proceedings of the aristotelian society 137th session
issue no. 1 volume cxvi 2015 / 2016
contractarianism as a political morality
benjamin sachs university of st. andrews
m o n d a y, 2 n o v e m b e r 2 0 1 5 17.30 - 19.15
the woburn suite senate house university of london malet street london wc1e 7hu united kingdom
This event is catered, free of charge & open to the general public
contact
mail@aristoteliansociety.org.uk www.aristoteliansociety.org.uk
Š 2015 the aristotelian society
biography Benjamin Sachs is a lecturer in philosophy at the University of St. Andrews. He has worked on issues in distributive justice, health care justice, coercion, normative ethics, environmental ethics, and the ethics of research on human subjects. He is currently interested in animal ethics and in addition is planning to write several papers that would together constitute an argument for contractarianism. editorial note The following paper is a draft version that can only be cited or quoted with the author’s permission. The final paper will be published in Proceedings of the Aristotelian Society, Issue No. 1, Volume CXVI (2016). Please visit the Society’s website for subscription information: www.aristoteliansociety. org.uk.
contractarianism as a political morality1 benjamin sachs Contractarianism initially made its mark, in the 17th century, as a sort of theory of everything in ethics. But gradually philosophers became convinced that there were resources available outside contractarianism for settling important moral questions—for instance, ideas of human rights and the moral equality of persons. Then Rawls revived contractarianism with a more modest aim—namely as a theory of justice. But even this agenda for contractarianism has been called into question, most notably by G.A. Cohen, who contends that we have other tools at our disposal for identifying the true conception of justice. So the question remains, how should contractarianism be construed if it is to provide answers to questions that cannot be answered in some other way? In my essay I offer a very simple answer: contractarianism should be construed as a political morality. I arrive at this answer by beginning with social contractarianism as a theory of everything and paring away the unappealing layers of social contractarianism so understood. I begin by describing what contractarianism is. Then I dispense with social contractarianism as a theory of state legitimacy, as a theory of interpersonal morality, and as a theory of justice. Finally, I distinguish political morality from the other alreadymentioned areas of morality and argue that contractarianism is a sensible theory of its grounds.
CONTRACTARIANISM initially made its mark as a sort of philosophical analogue of what scientists call the theory of everything. For Hobbes, the first well known contractarian, contractarianism was supposed to settle every important question in moral and political philosophy. It provided a theory of how individuals should treat each other, of how extensive the state’s power over the individual should be and toward what end it should use that power, and of the individual’s duty to obey the state and his/her right to overthrow the government. This kind of contractarianism was well suited to its historical context: 17th-century Europe. In that time and place, everything seemed up for grabs. The old feudal systems and the Catholic hegemony were collapsing and two diametric opposites— democracy and monarchical absolutism—were vying to become the next dominant political model. Furthermore, revolutions were rampant and it seemed there was nothing to hold back humankind’s underlying propenI would like to thank Lisa Jones and Sarah Broadie for an extremely helpful extended conversation on a previous draft of this paper and Theron Pummer for reading and commenting on that same draft.
1
4
benjamin sachs
contractarianism as political morality
draft paper
sity for settling differences through violence and brutality. So there was a place for something of the form of contractarianism—some theory of everything to explain how peace and order could be achieved in the face of disagreement and clashes of interests. But then contractarianism fell out of favor, and it is not hard to see why. Eventually, compelling non-contractarian answers to heretoforeopen questions became established both de facto and de jure, thus robbing contractarianism of its motivation. The spread of ideas of human rights and the moral equality of persons had much to do with this, as did the movement in philosophy toward the idea that a moral theory could be built on purely rationalistic grounds—a movement that arguably began with Bentham2—combined with the suspicion that contractarianism is necessarily quite speculative. Contractarianism, in the form of a theory of justice, was revived by Rawls. But even this more modest aim for contractarianism has been called into question, most notably by G.A. Cohen, who contends that we have other tools at our disposal, aside from contractarianism, for identifying the true conception of justice. So the question remains, what is contractarianism good for? How should contractarianism be construed if it is to provide answers to questions that have not been and cannot be answered in some other way? In this essay I offer a very simple answer: contractarianism should be construed as a political morality. A political morality is basically a morality constraining the state, though that definition will be made more precise as we go on. I arrive at this answer by beginning with contractarianism as a theory of everything and paring away the unappealing layers of contractarianism so understood. I begin, in §1, by describing what contractarianism is. Then, in §2, I dispense with contractarianism as a theory of legitimacy— a theory of the individual’s obligation to obey the state. In §3, I discard contractarianism as a theory of interpersonal morality. My final move, in §4, is to concede Cohen’s point that contractarianism is not plausible as a theory of justice. I conclude, in §5, by identifying what philosophical questions in the general area of moral/political theory remain, having removed so much from contractarianism’s purview. I contend that very important questions of political morality are left unanswered, I sketch how contractarianism addresses them, and I respond to worries about contractarianism’s ability to addressing them well. J.B. Schneewind, “The Divine Corporation and the History of Ethics”, in Essays on the History of Moral Philosophy (New York: Oxford University Press, 2010), pp. 149-69.
