Download Exploring Moral Theories: Scanlon's Contractualist Approach vs. Traditional Methods and more Lecture notes Ethics in PDF only on Docsity!
Nonlegislative Justification: Against Legalist Moral Theory^1 Liam Murphy If moral theorists who otherwise disagree all approach moral theorizing as a search for a set of desirable moral principles for the general regulation of behavior, then there is a sense in which they are all, as Parfit says, climbing the same mountain. But it is the wrong mountain. Morality should not be understood as hypothetical legislation; it is a mistake to set about constructing morality as if we were making law. Real legislators evaluate possible legal rules by considering the effects they would have. They can do this because enforcement and acceptance of law ensure a high level of compliance. Moral legislators have no reason to assume any particular level of acceptance; the effects of counterfactual acceptance of a principle are not morally relevant. The argument targets rule consequentialism and Scanlon’s official version of contractualism. The paper begins in a positive mode by arguing that a nonlegalist version of Scanlon’s approach, that seeks justification for conduct of such and such a kind in such and such circumstances by comparing the reasons in favor and the reasons others have to object, is a very attractive way to think about nonconsequentialist constraints on how we may treat one another.
- Reductionist Nonconsequentialism If we take for granted that familiar moral constraints on the treatment of other people cannot be explained away in terms of the beneficial consequences of adopting certain standing dispositions and deliberative rules of thumb, the question remains of what kind of explanation and justification of them is available. Can more can be said than that it is self-evident to “thoughtful and well-educated people” that certain ways of treating people are wrong?^2 As 1 (^2) W. D. Ross, The Right and the Good (Oxford: Oxford University Press, 1930), 41.
most moral philosophers are in this respect Rawlsians now, most answer yes. Moral theory starts but does not end with a statement of considered judgments; it aims to provide a set of principles or a moral conception that matches our considered judgments in reflective equilibrium. Even philosophers such as Judith Jarvis Thomson and F. M. Kamm, who, unlike Rawls, insist that some considered judgments about particular cases must be treated as fixed, nonetheless see the point of moral theory as providing explanation and justification for those beliefs. Thomson writes, invoking Socrates, that “while a cluster of beliefs may be a cluster of true beliefs.. ., knowledge that they are true requires knowledge of what makes them true.”^3 And Kamm, who insists that the first step in moral theory is the formulation of intricate moral principles as generalizations of “as many case-based judgments … as prove necessary,” holds that we cannot conclude that any principle is correct unless we find that it “expresses some plausible value or conception of the person or relations between persons.”^4 Thomas Nagel well expresses the standard view: “Common sense doesn’t have the last word in ethics or anywhere else, but it has, as J. L. Austin said about ordinary language, the first word: it should be examined before it is discarded.”^5 A rather different approach has been explored by T. M. Scanlon since “Rights, Goals, and Fairness,” in 1975.^6 His “reductive” strategy does not start from considered judgments about particular cases or proposed principles as data. The project has been to find an intuitively compelling unified account of the domain of nonconsequentialist principles a whole, reasoning (^3) The Realm of Rights (Harvard UP, 1990), 31. (^4) F. M. Kamm, Intricate Ethics (OUP, 2007), 5, 469. (^5) The View from Nowhere (OUP, 1986), 166. (^6) Reprinted in The Difficulty of Tolerance (Cambridge: Cambridge University Press, 2003), 26-41.
judgments about particular cases in reflective equilibrium, may in the end suggest to us a compelling nondeontic normative account of why those are the right principles, but it also may not. The reductive approach starts with such an account: there needs to be a master idea to identify and organize the non-deontic considerations that lead us to deontic conclusions. This is the sense in which the account is reductive, explaining all deontic principles by appeal to the same core considerations. In Scanlon’s contractualism, the master idea is justifiability of our actions to those they affect.^9 If the idea that wrongness is best understood and explained in terms of justifiability is not appealing, some other master idea is required for a reductive approach to be viable. A plausible master idea will have to navigate what Scanlon calls Pritchard’s Dilemma:^10 the idea must not be too external to morality, for example reducing it to enlightened self-interest, but nor must it simply state deontic principles seriatim, thus not producing a reductive account at all. Scanlon has long used the example of welfarist consequentialism as a possible competing master idea: the idea that wrongness of action must always in the end be explained in terms of the impact on people’s welfare. Scanlon mentions a second important appeal of a reductive approach: it provides “a way of explaining and interpreting incompletely specified exceptions”^11 to familiar nonconsequentialist principles. This certainly would be good to have. Especially the most important nonconsequentialist commitments, such as a right not to be killed, turn out to be (^9) Rainer Forst plausibly suggests that (at least as a moral theorist) Scanlon is better characterized as a “justification fundamentalist” rather than a “reasons fundamentalist”—as Scanlon characterized himself in Being Realistic about Reasons (OUP, 2014); see Forst, “Justification Fundamentalism: A Discourse-Theoretical Interpretation of Scanlon’s Contractualism” forthcoming in the Lauener volume. (^10) What We Owe to Each Other , 150. (^11) “Contractualism and Justification,” 8.
