Objectivity

Samuel Scheffler

  • Natural Reasons: Personality and Polity by S.L. Hurley
    Oxford, 462 pp, £40.00, January 1990, ISBN 0 19 505615 9

To many people it seems obvious that science is objective in a way that ethics is not. Without being able to characterise the contrast between the objectivity of science and the subjectivity of ethics as clearly or precisely as a professional philosopher might wish, many people would nevertheless regard it as absurd to deny that such a contrast exists.

The prevalence of this attitude testifies both to the enormous intellectual prestige of modern science and to the extent of contemporary scepticism about traditional ways of understanding the sources and authority of our moral values and precepts. These two factors are hardly independent, of course, since it is precisely the rise of a broadly scientific worldview that has, more than anything else, served to weaken the systems of belief in which people’s values and principles have traditionally been embedded. The recent growth of religious fundamentalism, which might at first seem to constitute an important exception to these trends, appears upon reflection to be better understood as a reaction against them, a desperate attempt to reverse them, and as such it provides additional evidence of their power.

There is another factor that also helps to explain why a contrast between the objectivity of science and the subjectivity of ethics has come to be taken for granted by so many people. Facts about cultural diversity have always been appealed to by those who doubted the objectivity of ethics. Never before, however, have those facts been so widely and so insistently disseminated. The technological developments that have, in this century, revolutionised travel and communications, have undeniably promoted a certain amount of cultural homogeneity in the process: but, at the same time, they have also made it virtually impossible for anyone to be unaware of the rich and continuing diversity of human religions, ideologies and cultures. Technology has made amateur anthropologists of us all, for it has given us no choice but to inhabit a world that is awash in evidence of its own pluralism. Since facts about cultural diversity have always seemed to many of those who confronted them to cast doubt on the objectivity of ethical values and principles, it is hardly surprising that in such a world these doubts should have grown more widespread. And since scientific beliefs, unlike systems of value, appear relatively immune to cultural variation, it is equally unsurprising that their objectivity has not seemed similarly under threat.

Interestingly enough, the awareness of pluralism also contributes in another way to the popularity of ethical subjectivism. To those who take pluralism seriously, subjectivism often seems to constitute the only humane and tolerant response. For subjectivism seems to imply that all of the various ways of life and systems of value that we encounter have the same kind of legitimacy: each is validated, for those who accept it, by the fact that it represents their preferred conception of how to live. No way of life is objectively superior or privileged. To claim otherwise may seem arrogant, narrow-minded or intolerant.

Now this train of thought, although seductive, is nevertheless confused. For when it is taken as a recommendation of subjectivism that it constitutes a humane and tolerant response to pluralism, humanity and tolerance are themselves being treated, not merely as subjective preferences, but as values worthy of allegiance. Our objection to inhumanity and intolerance is not that they don’t happen to appeal to us, but rather that they transgress authoritative values and norms. And this is just the type of thought we purported to be repudiating as narrow minded and intolerant when we embraced subjectivism.

The fact that we may seem to be driven toward subjectivism, not only by scepticism about our values, but also by those values themselves, helps to explain the powerful appeal of that doctrine. However, the idea that subjectivism about values receives support from certain of our own values is untenable, and the often unrecognised influence of that idea is symptomatic of a deep instability in contemporary attitudes towards morality. For, on the one hand, a broadly subjectivist understanding of morals has never been more widely accepted. Yet, on the other hand, we continue to make moral and evaluative judgments all the time, and those judgments purport to be communicating something other than our subjective preferences. For most people, moreover, the disposition to make such judgments is so deeply embedded in their personalities that they could not cease making them even if they wanted to, nor would they be likely to want to if they reflected on the many important ways in which such judgments help to structure human emotions and attitudes, as well as interpersonal relationships and social arrangements.

