T. M. Scanlon’s contractualism and more specifically his conception of morality narrowly conceived proscribe direct appeal to what he calls impersonal values, such as items of natural beauty or cultural goods. In the scope of what we owe to each other, one should appeal only to concerns arising from what bears directly on individuals, by reference to which they can reasonably reject one set of principles in favour of another. There is an overlooked ambiguity here in how we should understand the scope of “what we owe to each other.” Unpacking this helps us see more clearly how contractualism can respond to the challenge that it must accommodate the intuitive force of aggregating: that, when all other things are equal, we are required to save a greater number of people over a lesser. I explain the role personal and impersonal value have in Scanlon’s contractualism. I connect the focus on personal value with perspectivalism, the justificatory privileging of the perspective of victims. Perspectivalism, I suggest, is motivated by Scanlon’s constructivist methodology. Whether perspectivalism is adequate as the basis of contractualism is not independent of the question of aggregation. Given the ambiguity highlighted, I conclude that other forms of contractualism might safely ignore the demands of aggregation.
Le contractualisme de T. M. Scanlon et plus particulièrement sa conception de la morale étroitement conçue proscrit l’appel direct à ce qu’il appelle des valeurs impersonnelles, telles que les objets de beauté naturelle ou les biens culturels. Dans le cadre de ce que nous nous devons les uns aux autres, on ne devrait faire appel qu’aux préoccupations découlant de ce qui concerne directement les individus, par référence auxquelles ils peuvent raisonnablement rejeter un ensemble de principes au profit d’un autre. Il y a ici une ambiguïté négligée dans la façon dont nous devons comprendre la portée de « ce que nous nous devons les uns aux autres ». Déballer cela nous aide à voir plus clairement comment le contractualisme peut répondre au défi qu’il doit prendre en compte la force intuitive de l’agrégation : que, toutes choses étant égales par ailleurs, nous sommes tenus de sauver un plus grand nombre de personnes par rapport à un moindre. J’explique le rôle de la valeur personnelle et impersonnelle dans le contractualisme de Scanlon. Je relie l’accent mis sur la valeur personnelle au perspectivalisme, le privilège justificatif du point de vue des victimes. La perspective, je pense, est motivée par la méthodologie constructiviste de Scanlon. Que le perspectivalisme soit adéquat comme base du contractualisme n’est pas indépendant de la question de l’agrégation. Compte tenu de l’ambiguïté mise en évidence, je conclus que d’autres formes de contractualisme pourraient ignorer en toute sécurité les exigences de l’agrégation.
Suppose I propose that we should build a high-tech industrial park right on the slopes of Mount Snowdon, an opportunity for greater profits given the proximity of the sea; or suppose we consider whether the British Museum should simply be remodelled as a luxury hotel, well-situated for Continental financiers on their way to the City from St. Pancras. In each case, you might object that the vulgar plan misses the significant value that resides in the object under threat: in the one case, that of natural beauty; in the other, that of great cultural worth. And it may seem fairly straightforward to conclude that it is the value that resides in the Welsh countryside itself, and the value in the institution of the British Museum that is the ground of such a complaint.
T. M. Scanlon’s conception of morality narrowly conceived, or “Narrow Morality,” as I will refer to it in what follows, does not allow such direct appeal to values such as these in justifying any prohibition on principles that govern how we should treat each other. Morality more broadly conceived may allow for an appeal to what he calls impersonal values such as items of natural beauty or cultural goods in themselves. But in Narrow Morality, we are allowed to appeal only to concerns that arise from what bears directly on individuals, and by reference to which they can reasonably reject one set of principles over another. Considering whether a principle might be rejected on impersonal grounds—“that is to say, for reasons that are not tied to the well-being, claims, or status of individuals in any particular position” (Scanlon, 1998, p. 219)—Scanlon explains:
Many people, for example, believe that we have reason not to flood the Grand Canyon, or to destroy the rain forests, or to act in a way that threatens the survival of a species (our own or some other), simply because these things are valuable and ought to be preserved and respected, and not just because acting in these ways would be contrary to the claims or interests of individuals. Whether they are correct in thinking this is not, however, a question to be settled by an account of the morality of right and wrong; it belongs to morality in the wider sense and to the broader subjects of reasons and value. … To claim this—to claim, for example, that in destroying an ancient monument or tree I do no wrong to anyone—is not to claim that we have no reason (or even no moral reason in the broadest sense) not to commit such acts. It would even be natural to say that it would be wrong to destroy these things, using ‘wrong’ in the broad sense in which something is wrong if there is a very serious reason against doing it. But insofar as the value of these objects provides me with a reason to preserve them, it would be a misrepresentation of this reason to say that it is grounded in what I owe to others.Scanlon, 1998, p. 219
Narrow Morality may well prohibit redeveloping Snowdonia or refashioning of the British Museum, but if it does so, it does so not through an appeal to the value inherent in these objects, but rather by consideration of the concerns and interests that individuals invest in the beauty of the one and the cultural worth of the other.