extremely difficult to pin down at the level of casuistry. Kamm’s method of as many case-based judgments as prove necessary leaves many of us feeling that our ability to judge or even consider has reached its limit and that the only way to continue is to appeal to some principle or general method for thinking about such cases. On the face it, then, Scanlon’s reductive strategy is very appealing. It would be good to have something more to say than that treating someone in such and such a way is obviously wrong, and it would be good to have a method for resolving puzzling cases other than by appeal to “intuitions” about cases and principles that, for many of us, either don’t exist or can only be understood as the outputs of some imperfectly understood theory. I find Scanlon’s master reductive idea appealing as well: with justification as the core idea, we reason to nonconsequentialist moral principles by thinking about what kinds of reasons we could offer others in justification for our actions, taking into account our recognition of the reasons they also have to “object to being affected in certain ways.”^12 By contrast, the consequentialist reduction seems dogmatic and too substantive to serve as a starting point for thinking about how we may treat each other. And, looking in the other direction, Kant’s ambition in the Groundwork to derive the content of the moral law from nothing more substantive than the conditions of rational agency seems clearly hopeless.^13 So the thought is that treating people rightly means being able to justify my actions to them, by appeal to the reasons I have for acting in that way and taking into account the reasons they have to object. But it is a crucial part of Scanlon’s contractualism that this justification (^12) “Contractualism and Justification,” 7. (^13) On this, see Scanlon, “How I am Not a Kantian,” in Parfit, On What Matters vol. 2, (OUP, 2011), 166-39.
[W]hen we are considering the acceptability or rejectability of a principle, we must take into account not only the consequences of particular actions, but also the consequences of general performance or nonperformance of such actions and of the other implications (for both agents and others) of having agents be licensed and directed to think in the way that that principle requires.^15 That we must think about the implications of “having agents be licensed and directed to think in the way that that principle requires” means that what Scanlon has in mind are principles for the general regulation of behavior, generally accepted. Those more distant implications would include, for instance, the anxiety that would be caused if behavior were regulated by a principle that allowed one to be sacrificed, by force, for the sake of many others, even if anyone’s chances of being the one were very low.^16 Derek Parfit brings out the dramatic significance of this feature of contractualism. “[I]n the thought-experiment to which the Kantian Formula appeals, I would have the power to choose which principles everyone would accept, both now and in all future centuries. The principles I chose would be accepted by many billions of people.”^17 Principles for the general regulation of behavior require a wide field of justification, principles for cases like this in circumstances like these do not. Principles for the general regulation of behavior are also subject to design constraints that do not apply in the same way to principles for cases like this in circumstances like these. For the latter, my principle can be as complex as the case requires in the circumstances. Not so principles that can serve for the (^15) What We Owe to Each Other , 203. (^16) See Parfit, On What Matters , vol. 2, 210. (^17) Parfit, On What Matters, vol. 1 (Oxford: Oxford University Press, 2011), 382.