In view of these considerations, the obvious question is whether there is any chance of establishing the objectivity of ethics in the face of those considerations that seem to favour subjectivism. Until recently, philosophy in the analytic tradition had done little to encourage such hopes. The great analytic movements of logical positivism and linguistic analysis tended by and large to insist on a sharp distinction between facts and values, and in the heyday of those movements the orthodox view was that evaluative utterances were not even genuine assertions, let alone assertions of objective fact, but were instead either expressions of emotion, or something closely related. In addition, this view was taken to support a sharp distinction between ‘meta-ethics’, which was thought to consist exclusively in the analysis of moral language, and ‘normative ethics’, understood as involving the use of such language to make substantive moral judgments. Meta-ethics, according to the prevailing orthodoxy, was a subject of legitimate philosophical interest, but normative ethics, by most accounts, was not, since ethical statements remained expressions of emotion whether uttered by philosophers or non-philosophers.

Things began to change dramatically in the late Sixties and early Seventies. By that time the influence of positivism had waned considerably, and developments within the philosophy of language and other areas of the subject were widely seen as casting doubt both on the dominant analyses of moral discourse, and on the sharpness of the distinction between meta-ethics and normative ethics. As a result, normative ethical inquiry gained new-found philosophical respectability, and, against the background of the social and political upheavals of the period, it rapidly developed into a very active area of the subject. Almost overnight a new philosophical literature emerged dealing with topics like war, abortion and affirmative action, and this was followed soon after, especially in the United States, by a proliferation of university classes on ‘contemporary moral problems’ and ‘applied ethics’. New work in meta-ethics was slower to appear, but the resurgence of interest in normative issues, far from obviating the need for such work, made vivid the importance of arriving at some more satisfactory way of understanding our substantive moral thought and practice. Gradually, over the last decade or so, a new meta-ethical debate has begun to take shape, as several different positions have been articulated with increasing clarity and force. Although these positions differ sharply from each other, none of them relies on a rigid distinction between meta-ethics and normative ethics, and, accordingly, none of them conceives of its aims as restricted to the analysis of moral language. Instead, each sees itself as attempting to provide a broad understanding of the nature and status of ethical norms.

This does not mean that subjectivism has disappeared. The term ‘subjectivism’ can of course be understood in different ways, but if we use it to refer to the view that there are no objective ethical truths, then, far from disappearing, subjectivism has, in recent years, received new and more sophisticated formulations which are not vulnerable to the objections that undermined earlier versions. On the other hand, a variety of nonsubjectivist positions have also been forcefully advanced, and although subjectivism continues to be an important view which has many adherents, it can no longer be said to be the dominant position in English-language moral philosophy.

This is an important development, and Susan Hurley’s impressive new book represents a significant addition to the array of contemporary non-subjectivist outlooks. Unfortunately, although Hurley’s book is aimed at a wider audience than works of philosophy usually command, many prospective readers will find it largely inaccessible. In part, this is because Hurley has the explicit aim of drawing connections among a number of seemingly disparate areas of inquiry, including ethics, the philosophy of mind, the philosophy of language, decision theory, social choice theory, jurisprudence and political philosophy, and this leads her to spend a considerable amount of time moving back and forth among different bodies of professional literature, some of which are highly technical in nature. Moreover, this procedure, although admirable in its intent and often illuminating in its execution, is pursued so single-mindedly that it frequently impedes the flow of argument and distorts the structure of exposition. At times it seems that Hurley would prefer never to make a point directly if she can instead make it by summarising a debate in some other field and showing that it supports a conclusion analogous to the one at which she is driving. The problem is compounded by her highly professionalised style of writing, which is characterised by long, complicated sentences, and a level of academic jargon far in excess of the legal limit.

This is a pity, because Natural Reasons is a work of considerable intelligence and originality. It is far more original, in fact, than Hurley’s extensive appeals to the professional literature might lead one to believe. She develops an account of the nature of ethical judgments and ethical deliberation which is novel in important respects and which represents a first-rate contribution to the literature of moral philosophy. She links this account in interesting ways to a conception of self-determination and autonomy, and, in a long final chapter, she makes some intriguing suggestions about the implications of her account for democratic theory. Along the way, she has significant discussions of many issues, including an excellent treatment of akrasia (or weakness of the will), and of the nature of conflicts among reasons for action.

