Probably the most widely discussed divergence among forms of contractualism concerns the view of reason embodied in the core idea of rational acceptability. On the one hand, there are contractualisms of the Hobbesian type, in which the moral contract or agreement is seen as a question of the balance of interests, and within which reason or rational acceptability is to be understood in the instrumental, decision-theoretic sense. David Gauthier and, in a different way, John Rawls both defend such a view.Footnote 1 On the other hand are contractualists like Thomas Scanlon and Jürgen Habermas, who see the relevant agreements as a question of what is right, a consensus on the validity of a specifically normative claim and who, correlatively, resist the reduction of reason to its instrumental sense alone. Although I will have something to say about the nature of reason embodied in moral agreements, something in favour of the view shared by Scanlon and Habermas, I propose to carve up the domain of contractualist moral theory along different joints, in order to bring out the distinction between those who, like Scanlon—and the Rawls of A Theory of Justice Footnote 2—see moral agreement as the hypothetical projection of a solitary reasoner, and those who, like Habermas, insist that actual agreements must be the point of reference for our account of right. It is my aim here to present and defend the latter sort of view, which I will call ‘actual agreement contractualism.’ This is the view that norms arise from actual attempts by all those affected to arrive at legitimate terms for social cooperation. There are, of course, a variety of other moral theories which take actual interactions or discourses as basic but which are not contractualist in nature. (For instance, Virginia Held’s version of care ethics is one such view.) But I will consider here only the particular combination of the appeal to actual agreement with a contractualist account of moral justification. This combination is not unique to Habermas, but is also found in the work of other discourse ethicists, like Karl-Otto Apel and Seyla Benhabib, who disagree with Habermas on the details of his proposal. These family disputes will also be put to one side here, as a matter to be considered only if—as I will argue—the actual agreement approach to contractualism can in general be rendered plausible. In particular, this will mean rebutting the charge that the actual agreement approach must accept a conventionalist account of norms.
Although the issue comes up peripherally in his debate with Rawls, Habermas has not had much to say in defence of the role of actual agreements in his version of discourse ethics; nor, for that matter, does Rawls expend much effort in justifying his rejection of an actual agreement approach. Not so, however, with Scanlon, who makes for an ideal interlocutor here insofar as he not only shares with Habermas a view about the subject matter of justice, but because he has been much more explicit in his reasons for rejecting actual agreements. This, despite the fact that, as I will argue, several aspects of his own position are quite difficult to accept if ‘what we owe to each other’ must be understood as something which we each discover monologically rather than dialogically. After detailing the grounds internal to Scanlon’s approach that would favour actual agreement contractualism, I will turn to his objections to the idea of actual agreement; I will then offer a reply to those objections, before concluding with several reasons for seeing actual agreement contractualism as an attractive proposal, which concern the status of moral theory and the nature of obligation.
1. Why Scanlon Ought to be an Actual Agreement Contractualist
Scanlon defines his contractualism, which concerns the limited but central part of morality dealing with ‘what we owe to each other,’ as the view that moral judgments are “about what would be permitted by principles that could not reasonably be rejected, by people who were moved to find principles for the general regulation of behaviour that others, similarly motivated, could not reasonably reject.”Footnote 3 It is by appeal to the significance of this particular motivation that Scanlon chiefly distinguishes his position from competing forms of contractualism. But it is a point he shares with Habermas, who describes this motivation as an orientation towards understanding (that is, comprehension and agreement) or, specifically in the case of moral discourse, an orientation toward what is right. In a note, Scanlon observes both this similarity as well as the divergence on which I propose to focus:
Insofar as my view, like [Habermas’], interprets moral judgments as involving claims about the justifications that some agents could ask others to accept, rather than claims about what a single agent would have reason to want under specified conditions, it avoids in at least one way the charge of being ‘monological,’ which Habermas levels against some similar views. But he also says that ‘the justification of norms requires that real discourse be carried out, and thus cannot occur in a strictly monological form, i.e., in the form of a hypothetical process of argumentation occurring in the individual mind.’ In my view, while interaction with others plays a crucial role in arriving at well-founded moral opinions … reaching a conclusion about right and wrong requires making a judgment about what others could or could not reasonably reject. This is a judgment that each of us must make for him- or herself. The agreement of others, reached through actual discourse, is not required, and when it occurs does not settle the matter. In this respect my account may remain, in his terms, monological.Footnote 4
So, that is the point of contention. Yet there are at least three significant reasons internal to Scanlon’s argument that suggest the need, despite his stated position, for actual discourses.