general regulation of behavior. These have to be tractable and thus suitably broad. “[F]iner- grained principles will create more uncertainty and require those in other positions to gather more information in order to know what a principle gives to and requires of them.”^18 A principle for cases like this in circumstances like these just is a statement of my justification for treating you in a certain way.^19 By contrast, it is not immediately clear why a defense of a principle for the general regulation of behavior which allows me to treat you in that way should be thought of as a justification for my treatment of you at all. When thinking about principles for the general regulation of behavior we are thinking as legislators do, about what would, on balance, be the best rule for everyone to follow. We take into account all possible effects of general compliance, and we fashion the norms in such a way that they can efficiently guide behavior. None of that seems to be relevant when one person is trying to justify treating another in a particular way in particular circumstances. Scanlon writes that “[a]ccording to contractualism, thinking about right and wrong is in one respect like thinking about the civil and criminal law: it involves thinking about how there is reason to want people in general to go about deciding what to do.”^20 This is precisely the aspect of Scanlon’s contractualism that I am against.^21 (^18) What We Owe to Each Other , 205. (^19) The description of the circumstances might include information about me, such as my strength or wealth, so that a person with different strength or wealth might be subject to a different principle. These facts about me might affect the reasons I can offer another for or against being required or permitted to act in a certain way. (^20) What We Owe to Each Other 153. (^21) After noting, in the passage quoted above, that a justification of an act will ipso facto be the defense of a principle, Scanlon goes on to say: “There is a question (corresponding to the debate between act and rule utilitarianism) as to whether the justification for an action should appeal only to consequences of that act … or whether other considerations are also relevant. I will address that question in the next section.” What We Owe to Each Other 197. In the next section he discusses what I have called the wide field of justification which is clearly necessary if we are working out principles for the general regulation of behavior. But that section appears to assume that any principle successfully defended by me in my justification of my act to you will ipso facto be one that everyone will follow. Otherwise there would be no need to consider, as a general mater, “the consequences
claims: “An individual has a claim against the agent that the agent should comply with a candidate moral principle, just in case the personal interests of that individual make it reasonable for someone in his or her position to reject alternative principles for the general regulation of behavior.”^24 Wallace remarks that, in responding to Parfit’s insistence that Scanlon should drop his requirement that principles can be rejected only by appealing to personal reasons, Scanlon seems “surprisingly open” to the idea of allowing aggregation and appeal to impersonal reasons. Even Scanlon’s recent idea^25 that while the number of people affected by a principle cannot itself be a reason for rejecting a principle, it might be relevant to whether rejection is reasonable, seems to Wallace to weaken contractualism’s appeal: We were looking for an interpretation of moral reasoning that would elucidate the idea that moral obligations are owed to particular individuals, who have claims against the agent to compliance with the principles that determine them. Those elements are most firmly in place if what resolves the question of whether a given individual can reasonably reject a principle are the comprehensive implications of the principle for the life of that very individual, as compared to the similar effects of the alternative principles on the lives of other individuals.^26 But if that is what we were looking for, then we surely do not want what resolves the question of what someone can reasonably reject to include considerations of how general compliance with the principle would shape the social world, or considerations of (as it were) moral drafting, (^24) 180. (^25) In “Contractualism and Justification.” (^26) The Moral Nexus , 180.
having to do with limits to how fine-grained a principle for the general regulation of behavior can usefully be. The appeal of contractualism, I agree with Wallace, is that it focuses our thinking, when we are thinking about the directed obligations we have to others in virtue of their rights or claims to be treated in certain ways, to what seems to be at stake—a comparison of my reasons for acting in a certain way and their reasons to object. But if we are told that what we are doing is engaging in hypothetical moral legislation, that appeal is lost.^27 Consider this striking statement of Scanlon, connecting the contractualist idea to the value of human life: respecting the value of human (rational) life requires us to treat rational creatures only in ways that would be allowed by principles that they could not reasonably reject insofar as they, too, were seeking principles of mutual governance which other rational creatures could not reasonably reject.^28 As Kamm notes, this is Scanlon at his most Kantian.^29 Scanlon later aligns himself with Kant’s claim that the first and second formulations of the categorical imperative come to the same thing. Interpreting the “universal law formula” along contractualist lines (“What general principles of action could we all will?”), treating others as ends in themselves must be (^27) Clara Lingle, “Generic Objections, Actual Claims: A Problem for Contractualism as Wronging,” (unpublished ms.
- points out that if what I am calling a wide-field of justification is used in a theory such as Wallace’s, which appeals to contractualism to provide an account of wronging, or relational morality, it has the consequence that failing to follow certain principles wrongs everyone, not just the particular victim. As Lingle notes, Wallace acknowledges this consequence (at 207 and 219) but feels that it is compatible with the relational account. I agree with Lingle that if we accept that consequence the distinctive explanatory potential of the relational account is lost. If anyone who might benefit from compliance by others with a principle has a relational claim against those who fail to comply, then utilitarianism can be given a relational characterization. (^28) What We Owe to Each Other, 106. (^29) “Owing, Justifying, and Rejecting,” Mind 111 (2002): 325-54, 327.