This observation, I suggest, has a bearing on how we should understand Scanlon’s now notorious discussion of aggregative concerns and requirements on saving the greater number in chapter 5 of What We Owe to Each Other. I shall suggest that there is an overlooked ambiguity in how we should understand the scope of “what we owe to each other,” and this centrally connects to the overall ambitions of Scanlon’s theoretical approach. In turn, that highlights a question about what significance contractualism, as an approach to moral and political theorizing, should place on the allegation that any adequate account of morality should take account of the intuition that, when all other things are equal, we are required to save a greater number of people over saving a lesser number.
What I set out to do in this brief paper is to explain how the contrast of personal and impersonal values fits into Scanlon’s contractualism. I want to connect the focus on personal value with what we might call perspectivalism, the justificatory privileging of the perspective of victims. Perspectivalism, I suggest, can be understood as motivated by the form of constructivist methodology that Scanlon favours. Whether perspectivalism is an adequate account of the basis of contractualism more generally, though, partly turns on the question of aggregation. In consequence, with the ambiguity clearly in view, I’ll suggest that the fortunes of contractualism do not so stand or fall on how we respond to allegedly aggregative concerns.
Consider two different glosses we can put on Scanlon’s slogan about Narrow Morality, that it concerns what we owe to each other. We might read this so:
WWOtEO1: The collection of benefits, harms, injuries or rewards we each distribute among us all.
And we can contrast this with a rather different gloss, equally central to Scanlon’s own concerns:
WWOtEO2: What is required and permitted of us deriving from the distinctive kind of justifications of our actions toward each other, where permission is filtered through what each can reasonably reject of the claims of everyone else.
The first of these picks out the set of concerns, goods, and rights that fall within the scope of Narrow Morality. The second definition, on the other hand, directs our attention to a particular procedure, or set of justifications, from which permissions, rights, goods, and penalties might be derived. As I read Scanlon’s discussion of these matters, an ambition of the book is that in the end these two definitions should coincide: the set of goods and rights properly assigned to individuals are those derived by the procedure of reasonable rejection. But that raises a question that I’ll pursue below: What if we question the scope of the derivation, but still accept the idea that a portion of morality connects directly with what we should give to, and can demand from, each other?
Why should Scanlon rule out appeal directly to the beauty of Snowdonia or the worth of the British Museum in explaining the prohibition on speculative redevelopment? One answer is offered in his discussion of well-being and what we might call Welfarism, the treatment of well-being as a “master value.” In criticizing such a conception of well-being, Scanlon rejects an associated methodology: that we should identify the foundational values of morality and derive our principles on the basis of these (Scanlon, 1998, chapter 3, particularly p. 141ff). Scanlon objects to such foundationalism: we should not suppose that we can give a complete overview of the grounds of our moral verdicts that make no appeal to other normative notions. In its place, he suggests that we need a holistic and piecemeal approach to moral theorizing that looks to the local arguments we are prepared to give in favour of, or in opposition to, a particular verdict. This will not be uninformatively circular, he assures us, if the system as a whole has appropriate virtues.
Buried in this rejection of foundationalism is an additional methodological element beyond the appeal to holistic properties of theories, and a preparedness to deal with these matters piecemeal; we might label this additional constraint perspectivalism. If our systematic account is to look to the justifications we offer in local disputes, there remains the question whose arguments need to be taken into account. For example, one might think that the torture of young children is something that we can all easily recognize as a wrong, something to be annotated as such in any acceptable account of rights and wrongs (Thomson, 1990, p. 30, n. 19). But in the context of Scanlon’s approach, another question comes to the foreground: whose objection to this behaviour should be seen as fundamental to its rejection? If it is simply the recognition of the badness of this situation by just any one of us, or all together, then nothing, so far, indicates any difference in this kind of harm from the despoliation of Snowdonia: in both cases we find, on the one hand, something of a recognizable value and, on the other, actions that fail properly to respect that value. If ruining Snowdonia is ruled out by an impersonal value, one might suppose that the torture of children too is ruled against by something impersonal.
Note that for Scanlon it won’t do to make the difference reside in the idea that the values in play with young children concern the welfare of persons. This is not because one may not consider infants, being still before the age of reason, properly persons. It is rather that the rejection of foundationalism in moral theory prevents us from privileging any particular source of value as having a distinctive role here. The insistence that persons are special and, in turn, so too are their interests is no less a form of the foundationalism under question than making an eloquent ecological plea for the significance of Snowdonia is.
