyer, however, these problems seem to be primarily the concern of scientists generally or of particular scientists in competitive research programs and much less urgently a matter of fiduciary responsibility to a nameable client. Historically, some part of the concern over codes of ethics within professions has been about such intra-professional problems. But another and arguably the more central part has been about fiduciary responsibilities. In particular, the putatively moral argument in defense of self-regulation was that doctors and lawyers must be inculcated with the relevant norms of service to their clientele. Similarly, the corrective for scientists who might have succumbed to conflict of interest may not be to address their motivations, but only to address flaws in their science: poor research methods, bad statistical tests, apparently falsified data, and so forth. These are flaws that can be motivated by anything from venality to sloppy character traits and that may affect the work of any researcher, not only of researchers working against a background of conflict of interest. Is my messy desk a sign of unethical professional practice? Is your cavalier record keeping unethical? Going

R USSELL HARDIN 234 after such practices as this may not be an efficient or sensible device for controlling conflicts of interest, because the number of scientists caught up in claims of sloppy practice might grossly swamp the plausible number of those genuinely guilty of succumbing to conflict of interest. The device that seems to work well enough for lawyers and doctors is merely to prohibit action under certain conflicts of interest. This would be grossly destructive of the scientific enterprise. Perhaps tightened scrutiny of data and methods for those doing research against a background of conflict of interest is in order. But this would be a major institutional change that would require extended public debate among scientists and their employers and funders before the inherent inequalities of it may be accepted. In the meantime, scientists acting against a background of conflict of interest are surely less prone to successful self-regulation than are other scientists. I will not discuss views that scientists should or should not have a moral obligation to advocate policies that their science might support. But there is a background issue that makes the discussion of professional ethics for scientists relevant. In accounts of the ethics of science, we may treat practicing science as an institution of sorts. It has an imputed purpose, which is, roughly, finding the truth about vast classes of causal relations. From there we may choose to argue that truth is itself a moral purpose, as Peirce may have thought, or we may say that truth contingently serves some other purpose which, typically, is moral. The enterprise of science is especially valuable in general if it produces results that are “true” so that policy can be based on them. But this makes the focus of an inquiry into the central ethics of the scientific enterprise indirect. If policy is good, then the use of science is good. If we can generally expect policy choices to tend toward what our best scientific results suggests and we can generally expect the policies to be relatively good, we want science to proceed as though the point of the enterprise itself were to find truths about relevant matters. In this indirect justification of science, truth or the search for truth may often not be justified directly by the specific purpose of the search. We do not need to know its specific payoff, although we might undertake a line of research—the Manhattan project or the search for mastery of AIDS—with a relatively specific end in view. Hence the moral defense of science depends essentially on a functional argument. To say that x is good because it leads to y is to imply that y is good in its own right. Some contemporary critics of science and modern society hold that what science leads us to these days is not generally good, is perhaps even disastrous. If one accepts that view, one must say that science, or at least some parts of it, is not morally justified. It is only at this level that

E THICS IN BIG SCIENCE 235 a judgment of the goodness or rightness of organizing science as the quest for truths can be argued. To argue for the morality of the practice of science is tacitly to assume that the results of applying science are apt to be good.