principle for this case and circumstances, I should be prepared to follow it when relevantly similar cases and circumstances arise in the future. It is also relevant to consider hypothetical variations on the case and the circumstances to see how a suggested principle would fare in such a situation. This is not because we are looking for a principle to cover both situations—a principle of sufficient breadth to be workable, teachable, and enforceable that will necessarily have to smooth over some relevant differences. The point of considering hypothetical variations, once again akin to common-law reasoning, is to help us sharpen our understanding of the reasons we are bringing to bear in the actual case and circumstances. What is not relevant is the impact of the principle in counterfactual circumstances of its general acceptance. Principles for acts like this in circumstances like this do have explanatory power. They help us identify the reasons that people may reasonably advance in favor of treating others in some way, and the reasons the others may reasonably advance in objection. As an example, take the issue of responsibility for reliance that one has invited or encouraged. A possible principle, R, is that if one person invites another to rely on their stated intentions, and the other person does rely, then the first person must do what they can to prevent that reliance coming at a loss.^33 We could just eyeball this principle and see if it seems plausible. But approaching it via Scanlon’s reductionist method seems much more illuminating. The relying person has obvious grounds to reject a principle permitting the reliance-inviter to walk away from the induced losses. And rejection of R by the inviter would not be appear to reasonable: Yes, it is now inconvenient and perhaps costly for the inviter to take responsibility for the losses, but the (^33) This is a more restrictive, and I think, more plausible principle than Scanlon’s Principle L, What We Owe to Each Other , 300-301.
relier has the losses just because of what the inviter did, and the inviter got themselves into this by doing something they didn’t have to do.^34 The great merit of Scanlon’s reductionist project is to have us focus on this kind of comparative evaluation of reasons, rather than going directly—blindly—to a verdict on the wrongness or not of the action or the soundness of the principle.^35 Removing the factor of the impact of general acceptance of a principle on the social world will likely mean that thicker or more moralized reasons than those having just to do with relative burdens will have to be invoked.^36 But this appears likely to be necessary even under the model of hypothetical legislation.^37 This is not the place to try to establish the fruitfulness of the method of nonlegislative justification across the board, but it is not obvious that it will fail. What I do believe to be clear is that the appeal of the method of justification is lost if it is understood to involve hypothetical legislation. In embracing a nonlegislative form of Scanlon’s method of justification, I have been discussing throughout deontological constraints, or what Thomson calls the realm of rights. In What We Owe To Each Other , Scanlon clearly has in mind a theory with broader scope, something like an account of impartial moral rules governing interactions among humans; some morally significant matters such as friendship and the treatment of animals and nature are (^34) Wallace, 181, goes through this argument, but wrongly takes it as explaining why promises bind. It cannot do this, as promises bind even in the absence of reliance. (^35) As indicated above (ms. 3) there is room in Scanlon’s method for taking intuitive plausibility judgments into account as a kind of secondary check on possible principles. (^36) For example, Rahul Kumar offers as a reason someone might reject a principle requiring sacrifice whenever this will benefit another more, the reason that this will greatly reduce the control people have over their lives. “Defending the Moral Moderate,” Philosophy & Public Affairs 28 (1999): 275-309, 298. Lingle argues that if appeal to what the social world would be like under full compliance is blocked, the contractualist method of justification will simply rehearse existing deontological intuitions. I am not as pessimistic as Lingle, but the danger is clearly real. (^37) See Scanlon, “Contractualism and Justification.”
The deeper issue is that the domain of impartial responsibility for the welfare of others is distinctive and distinct from the domain of deontology,^44 and is simply not apt for the method of justification. A full discussion is not possible here, but two features of principles of beneficence are worth emphasizing. First, they are principles of imperfect duty. A person in need cannot complain if I exhaust my resources helping another person in similar need. And if I do nothing, though both people could rightly say that I am acting wrongly, neither of them could complain that my treatment of them as individuals is unjustified. Second, principles of beneficence are agent-neutral. Agent-neutral principles are best understood as applying to us collectively.^45 The method of justification by each to each is out of place.
- The Wrong Mountain: Legislation, Real and Imagined Scanlon’s contractualism is of course not the only moral theory that employs the model of morality as hypothetical legislation; the other main example is rule consequentialism. Defenders of that view are comfortable, vocal even, about the legislative analogy: we are asking what moral code would be best.^46 What Parfit calls Kantian contractualism, a view based on Scanlon’s contractualist reading of the universal law formula of Kant’s categorical imperative, also fits the bill.^47 And of course, whether or not that is the right way to interpret the universal law formula, Kant’s language would seem to cast him as the philosophical father of the whole school: always moral law ( Gesetz —a statutory provision) rather than principle or rule or norm; (^44) See Thomas Nagel, The View From Nowhere , chap. IX. (^45) See Murphy, Moral Demands in Nonideal Theory (OUP, 2000). (^46) See Richard Brandt, A Theory of the Good and the Right (Oxford: Oxford University Press, 1979), Brad Hooker Ideal Code, Real World (Oxford: Oxford University Press, 2000). (^47) On What Matters vol. 1, § 52.