What Scanlon’s methodology requires is that we focus on the features of local disputes about what to do, or about what we should rule out as impermissible: contexts of providing or rejecting justifications for courses of action. Against this background, one might surmise that the claims of well-being arise as significant not in virtue of some kind of foundational value—that the welfare of infants, or puppies, is paramount—but precisely because, as matters stand, victims protest of their injuries (that is, of the damage done to their well-being) or others protest on their behalf, standing in for the perspective that the infants would have. Rather than taking any value as foundational, Scanlon takes as primitive the form that justification takes: the essential elements in providing and testing a justification for any course of action. In doing this, he privileges the perspectives of those who formulate justifications and those to whom justifications are given. The justificatory privileging of the perspective of victims is what brings out the contrast with the despoliation of Snowdonia, and what marks the bounds of Narrow Morality. The torture of young infants belongs properly within the scope of Narrow Morality, if we can extend the concern for victims to them, even if they are not yet actually able to occupy the position of justifiers or justified to agents.
Scanlon’s principal objection to treating well-being as a master value, as found in Welfarism, is the commitment—that allegedly comes with it—to foundationalism. We have no reason to suppose an adequate account of our moral practice can be provided through a re-description of the piecemeal parts of our moral scheme in terms other than those of right, wrong, entitlement, or prohibition: these terms provide appropriate justification for the consequences of normative classifications.
So, even if Scanlon rejects the foundationalist ambitions of utilitarianism as misguided, that leaves open the possibility that he might still agree with Welfarists about what is of significance in each case of right or wrong. Leaving aside the foundationalist ambition, it might turn out that rightness and wrongness always turned on the benefit or harm to some person’s or persons’ well-being. But, in fact, Scanlon disagrees with Welfarism here too. Welfarism is focused distinctively on questions of benefit and harm, and different forms of Welfarism debate what can be considered part of one’s welfare. Although Scanlon doesn’t deny that harm and benefit generate reasons for us, he rejects the idea that these could exhaust our concerns in justifying a principle as correct or in rejecting a principle as unreasonable. Most notably, Scanlon insists that we all have an interest in questions of fairness as well as questions of welfare. The complaint of unfairness, he highlights, is not reducible simply to a second-order complaint of harm:
It is no doubt particularly clear that individuals typically have strong reason to want to have certain benefits, and to want to avoid pain and injury. Perhaps this claim can be generalized to cover anything that affects “how well one’s life goes.” But these are not the only things that people have reason to want and to object to being deprived of. …For example, … it is reasonable to object to principles that favor others arbitrarily. A principle that favors some in this way will often deprive others of benefits and opportunities they have reason to want. But why should these concrete disadvantages be the only grounds for objecting to such a principle? It would be circular for contractualism to cite, as the reason that people have for objecting to such principles, the fact that they are wrong according to some noncontractualist standard. But we need not choose between objections of this kind and objections based on loss of well-being. We have reason to object to principles simply because they arbitrarily favor the claims of some over the identical claims of others: that is to say, because they are unfair. In the process of moral reflection that contractualism describes, this provides a perfectly understandable reason for finding partial principles objectionable, a reason that does not depend on a prior idea that such principles, or the practices they would permit, are wrong.Scanlon, 1998, p. 216, emphasis added
The complaint of unfairness goes beyond a second-order complaint about the distribution of harms and benefits. Should we then think of fairness as an impersonal value? If the key contrast between the natural beauty of Snowdonia and torture were a matter of how individual persons are affected with respect to welfare, one might surmise that fairness could not be a value embodied in individuals as such. Like natural beauty, it might be thought of as the kind of value that stands apart from, and needs to be abstracted out of, more basic concerns that all organisms pursue, and which play a central role in understanding welfare. We might think that morality broadly understood needs to bring into harmony such further values with any concern we have with welfare. As I have already noted, this interpretation doesn’t really fit the way Scanlon sets up the issue. He is unlikely to mark out the realm of personal value in quite this way, since he is critical of the foundationalist ambitions of utilitarianism and of Welfarism; so, fairness could hardly earn its keep by being introduced as yet some further matter of import that we needed to recognize.