2
5
benjamin sachs
contractarianism as political morality
draft paper
i. characterizing contractarianism
Contractarianism, as I understand it, begins with a state of nature thought experiment. The state of nature is a situation in which there are no generally effective checks on individual behavior. This may be because there is no widespread recognition of moral constraints, or rather simply because however much individuals might agree on what the moral constraints are they nevertheless cannot rely on each other to obey those constraints and agree about which actions constitute violations. The second essential element of contractarianism is, of course, contracting. The idea is that rational individuals would perceive the state of nature as a non-ideal circumstance and therefore attempt to avoid it, and would do so by contracting with each other to create some publicly recognized authority. This authority could take the form of a list of rules, a sovereign ruler with unlimited discretion and a monopoly on coercive force, or anything in between. I’m interested in what moral/political question is supposed to be settled by this contracting, according to contractarianism understood the way I’ve described it. The contention I intend to support in what remains of this essay is that contractarianism is most plausible when we take the contracting as settling only questions of political morality.
ii. legitimacy
We begin, however, by examining contractarianism as a theory of legitimacy. The question of the state’s legitimacy is often understood to be the question of the individual’s obligation to obey the state’s dictates.3 So understood, Hobbes and Locke undoubtedly used contractarianism as a theory of legitimacy—that is, a theory explaining under what circumstances, and why, the individual is obligated to obey the state. It’s also been argued that Rousseau used contractarianism as a theory of legitimacy.4 Gregory Kavka, Hobbesian Moral and Political Theory (Princeton, N.J.: Princeton University Press, 1986), p. 22; A. John Simmons, “Philosophical Anarchism”, in Justification & Legitimacy: Essays on Rights and Obligations (Cambridge: Cambridge University Press, 2001), pp. 102-21 at 106.
3
David Lay Williams, “Ideas and Actuality in the Social Contract: Kant and Rousseau”, History of Political Thought, Vol. XXVIII, No. 3 (Autumn 2007): 469-95; Nenad Miscevic, “In Search of the Reason and the Right—Rousseau’s Social Contract as a Thought Experiment”, Acta Analytica 28 (2013): 509-26 at 510.
4
6
benjamin sachs
contractarianism as political morality
draft paper
The basic idea of contractarianism as a theory of legitimacy is that individuals are obligated to obey the state because that’s what they did agree to or would agree to through contracting. But this is a seriously deficient theory of legitimacy no matter how it gets worked out. The problems here are familiar and much discussed, so instead of trying to add anything new I will simply review what seem to me to be the key objections. David Hume was perhaps the first to point out the most obvious problem, which is that most or perhaps all current states (Hume said all current states) were not founded through contracting.5 Furthermore, even if current states were formed through contracting, that fact would seem to be irrelevant to the question of whether the current denizens of the state are obligated to obey its dictates, except in the rare case of a new state where the current denizens are the founders. Contractarianism has room for maneuver here if we are willing to be accommodating regarding what it takes for a theory to qualify as contractarian. There may never have been any contracting in the sense of a group of individuals coming together and striking an agreement to form a state vested with a certain measure of authority over their behavior, but we may be able to at least understand individuals as consenting to the authority of the state. Admittedly, such an action might not really qualify as contracting, since it isn’t reciprocal in the way that contracting is generally supposed to be. But if we construe the underlying spirit of contractarianismas-a-theory-of-legitimacy as the idea that a state gains its legitimacy over the individual through an act of that individual’s will, then the consent theory would still qualify as contractarian. Nevertheless, problems persist, as Jean Hampton has nicely illustrated. There are four ways of understanding the connection between the individual agreeing to something and her having an obligation to obey the state, but there is trouble no matter which understanding the contractarian plumps for.6 We have seen the first theory and the problem for it already— namely the theory of the contract as a historical event and the worry that no such events ever occur. The second theory proposes that currently living individuals actually consent to be subject to their state’s authority; its problem of course is that this is simply not true, as most people have never been invited to register their consent or dissent. The third theory is that currently living individuals provide tacit consent, through their actions, to the state’s authority, while the fourth theory is that currently living indiDavid Hume, “Of the Original Contract”. Michael Lessnoff emphasizes this problem as well in Social Contract (Basingstoke, Hampshire: Macmillan, 1986), pp. 87-90.
5
Jean Hampton, Hobbes and the Social Contract Tradition (Cambridge: Cambridge University Press, 1986), pp. 266-8.