1.1. The Dynamism of Contractualist Justification
Central to the idea of contractualism is the rejection of any theory privileging some particular good as the currency of moral deliberation, including even such complex candidate goods as Amartya Sen’s “capabilities.”Footnote 5 As long as moral reasoning is defined by the search for principles for the distribution of some such substantive good, it is at least easier (if still not exactly easy) to see how monological decisions are possible provided the availability of adequate information. But if there is no such good by which all participants are or ought to be oriented, it seems prima facie impossible for a solitary reasoner to determine what others could not reasonably reject: we would need to know what sorts of things others want and value in order to answer such a question; and this is presumably something others will have to tell us themselves.
Scanlon does not think so; but it is unclear that his reasoning on this point is actually consistent with a monological decision-procedure. Part of his solution to the problem of the plurality of goods and values is found in this appeal to the idea of ‘generic reasons,’ which I will discuss below. But consider, first, the following: Scanlon concedes that, absent some privileged good, there “is no fixed list of ‘morally relevant considerations’ or of reasons that are ‘morally excluded’” so that the “aim of justifiability to others moves us to work toward a system of justification that meets its demands, and this leads to a continuing process of revising and refining our conception of the reasons that are relevant and those that are morally excluded in certain contexts.”Footnote 6 In other words, Scanlon proposes to deal with our inability to anticipate what will be relevant to the moral deliberations of others by accepting—quite properly, I think—that we must conceive of the project of justification dynamically, that our account of morally relevant reasons is, at any particular moment, fallible. Yet his manner of describing this dynamism renders the impetus for change mysterious, or solipsistically displaces it onto the ‘system of justification’ itself. As William James puts it, it is the “cries of the wounded” that inform us of our errors and presumptionsFootnote 7 (for instance, that men can reliably anticipate what women could not reasonably reject). For the contractualist, it is specifically our inability to satisfy their demands for justification that spurs us to revise our conception of the morally relevant reasons. The upshot is that the process through which we come to recognize our system of justification as needing more work is left utterly mysterious, unless we see it as rooted in actual interaction, actual attempts at agreement. Thus, Scanlon’s own position on the essentially open, revisable nature of the concept of moral relevance seems to point to actual discourses as the source of critical revision.
1.2. Why We Care About the Reasons of Others
Scanlon’s monological account of moral reasoning leads him to endorse an unattractive view of why we care about the reasons of others, one which is like Kant’s in routing the justification for my duties to others through the question of my own self-consistency.Footnote 8 For Scanlon as for Kant, the fundamental reason we must be concerned with the reasons of others is the fact that reasons (‘maxims,’ in Kant’s formulation) simply are general rather than personal. As a result, Scanlon declares, we have “wholly self-regarding reasons for having views about the correctness or incorrectness of the judgments people make about the reasons they have, since these judgments imply conclusions about the reasons we have.”Footnote 9 Basing our fundamental moral concern on self-regarding reasons strikes me as prima facie unattractive; but I will try to justify that judgment later, in conclusion. For now, consider the additional reasons Scanlon compiles to explain our concern for the reasons of others: (a) because these others may be better positioned, epistemically, so that I have reason to value their reasons as a kind of valid expertise; (b) because the reasons held by others may “represent an emerging consensus” that may affect me as, for instance, when there is a new development in my professional field; (c) because shared reasons/beliefs are constitutive of some kinds of group life so that, if members of my group begin to diverge in terms of the reasons they recognize, this concerns me as a question of social integration; and, finally, (d) because the reasons recognized by others may affect my moral standing among them, whether I count for them as a being intrinsically deserving of respect, or only as a fellow in-group member, or perhaps not at all.Footnote 10
These are all persuasive considerations, which cast valuable light on the fundamentally social nature of justification. Yet none of them seem to me to capture why it is that other people’s reasons necessarily matter to us. If, as actual agreement contractualism insists, moral discourses arise among those affected by a particular proposed norm, then there is a much simpler and more intuitive reason why I must care about the reasons of others: because those reasons (even if I do not share them) must figure as premises in my argument or attempted justification. Other people’s reasons matter to me because it is to those other people that I am attempting to justify my position, by deriving it from reasons or premises they accept. If being a person has an important connection with the capacity to act on the basis of reasons (a view which I share with Scanlon), then according to actual agreement contractualism, I must care about the reasons of others precisely because they are persons, that is, because the reasons they accept guide their conduct in social contexts (more on this in Section 3, below). The same point emerges in the inverse from Scanlon’s third and fourth reasons above, dealing with (c) social integration and (d) moral standing. These points anticipate his own argument for the special importance to us of the failure to recognize moral reasons: namely, because “the reasons that a person recognizes … affect the range of relations we can have with that person” and, in the case of a failure to appreciate moral reasons specifically, this entails “a failure to see why the justifiability of his or her actions to us should be of any importance.”Footnote 11 Yet, contra Scanlon, this last failure matters to us not in abstracto, but as people potentially affected by the behaviour of the person in question. What we want, should we think ourselves confronted with such a person, is for them to justify to us the norms that guide their conduct, not the odd assurance that they will consider, on their own, the justifiability of their actions to us, speaking for us and for everyone else as they go. The latter is not just a formula for moral paternalism: as I have already argued, it goes to the heart of what we are willing to take as evidence of a good reason, evidence that revision to our reasons is not needed. If someone whose conduct affects me claims to be acting justifiably, then my rejection of that claim counts pro tanto as evidence to the contrary; it must then be possible for the proponent to offer some account of my error, or of my non-communicative orientation, if it is to be believed that my rejection is unreasonable. This speaks again to how we make sense of the idea, endorsed by Scanlon, that moral justification is a fallible process.
Thus, we care about whether others care about what we owe to each other because we are others to them, and we want to be respected as persons. The additional reasons offered by Scanlon in support of this concern all differ from his ‘fundamental’ self-regarding reason in that none of them are plausible outside of actual interactions with others: if not through such interactions, how could I come to possess the knowledge reflected in reasons (a) through (d)?
1.3. Justification as a Form of Mutual Recognition
In What We Owe to Each Other, Scanlon claims that the relation in which we stand to others who appreciate moral reasons is one of “mutual recognition,” and argues that “[s]tanding in this relation to others is appealing in itself—worth seeking for its own sake.”Footnote 12 And, indeed, he adds that charges of injustice made against us are so moving because of their implications “for our relations with others, our sense of justifiability to or estrangement from them.”Footnote 13 It is difficult to square these claims with the monological account of moral reasoning. I assume that Scanlon does not intend to say that, in recognizing moral reasons, we stand in hypothetical relations of mutual recognition. But there is something weirdly Leibniz-ean about the view that, if we each separately commit ourselves to acting only on the basis of reasons which others hypothetically could not reasonably reject, then we will all stand in actual relations of mutual recognition to one another. There are good theoretical and empirical grounds for asserting that participation in actual discourses, in which one’s interlocutors charitably consider and respond to one’s arguments, offers a kind of substantive recognition for one’s personhood (and this, whether or not one’s own position carries the day).Footnote 14 But it is quite unclear how a hypothetical discourse could do this, how the self-regarding concern for justifiability on the part of others could offer us the experience of mutual recognition. It is similarly difficult to see how our unity with others, our non-estrangement from them, could be the product of our own private, hypothetically-grounded good conscience. In fact, much of Scanlon’s discussion of the repertory of moral life—including guilt, apologies, and forgiveness—only makes sense when seen as part of the effort to repair actual and damaged social relations.
2. Scanlon’s Objections to Actual Agreements
There are, then, a variety of reasons internal to Scanlon’s account which suggest that he, too, should—and, even at times, despite himself, does—think of contractualism in terms of actual processes of justification to others. There would be reason to resist these internal tendencies only if there were weighty considerations speaking against a focus on actual agreements. So, it is to Scanlon’s objections that I now turn.
First, though, I should note that Scanlon admits—in another note on Habermas—that actual processes of argumentation are often the best way to prevent the distortion of our real interests by others; but he insists that moral judgments nevertheless remain a question of what others could or could not reasonably reject, and that while knowing “what they do or do not actually reject may be useful in arriving at such judgments … [such knowledge] is not sufficient for them.”Footnote 15 Scanlon also concedes that there are some particular cases of morally-relevant conduct in which actual agreement is important: for instance, when we want consensus about or permission for some act that will affect others and where it would be wrong (judged so monologically, it would seem) not to seek such advance agreement. But he again insists that this represents a special subclass within morality and not a definition of moral right itself; and, he adds, “the significance of this kind of agreement should be clearly distinguished from the ideal of hypothetical agreement which contractualism takes to be the basis of our thinking about right and wrong.”Footnote 16 Why, then, these caveats notwithstanding, does Scanlon reject the actual agreement approach to contractualism? As I read him, there are two main reasons: (1) because he believes that basing our account of normative rightness on the outcome of actual agreements risks conventionalism and/or an objectionable form of relativism; and (2) because he regards the scope of proper moral deliberation as incongruous with the scope of conceivable, actual discourses.