We can see how fairness properly belongs within the personal realm of concerns if we make more explicit the role that different perspectives play in settling the moral claims on us. Fairness finds its place among personal perspectives on what it is reasonable to rule out: victims may complain that they have been wronged through being treated unfairly, even if no harm to their welfare can be identified independent of the insult of unfairness. Moreover, we may reasonably reject some course of action given its unfairness, even if no one as such is harmed, and no individual is the victim or is wronged. Indeed, one may take fairness as precisely the example that Scanlon wishes to highlight of that aspect of our ordinary justification of courses of action that cannot be accommodated within a foundationalist methodology. It is something of concern to those who seek to justify a course of action or reject the actions of others; yet it is something that can be framed only in normative terms, as he goes on to observe, just after the lines quoted above:
It seems to me an important strength of contractualism that, in contrast to utilitarianism and other views which make well-being the only fundamental moral notion, it can account for the significance of different moral notions, within a unified moral framework, without reducing all of them to a single idea.Scanlon, 1998, p. 216
One might then see, lying behind the discussion of circularity, the faintest sketch of a constructivist account. We need only suppose that all agents start with each having a concern for welfare, their own plus that of their loved ones, and a concern to justify their actions to anyone else equally disposed to justify their behaviour. Fairness figures both as a concern with the effects on well-being of unduly favouring some over others, but also as an interest with reasonable procedures, a concern we might consider inseparable from the focus on reasonable justification to individuals. This second concern is still placed by Scanlon in the realm of personal reasons, precisely because it has to do with the status of individuals within a pattern of justification:
In discussing fairness … I maintained that it is sufficient ground for rejecting a principle that it singles others out, without justification, for a privileged moral status. Even if these reasons do not have to do with well-being, however, they are still what might be called personal reasons, since they have to do with the claims and status of individuals in certain positions.Scanlon, 1998, p. 219, emphasis added
Scanlon insist about reasons of fairness being in one sense personal. However, what is distinctive is that concerns of fairness arise within the structure of justification of individuals to each other. Out of this pattern, we will have the reasons of morality narrowly conceived, grounded either in what people care about or, as the case of fairness illustrates, in what they care about in the structure of justification itself (such as treating alike, alike, for example).
We started the discussion with the methodological constraint, perspectivalism, the privileging of the perspective of victims in understanding the force of moral claims. Setting perspectivalism up as in opposition to foundationalism, and providing an alternative approach to moral theorizing, one might also suggest that this constraint is closely tied to a constructivist methodology in such theorizing. What Scanlon identifies as Narrow Morality is of interest to us not so much practically—that is to say, as part of a substantive question about what we should care about (although that may be of concern)—as theoretically, as making us see what the distinctive nature of moral concerns are in the first place. It shows us what we can establish about the moral, while making the minimum assumptions about the activity we are engaged in.
And putting things together in this way brings into focus the question about Scanlon’s treatment of aggregation that I want to raise in this brief discussion. It is not at all obvious that the only sense we can make of a contractualist perspective on social policy is via Scanlon’s formal ambitions of framing a constructive justification of the various elements of Narrow Morality. That is, in other terms, it may suit us for some purposes just to focus on the first definition of what we owe to each other, WWOtEO1, as the basic remit of measuring the correctness of such policies. In contrast, given a constructivist gloss, in keeping with WWOtEO2, Scanlon’s story gives an account of why impersonal values should have no role in Narrow Morality. Without this additional constraint, we have no clear-cut basis to rule in or out of our deliberations further values other than those of welfare or fairness. And what I’ll suggest in the discussion that follows is precisely that the constructivist gloss on personal values skews Scanlon’s presentation of the problem of aggregation.
In What We Owe to Each Other, Scanlon resists consequentialist theories that aggregate the sum of welfare. He requires that we adopt principles that no one has a reason to reject: we are to justify principles to each of the people they affect. Indeed, one might surmise that the original impulse against impersonal values is to resist theories that would require that the aggregation of modest advantages to a large number of people outweigh the severe cost to each of a much smaller number of people. This is the sort of unintuitive result of utilitarian theories highlighted by Scanlon’s much-discussed example of Jones and the football fans.
The moral Scanlon wishes to extract from the example of Jones is that we should not find the aggregation of lesser harms of the various and numerous viewers sufficient to outweigh the more serious concerns of the single Jones. This shows that at least some commonly proposed aggregative principles are counterintuitive, and so likely to be rejected within a systematic theory. But it doesn’t in itself show that aggregative concerns never have any force over us. Consider: in a case where an identical severe harm threatens to befall two groups of people unequal in number, no one individual would seem to have any concern greater than that of any other. In the kind of examples Scanlon considers, each out of three people faces near-certain death if not rescued: we can imagine them stranded one on one rock, and two on another, and a lifeboat with time to rescue only one of these two groups as the tide rises.
Now Scanlon himself feels the intuitive pull of a requirement that the person in charge of the lifeboat should strive to save the larger-numbered group, all else being equal. While he feels that pull, he wishes to resist any explanation of it in terms of aggregative principles. As with his earlier discussion of well-being, he is concerned to derive an account of the relevant values out of the pattern of argument that individuals can offer for or against the rejection of various policies. Where an identical severe harm befalls all members of the two groups, individual reasons are balanced. If one can save either group but not both, there is no welfarist concern that any given individual can appeal to as a victim in order to back up a complaint against a principle that permitted rescuers to save the others rather than him or her. In each case the well-being of some individual will be sacrificed, and no individual has a greater claim than any other individual. So individual claims seem, at first sight, not to favour the saving of the larger group over the smaller, but rather (as John Taurek originally proposed) an indifference between them. Moreover, direct aggregation of claims (or “reasons” as Joseph Raz has it) is not permitted, as far as Scanlon is concerned, given the structure of justification.