6
7
benjamin sachs
contractarianism as political morality
draft paper
viduals would, hypothetically, consent if they were invited to do so. But both of these theories confront the same problem: tacit and hypothetical consent don’t seem to have any moral force. It seems plausible to suppose that we are bound to do what we have actually agreed to, but the moral considerations underlying this thought lend no support at all to the idea that we are bound to do what we would agree to do7 or what our actions suggest we are content with.8,9 Of the four versions of contractarianism-as-a-theory-of-legitimacy, the only one that seems potentially salvageable given the objections is the version that makes legitimacy contingent on actual consent. The other three are (at least) morally deficient; they inflate the moral import of their respective varieties of consent. Actual-consent theory, by contrast, is undermined only by the facts, so it is open to an actual-consent theorist to stick to her moral claim that legitimacy requires actual consent and simply concede that all current states are therefore illegitimate.10 Ronald Dworkin, “The Original Position”, In N. Daniels, ed., Reading Rawls (Stanford University Press, 1989), pp. 16-52; Williams, “Ideas and Actuality in the Social Contract”, p. 479.
7
The reader might at this point worry that this is not a fair catalogue of the contractarian’s options, since it omits Rawls’s theory of legitimacy, and Rawls is certainly a contractarian. In response, I would insist that there is nothing contractarian about Rawls’s principle of legitimacy, which runs as follows: 8
Our exercise of political power is proper and hence justifiable only when it is exercised in accordance with a constitution the essentials of which all citizens may reasonably be expected to endorse in light of principles and ideals acceptable to them as reasonable and rational. (Rawls, Political Liberalism (New York: Columbia University Press, 1996), p. 217.) The reason this principle isn’t contractarian, even under the broad conception of contractarianism with which we have been operating, because it makes no use of the idea of contracting. Indeed, Rawls calls it “the liberal principle of legitimacy”, suggesting that liberalism is the driving idea behind it, not contractarianism (which is a version of liberalism). My argument in this section is aimed narrowly at contractarian theories of legitimacy, as opposed to being aimed more broadly at theories of legitimacy that make use of the idea of acceptability from the viewpoint of the people who are ruled. For further objections to contractarianism as a theory of legitimacy, see Margaret Gilbert, A Theory of Political Obligation (New York: Oxford University Press, 2006), Chs. 5, 10). In addition, Russell Hardin offers various objections to contractarianism, some of which apply to contractarianism as a theory of legitimacy, in “Contractarianism: Wistful Thinking,” Constitutional Political Economy 1 (1990):35-52.
9
Simmons, op. cit. and “Justification and Legitimacy” in Justification & Legitimacy: Essays on Rights and Obligations (Cambridge: Cambridge University Press, 2001), pp. 122-57. Williams (“Ideas and Actuality in the Social Contract”) defends Rousseau’s actual-consent theory on moral grounds, and so we can speculate that he may think that
10
8
benjamin sachs
contractarianism as political morality
draft paper
Unfortunately contractarianism, so understood, would be quite anemic, so I will set aside this interpretation of it. I want to determine whether contractarianism can be understood as providing an answer to a question that is urgent both philosophically and practically. A theory of legitimacy that, like actual consent theory, can’t tell us that North Korea is less legitimate than other states, is of little use in our practical project of determining how to respond to the state’s dictates. iii. interpersonal morality
Having dispensed with contractarianism as a theory of legitimacy, we now move on to consider contractarianism as an interpersonal morality, or what we would ordinarily simply call “morality”—i.e., the set of moral obligations and permissions that apply to moral agents as such in their dealings with other individuals. My argument that contractarianism should not be deployed as an interpersonal morality rests on two key premises. First, contractarianism is a deeply flawed interpersonal morality. Second, the idea, put forward by various contractarians over the years, that there is some aspect of interpersonal morality that cannot be accounted for outside a contractarian framework, is mistaken. I defend these two premises respectively in what follows. Contractarianism as an interpersonal morality is endorsed in the works of 20th-century theorists such as John Harsanyi,11 David Gauthier,12 Gerald Gaus,13 Jan Narveson,14 and Jeffrey Reiman.15 Its intellectual forebear Rousseau should go this route. John C. Harsanyi, “Can the Maximin Principle Serve as a Basis for Morality? A Critique of John Rawls’s Theory”, in Essays on Ethics, Social Behavior, and Scientific Explanation (Dordrecht: D. Reidel, 1980), pp. 37-63. One could be forgiven for thinking that Harsanyi is not using contractarianism as an interpersonal morality, given his concern to extract the preferences we have “for society” or for welfare functions “of society”. His habit of using the “of society” qualifier could lead one to think that whatever moral framework Harsanyi is trying to construct, it is supposed to apply within societies only, which would suggest that it’s not an interpersonal morality since interpersonal morality is now generally agreed to apply between any two moral agents regardless of whether they’re members of the same society. However, there is no reason Harsanyi needs to use “society” in its everyday sense; for him it can denote any group of individuals each of whom has a welfare. So we can, and should, interpret Harsanyi’s “society” as consisting of all individuals with a welfare.