TRUTH AS THE IDEAL OF SCIENCE Truth is typically a good guide to what a scientist should say. For example, in debates on the regime for controlling severe pollution of the Mediterranean, scientists often took a universalist stance and spoke about what regulations were desirable for the entire basin. Their models and assumptions were inherently apolitical. Political leaders often took a nationalist stance and opposed standards and methods of measurement that might put the burdens of controlling pollution on their nations. The politicians often played as though to win against other nations. The scientists more often tended to speak for a truth on which all nations should coordinate. In discussions of the commitments of scientists, the commitment to truth is often treated as normative, as though scientists were an odd breed of humans who are driven by values rather than by interests. This is the apparent intent of the claim of Peirce, quoted above. But the claim that scientists are motivated to seek the truth is not merely a normative claim and, indeed, normative commitments might be relatively minor in the concern for truth. In many contexts it is especially credible that scientists care deeply about truth because it is virtually a claim of self interest. It is in the interest of any given scientists to find the truth because other scientists are likely to show them wrong if they do not, as in the case of claims made for cold fusion that could not be replicated by others. Career prospects will tend to turn on getting it right. Truth, interest, and passion typically cohere. Perhaps, as Dennis Flanagan says, abuse is typically a matter of pathology. 5 Scientists are, in this respect, not unlike lawyers. Lawyers have long gone through the Justice Department and on to major law firms or to corporate legal staffs. Critics of this practice suppose lawyers seeking a comfortable home after a few years in, say, the Antitrust Division of the Justice Department will be careful not to hurt future employers too badly. But future employers are not financially interested in merely rewarding those who have been kind to them in the past but who are no longer in a position to be kind. They are interested in hiring those lawyers who have demonstrated the most inventive competence in the antitrust laws. The lawyer who does best after leaving Justice, therefore, is likely to be the lawyer who did best in serving Justice first.

R USSELL HARDIN 236 Unfortunately, many issues are divisive for relevant scientists, who firmly disagree about causal effects or the facts on which models are based. Politicians, public interest groups, and corporations may naturally choose those scientists whose science is most congruent with their interests. For example, the tobacco industry may ferret out and hire scientists who say there is no proven association between smoking and cancer. Hence, scientists appear to be aligned with interests, as though their science were determined by their association with particular interests. As a result, the public may come to suspect scientific expertise that supports any policy position. A thousand scientists might agree with the analysis that recommends regulation or labeling of substance X, but one might be found to disagree, and that one may be given a platform by an affected interest. There is a perhaps worse problem in the public’s inability to assess the expertise of conflicting experts and the fact that it cannot typically even know that the experts on one side might overwhelmingly outnumber those on the other side or have far greater scientific prestige. It is now a fairly popular assertion that science is as divided as, say, history or sociology. This might merely be a sad implication of the way news media often give seemingly equal time to opposite views even though science metaphorically does not give equal time to the views of tiny, sometimes disreputable groups of scientists who are advocates of commercial interests, such as the tobacco industry or the lead industry. Hence, a large part of the reason for developing canons of professional ethics is that relevant publics cannot know enough to judge the reliability of the professionals. Since the public—and public officials—cannot adequately monitor or police professional behavior, the professionals themselves take on much of the task. If truth worked its influence without fail, there might be no debates over ethics in science today. But it does not, and science may now be on the way to professional self-regulation without any tradition to build on or, perhaps more likely, on the way to regulation by public agencies that would trump professional self-regulation. The core problem in this transition is conflict of interest. Conflict of interest has been an issue in medical practice, but among the traditional professions it has been most important in the practice of law and therefore in legal ethics. Legal practice, however, may be a bad analog for science.

CONFLICT OF INTEREST Typically, a scientist has a direct interest in being on the side of scientific truth because career and opportunities depend on the quality of one’s

E THICS IN BIG SCIENCE 237 past scientific judgments. This interest could be overridden by support from an interest that wants the science to be other than what the scientist thinks it is. But this problem is not parallel to that of the lawyer with a conflict of interest. If it is problematic, the lawyer’s conflict is directly with respect to the lawyer’s client. A scientist’s questionable practice is to join in supporting a client’s interest even against the science. The scientist’s conflict is, metaphorically speaking, with what she should believe to be the truth. Unfortunately, we cannot stipulate or identify the truth in the trivial way we might identify a lawyer’s stock holdings. It is through debate and disagreement that we are likely to achieve better understanding of the science. Hence, we must often expect that there will be divergence of views on a scientific matter. If scientists function to produce knowledge for the collective benefit, that fact recommends a form of competition between ideas. We therefore often want to encourage divergence of views, so we cannot say the odd scientist is out of line for holding a minority view. We would not generally want to have a formal structure of scientific opinion to rule on what arguments or approaches meet some ethic of truth and which ones must be ruled out. Scientists working for industry in developing profitable technology are subject to control by truth through standard market forces. It is only when the technology has unaccounted external effects or hidden direct effects that their role begins to affect the larger public in ways that may involve conflict of interest. The hidden direct effects are analogous to the harms lawyers and doctors may inflict on their clients without their clients knowing. Doctors and biological researchers who work for relevant industries may tell us that tobacco does not cause lung cancer or that low-level exposure to lead does not cause dementia in children. But we cannot say definitively that a scientist supported by a particular, partisan interest must have let her research be influenced by that interest. An independent scientist with a dissident view might well become supported by an interest that finds the scientist’s views congenial. For example, Claire B. Ernhart and Sandra Scarr were early critics of the work of Herbert L. Needleman showing harm to children from very low levels of lead exposure. They subsequently were supported by the lead industry in their further research and even directly paid for consulting. A US Federal Court recently ruled that gifts given by lobbyists to public office holders could not prima facie be considered illegal. 6 Rather, it must be shown that a gift resulted in a favor. The court’s proposed standard is hopelessly too rigid and, if allowed to stand, must effectively overrule the point of the legislation. That point was to block