Again, as has been the object of much discussion in the further literature, Scanlon seeks to draw on resources that he finds in Frances Kamm’s writings to offer an explanation of why the requirement to save the greater number arises (See Kamm, 1993, p. 116-117). In essence, his proposal is that we need a different procedure for cases in which we can save either of two people, one-on-one cases, and two-on-one cases such as that facing the person in charge of the lifeboat. The additional person on the rock with two people will have a complaint if we adopted the same procedure for a one-on-one case and for a two-on-one case.
Most commentators (often coming from very different perspectives) have found what Scanlon has to say here unsatisfactory. But what I want to do here is to abstract away from that specific dispute and reflect on the terms in which the debate has been framed in the light of the concerns we have raised so far. For if Scanlon is to provide a principle in this case consistent with the constraints he imposes on a systematic moral theory more generally, then he needs to find a source of justification that arises from, so to speak, a victim’s or rescuer’s perspective.
That, I take it, is how we are to understand his proposal that in the larger group, the surplus person has a complaint that he or she has not been taken into account. Were such a complaint appropriate, then one could understand its force as arising, as with fairness, through a concern with the proper form that justifying to others and being justified to should take. In this way, the consideration would be appropriately individualistic in the sense that Scanlon endorses. And, once the proposal is seen in this light, whatever its other difficulties, it should be clear that no appropriate procedural ground has been identified. For the concern that the additional person has not been taken into account rests on the thought that the rejected verdict (“Save either group”) is the same whether the competing groups are one-on-one or two-on-one. But that the outcome of a deliberative procedure is the same when the numbers are different does not suffice to show that the considerations that lead to the outcome are the same in the two situations. Rather, we need some sketch of the reasoning that leads to the decision. And, if the complaint is at root a procedural one, then we need to have highlighted what the procedural failing is in coming to the same conclusion in the two-on-one case, as the one-on-one case. And, at this point, one would be rightly suspicious that the needed material can be provided only by an illegitimate, aggregative thought: namely, that there is one on one side and two on the other, and that this should generate a different decision. And that thought is precisely what the argument seeks not to assume.
Faced with the difficulties of Scanlon’s account, some, following Derek Parfit, are quick to suggest that contractualism should drop the “individualist restriction” and simply embrace aggregation. But for those who hold that justification to individuals is at the core of the appeal of contractualism, that move is profoundly unattractive. Another way of addressing these difficulties, but one that holds on to the individualist restriction, is to adopt a narrower understanding of what we owe to each other: namely, an understanding of it as a set of fundamentally relational requirements, or as involving a second-person perspective. (See Wallace, 2019, and Darwall, 2006.) According to such relational conceptions, morality involves a set of requirements that are connected to claims others have against us, things that we owe them to do.
Within this approach, we can explore thoughts about the significance of the number of people that our actions may affect. What the approach relies on, though, is a pattern of binary relations between individuals: each owing concern to the other. So depicted, the view does not focus on structural considerations that go beyond these duties of one to another. It is unclear how fairness as such, as opposed to the distribution of harms and benefits it gives rise to, fits within it. But, as our discussion above has highlighted, it may well be that just such structural concerns take centre stage in our moral concerns. It is plausible, that is, that there is such a thing as pure procedural fairness. Therefore, the question is whether the constraint of looking solely at the mutual relations among pairs leave one sufficient resources to offer a plausible explanation of the duty to save the greater number: one that holds on to the individualistic restriction, but also avoids sidestepping this structural aspect of what we owe to each other.
Let us take stock. Stepping back from Scanlon’s specific initial proposal in What We Owe to Each Other, we can discern the following structure in the position he recommends. The personal claims that each has constitute the starting point in the web of Narrow Morality. These personal claims are the potential costs (in terms of harms and lost opportunities of benefit) that accepting a given principle may entail. But, in addition, we have seen that there are further concerns that cannot easily be analyzed just in terms of personal reasons, as illustrated by the value of fairness. Fairness clearly is a structural consideration, and one that arises in the context of the practice of justifying principles to each other. In considering the acceptability of a principle, an individual may be moved by the concern that it would be unfair to that individual, or to others, that the act in question be permitted. But nothing we have said so far shows that this is the only such structural concern that could arise in the context of mutual justification. So, we could seek to locate Scanlon’s defence of aggregative intuitions not simply at the level of personal reasons, all of which seem to balance in the relevant cases, but in some suitable structural concern that would explain why it would not be reasonable to permit a policy of saving the lesser number.
And, as we have seen, Scanlon emphasizes the possibility of such structural considerations in his rejection of a Welfarist account of the grounds of reasonable rejection. It should be no surprise, therefore, to find Scanlon further concerned with pattern-related, or structural, features in his most recent discussion of intuitions about saving the greater number and the attraction of aggregative principles. In a recent article, “Contractualism and Justification,” Scanlon withdraws his earlier proposal about how to vindicate saving the greater number, but still seeks to affirm the principle, this time by appealing explicitly to structural concerns.