11
12
David Gauthier, Morals by Agreement (New York: Oxford University Press, 1986).
Gerald Gaus, The Order of Public Reason: A Theory of Freedom and Morality in a Diverse and Bounded World (New York: Cambridge University Press, 2011), and “Why the Conventionalist Needs the Social Contract (and Vice Versa)”, Rationality, Markets and Morals 4 (2013):71-87.
13
Jan Narveson, “Social Contract: The Last Word in Moral Theories”, Rationality, Markets and Morals 4 (2013):88-107. 14
15
Jeffrey Reiman, Justice and Modern Moral Philosophy (New Haven, Conn.: Yale Uni-
9
benjamin sachs
contractarianism as political morality
draft paper
may be Thomas Hobbes, though that is controversial and I will not try to settle the matter here, as my concern is not historical.16 As to whether contractarianism is an adequate interpersonal morality, I’ve argued elsewhere, in detail, that it is not.17 The argument, briefly, runs as follows. First, moral theories, in general, have an explanatory job to perform; specifically, they must not only identify which actions are wrong but also identify what it is about them that makes them wrong. Second the contractarian explanatory story is substantially impersonal. When an action is wrong, the explanation of this, according to contractarianism, is that it would be forbidden by the contract that would be agreed upon in the state of nature. This explanation is impersonal in that no particular individual, aside from the perpetrator, is mentioned in it. Now the key move: Note that the explanatory story that a theory tells determines what that theory says about wronging—where “wronging” is a kind of wrongdoing that has a direction in that it is something that one individual (the perpetrator) does to another (the victim) and that has further implications in terms of the appropriateness of resentment, apology and forgiveness. Depending on what the explanation of a certain instance of wrongdoing is, we can determine whether that instance of wrongdoing is an instance of wronging. The payoff of all this is that since the contractarian explanatory story for each instance of wrongdoing is impersonal, then according to contractarianism no wrongdoing counts as wronging, since wronging is a personal relation between a perpetrator and a victim. Therefore, contractarianism is an inadequate theory of interpersonal morality, as some violations of interpersonal morality are indeed cases of wronging (rape, for instance). One possible response is that contractarian explanations of wrongdoing do indeed mention individuals other than the perpetrator, namely all the parties to the contract. The contractarian idea is that they are all victims in that a contract to which they were party has been violated. But the problem here is that there are certain individuals who cannot be sensibly represented as parties to the contract yet can be victims of wronging, namely non-rational-but-sentient individuals. Pinning the objection to contractarianism on this point makes the objection less all encompassing, but still absolutely devastating. If a theory cannot say that the savage beating of a horse by a rational person constitutes wronging, then that versity Press, 1990), Chs. 1-2. For an interesting take on the issue, see Gregory Kavka, Hobbesian Moral and Political Theory, Ch. 9.
16
17
Sachs, unpublished book manuscript, Ch. 2.
10
benjamin sachs
contractarianism as political morality
draft paper
theory is significantly lacking. Note, finally, that if we treat contractarianism only as a political morality, which is what I later will suggest we should do, the impersonal nature of the explanatory story it tells is no longer a problem. Suppose, for instance, that the state fails to dictate which side of the road its denizens should drive on, and that this leads to many fatal car accidents. I assume we all accept that this constitutes wrongdoing on the state’s part. Contractarianism as a political morality will say that this constitutes wrongdoing on account of its constituting a failure on the part of state to do what the contract demands that it do (solve life-threatening coordination problems, perhaps). This wrongdoing is a wrong to all parties to the contract equally, if anyone. I see nothing unreasonable about such an implication. This completes the first part of my argument against contractarianism as an interpersonal morality. We move on now to the second part, which addresses the question of whether there is some feature of interpersonal morality that only contractarianism can account for. The classic motivation for contractarianism, on the moral as opposed to the political side of things, has been its promise of showing how the constraints of interpersonal morality—such as constraints against killing, enslaving, and encroaching on the property of, other individuals—can align with the demands of self-interest. The idea is that a contract that sets certain ground rules of behavior that would, if generally obeyed, be such a great good that it is in one’s self-interest to obey such a contract in general. The dividend in terms of general peace and prosperity would be so enormous that the self-interested payoff of any particular violation of the rules would be outweighed by the expected disutility of its precipitating a breakdown in general adherence to the rules. This is certainly the motivation for Gauthier, as well as for Hobbes18 (though as I admitted already it’s controversial whether Hobbes intended to make contractarianism the basis of interpersonal morality). This motivation will be, I take it, generally acknowledged as outdated. The idea that the demands of interpersonal morality must be vindicated by showing that they overlap with the requirements of self-interest is based on a discredited assumption that the demands of self-interest are somehow foundational and beyond question. We now believe, by contrast, that the entirety of the normative domain is subject to certain forms of skepticism, though of course there is no general agreement as to the extent to which this skepticism can be defeated. 18
Kavka, Hobbesian Moral and Political Theory, Ch. 9.
