On occasion we are faced with acute moral choices – whether to join the Resistance or stay at home and care for our widowed mother; whether to run off with Vronsky or remain with Karenin. But largely, morality shapes our lives in ways we don’t even think about, in fact it does so partly by excluding certain options from our thoughts. Most of us, for instance, wouldn’t even consider (a) threatening to expose a colleague’s adulterous affair to his wife unless he votes our way on a contested appointment or policy issue; (b) extracting some cash from the pocketbook of an interior decorator as she inspects our house, because we think she is overcharging us; (c) stealing a kidney for a friend who needs a transplant; (d) selling all we have and giving it to the poor. It isn’t that we weigh the pros and cons and determine that the cons outweigh the pros. These things are not on the menu of options among which we feel we must choose. Such exclusions, as well as restrictions on what may legitimately be taken into account in some decisions but not others (prohibitions against nepotism, for instance), typify the complexity of moral standards and suggest that an accurate account of morality and its role in life will not be simple.
Thomas Scanlon’s understanding of this complexity and of its sources in the variety of human relations and values is one of the virtues of this illuminating book. To say that it is long awaited would be an understatement. Scanlon has been one of the most influential contributors to moral and political philosophy for years, but with overdeveloped diffidence has never published even a collection of his most important essays. The appearance of his first book, a complex and powerful argument for the moral theory first sketched in his essay ‘Contractualism and Utilitarianism’, is a philosophical event. Scanlon sets out an understanding of the nature and content of morality that is both original and credible, and he makes a strong case for its advantages over rival theories. The careful attention he gives to alternatives also provides an accurate picture of the current state of the field.
The book is about morality rather than politics, though its general method can be applied to the political domain, where some of the most heated moral arguments and controversies take place. Recent work in political theory is more widely known, but moral philosophy has been an intensely active field over the past three decades, and Scanlon’s theory addresses a number of its central questions: first, the question of the objectivity or truth of moral claims, their relation to reason, and whether or not they should be regarded as in some sense relative or subjective; second, the question of the kind of concern or respect for persons that is at the foundation of morality – what kinds of motive it calls on when it requires us to forgo certain means that would advance our personal aims, and how much it can ask that we sacrifice for the sake of others; third, the question of how, and to what extent, individual rights, liberties and prerogatives are morally shielded from encroachment in the name of the general good; fourth, the question whether modest advantages to each of a large number of people can be aggregated to outweigh a large cost to each of a much smaller number, for purposes of moral justification – a besetting problem for the intuitive acceptability of utilitarianism.
Scanlon’s answers to these and other questions are presented in a theory of right and wrong that gets some support from particular moral intuitions but which is also deeply unifying, foundational and systematic. It is not a general theory of value and the ends of life: Scanlon believes that morality – the standard of right and wrong in our dealings with other persons – forms a distinctive and uniquely important subpart of ethics more broadly conceived. That is the significance of his title.
The central claim is that the motivational source of morality is something quite different from the impartial universal benevolence most naturally expressed by a utilitarian system – a system whose ultimate standard is the maximisation of aggregate well-being. In fact, he sets himself against the natural but simplistic idea that well-being is the dominant value, or that any other measure of the good, conceived as an end to be promoted by everyone, is the basic form of value. Value takes many forms other than that of something to be promoted or maximised. One would not, he observes, show an appreciation for the value of friendship by betraying one friend in order to make several new ones.
Morality, too, is not identified with promoting the good – human happiness, for example. Its motivating aim, according to Scanlon, is a certain kind of relation with our fellow human beings, the relation of being able to justify our conduct to each other, as individuals, in what he describes as a form of ‘codeliberation’. That is how we show our appreciation of the distinctive value of persons – not by promoting a collective human good in which the interests of a minority may be outweighed by the greater aggregate interests of a majority.
The big question about such a proposal is where we are to find the standards that will enable us to justify our conduct to one another. Doesn’t this just postpone the search for the right standards by one step? The originality of Scanlon’s answer, and what will arouse the most critical resistance, is that he thinks the search for conditions of mutual justification will itself lead us to the right standards, by combining diverse reasons in an appropriate framework for the identification of acceptable principles. But for such a method to succeed, it must uncover forms of justification that avoid circularity; i.e. avoid appealing surreptitiously to precisely those moral principles that the process of mutual justification is supposed to warrant. It must appeal to some thing more fundamental.