Hurley’s response to the naive contrast between the subjectivity of ethics and the objectivity of science is to adopt what she describes as a ‘divide-and-conquer strategy’. She disentangles four distinct theses that might be intended by the claim that ethics is subjective. Three of these, she argues, are false. The fourth may or may not be true, but its truth would not in any case undermine the objectivity of ethics or the possibility of ethical knowledge. Subjectivism proper, as she construes it, is the view that preferences are conceptually prior to values, and that the latter are determined by the former. Non-cognitivism is the doctrine that ethical judgments serve to express preferences rather than beliefs, and that they are therefore incapable in principle of being either true or false. Scepticism is the position that ethical knowledge is impossible. And scientific anti-realism is the view that evaluative concepts have no role to play in the kinds of scientific theories that explain how the world works.

Hurley’s argument against subjectivism proper appeals to conceptions of interpretation drawn from the philosophy of mind, and in so doing it rests heavily on the work of Wittgenstein and the Berkeley philosopher Donald Davidson, among others. The aim of the argument is to show that preferences cannot determine values in the way that the subjectivist supposes, because values and preferences are conceptually interdependent. That is, it is impossible to identify a person’s preferences independently of his values, for our judgments about what preferences people have depend in part on our shared sense of what values they can intelligibly be understood to accept. This argument is developed in the context of formal decision theory, where the problem of interpreting people’s preferences in the light of their behaviour arises in a very clear-cut way, and where, accordingly, the role of values in constraining such interpretations can be readily illustrated.

Against non-cognitivism, Hurley argues, with reference to work in the philosophy of language, that the interpretation of ethical judgments as expressing preferences rather than beliefs cannot do justice to the logical character of an important class of valid inferences from premises containing ethical terms. These cases are said to support the cognitivist position that ethical judgments are best understood as genuine assertions which express beliefs and which purport to be true.

A sceptic could concede this point, but insist that the ethical beliefs expressed in our judgments are all false, or, at any rate, that these beliefs never amount to genuine knowledge. Hurley deploys her own account of the nature of ethical judgments against this kind of scepticism. According to her, when we say that one ought to perform a certain act, we are to be understood as claiming that the act is favoured by that theory, whatever it is, that provides the most coherent account of the specific reasons or values that apply to the options at hand. Suppose, for example, that we must decide between two options, one of which is supported by considerations of fairness and equality, while the other is supported by considerations of liberty and autonomy. If we decide that we ought, all things considered, to choose the first option, then what we are claiming is that that option is favoured by whatever theory provides the most coherent account of the relations amongst fairness, equality, liberty, and autonomy. Thus deliberation about what one ought to do is a kind of theoretical activity; it is the quest for a coherent theory. The relevant type of theorising can be quite complex. Hurley illustrates the process with an elaborate legal example, and her view is notably indebted to the jurisprudential writings of Ronald Dworkin.

A sceptic might claim that our beliefs about what ought to be done are all false because it is impossible that there could be a coherent account of the relations among our specific values. Hurley considers and rejects an argument along these lines which attempts to establish the impossibility of a coherent relation among an individual deliberator’s values by analogy to impossibility results in social choice theory. A more common and influential form of scepticism begins by maintaining that the correct explanation of why people have their ethical beliefs does not in itself make any use of ethical concepts or propositions. Perhaps there is a satisfactory explanation in psychological terms, or perhaps some sociobiological explanation is correct. In any event, the sceptic continues, since the truth of our ethical beliefs is not what explains why we accept them, we would accept them even if they were false. And if that is so, then these beliefs do not satisfy the conditions for genuine knowledge. Against this type of scepticism, Hurley argues that the possibility of explaining our acceptance of ethical beliefs without using ethical concepts would not by itself show that those beliefs were insensitive to the truth, or that we would accept them even if they were false. Thus the sceptical argument provides no general reason for doubting the existence of ethical truths or the availability of ethical knowledge even if we assume that some such explanation is correct.

The idea that some such explanation is in fact correct comes very close to the doctrine of scientific anti-realism. Hurley is content to remain agnostic with respect to the truth or falsity of that doctrine because, she believes, its truth would not reinstate subjectivism, non-cognitivism or scepticism, nor would it invalidate the type of interpersonal understanding to which ethical concepts and judgments contribute. Ethical concepts may not figure in scientific explanations, but scientific explanation is not our only source of understanding or objective knowledge.