R USSELL HARDIN 238 conflict of interest itself, as the Code of Ethics for the City of Chicago and many others do. For lawyers, we also could not say definitively that a particular lawyer working against a background conflict of interest must distort her efforts against the interests of her client. Instead we merely rely on a general tendency to bar even the possibility of such distortion. In a vapid moment of theorizing, we might similarly think to bar all support of scientists that might entail a conflict of interest, just as lawyers are prohibited from taking on clients with whom they have conflicts of interest. An immediate consequence of this policy might be a severe reduction in funding for science and a radical decline in the quality of technological change. No scientist could work for industry, perhaps not even for government. All research would have to be funded by individual scientists themselves or by disinterested foundations, including universities. This consequence of a strong rule against conflict of interest argues against such a rule. Hence, it seems likely that we would not choose to have scientists treat conflicts of interest the way we would want lawyers to do. If we went through many other principles in the code for lawyers, we might similarly think them not to apply to scientists. Yet it does not follow that scientists should not be bound by a code of ethics or that the general principles behind the code should not be analogous to those behind the lawyers’ code. The more general principle behind the conflict-of-interest rule for lawyers is to enhance the prospects that lawyers will serve the interests of clients. The role of some scientists, at least, is quite similar to that of lawyers: Their activities are supported only because it is supposed they will benefit a relevant clientele. This clientele may be the relatively grand public at large rather than a particular party to a suit or a criminal trial. 7

SCIENCE IN THE PUBLIC INTEREST Public support entails some public determination of the content and direction of scientific research. Partly this is just a causal claim: If the public had put up enough money for it, we would even have a superconducting super collider, as we have an unprecedentedly large attack on such diseases as cancer and AIDS, or the quick mapping of the human genome. If the public does not put up the money, we will not have these things, at least not on the current scales. More generally, there may be a filtering effect of public support. What the public is likely to fund is what scientists may therefore tend to prefer to do, other things equal. The scientist who would like to do any one of a dozen things might choose happily to do the one of these that attracts funding.

E THICS IN BIG SCIENCE 239 But in part the claim of entailment is not merely causal, it is a moral or political claim: The public is unlikely to put up large sums of money without taking an interest in determining the content and direction of the funded research. The public may therefore demand accountability for scientific quality. Scientists might perform their own self-regulation, as lawyers, doctors, and other professionals do to a large extent. But they may also fail in some instances and thereby provoke public legal oversight. Here the major difficulty is whether scientists should go public over hidden effects of their institutions’ work. A lawyer involved in a case in which her colleague violated the ABA’s Model Rules might be required to report the violation to the court. Some scientists evidently feel a personal moral obligation to report what they consider to be unethical behavior of colleagues, employers, and others, including even unconnected others whose work they think flawed. Some are even prepared to go public simply on the ground that a particular development does not serve the public or relevant specific interest. In some cases, the urge is great enough to override career and financial interests. If there were a compelling and reasonably well enforced code of ethics in place for research scientists, open criticism of an employers’ misuse of science might be made much easier and employers might find it far less in their interest to suppress or fire internal critics.