Here is how Scanlon formulates his contractualism: ‘It holds that an act is wrong if its performance under the circumstances would be disallowed by any set of principles for the general regulation of behaviour that no one could reasonably reject as a basis for informed, unforced general agreement.’ The idea is that if our aim is to be able to justify our conduct to others, we will want it to conform to principles that none of them could reasonably reject, because then everyone who shares our interest in justification would in effect be prepared to license what we do insofar as it accords with those principles. If we deliberately do something that is in this sense wrong, we are in effect saying we don’t care about its admissibility to reasonable others.
The term ‘contractualism’ should not mislead: no actual contract is supposed to give rise to moral principles, only an imaginary agreement, by persons imagined to be both reasonable and motivated by the desire for such an agreement. This is in the tradition of Kant’s categorical imperative, which also tests principles of conduct by their hypothetical acceptability from all points of view, suitably harmonised. And as with Kant’s method, the application of Scanlon’s contractualism requires further value judgments, since the question of what constitute reasonable grounds for rejection of a principle is an irreducibly normative one.
The nerve of Scanlon’s position is that reasonable grounds for rejecting a principle come from the points of view of distinct individuals rather than from any collective or impersonal point of view. Utilitarianism would require us to accept principles that maximise the expected sum of human well-being, and reject those that do not, because the point of view from which acceptance or rejection is determined is that of impartial benevolence towards all. Scanlon, by contrast, believes one could reasonably reject certain principles that would maximise total well-being, in favour of other principles that would produce a lower expected total but that have other virtues – they are less unfair, they do not impose such severe burdens on anyone, they do not require the abandonment of important values not reducible to well-being. The reason ableness of an individual’s rejection of a principle depends on his taking the points of view of other individuals into account, but it does not depend on conformity to the verdict of an external point of view which is not that of any individual.
Here is one of his examples:
Suppose that Jones has suffered an accident in the transmitter room of a television station. Electrical equipment has fallen on his arm, and we cannot rescue him without turning off the transmitter for 15 minutes. A World Cup match is in progress, watched by many people, and it will not be over for an hour. Jones’s injury will not get any worse if we wait, but his hand has been mashed and he is receiving extremely painful electrical shocks. Should we rescue him now or wait until the match is over? Does the right thing to do depend on how many people are watching – whether it is one million or five million or a hundred million?
Scanlon thinks we shouldn’t wait, and that his contractualist approach explains why. The agony of Jones is vastly greater than the frustration any one of the viewers would feel at the interruption, so none of them could reasonably pose an objection to being deprived of 15 minutes of the game, merely to relieve Jones. And in Scanlon’s model, it is only individuals whose objections can knock out a principle. There is no collective point of view that combines the frustration of all those viewers (a billion watched the Final between France and Brazil last summer), and by reference to which Jones’s pleas for rescue can be reasonably rejected – or even be counted unreasonable. If, on the other hand, Jones could be rescued immediately only by a manoeuvre that would kill Smith, also pinned down by the equipment, then it wouldn’t be reasonable for him to object to Smith’s being freed first. The comparisons that determine what is reasonable must, according to Scanlon, be individual rather than collective.
The same applies when we are evaluating moral principles in advance. We have to imagine their prospective impact on the lives of individuals, and if a proposed principle would generate reasonable individual complaints more severe than the alternatives, it is to be rejected. For example, suppose that in the course of the construction of a new movie theatre in New York, an accident injures a pedestrian as severely as Jones has been injured. Here we have to compare the burden on any individual of a general slight risk of injury from construction accidents with the burden on each of those same individuals of ruling out all construction in cities, even with high but not fool-proof levels of care to minimise the risk. As Scanlon points out, it is clear that no one could reasonably reject a rule allowing construction projects with due care – not even someone injured as a result. This is not because the aggregate pleasure of the moviegoers outweighs the agony of the victim, but because a ban on construction would be pervasively and certainly constraining for practically everyone, including those who know it would save them from a small chance of being the victim of a construction accident.