On the new view, if one can save two people on Left Island or five people on Right Island, but cannot save both groups, we still suppose that one is rationally or morally required to save the people on Right Island, if all other considerations are balanced. However, Scanlon now suggests that restricting the reasonableness of rejection solely to the personal reasons an individual has to accept or reject a principle in pairwise comparison with another individual is mistaken. Impersonal reasons and aggregative reasons may not be an individual’s grounds for rejecting a principle, they may still explain whether the rejection is reasonable or not. The new view allows for the possibility that what makes it unreasonable to reject a principle is sometimes not just the costs that alternative principles involve for an individual, but rather a broader range of factors. In the example we are considering, while a principle to save the lesser number (i.e., save those on Left Island) would be reasonably rejected by those on Right Island, since such a principle would be insensitive to the number of reasons in play on Right Island, it would be unreasonable for those on Left Island to reject a principle to save the greater number (i.e., those on Right Island). We would, that is, take it into account that there are more reasons counting in favour of one course of action than the other. Given the incompatibility of the two courses of action, this has the consequence that it would be impermissible to save the Left Islanders.
The question is settled at the level of reasons by relying on the number of reasons. Whether the number or balance of reasons across individuals is the crucial structural element in the situation is a moot point. If one is unhappy with the original response, that is, it remains to be seen whether the new one convinces any better. But the important feature of the new view to note is that, as with fairness, the reasonableness of rejection is now explicitly settled by relying on considerations other than solely individual costs and benefits.
Now, suppose that Scanlon fails even in this revised version to find a suitable procedural source for the relevant requirement of saving the greater number. And suppose that, with Scanlon, you feel the force of the intuition that saving the greater number is required of us. Is that sufficient to show the failure of contractualism? It is at this point that I think it useful for us to distinguish between contractualism as a general model of moral and political thought, and the constructivist ambitions that Scanlon sketches in What We Owe to Each Other that give rise to his distinctive form of contractualism. It may well be that without some such further justification, Scanlon’s constructivism will be frustrated, given a commitment to the principle of saving the greater number. Recall that Scanlon seeks to provide a noncircular account of what we value in Narrow Morality by providing an overall systematic theory of how we justify one to each other. The kinds of things that actual or potential victims can appeal to on their own behalf are taken as basic. This is in contrast to looking outside of the practice of giving or receiving justifications for something more basic that can reveal the correctness of our claims. In this context, if we think the two in the larger group together have a complaint against the permissibility of saving the other one, then we can find little ground for this in the concerns Scanlon is prepared to take as foundational.
On the other hand, if by contractualism we mean the more general methodological focus on the need to weigh social policies against the concerns and claims of each individual, rather as our first conception of the extent of Narrow Morality suggested, then the verdict is not so clear cut. We are already aware that the kind of claim that Snowdonia has over us, or that the British Museum has, is something that goes beyond what Scanlon conceives of as Narrow Morality. Why then shouldn’t we be as prepared to recognize, in any intuition in favour of the greater number, that we are moved by further concerns as well: that we are, for example, moved by thoughts of lumpen humanity? That is, why shouldn’t the seemingly aggregative intuition that Scanlon’s interlocutors are moved by show that we are responsive to values that do not fit within Scanlon’s constructivist framework? Perhaps the sheer volume of human life is of concern to us, and the proper way to respect that value is to make sure that sufficient human flesh is living.
Putting matters this way no doubt caricatures the concerns of those moved by the idea that we are required to save the greater number. But teasing aside, the formal point remains: for all that has been said about the claims that each has on others, grounded in their personal concerns, nothing about the original picture of contractualism requires that these be all that matters in our moral arbitrage. Since we recognize impersonal values in such things as Snowdonia and the British Museum, perhaps we should think of the impulse towards aggregation here as a symptom of some further value, in addition to and not competing with personal concerns, which naturally aggregate.
The debate around Scanlon’s discussion of the choice between groups has tended be focused on the need for some aggregative principles to do justice to an intuition of the requirement on saving the greater number. In that context, it has often been suggested that the failure of Scanlon’s argument is a decisive case against a contractualist account of morality.
Even on Scanlon’s own terms, such a conclusion would be hasty. Although Scanlon wishes to focus on Narrow Morality in What We Owe to Each Other, the discussion admits that there is a broader range of concerns that we have, and values to which our deliberations answer. One defeats Scanlon’s account of morality only where one also considers all the possible ways in which he can accommodate this intuition in combining morality more widely with morality narrowly conceived.
Our discussion suggests that the state of play is even messier, however. For, strictly speaking, Scanlon seems moved not simply by contractualism per se but rather by his constructivist ambitions in providing a contractualist account of Narrow Morality. It is in the latter context that an appropriate range of personal values is acknowledged when asking how individuals (potential victims, putative violators) can arbitrate the policies that bear on their lives. Constructivism can only recognize the importance of the greater number if this consideration can be derived from this pattern of justification.