11
benjamin sachs
contractarianism as political morality
draft paper
Granted, it may seem as though self-interest is in an advantaged position as compared to other domains of normativity, because it seems easier to show that self-interest can do one of the things that normative domains are supposed to do: generate reasons for action. If we accept the Humean internalist position on reasons, namely that in order for an agent to have a reason to φ it must be possible for that agent to come to be motivated to φ, then we have to ask whether agents can in general be motivated to pursue their own self interest. And it’s obvious that they can. However, accepting all this, it’s still not clear that self-interest is in a better position than other normative domains. The idea that agents are, as a psychological fact, motivated by nothing other than self-interest is another plank of the contractarian position that has become dislodged over time. The weight of the evidence simply no longer supports this position. The second most influential motivation for contractarianism as a theory of interpersonal morality is its promise of showing how an individual’s obeying the commands of interpersonal morality is compatible with her autonomy. (Reiman’s contractarianism seems to be motivated this way.) There is an old worry about morality, namely that it confronts us as an alien source of demands, and contractarianism seems to be well situated to answer this concern since it represents moral demands as self-imposed. This evident virtue of contractarianism comes at a cost, however. How a moral theory represents the source of its demands has implications for what explanatory story it must tell. Consider, for instance, theistic ethics. For those religions that represent moral demands as having a voluntaristic source—that is, as being determined by God’s will—wrongdoing is classically described as a rupture in one’s relationship with God. It is what one does vis à vis God that explains the wrongness of one’s action, and so if the action wrongs anyone, it wrongs God. Analogously, insofar as contractarianism is the view that moral requirements are actions that one’s own autonomy demands of oneself, any instance of wrongdoing will have to be classified as wrong on grounds of its rupturing one’s relationship with oneself. Consequently, the only potential victim of one’s own wrongdoing will be oneself. Now one might object that the theory critiqued just now isn’t really contractarianism at all, but rather Kantian constructivism. Contractarianism, one might say, represents the demands of morality as autonomously agreed to, which is a very different thing from Kantian constructivism’s representation of moral demands as being grounded in respect for autonomy. Unfortunately for contractarianism, it’s only the more Kantian inter-
12
benjamin sachs
contractarianism as political morality
draft paper
pretation of the view that can deliver the promised virtue of representing the demands of morality as self-imposed. Mere autonomous agreement to do something, which is what non-Kantian contractarianism says is the source of moral requirements, doesn’t make the resulting moral requirement to do the agreed-to thing self-imposed. The act of agreement creates a moral requirement only when there are at least two individuals engaging in that act. By contrast, where there is only one act of “agreement” the result is a resolution or a commitment but not a requirement. So under the non-Kantian interpretation of contractarianism the moral requirements incumbent on the agent are no more the agent’s creation than they are the creation of the other party to the agreement, and therefore they do not qualify as self-imposed. I conclude, then, that contractarians have not managed to demonstrate that there is some important aspect of interpersonal morality that cannot be accounted for outside of a contractarian framework, at least not without incurring serious drawbacks. Although contractarianism might succeed in showing that the demands of morality coincide with those of self-interest, this is no great step toward accounting for their normative force. And although Kantian contractarianism can show that moral demands are self-imposed and thus not alien to us, this comes at the cost of implying that all wrongdoing is an offense only against oneself. * * * In §2 we determined that contractarianism is not best construed as a theory of legitimacy—that is, a theory of what obligations the individual has to the state—and in this section we concluded that contractarianism is flawed as a theory of what obligations individuals have to each other. From this we can conclude that contractarianism should not be understood as the grounds of any of the obligations that individuals have. The obvious alternative is to interpret contractarianism as a theory of the obligations that the state has. I investigate this alternative in the next two sections. Before doing so, however, there is one more question about interpersonal morality that we should ask: If contractarianism is to be construed as a theory of the state’s obligations, what should a contractarian, so understood, say about interpersonal morality? My view, which I will explain in §5, is that contractarianism works as a theory of the state’s obligations only if there is some true theory of interpersonal morality. Therefore, while it is sensible to subscribe to contractarianism (or so I will argue), it is not sensible to subscribe only to contractarianism. We might, for instance, emulate Locke (and Hobbes, on some interpretations of him) who