2.1. Actual Agreement Conventionalism?
While it “would be pleasant,” says Scanlon, to live in actual agreement with others, and while “it is unpleasant to be in conflict with those around us,” morality will at least in some cases require that we reject “the prevailing consensus about what is and is not justified”; therefore, actual agreements cannot be the standard of right.Footnote 17 For instance, some individuals might, through deformed processes of socialization, have internalized a sense of the inferior value or worthiness of their own needs and so be “quite happy with, and even grateful for, much less than is their due.”Footnote 18 Indeed, a focus on actual agreements opens to the door to “objectionable forms of relativism” according to which entire societies might be unjustly treated on the basis of the observation that “they do not value privacy, or individual liberty, or even life, in the way that we do.”Footnote 19 Although Scanlon rightly notes that such absurd generalizations are descriptively problematic, he nevertheless takes seriously the possibility that groups of individuals could become accustomed, in some way, to injustice and violence and so be willing to agree to norms which failed to eliminate these wrongs. In Scanlon’s view, hypothetical contractualism protects such people from what they might actually agree to or, rather, what they might fail to reasonably reject.
Finally, Scanlon argues, there are cases in which, given the character of the individuals involved, the reasons most likely to generate actual agreement could nevertheless be reasonably rejected. For instance, he imagines an effort by a group of neighbours to arrive at an agreement over common water usages with a powerful and irascible landowner. It would be irrational for the neighbours to advance terms that correspond to their genuine claims if doing so would foreseeably result in arriving at no deal at all. Questions of character and power are an inescapable part of the vicissitudes of actual agreements, according to Scanlon, and another reason to sharply distinguish pursuit of them from the question of what could not reasonably be rejected. The latter represent “not a judgment about what would be most likely to advance their interests or to produce agreement in the actual circumstances or in any more idealized situation, but rather a judgment about the suitability of certain principles to serve as the basis of mutual recognition and accommodation.”Footnote 20
2.2. The Problem of Scope and Inclusion
Contractualist morality, according to Scanlon, requires that we be motivated by “the thought of our [hypothetical] relation to a large number of people, most of whom we will never have any contact with at all.”Footnote 21 Indeed, the demand to consider the reasons of every human being, present and future, is so great that it threatens us with ‘moral overload,’ a problem resolved by Scanlon through the reassurance that we need only consider “representative cases” while entertaining only those objections “that a person could raise while recognizing the force of similar objections by others.”Footnote 22 The salient point here is that the scope of those included within contractualist moral reasoning is not, Scanlon says, limited to those affected by our actions (even if, naturally, they are the first to come to mind when we ask after the justification of our actionsFootnote 23). Often, in fact, we simply have no way to know who will be affected, and in what way, should some candidate norm or principle count as justified. So, we have no choice but to focus our consideration on how representative persons might be affected, given “commonly available information about what people have reason to want,” which Scanlon terms “generic reasons.”Footnote 24
3. A Response: The Dual Character of Actual Agreements
The central error in Scanlon’s treatment of the idea of actual agreement is the misleading way he links the distinction between actual and hypothetical agreement to two additional distinctions: the descriptive versus the normative and the strategic (rational) versus the communicative (reasonable). Connecting up these pairs without substantive argument, Scanlon presents actual agreement contractualism as confined to a descriptive approach to actual processes of argumentation in which individuals pursue the maximization of their own interests; by contrast, in his view, only a contractualism directed toward what reasonable, communicatively-oriented people could agree to has the critical or normative resources to condemn the distortions of ‘actual agreement conventionalism.’