SCIENCE AS COLLECTIVE ENTERPRISE Perhaps the most striking feature of public support of science is the remarkable change in the structure of scientific practice. A popular image of science in earlier times is of the individual researcher seeking truth and perhaps fame, especially including recognition by peers. The Manhattan Project signaled dramatically how different the contemporary reality is. The bulk of science is now done by big research teams, working with big sums of money, and housed in large institutional infrastructures. A similar change in the structure of medical delivery has radically affected the understanding of medical ethics, which no longer resembles the quaint principles of the Hippocratic Oath. The Hippocratic Oath assumed that doctors would deal with anyone in need; modern institutional medical agencies first must decide whom to serve and whom to exclude. Once such a choice gets in the way, it is hard to see the principles of medical care as strictly about what a universal doctor should do. Institutional and technological change in law have not been as dramatic as in medicine. But law has been

R USSELL HARDIN 240 bifurcated into big-firm law and more nearly solo-practice law, with the bulk of the ABA’s Model Rules now concerned with big-firm law. In the contemporary settings of medical care and law, much of the moral concern is with functional efficacy of various behaviors, where the function is service of some relevant kind as defined or constrained by institutional possibilities. For law and medicine, the focus of service continues to be on clients and patients, although there are increasingly intrusions of concern for the larger public, especially in medicine as it consumes ever more of the domestic product (about 13% in the US currently) and a huge fraction of national budgets. For science there is hardly any reason for concern with service to particular clients, who, after all, are generally firms substantially capable of overseeing their own scientists’ work. Indeed, we may even worry that the scientists in a firm are at risk whenever their truths seem to run against the interests of the firm. The principal concern in the ethics of the scientists themselves is with service to the larger public, service that may conflict with both scientists’ and their firms’ interests. Much of scientific work is a collective or social enterprise. What David Bella, a civil engineer, writes of engineers, we may say more generally of all scientists. They depend on socially created knowledge, most of which they are not individually able to test or confirm. Hence, Bella writes, “Professional integrity involves more than personal integrity.”8 There must be institutions or social processes that have integrity if individual scientists are to have any integrity in the eyes of those unable to judge the science directly and even in the eyes of the scientists themselves.

SELF-REGULATION AND PUBLIC REGULATION Regulation of scientists’ ethics could take three quite different forms or some combination of these. First, there might be public regulation by an independent public body with some policing and sanctioning power. Second, there might be professional self-regulation by a professional organization of scientists themselves. Third, there might be nothing more than individual self-regulation by the scientist and by spontaneous claims by other individual scientists. If the third regime were self-enforcing, there might be no urge to pursue either of the other two regimes. Is the search for truth self-enforcing? To some large extent we must suppose it is, as argued above in the claim that it fits scientists’ self interest, because no scientist will want to be shown wrong, perhaps especially not deceitfully wrong. But this does not seem to be conclusive in every case. We know of far too many cases of scientists who were willing to be