13
benjamin sachs
contractarianism as political morality
draft paper
not only endorsed contractarianism in political philosophy but also natural rights theory in interpersonal morality; alternatively we might emulate Kant, who endorsed contractarianism in political philosophy but also the categorical imperative in interpersonal morality. iv. justice
But first, we need to understand which aspect of the state’s obligations contractarianism is best understood as settling. In this section I investigate the idea that contractarianism should be understood as a theory of the state’s obligations of justice—or, to put it more succinctly, as a fundamental theory of justice. Rawls was probably the first theorist to put forward contractarianism explicitly as a theory of justice,19 and the contemporary trend in contractarian thought certainly leans in this direction,20 though there are very few contemporary thinkers who have, like Rawls, offered a completely worked out theory of contractarianism as a fundamental theory of justice.21 In any event, the basic idea is that the content of the contract settles what justice fundamentally requires the state to do. So if the contract secures freedom of expression for everyone, then the state has a fundamental obligation of justice to allow individuals to express themselves. G.A. Cohen rejects this use of contractarianism.22 His argument is long and complex and unfortunately I can do no more than summarize it here. Briefly, it runs as follows: First, ultimate normative principles—principles regarding what an agent ought to do—are fact-insensitive. In other words the fundamental truths about what individuals ought to do are not sensitive to what the non-normative facts are. Second, the principles that emerge from the contractarian thought experiment are fact-sensitive, because the contractors are stipulated to be in possession of knowledge of the facts and their knowledge of those facts influences their thinking regarding which principles to advocate. Cohen takes Rawls as his target here, noting that Rawls allows his contractors to have knowledge of basic sociological and psychological facts as well as knowledge that the circumstances of justice obtain.23 But, as Cohen notes, the claim applies 19
John Rawls, A Theory of Justice (Cambridge, Mass.: The Belknap Press, 1999).
20
As noted in Hardin, “Contractarianism: Wistful Thinking”.
21
Michael Lessnoff’s Social Contract might be the only other example.
G.A. Cohen, Rescuing Justice and Equality (Cambridge, Mass.: Harvard University Press, 2008), Chs. 6-7.
22
The circumstances of justice, briefly, are 1) the fact that resources are scarce, but not too scarce, and 2) the fact that individuals are much more concerned to promote their
23
14
benjamin sachs
contractarianism as political morality
draft paper
to contractarianism per se.24 This is the case because a universal feature of contractarianism is that it imagines the contractors as being given the task of agreeing on principles to regulate some domain of affairs, and we cannot sensibly imagine the contractors having preferences on how some domain of affairs should be regulated if they know nothing about how the domain works. From these two premises we may conclude that contractarianism does not yield fundamental principles. Third, justice is just one normative domain among many. Fourth, the principles that emerge from the contractarian though experiment will inevitably be sensitive to the entire range of normative considerations. This, again, is due to the practical nature of the contracting situation.25 The contractors are given the task of drafting principles to regulate some domain of affairs, and it would run counter to sensible psychological assumptions to depict the contractors as caring only about justice and therefore designing principles that cater only to justice and to no other normative agenda. From these two premises Cohen concludes that the principles inferable from the contractarian thought experiment are not principles of justice. Putting the conclusions of the foregoing two arguments together, Cohen arrives at his overall critique of contractarianism as a fundamental theory of justice: Contractarianism yields principles that are neither fundamental nor about justice. I believe that this argument is sound. Others deny this, of course, but I lack the space here to defend Cohen against objections. Instead, I will simply assume that he is right. v. political morality
From §2-3 we were able to infer that if contractarianism can be made plausible at all, it will have to be as a theory of the state’s obligations. But if the conclusion of §4 is correct, contractarianism is not well placed to yield a theory of the state’s obligations of justice. Which of the state’s obligations, then, is contractarianism suited to reveal? own interests than they are to promote the interests of others. Ibid., p. 294. Actually, Cohen aims his critique at all constructivists, where ‘constructivism’ is the view that “a principle gains its normative credentials through the being the product of a sound selection procedure” (ibid., p. 274). Contractarianism is, as Cohen notes, a version of constructivism.
24
On the practical nature of the contracting situation and why it is important to distinguish the practical question confronting the contractors from the theoretical question confronting you and I, see ibid., pp. 276-7.