3.1. Description and Normativity in Actual Agreements
Let us consider the descriptive-normative issue first. In addition to the instances already cited above, we see the false contrast on Scanlon’s part in his rejection of the attempt to root the obligatory character of promises in a “social institution” of promising; “when I say, ‘I promise to help you if you help me,’” says Scanlon, “the reason that I suggest to you that I will have for helping is my awareness of the fact that not to return your help would, under the circumstances, be wrong: not just forbidden by some social practice but morally wrong.”Footnote 25 We see it again in his distinction, drawn early on in What We Owe to Each Other, between “operative reasons”—which call for a descriptive explanation regarding why X believes that p—and reasons in the “standard normative sense”—which ask, evaluatively, whether there are indeed good reasons for X to believe that p.Footnote 26 The falsity of the distinction is a product of the erroneous premise that we can understand in a purely descriptive attitude what people take as their operative reasons for normative conduct, what people regard as the reasons underlying their social practices and accepted norms.
Habermas’ rejection of this idea originates in his criticisms of positivist social science: a norm, he observes, is a claim to rightness, to legitimacy, and not simply the positing of a factual consensus.Footnote 27 The ‘operative reason’ of someone acting on the basis of a norm, or seeking to defend that norm in actual discourse, is its claim to rightness, an account of which is—contra Scanlon—not identical with an explanation of how they came to form their belief in the validity of that norm (a consideration which is prima facie irrelevant in discourse). For example, imagine that cultural respect was a norm in Canada. First of all, that does not mean that, descriptively, all or even most people abide by it as a matter of fact.Footnote 28 But, more importantly, the meaning of the norm is not: ‘Canadians have agreed that no one should be marginalized on the grounds that their cultural practices differ from those of the majority’ (the positing of a factual consensus). Rather, the content of the norm is: ‘persons are deserving of cultural respect, irrespective of their majority/minority status.’ The claim to validity is, in other words, not identifiable with the context in which it is raised nor with the particular form in which it is reflected in a de facto consensus. This is precisely what makes it possible for the social critic to appeal to the same norm in objecting to its distorted institutionalization, distortions to which even those who genuinely endorse the norm may be blind.
Contractualists of all stripes agree that norms claim to regulate conduct according to principles which reflect the generalizable interests of those affected. It is obvious that de facto agreements do not always do this. But what we are faced with in such cases is a pseudo-consensus regarding some norm. If U.S. slavery would have been supported by plebiscite in the South in the 1850s, we do not understand that social fact until we see it as the product of spurious race science, propagandistic distortion, socially-induced fear, and so on. These are not evaluations imposed upon the social fact from the outside, as the positivist would maintain: that it is a pseudo-consensus is the social fact. Consider a case analogous to the claim to moral legitimacy: political legitimacy is, in most developed states, tied to a claim of democratic legitimacy. Because of this, we recognize that a stolen election is not simply an election which we happen to find wanting by a set of external criteria: it is a fraudulent- or pseudo-election.Footnote 29 The specific claim raised immanently by every purported consensus, and by each proposal within a discourse that a given norm may serve as the basis of a consensus, is summarized by Habermas under the heading of principle (U), which states:
For a norm to be valid, the consequences and side effects that its general observance can be expected to have for the satisfaction of the particular interests of each person affected must be such that all affected can accept them freely.Footnote 30
It is important to note that everyone must be able to accept the impact on each person: in other words, there is no presumption that each person is an infallible interpreter of his or her own needs and interests. The actual agreements that trouble Scanlon and motivate his rejection of actual agreement contractualism fail to satisfy this principle. But that also means, if Habermas is right, that they fail to satisfy the very principle the agreements themselves implicitly assert, which is why a de facto consensus that emerges in such conditions is a pseudo-consensus. Actual agreements, in raising the claim to normative validity (and so implicitly asserting [U]), also provide the immanent or internal point of reference for criticism. So, when Scanlon worries that some individuals may have internalized a demeaning view of their own needs or become reconciled to violence and injustice, he is saying that even though these individuals might agree to the manner in which they are affected by the relevant norm, he cannot so agree. That’s good enough to upend the claim to actual agreement.Footnote 31
The means by which we vindicate a challenge to someone else’s interpretation of his or her own needs raises difficult questions for both Scanlon and for the actual agreement approach, although it is not clear that Scanlon is sufficiently sensitive to these difficulties. The nature of his objection to actual agreements, when these are (mis)understood in a conventionalist sense, demonstrates his recognition that we do indeed sometimes find it necessary to challenge others’ interpretations of their own needs. If we adopt a hypothetical approach like Scanlon’s, this might even occur in a great number of cases. A proper sensitivity to the potential for paternalism involved in these challenges recommends that, as with psychoanalytical insight, we regard the ultimate or full vindication of charges of self-deception or self-misinterpretation as dependent upon the acceptance of or appropriation by the one challenged. If I believe that you subordinate your own needs to an unreasonable desire for approval, then that diagnosis counts as redeemed only if you ultimately come to see it as valid, internalize it as offering you a clarifying insight into your own psychology. By contrast, a principle that allowed others to prevail unilaterally, and in the privacy of their own deliberation, over the interpretation of our needs and their reasonableness—as the hypothetical approach seems to do—is ‘reasonably rejectable’ by those potentially affected. Scanlon’s position indeed appears to allow this: that I can have made a morally unimpeachable decision to act in a way that affects you, in ways to which you, in fact, object, because I have determined—without any necessary effort to convince you—that you are deceived in your interpretation of your own needs and interests. This is also why the actual agreement approach, even where it involves a critical rejection of the actual agreement of some of those affected, does not finally collapse back into the hypothetical: that is, because the actual agreement contractualist does not accept, as the hypothetical contractualist does, that purely monological rejections on behalf of others can figure in or count as cases of normal or completed moral justification.