legislating for a kingdom of ends. The Kant-inspired discourse-theoretic theories of Habermas and Forst work with the same model. Habermas’s discourse principle is a general principle for the justification of all norms, be they legal or moral: “Only those norms of action are valid to which all possibly affected persons could assent as participants in rational discourses.”^48 As Forst elaborates, norms from different “contexts of justification”—ethical, moral, political, legal—will require different kinds of justification, but the result is the same: valid general norms.^49 Nicholas Southwood’s Habermas-inspired “deliberative” contractualism is cast as method for determining a “common code” to live by, thus appropriating the explicitly legalist language of rule-consequentialists.^50 Not all these legalist theorists are offering constructivist or reductive theories in Scanlon’s sense. Brad Hooker, for example, falls in the more traditional camp of offering rule consequentialism as a moral conception that matches our considered judgments in reflective equilibrium.^51 And on the level of substantive detail too there will be considerable divergence. But in so far as they all share a conception of moral theorizing as a search for an ideal set of moral principles or rules for the general regulation of behavior, then there is a sense in which they are all, as Parfit says, climbing the same mountain.^52 In my view, all of these legalist moral theories are fundamentally flawed. The model of morality as hypothetical legislation should be rejected wherever it is found. (^48) Between Facts and Norms , trans. William Rehg (Cambridge, Mass.: MIT Press, 1996), 459. (^49) Contexts of Justice (Univ. of California Press, 2002) , The Right to Justification. (^50) Contractualism and The Foundations of Morality (OUP, 2010) 102. (^51) Ideal Code , Real World. (^52) On What Matters vol. 1, § 64.
enforced, and checking to see whether anyone in each society would have personal grounds for rejecting the relevant list of rights. The legal rules or principles that enshrine the rights need to be well drafted, of course, with an eye to ease of use by their intended audiences (legislatures, courts, officials). The wide field of justification and the importance of drafting are obvious in this case. It is arguable that the legislative approach is very well suited to the task of figuring out how to put rights into law—how to figure out what legal rights there should be. Scanlon notes that his reductionist approach to moral rights developed out of his reflection on the right to freedom of expression, a political-legal right.^53 The crucial point is that when considering both direct and systemic effects law-makers can assume a high level of compliance with rules they enact. Most of the time, and for most people, both private citizens and state officials, the existence of sanctions for noncompliance will incentivize compliance. In some cases, where the person involved is an official high up in the executive branch of government, direct coercive enforcement may not be present; but even then, if institutions are well-structured, there will be other pressures encouraging compliance. When thinking about direct and systemic effects, then, legislators have good reason to assume a high level of general compliance. The systemic effects must be considered because those effects will be real. Now consider the hypothetical legislation of moral principles or rules. We must think about what the world would be like if everyone followed or accepted a certain principle or rule and consider how well off people would be, or the grounds individuals would have to reject (^53) The Difficulty of Tolerance , xx; “Contractualism and Justification,” 9.
principles.^54 It is not obvious what the relevance of the effects of compliance with or acceptance of a principle is if people will not, in fact, comply with or accept the principle. Thompson asks, during a discussion of property rights: Is it so much as within reason to suppose that the fact that adopting a certain set of rules would be efficient in one or another of the ways I indicated, or in any way at all, makes it the case that those rules among them that govern property rights are already in force?^55 Let us start with an even more extreme example. It is obvious that zoning law in New York City does not enforce independently existing moral rules. You can’t make sense of the rules other than by seeing them as part of a scheme of urban planning that will be good or bad because of its effects—does it make life in the city more affordable, efficient, beautiful, integrated, and so on. Real legislators can reason as follows: It would be good if there were a set of legal rules that required such and such behavior, because such and such behavior would have good effects; therefore, we now enact those rules. They are entitled to reason in that way because they are making law. When you make law you aim to change the world by affecting behavior. Legislators in well-functioning polities have the power to affect behavior because for the most part people generally comply with the rules they make. Not so the hypothetical moral legislator. Suppose I rightly conclude that a certain zoning code of my devising would be best in terms of consequences, or justification; suppose also that it is compatible with but more restrictive than the existing legal code. It would be mad to point (^54) Like Scanlon, I will from now on focus on these two possibilities. (^55) The Realm of Rights, 332 - 3.