25
15
benjamin sachs
contractarianism as political morality
draft paper
Suppose I were to suggest that contractarianism yields the facts as to what the state is all-things-considered morally required to do. One would surely think I’d missed the point of Cohen’s critique. Justice is an aspect of morality, and therefore if contractarianism cannot reveal the requirements of justice then ipso facto it cannot reveal the requirements of morality all-things-considered. We can be pretty much certain that this is precisely what Cohen would have said in response to such a proposal. We know this because although Cohen never responded to that proposal, he does respond, in Rescuing Justice and Equality, to a similar proposal: to construe contractarianism as yielding the all-things-considered correct rules for regulating society. And what he says against that proposal is that although the correctness of a rule of regulation is a function of many things, of which its adherence to justice is only one, such principles should serve justice “as much as principles in all due reason should and can.”26 Therefore, we cannot know which rules of regulation are best if we don’t know what justice requires.27 Consequently we make a mistake if we skip straight to the question of what the all-things-considered correct rules of regulations are. Similarly, Cohen would no doubt maintain that we make a mistake if we skip straight to the question of what morality requires of the state all-things-considered. Note, however, that justice is an aspect of interpersonal morality, which is to say that individuals have obligations of justice to each other. Given this, the contractarian can apparently construct an adequate rejoinder to Cohen’s objection by making three moves. The first move is to adopt, in addition to contractarianism, some theory of interpersonal morality—specifically, one that concedes the existence of obligations of justice. The second move is to assert that the state is bound by the requirements of interpersonal morality. Neither of these first two moves is ad hoc. If contractarianism is not itself supposed to be a theory of interpersonal morality, as I argued in §3 that it should not, then there is nothing stopping somebody who accepts contractarianism as a political morality from adopting some other theory of interpersonal morality. And it is quite plausible to say that the state is bound by the requirements of interpersonal morality. It makes sense as a matter of theory since the state is an agent, and the requirements of interpersonal morality are supposed to apply to agents as such. (Or, if one is discomfited by the metaphysics of ascribing agency to an abstract entity, we can at least say that the state is an entity through which flesh-and-blood agents exercise their agency.28) 26
Ibid., p. 305; see also p. 284.
27
Ibid., pp. 266-7, 286.
Likewise, if one is discomfited by saying that a state can do wrong, then we can say instead that the state is an entity through which flesh-and-blood agents—in this case,
28
16
benjamin sachs
contractarianism as political morality
draft paper
And it has comforting implications for cases. Assuming, for instance, that interpersonal morality contains a ban on killing except in self-defense, interpersonal morality yields a sensible explanation for why it is wrong for the state to launch wars of aggression. Unfortunately, the first two moves, when combined, leave us with an apparent contradiction. They imply that something other than contractarianism tells us what the state is morally obligated to do, and yet I proposed earlier that contractarianism is best understood as telling us what the state is morally obligated to do. However, the apparent contradiction can be resolved as long as we make the third move: maintaining that 1) something other than contractarianism, namely the correct theory of interpersonal morality, settles what the state is obligated-qua-agent to do, and 2) contractarianism settles what the state is obligated-qua-state to do. Again, there is nothing arbitrary about this move. It makes no less sense than saying that there are some obligations we humans have simply because we are agents while there are other obligations we have because of certain roles we occupy (parent, friend, teacher, e.g.). I began this section with the idea that contractarianism tells us what the state ought to do all-things-considered. We anticipated that Cohen would respond by objecting that we cannot know what the state ought to do all-things-considered until we first know what justice requires of the state, and of course contractarianism cannot tell us what justice requires of the state (as I admitted in §4). I responded by suggesting that since the state is both an agent and a special kind of agent—a state—it is entirely plausible to think that it will take two moral theories to give us the full story as to what the state is obligated to do. Obligations of justice are obligations of interpersonal morality and surely the state is constrained by interpersonal morality. But what obligations the state has because it is a state is a separate matter, and contractarianism could well be the right theory of those obligations even if it would be defective as a theory of justice. I can now state my thesis: Contractarianism is most plausible when it is construed as political twice over, in virtue of taking as its subject matter the state’s obligations and in virtue of taking as its subject matter obligations-qua-state. So, contractarianism is most plausible as a theory of the state’s obligations-qua-state. Or, as I prefer to say, it should be understood as a political morality. In what remains of this section I anticipate and respond to five possible objections to what I’ve said in this section. First, one may believe that Cohen’s worry about contractarianism failing to distinguish justice judges, legislators, bureaucrats—can do wrong.
17
benjamin sachs
contractarianism as political morality
draft paper
from other normative considerations still applies given my suggested version of contractarianism. For even that version of contractarianism treats the question of political morality as settled in one fell swoop by the contractarian thought experiment, while someone like Cohen might believe that political morality is a function of many values, each of which must be analyzed separately. Second, it may be argued that I’m still missing the point of Cohen’s fact-sensitivity critique, which is that all the foundational facts about what the state should do, including the foundational facts of political morality, are fact-insensitive, and therefore contractarianism can reveal none of them since its state of nature thought experiment is factsensitive. Third, it may seem obvious that the state is not bound by the constraints of interpersonal morality, since the state’s modus operandi is to use threats (of imprisonment, e.g.) to force people behave in certain ways—a sort of behavior that interpersonal morality would deem wrong for you and I but seems entirely appropriate for a state. Fourth, one might doubt that a political morality could fail to imply a theory of legitimacy, thereby undermining my claim that contractarianism could be a compelling political morality without being a compelling theory of legitimacy. And fifth, it may be thought that my defense of contractarianism is a Pyrrhic victory, since I’m allowing that a great deal of what the state morally ought to do is settled by something other than contractarianism. The first objection, again, is that contractarianism errs in treating political morality as settled lock, stock and barrel by the state of nature thought experiment as opposed to requiring a separate investigation of its (presumably) multiple elements. To understand whether this objection succeeds, I need to say more than I’ve yet said regarding the contractarian’s position on the grounds of political morality. I do so in what follows. I’ve suggested that contractarians should construe political morality as role morality—specifically, the morality of being a state. And obviously the role in question should be conceived of as socially constructed as opposed to natural. Having said this much, the contractarian can go on to point out that traditionally we think of the role morality associated with any socially constructed role as being grounded in the purpose of the role, which in turn exists in virtue of intentional human action. Consider the attorney role. Why is it a part of the role morality of an attorney to act in the best interests of her clients? Because attorneys have the purpose of playing the role of advocate in our adversarial justice system, and they wouldn’t be good advocates if they didn’t act in the best interests of their clients. And of course the adversarial justice system is an intentional human creation.