Scanlon’s complaints about “objectionable forms of relativism” seem to be based on something like the following modus tollens:Footnote 32
(i) If actual agreement contractualism was correct, then we would have to concede that there would be nothing wrong with, e.g., human slavery in a society in which all accepted this institution;
(ii) But this consequence is unacceptable;
(iii) Therefore, this sort of contractualism must be mistaken.
But, as I’ve been arguing, the major premise fails because the reasons given in pursuit of agreement are claims to normative validity. There is no possibility of approaching them purely descriptively: rather, understanding a given apparent consensus requires that we determine whether it is indeed reasonable and justified or the product of various forms of unreasonable exclusion, distortion, and so on. Since human slavery cannot be reasonably justified—it was not the product of genuine agreement nor would it be today—the first premise is a non-starter: the actual agreement contractualist condemns the imagined consensus as a false one. The history of the struggle against racism and sexism is quite incomprehensible without conceding that the normative concepts used to justify the status quo—‘All people are children of God,’ ‘All men are created equal,’ and so on—are understood as making claims that transcend their local interpretation and empirical institutionalization (which, once again, points us to the real source of the dynamism of justification).
The above argument also positions the reasoner in a contradictory manner: we are to consider the proposed scenario from a God’s eye view, so that our own other-regarding rejection of slavery is not taken to challenge the premise that ‘all accept this institution.’ It will not do simply to insist that we are not members of the society envisaged, since that would wrongly presuppose that moral validity is confined to specific societies. In fact, it is not possible to construct a valid scenario in which our moral intuitions are violated but in which all those affected see no violation—because, ex hypothesi, we see the violation. This is not to deny that there are such things as moral blindspots, but that we can have no idea what they are. There are certainly no grounds for thinking the hypothetical agreement contractualist has less reason to be concerned that moral blindspots might distort his or her reasoning; indeed, Scanlon has already admitted that it often takes actual discourses to reveal them.
The view of norms defended here, as immanently raised claims to validity, is not peculiar to Habermas: in addition to other discourse ethicists already mentioned, equivalent views are found outside contractualist moral theory in the work of both Kurt Baier and Philip Kitcher.Footnote 33 Baier describes the “moral order” as a reflective social order dedicated to asking whether the moral directives contained in its mores actually deserve to be recognized in the way they claim.Footnote 34 The role of the moral critic and reformer is to engage in internal criticism of the accepted guidelines and suggest alternative arrangements that would allow us to better fulfill the moral enterprise. Kitcher ties the ‘ethical project’ to remedying altruism failures and enabling social cooperation with a minimum of conflict and violence; he too sees the moral critic as engaging in immanent criticism—the reference point for which is the function of morality—of the manner in which the project is currently institutionalized.Footnote 35 In short, both Baier and Kitcher see norms as making a context-transcending claim to validity which is tied to the function of morality in such a way that understanding this claim simultaneously means evaluating its truth or adequacy.