E THICS IN BIG SCIENCE 241 wrong in order to serve a particular master. During the Nazi era, some German scientists were willing to discount the scientific findings of Jews as wrong because they came from Jews. Many people today, including relevant scientists, obviously disbelieve the claims of scientists who work for the tobacco industry or for various industries that are thought to harm their workers through exposure to various substances. But for particular scientists who are firmly committed to scientific reputation, it may not matter very much that they are found wrong only years later rather than immediately. They wish not to be found wrong at all. Why is truth not self-enforcing for commercial firms? They depend for their income on buyer expectations and understandings. When there is uncertainty in buyers’ minds about the claims of a manufacturer and the manufacturer’s critics, at least some of them may conclude in favor of the manufacturer. In this case, the scientists who advise a firm to pursue its interest may be pursuing their own interests rather than the scientific truth. Much of current research is focused on matters of great uncertainty, so that the criterion of truth is indeterminate. Hence, there is room for honest disagreement and therefore incentive for dishonest disagreement. For a firm such as a cigarette producer, the difference between letting the public record assume an association of smoking and lung cancer and contesting such an assumption might well involve billions of dollars in earnings. Self-regulation through a professional organization’s code of ethics and public regulation through a formal, legally empowered agency might differ substantially. It is commonly asserted that the ABA and AMA impose few serious sanctions, that they tend to view their members as part of the club to be protected against complaints from outsiders. A public agency might be considerably more aggressive. In addition, scientists might tend to enforce more rigorously against less prestigious scientists. A public agency might have less respect for scientific prestige and might even find it especially newsworthy or career-enhancing to sanction major scientists. We are now in an era when high-profile regulation of scientists by public officials and agencies has become common. Unfortunately, such regulation may be disruptive, relatively inegalitarian, and highly distorted, as in Congressman John Dingell’s crusades against particular supposed miscreants. Although public regulation may well grow in breadth and significance in the immediate future, much of the regulation of ethical behavior by scientists is likely to remain subject to little more than selfregulation by individual scientists and their individual detractors. In general, as argued above, we must conclude that the content of the moral code of a professional must be institutionally determined. It

R USSELL HARDIN 242 would have been implausible that an individual doctor would have reached moral conclusions fully consonant with the current AMA code. Yet, it is also true that the behavior stipulated in that code is now generally right for the chief reason that it is in that code. 9 The code is not only explicable sociologically, its morality is crafted socially. The ethical principles for the governance of medical practice are in part socially created. They respond, among other things, to strategic interactions between doctors in various specialties, hospitals, patients, insurers, government, and others. That is why it only makes sense to construct them sociologically, not purely by abstract deduction. Suppose there were no ethics code for doctors and you are a doctor. What ethical principles should you follow? It might seem that a first answer is that it would be good to do more or less the same things specified in the actual code. But this answer is incoherent. It becomes right to do all of what the code prescribes only on the assumption that various others, also governed by the code, are doing their parts. You acting alone cannot assure any such thing. Moreover, you could not possibly justify spending the time to work out the principles of the code that creating the code has required from large numbers of people with varied experiences working over several years. Indeed, it would violate an obvious ethical consideration for you to take out all that time rather than to deliver health care to those in need, which is what you are actually educated and trained to do. In the best of all institutional settings, there would be specialists to determine the morality of various practices. So what should you do? Obviously, you should behave in ways that you can reasonably expect to affect health beneficially and not in ways that would harm health. You should generally avoid conflicts of interest— for example, you should not own the hospital, laboratory, or drugstore to which you refer your patients. You might not be able fully to avoid the conflict of interest inherent in your recommending to a patient that she have an operation from you or in other contexts without getting in the way of your care for the patient. And it might not even occur to you that prescribing drugs to be sold by your office involves a conflict of interest (chiropractors, for example, may not see the conflict of interest in their very profitably supplying patients with so-called supplements that they push very hard). But if someone points out the conflict and explicates its logic, we may reasonably doubt your honesty if you insist there is no conflict. Many of the rules of the AMA code might not be mandated by clear reason (few of them are even if we suppose a relatively simple moral theory), and we ought not conclude that you should deduce them on your own and act by them even though the AMA has not promulgated them.