Suppose, then, that contractualism is not simply to be identified with Scanlon’s constructivism. A more modest contractualism needn’t be worried at all that there are things of value that impose requirements on us beyond the requirements that other people can directly impose on us. They may happily grant that the natural beauty of Snowdonia imposes a direct constraint on any acceptable social policy, independently of our concern about how others’ interest in that beauty may generate a complaint. By analogy, then, they may interpret the intuition that one is required to save the greater number if they think this a correct demand, as showing that there are additional values to be respected in our social policies: as I put it above, they may recognize the value of lumpen humanity.
Note that to see lumpen humanity as a value, and hence as imposing constraints on our actions or demands on our resources, is not yet to agree to any aggregative principle as such, or even to allow that the concerns of individuals can be overridden by our respect for this important concern. If the claims of individuals balance against each other, and the stringent demands of urgent need of one person cannot override those of any other, as is suggested in the initial set up of the lifeboat, then the value of lumpen humanity will have no greater power to override. Rather, the proposal here is that, with all individual concerns balanced, the requirement to act in one way rather than another arises through the existence of a further consideration bearing on our action, where otherwise there would be moral indifference.
We are now in a position to draw three tentative conclusions.
First, the suggestion that the value of lumpen humanity might play a role within the scope of what we owe to each other is made in a friendly spirit towards those who feel the force of the intuition that it would sometimes be wrong not to save the greater number, particularly in cases of rescue between groups unequal in number and facing the same injury. Like me, some will be less moved by this intuition and seek to depart from consequentialism even in these cases. However, this is not the only context in which to apply the suggestion that we might understand the scope of what we owe to each other in the first, rather the second of the readings suggested at the outset. If we read the scope of what we owe to each other merely as including the benefits, harms, rewards, and so forth that we distribute among ourselves, and allow impersonal reasons, then we have a direct way of arguing that we ought to preserve Snowdonia or the integrity of the British Museum, without having to go through the detour of concerns and interests that individuals rest in the beauty of the one and the cultural worth of the other.
The more interesting question, though, is in what sense this thinner conception of contractualism, one without the constructivist ambition, remains genuinely a form of contractualism. What does constructivism distinctively offer? Its broader ambition is great indeed: to provide an overview of the grounding of a large area of our ethical thought in the context of a need that human beings have to justify and explain their actions to each other. More specifically, it offers a substantial and interesting way of drawing the contrast between personal and impersonal value. For the constructivist, as we have seen, not only welfare need be counted as among personal concerns, but we can also include fairness or justice as a principal concern for individuals, to be contrasted with other things that stand outside of their immediate interests. So, if we give up constructivism, we can no longer explain why fairness is a distinctively contractualist concern, in contrast to natural beauty or social goods.
We might wonder what, if anything, remains of contractualism, if we don’t have a sufficiently rich and distinctive set of contractualist values to ground political debate. I assume that no contractualist can ignore the distinctive claims of fairness or justice alongside individual welfare. And I take it the answer to the scope and content of contractualism remains within political theory rather than within the ambition of ethical thought, since we have given up the ambition of giving an overview of what makes our values, values. Instead, we have the very narrow constraint that the context of formulating and justifying social policies is one in which policies are justified to individuals. The concerns that we draw on in providing these justifications, and in balancing the claims of one against others, will partly draw on what is of value distinctively to those individuals, but plausibly can extend beyond it.
Whether this hallmark is too thin to provide a characterization of contractualism turns on how significant this constraint will turn out to be. The history of political theory suggests that as thin as the constraint is, it is nonetheless one difficult to do justice to. Most recently, many authors have urged, following Derek Parfit, that we should drop what he has called the “individualist restriction,” and perhaps redefine a contractualism free of this constraint. I suggest this is the wrong move entirely. Particularly in the political context, or that of social policies that deal with distribution of benefits and harms or with the imposition of risks inherent in any human activity, insistence that we justify our decisions to individuals leads to very different arguments and sometimes to very different policies. How that insistence on justification to individuals leads to a substantive contractualism is a matter for discussion elsewhere, however.
For a good discussion of the “Impersonalist Restriction” in Scanlon, see Parfit (2011, p. 214ff).
For a particularly clear argument that Scanlon’s treatment of the duty to save the greater number in cases of rescue highlights problems with his contractualism, see Raz (2003).
Frances Kamm seems to adopt a different take on the place of the value of fairness in her review of What We Owe to Each Other, although she is not entirely explicit about it. She writes: “We cannot present as a reason for rejecting a principle that the acts it licenses are wrong, since in Scanlon’s theory wrongness is only a property an act has once any principles licensing it are shown to be reasonably rejectable. However, moral considerations other than wrongness (or rightness), such as unfairness, can be given as a reason for rejection” (Kamm, 2002, p. 338, emphasis added).