How, on Hurley’s view, are we to understand facts about cultural diversity and pluralism? Someone might object that if people from different cultures have different specific values, then when they seem to be disagreeing with each other about what ought to be done, Hurley’s ‘coherence view’ requires her to say that they are in fact making claims about the relations among different sets of evaluative considerations. And this implies that they are not actually disagreeing at all, but are instead addressing different questions. Thus, someone might argue, far from providing an objective basis for adjudication among conflicting moral beliefs, Hurley cannot even recognise the reality of such conflicts.

In response, Hurley acknowledges, and indeed insists, that if certain people have specific values that are so alien to us as to seem unintelligible, then the type of conceptual common ground necessary for substantive disagreement with them may be lacking. A background of conceptual agreement is required if disagreement about particular issues of substance is to be possible. Usually, however, when we find ourselves in apparent disagreement with others about what ought to be done, this requirement is met. Rarely, in such cases, do we find the specific values that our opponents appeal to unintelligible. Indeed, unless we can interpret them as having many of the same values that we do, and as participating in many of the practices whereby we treat some types of consideration but not others as providing reasons for action, we may not be in a position to regard them as having values or as acting for reasons at all. Thus many apparent disagreements about what ought to be done – Hurley mentions disputes about vegetarianism, political terrorism and mercy killing – are just what they seem to be: genuine substantive disagreements. They are made possible both by substantial agreement in our specific values and ‘reason-giving practices’, and, crucially, by the fact that our general concept of what ought to be done is an ‘essentially contested concept’. In other words, its application is answerable to all of our specific values and thus to an entire array of potentially conflicting criteria, and this makes deliberation, argument, and disagreement about what ought to be done both possible and unavoidable. It does not, however, preclude the possibility of resolving such disagreements by theoretical reflection, although there may be some instances in which two different theories give equally coherent accounts of the specific values that apply to a particular case, and in which, accordingly, two different answers to the question of what ought to be done may both be equally good.

Pending a more detailed analysis of at least some actual moral controversies, this response cannot be regarded as fully satisfactory, for the following reason. Different religions, ideologies and cultures are distinguished largely by the ways in which their values and reason-giving practices, although overlapping to a considerable degree, nevertheless differ from each other. For Islamic fundamentalists the fact that a book is regarded as blasphemous may be a reason for banning the book and killing its author; for Western liberals it is a reason for neither. Each group may find the other’s values profoundly alien, although not, perhaps, unintelligible. Do the two groups genuinely disagree about whether the author ought to be killed? If so, is there any objective basis for resolving the disagreement and saying that one side is right and the other wrong? Any putative defense of ethical objectivity that fails to provide affirmative answers to both of these questions will be open to a charge of false advertising. Yet it needs to be shown that affirmative answers are available on Hurley’s view. For on that view, a claim about what ought to be done is a claim about which course of action is favoured by the theory that provides the most coherent account of the specific values relevant to the case. And the problem in this case is that Islamic fundamentalists and Western liberals have different, albeit overlapping, sets of values that they regard as relevant. Thus if Hurley’s view is to yield affirmative answers to our questions, we need to know what reason there is for supposing that some one theory provides the most coherent account of both sets of values, despite the differences between them.

It should also be noted that some people would regard Hurley’s conception of ethical objectivity as insufficiently robust even if it did support affirmative answers to both of our questions. For Hurley’s understanding of objectivity depends on the idea that the ‘existence of certain shared practices, any of which might not have existed, is all that our having determinate reasons ... to do anything rests on’. And some people will feel that this represents an intolerably deflationary view of ethical considerations and their role in our lives. These people may find themselves drawn to one of the other non-subjectivist positions now being developed: perhaps to some version of a Kantian position, or to a view that makes stronger claims than Hurley about the explanatory role of ethical considerations.

Of course, the fact that Hurley’s view is unlikely to satisfy everyone is, in itself, hardly surprising. What may be more surprising, to those unfamiliar with the recent history of moral philosophy, is the fact that philosophical criticism of her position is at least as likely to come from those who regard it as insufficiently objectivist as from those who regard it as too objectivist. Like Hurley’s book itself, this testifies to the growing vitality of non-subjectivist thought within academic moral philosophy. Since we urgently need a better way of understanding our own values than is afforded by the naive subjectivism that so often presents itself as common sense, this development is greatly to be welcomed.