E THICS IN BIG SCIENCE 243 In the present system of science in the United States, an individual scientist is commonly in the state of the doctor without an AMA code. Many scientific associations have, of course, adopted codes of ethics, but these are sometimes nominal codes not much advanced beyond the Hippocratic oath. And engineering associations have long had professional codes that might seem to be largely codes of ethics rather than merely codes for how properly to construct a road or a building. Some of the engineering codes are quite elaborated, clear, and specific in their terms. No doubt, this is because engineers are much more like lawyers and doctors in that they principally serve nameable clients. There may be pure engineering research, but the bulk of engineering is practice. Yet, scientists are very much like doctors and lawyers in that their working lives are highly institutional. Therefore, their morality should be determined in large part with reference to their institutional settings, just as the recent revisions of the main legal and medical ethics codes respond to institutional constraints. We may now be in the stage of transition between the primitive conditions of full self-regulation and legalized regulation by public agencies. While doctors and lawyers are under legal as well as professional self-regulation, they have managed to keep the latter dominant. Apart from some engineering groups, scientists have generally failed to create capacities for professional selfregulation, and they may have waited too long to preempt direct public regulation. Public regulation of doctors has recently been dominated by concerns with the costs and distribution of health care. Public regulation of scientists may quickly assume much of the panoply of issues that might seem to matter from other regulatory regimes covering, for example, doctors, lawyers, and corporations. Oddly, at the moment, it may be true that a number of public officials know more about the issues than do virtually any practicing scientific researchers. As scientists’ professional organizations attempt to draft principles of ethics for their professions, they would be well advised to turn to government officials, traditional professions, and academic students of professional ethics for expert advice. That might seem to be historically embarrassing. Generally, public agencies concerned to regulate medical practice have been much more deeply dependent on the expertise of the professionals to be regulated. Note that doctors without a formal code must combine an attitude of service with a concern for the truth of what serves. Scientists may generally have been able to act reasonably with no more than the natural confluence of individual interest with the truth. But in the age of institutionalized science, with career stakes outside the accumulation of scientific findings and with institutional interests often directly conflicting with truth,

R USSELL HARDIN 244 this formula is no longer so clearly adequate—even though the self interest that scientists have in not having their findings ridiculed or dismissed by other scientists may continue to be the chief regulator of scientists’ behavior. Now incentives must be reinforced from outside the scientist’s personal research arena or the scientist must be normatively governed by a desire for truth. In big science, there is little doubt that a formal regulatory system outside the scientist’s research arena is institutionally easier to design and implement than is the inculcation of an attitude of service. 10 Russell Hardin, New York University, New York, NY; [email protected]

NOTES 1. Charles Peirce, Scientific Metaphysics, vol. 6 of Collected Papers of Charles Sanders Peirce, ed. Charles Hartshorne and Paul Weiss (Cambridge, Mass.: Harvard University Press, 1935), 3. 2. See further, Russell Hardin, “The Artificial Duties of Contemporary Professionals,” Social Service Review 64 (December 1991): 528-541. 3. Some years ago, a corporation reputedly gave up on trying to sue Exxon because all major law firms in the United States had Exxon as a client and they claimed they could not sue it without conflict of interest. One might suppose they could also not choose not to sue it without conflict of interest. 4. Bernard Barber, The Logic and Limits of Trust (New Bruns-wick, N J: Rutgers University Press, 1983). 5. At the Wingspread conference, October 1992. 6. San Francisco Chronicle, 5 November 1998. 7. There is a further conflict of interest that may be of major significance in some scientists’ lives: the conflict of interest of employers, such as universities and perhaps corporations, who may find it in their interest to sanction scientists to protect themselves. 8. David A. Bella, “Engineering and Erosion of Trust,” Journal of Professional Issues in Engineering, 113 (April 1987): 117-129, esp. 117f. 9. See further, Russell Hardin, “Institutional Morality,” inThe Theory of Institutional Design, ed. Robert E. Goodin (Cambridge: Cambridge University Press, 1995), 126-153.

E THICS IN BIG SCIENCE 245 10. This paper partly grows out of a Wingspread conference, “The Moral Role of Scientists,” sponsored by the Midwest Consortium on International Security Studies. I especially thank: Marian Rice for organizing that conference; David Breeden, Michael Davis, and Ben Horowitz for extensive critical comments on my remarks there; the John D. and Catharine T. MacArthur, the Johnson, and the Russell Sage Foundations for their support of this project; and the Mellon Foundation for generous general support. Most of all I thank Robert K. Merton for probably the best seminar I have ever had on a paper. That session reminded me of the reputed remark of President Kennedy that one of his crowds of Nobel laureates at dinner in the White House represented the greatest average intellectual power seen there since Thomas Jefferson dined alone. My participation in the seminar might have been regrettable for the record books, but there could hardly be a seminar with no audience.