See Scanlon, 1998, p. 235: 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 fifteen 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? It seems to me that we should not wait, no matter how many viewers there are, and I believe that contractualism can account for this judgment while still allowing aggregative principles [in cases of rescue of one of two groups of people facing severe harm and unequal in number].
The original challenge to aggregation in cases of rescue of one of two groups unequal in number is in Taurek (1977). See also Anscombe (1967). Scanlon proposes a contractualist account of a duty to save the greater number in his discussion of the structure of contractualism (1998, p. 229-241).
I have had my say in Munoz-Dardé (2005).
I develop this argument in greater detail in Munoz-Dardé (2005).
For an illuminating discussion of resources within the relational conception to envisage cases of rescue such as those discussed by Scanlon, see Wallace (2019, chapter 6). Wallace proposes that there might be a division of labour between individual morality and democratic decision making. When it comes to public policies, he suggests, we all have an ex ante interest in the adoption of policies that would save larger groups in cases of urgent emergencies and rescue. And the thought in favour of this is that on decision-theoretic grounds, one is always more likely, all other things being equal, to be in the larger group in any situation.
The paper on which this article is based was presented at a workshop devoted to metaethics at the Université de Montréal: “La métaéthique et ses implications normatives.” The workshop was organized in honour of Ruwen Ogien, who died in May 2017: with him, the French analytical philosophy community has lost one of its most distinctive and original voices. My own text focuses on one of Ogien’s philosophical concerns—namely, consequentialism. See, in particular, his small joint-authored book with Christine Tappolet (Ogien and Tappolet, 2008). Chapter 3 in that book, “Peut-on réduire les valeurs aux normes?,” and, in particular, p. 80ff contain an enlightening overview and discussion of Scanlon’s anticonsequentialism, and the role of values in his contractualism.
Let me add two words about my personal relation to Ruwen Ogien. Since we first met at a workshop at Cerisy-la-Salle in 1998, I have taken every opportunity to hear or read Ruwen: he was always illuminating, and he made philosophy fun. We may have disagreed about the appeal of consequentialism, but our views coincided about much else. I often found myself inspired by his positions, particularly in what I call “Regulation of Intimacy”: his anti-authoritarianism and antipaternalism shone through in his Penser la pornographie, (Paris, PUF, 2003) and in La panique morale (Paris, Grasset, 2004). They are also at the centre of his own desperately humorous and perlucid analysis of the practices and language surrounding disease written during his own fatal illness: Mes Mille et Une Nuits : la maladie comme drame et comme comédie, Albin Michel, 2017.
For their comments on earlier drafts of this article, I am grateful to audiences at the Université de Montréal, the workshop on Aggregation and Risk at the LSE, the workshop in Practical Philosophy at the University of York, and my graduate seminars at the University of California, Berkeley, and UCL and, in particular, to Nick French, Johann Frick, Joe Hoton, Mikayla Kelley, Dan Khokhar, Niko Kolodny, Mike Martin, Rowan Mellor, Christian Piller, Wlodek Rabinowicz, Joseph Raz, Christine Tappolet, and Nandi Theunissen.
- Anscombe, ‘Who Is Wronged?’ The Oxford Review, no. 5, Trinity, 1967, p. 16-17.
- Darwall, Stephen, The Second Person Standpoint: Morality, Respect, and Accountability, Cambridge Mass., Harvard University Press, 2006.
- Kamm, Frances, Morality, Mortality, 2 vols, Oxford, Oxford University Press, 1993.
- Kamm, Frances, ‘Owing, Justifying, and Rejecting’, Mind, vol. 442, 2002, p. 323-354.
- Munoz-Dardé, Véronique, ‘The Distribution of Numbers and the Comprehensiveness of Reasons’, Proceedings of the Aristotelian Society, vol. 105, no. 2, 2005, p. 207-233.
- Ogien, Ruwen and Christine Tappolet, Les concepts de l’éthique. Faut-il être conséquentialiste?, Paris, Herman, 2008.
- Parfit, Derek, On What Matters, vol. 2, Oxford, Oxford University Press, 2011.
- Raz, Joseph, ‘Numbers with and without Contractualism’, Ratio, vol. 16, 2003, p. 346–67.
- Scanlon, Thomas M., What We Owe to Each Other, Cambridge, Mass., Harvard University Press, 1998.
- Scanlon, Thomas M., ‘Contractualism and Justification’, forthcoming in a volume of papers on Themes from Scanlon from the September 2016 Lauener Conference, to be edited by M. Frauchiger and M. Stepanians.
- Taurek, John M., ‘Should the Numbers Count?’, Philosophy and Public Affairs, Vol. 6, No. 4 (Summer, 1977), p. 293-316.
- Thomson, Judith, The Realm of Rights, Cambridge Mass., Harvard University Press, 1990.
- Wallace, R. Jay, The Moral Nexus, Princeton, Princeton University Press, 2019.