Let me move to a more difficult example and ask how Scanlon’s contractualism would handle the question of whether, in the present global situation of inequality and indifference, a well-off person who wished to do the right thing would have to devote most of his energy and resources to combating the acute misery that exists in the world (with the implication that most of us in the rich countries are living morally unacceptable lives). Utilitarianism makes it difficult to avoid this conclusion, and it is accepted by some utilitarians, like Peter Singer. There are so many people you could save, each at a modest cost.
Scanlon mentions this problem without offering an unqualified answer, but here is a suggestion. While no one could reasonably reject some requirement of aid from the affluent to the destitute, the cumulative effect on an individual life of an essentially unlimited requirement to give to those who are very much worse off than yourself, whatever other affluent people are doing, would simply rule out the pursuit of a wide range of individualistic values – aesthetic, hedonistic, intellectual, cultural, romantic, athletic and so forth. Would the certain abandonment of all these things provide reasonable ground for rejection of a principle that required it – even in the face of the starving millions? The question for Scanlon’s model would be whether it could be offered as a justification to each one of those millions, and my sense is that perhaps it could, that one could say: ‘I cannot be condemned as unreasonable if I reject a principle that would require me to abandon most of the substance of my life to save yours.’
This sounds hard, and I am not sure whether Scanlon would accept it. But if he did, it would illustrate two important things about the method. First, as already indicated, it resists aggregation of the value of all the lives I could save by radically transforming mine, and makes the reasonableness of my rejection depend on a one-to-one comparison. Second, it gives a result with regard to the demand for self-sacrifice different from the result it gives with regard to principles governing the conduct of impartial third parties. In the latter type of case, Scanlon holds that it is right to save the larger number, when the threatened losses are comparable (as they are not in the case of poor Jones v. the soccer fans). If, for example, a disinterested third party somehow had to choose between preventing the loss to me of all the resources and opportunities that permit me to lead an agreeable life, and saving numerous other people from starvation, it is clear that no one could reasonably reject a principle requiring him to save the greater number.
Differences of this kind, depending on the relations of the actors and victims, are common in the morality most of us intuitively take for granted. That morality treats very differently (a) the choice of a disinterested third party about which of two groups of people to rescue and (b) the choice of one party to rescue himself or a loved one as opposed to some strangers, or (c) the choice of a third party whether to harm someone not otherwise in danger as a means of saving someone else from a greater harm, or several other people from comparable harm. Potentially, the contractualist focus on individual points of view may be able to shed light on these complex standards.
As an example of (c), it seems clear that it would be wrong to cut up a healthy person to provide organ transplants that would save the lives of five other people, and that a society which condoned such a practice would be monstrous: but why? Scanlon might say that any principle permitting this would, reasonably, be rejected by everyone, in advance of their knowing the likelihood of their needing an organ transplant, simply because it is essential for each person’s secure sense of self that the possible usefulness of his body parts to others should be ruled out of consideration absolutely. This is not actually circular, because it rests the general principle of bodily integrity on the vital importance of the sense of bodily inviolability for each individual. But it may seem a bit wobbly as a justification, since the ground, in a normative judgment about what it is reasonable to reject when bodily integrity conflicts with probable increases in ex ante life expectancy, seems just about as uncertain as the answer to the original question, and not so different from it in character. In the face of the uncertainty that opens up when we try to apply the contractualist method to difficult cases with conflicting individual points of view, those who want answers may conclude that only a move back to a more impersonal level can determine what it is and is not reasonable for individuals to reject, and that the contractualist framework is an unnecessary detour.
Others will find such cross-currents and uncertainties true to the complexity of the moral life. Or perhaps the right thing to say, with Aristotle, is that one should not demand from the philosophical treatment of a subject more certainty than the subject admits. Scanlon’s method is highly controversial, and so is its application to specific questions. It requires not just the plugging in of factual premises, but moral thinking all the way down. Its value lies not in providing a decision procedure but in identifying a very specific type of moral reasoning, and a special set of questions that must be posed in the course of it. Through its structure, Scanlon’s method tries to explain what moral questions are questions about. If such an account succeeds, one of the things it will explain is why some of those questions are so difficult.