So much for Scanlon’s overdrawing of the descriptive-normative contrast. It is easier to deal with the issue of his linking actual agreements with what is rational while tying hypothetical agreements to what is reasonable. It is easier because there is no argument with which to contend here: perhaps it is thinking of a position like Gauthier’s that leads Scanlon to treat these distinctions as paired; but he offers us no reason to think they must or ought to be so. And, indeed, we would have good reason to be sceptical of Scanlon’s own position if it wasn’t clear that he believes an orientation to what is reasonable, to justifiability, is not after all so very rare in actual practice.Footnote 36 It is worth observing, however, how normative issues arise here as well and in ways which anticipate Scanlon’s second objection regarding moral scope. For instance, Scanlon rejects the idea that the aim of contractualist morality involves “securing the benefits of [actual] cooperation” and contrasts this to the aim of hypothetical justifiability. The latter, he believes, includes within its scope persons who, because they can neither harm nor benefit us, will be excluded from consideration under a scheme of cooperative benefit.Footnote 37 Similarly, Scanlon seems to think that actual agreement contractualism will face ‘strongman’ problems of the Kantian type, in which persons rationally reject a principle or norm requiring mutual aid because they correctly reason that they are unlikely to need it themselves.Footnote 38 But the point of actual agreement contractualism is that moral norms arise from the attempt to secure legitimate (that is, normatively valid) terms for social cooperation, so there is no reason to accept Scanlon’s reduction of this aim to a form of strategic bargaining or negotiation. We do not need to retreat to hypothetical, monological reasoning to explain why a fallacy of irrelevance is committed by anyone who, in an actual discourse regarding legitimate terms, argued thus: ‘I disagree that one ought to help others because I don’t believe that I will likely need help.’ Scanlon himself has rightly insisted on the public character of genuine reasons which excludes such an argument.
3.2. The Problem of Scope and Actual Agreements
This issue of scope is the basis of Scanlon’s second objection to actual agreements: that is, the lack of congruence between the scope of those to whom my actions should be justifiable and those with whom I could conceivably engage in actual processes of justification. There is undoubtedly something to this: if justification appeals to all those affected by some candidate norm, then we are faced with hard problems in identifying who is affected and in imagining how an actual process of justification could be carried on among such a community. Actual agreement contractualists must accept that full participation is a regulative ideal. With respect to issues of intergenerational justice and those otherwise incapable of participation, the actual agreement contractualist has recourse to the same mechanisms of ‘trusteeship’ as does Scanlon’s hypothetical version, though with the added benefit, which is unavailable to Scanlon, of being able to insist that, at least in many cases, legitimate exercise of ‘trusteeship’ is subject to empirical qualifications.Footnote 39
Still, Scanlon has a point here, which actual agreement contractualism must consider. At the same time, it is not clear that the criticism really targets something peculiar to actual agreement contractualism, so that it constitutes a reason to prefer other approaches which are not similarly vulnerable. Consider that, despite its prominence in discussions of global justice, consequentialism certainly faces serious and parallel problems of calculability and prediction in a globally interconnected society. Recall Scanlon’s concession that the aim of justifiability, simpliciter, is so overwhelming that we must opt instead for the consideration of ‘representative cases.’ But it is quite unclear to what degree this simplifies our task. Indeed, it is precisely the same set of cases that cause trouble for both approaches: where the scope of those affected remains uncertain, the identification of relevant representatives must be equally problematic. If we are uncertain who is likely to be affected, then we can have no good grounds for assigning representatives, that is, unless we dramatically homogenize the interests and values of all persons so that anyone is more or less representative of everyone. But with that, we will have undone Scanlon’s claim to recognize the basic plurality of goods and values which is central to the distinctiveness of contractualism vis-à-vis distributive theories of justice.
Moreover, a monological determination of ‘representativeness’ seems vulnerable to the worries about paternalism already discussed. There is a real and vexing problem here which concerns the design of moral institutions and which merits careful attention. But the obstacles faced by actual agreement contractualism in this area are not substantially different than those faced by political democracy: that full participation of a citizenry in debates of national importance should prove unattainable in the event does not justify the resolve on the part of a philosopher-king to answer these questions monologically. Indeed, the degree of the legitimacy of the decisions reached is in large measure (though not exclusively) a function of the degree of participation: there is no shortcutting of the actual process while reaching the same result (namely, legitimate decisions).