The most basic level of normative thought concerns the reasons people have – reasons for acting or refraining, for rejecting or accepting a principle, for taking into account or excluding from consideration other reasons in this or that context, and so forth. These are not just reasons ‘for’ and ‘against’ doing something. Scanlon makes an illuminating comparison between reasons for acton and reasons for belief:
We all recognise that reasons for belief do not … simply count for a certain belief with a certain weight, and deciding what to believe is not in general simply a matter of balancing such weights. There certainly are cases in which deciding what to believe is a matter of ‘weighing’ evidence for and against the proposition in question, but this is so only because our other beliefs about the nature of the case identify those considerations as relevant for a belief of the kind in question. In general, a given consideration counts in favour of a certain belief only given a background of other beliefs and principles which determine its relevance … Because of these connections, accepting a reason for or against one belief affects not only that belief, but also other beliefs and the status of other reasons … My claim is that reasons for action, intention, and other attitudes exhibit a similarly complex structure. I do not mean to deny that deciding what to do is sometimes a matter of deciding which of several competing considerations one wants more or cares more about. My point is rather that when this is so in a particular case it is because a more general framework of reasons and principles determines that these considerations are the relevant ones on which to base a decision. Much of our practical thinking is concerned with figuring out which considerations are relevant to a given decision, that is to say, with interpreting, adjusting and modifying this more general framework of principles of reasoning.
Reasons at a multiplicity of levels are what shape our conduct and our morality. Scanlon takes the existence of reasons as basic, and indefinable in terms of anything else. They are not, in particular, reducible to desires or motives, and the reasons we have do not derive from desires.
We find out about what reasons we have by thinking about it. Since reasons are general, we can test the plausibility of a hypothesis about what we have reason to do or want by considering the credibility of its implications for other cases. Deciding what we have reason to do, like deciding what we have reason to think, is what makes us rational beings, and for a rational being, recognising a sufficient reason to do something can by itself motivate that action.
So Scanlon is what would be called a realist about moral truth, but his realism has no metaphysical implications. It falls entirely within the realm of reasons and morality, and rightly avoids the strategy of reducing these to anything else more ontologically or scientifically ‘respectable’. As he puts it:
The question at issue is not a metaphysical one. In order to show that questions of right and wrong have correct answers, it is enough to show that we have good grounds for taking certain conclusions that actions are right or are wrong to be correct, understood as conclusions about morality, and that we therefore have good grounds for giving these conclusions the particular importance that we normally attach to moral judgments.
His defence of objective truth in morality is therefore to be found in his substantive moral theory, and the arguments he offers for particular results – many of which I haven’t touched on, in particular the detailed discussions of promises and of the conditions of responsibility.
There is room in Scanlon’s theory for a degree of relativism, in two senses. The first is what he calls ‘parametric universalism’, according to which the appropriate ways to show respect for certain general values such as privacy or loyalty will vary with different social conventions or traditions. The second is that people in different social circumstances or from different traditions may have reasons to accept or reject different principles. Some things, like killing people because of their membership in an ethnic or religious group you don’t like, are wrong everywhere, whatever people may think. But other things may be wrong in one culture but not in another, because of different conceptions of personal honour, for example. Neither form of relativism is inconsistent with the objective correctness of moral judgments. These are just ways in which morality includes some relativity in its content.
Neither, of course, does realism about morality mean that the truth is what we now believe. As in any field where we are trying to get things right, or less wrong, we can never say that we have reached a point where openness to further revision is no longer necessary. As Scanlon observes, ‘Working out the terms of moral justification is an unending task.’ Moral philosophy should be interested in answers, but even more in fully understanding the questions. To this aim Scanlon has made what I believe will be an enduring contribution. Philosophy, like everything else, proceeds by the comparison of alternatives. Scanlon has presented a distinctive conception of the nature of morality that is compelling in itself, but that will deepen the understanding even of those who are not persuaded by it.