Finally, there seems to be a sense on the part of Scanlon and others that actual agreements on the topic of right are somehow more difficult to imagine than actual bargains reflecting a fair balance of interests, which perhaps explains the tendency to identify the pursuit of actual agreement with what is rational rather than what is reasonable. But this judgment is an extraordinary one. Class societies—which is to say, all societies—are founded on the institutionalization of conflicting, antagonistic interests under the cover of purportedly shared normative principles (for instance, equality, freedom, etc.). To think agreement on the imperatives of the latter more difficult to attain than a fair or just balance of the former seems to me entirely utopian and to fly in the face of the evidence.Footnote 40 The real challenges for actual agreements are problems of enforcement: namely, institutionalized or even socially-induced akrasia, and dealing with obfuscating and bullshitting (pseudo-communicatively-oriented) participants.Footnote 41
4. Conclusion: Why You Should Be an Actual Agreement Contractualist Too
So much for Scanlon’s worries: actual agreement contractualism is not an invitation to conventionalism or relativism; and the problems it faces regarding inclusion and scope, while real, are not substantially different from or more serious than those faced by any other normative theory (political or moral) according to which everyone matters. Moreover, the dynamic element of contractualist justification as Scanlon portrays it, the fact and nature of our concern for the reasons of others, and the role of justification in establishing relations of mutual recognition, all speak in favour of the actual agreement approach. But there are two additional considerations which, although rather general, I would like to suggest recommend the actual agreement version of contractualism: first, its implications for our account of the function of morality and of the moral theorist; and second, its consequences for our account of obligation.
4.1. What is Morality and What are Moral Theorists For?
We recognize, I think, that the term ‘morality,’ when used without modifiers (e.g., ‘Nazi morality’), does not require scare quotes. That is to say, we recognize that morality is not a theoretical construct, something which comes into view only from the perspective of the social-scientific observer; nor does it exist simply in the minds of the philosophically informed as a device for making decisions of a peculiar sort. Since Plato, philosophical reflection on morality has presented itself as a replacement for or corrective to existing social custom and, therefore, as the basis of a possible, alternative social practice, though one which is a continuation, development, or refinement of that existing practice.Footnote 42 In other words, philosophically-informed morality is supposed to do (some part of) what social custom aimed to do, but better. Moral theory is, as Baier recognizes, what happens when we become reflective about the claims to rightness contained in our mores or customsFootnote 43: for ‘western culture’ (which does deserve scare quotes), at least as far as we know, this happened with Plato (or, if you prefer, with Socrates). In the face of growing discontent with moral theory as a form of autonomous puzzle-solving, or as a series of exercises in intuition-pumping, actual agreement contractualism offers an attractive answer to the question of the purpose of moral theorizing: to help articulate an effective yet reflective equivalent for social custom, one which, by operating through internal critique, deranges actual moral practice as little as possible. To repeat, since actual moral practice concerns the attempt to reach legitimate (morally valid) terms for social cooperation, there is a role for the moral theorist to play as internal critic of the process of agreement formation, as the identifier of biases, distortions, exclusions, and illegitimate power that allow us to call out a given agreement as a pseudo-consensus.Footnote 44 This is, in fact, what the major and clear examples of moral progress in history have looked like. It also relies on the sorts of argumentation skills that we (‘we’?) regularly claim justify the place of philosophy in modern universities. Obviously, the fact that actual agreement contractualism offers an attractive purpose for moral theorists does not itself constitute proof of the validity of that theory; but I hope the foregoing has already rendered the actual agreement approach at least plausible.
4.2. Obligation and Intersubectivity
I mentioned earlier that grounding a concern for the reasons of others in self-regarding reasons strikes me as unattractive: I return to that now. Baier complains that, quite apart from the normative problems plaguing divine command ethics, the latter misrepresents the nature of moral wrongness by interpreting proscribed actions as sins against God. That is, according to divine command theory, sins are not wrong because they wrong another person; and one’s salvation is not dependent upon anyone else’s salvation.Footnote 45 God is the middle term in all ethical judgment, according to such a view. There is a kind of echo of this misrepresentation in theories which, like Scanlon and Kant’s, attempt to offer us exclusively or fundamentally self-regarding reasons for caring about the reasons of others. Our obligations to the other people affected by our actions and decisions are detoured through our obligation to, or self-regarding reasons to care about, ‘justifiability’ or the consistency of our maxims. But, as Baier adds, “morality is not a private matter. It is not solely our own business whether we do the right thing or not.”Footnote 46 If Baier’s complaint is a good one, then it is an additional virtue of actual agreement contractualism that it grounds our obligations to engage in justification in our actual relations with actual other persons: it is to those affected by our actions, primarily, that we owe reasons, and we owe them because the others demand and deserve them.