18
benjamin sachs
contractarianism as political morality
draft paper
So if political morality is indeed role morality, then in order to determine the contents of political morality we need to discern the purpose of the state, which in turn requires understanding what intentional human activity grounds its existence. This is exactly what a state of nature thought experiment is used, by the contractarian, to provide. The thought experiment is supposed to provide a rational reconstruction of the activity on the part of rational agents that leads to the existence of the state and gives it its purpose.29 From all this the contractarian will conclude that an investigation of political morality has to start with a state of nature thought experiment. But the contractarian need not go on to say either of the following two things: 1) That political morality does not contain distinct constituent parts—i.e. is not pluralistic; 2) that the state of nature thought experiment settles on its own the entire content of political morality. The contractarian can allow that political morality is pluralistic, even irreducibly pluralistic. But the way the contractarian can allow for that is by telling a plausible state of nature story on which the contractors want the state they’re creating to accomplish an irreducible plurality of things. So perhaps the contractors would be concerned that their well-being be augmented, and also concerned that their natural rights be protected, and perhaps neither concern would be reducible to the other.30 Furthermore, the contractarian can allow that her thought experiment does not settle in one fell swoop the entire contents of political morality. This is true for at least two reasons. First, the contractors might resolve that certain matters of political morality are to be settled by democratic procedures. And second, the contract might contain certain concepts that are left somewhat vague, in which case it would be up to us, as interpreters of the contract, to reason out its implications. The takeaway from all this is that the state of nature thought experiment is, for the contractarian, the starting point of political morality. For the contractarian, every truth of political morality is grounded somehow I dedicate a whole other essay to explaining how this works. See my “Teleological Contractarianism”, unpublished manuscript.
29
The contractarian might even argue that the contactors would care to see justice secured. It could be, in other words, that the contractors would want the state to take on the role of enforcing that part of interpersonal morality known as ‘justice’. The state would then be obligated twice over to secure justice. It would be obligated-qua-agent to do this, just as every other moral agent (we are assuming) is so obligated, and also obligated-qua-state to do the same thing. (Even Cohen would have no objection to using the state of nature thought experiment to establish that the state is obligated-qua-state to secure justice. What he objects to is using the state of nature thought experiment to establish the content of justice.)
30
19
benjamin sachs
contractarianism as political morality
draft paper
in an aspect of the thought experiment. But this grounding may be only partial, and so the contractarian need not and should not say that her thought experiment settles the entirety of political morality. Next, the fact-sensitivity objection. To evade this criticism, the contractarian should, I propose, make use of the idea advanced earlier of political morality as a role morality. What the contractarian should maintain, I suggest, is that contractarianism can reveal the principles that are foundational as far as political morality is concerned, while admitting that it cannot reveal any principles that are foundational strictly speaking. For the sake of illustration, consider again the practice of law—our earlier example of an area of activity that has its own role morality. Suppose someone were to propose that the principle requiring attorneys to act in the best interests of their clients is foundational within the role morality of practicing law. This is a reasonable proposal, as it’s not clear what normative principle for the attorney qua attorney could be more fundamental. This is not to say, however, that the proposed principle is foundational strictly speaking. Rather, it seems to draw support from the idea of the attorney as occupying a certain role within the justice system, namely the role of advocate, plus the principle that individuals should do the things that it is their role to do. Note that this latter principle is not a principle for the attorney qua attorney. Rather, it is perfectly general principle that applies to every moral agent. So the principle, ‘attorneys should act in the best interests of their clients’, is plausibly regarded as foundational within the role morality of the attorney but not foundational strictly speaking. Furthermore, it is fact-sensitive. It is based on the idea that the role of an attorney is to be an advocate, which itself is contingent on the various facts about human intentions and that make it the case that there is this practice known as the justice system and a role within that practice known as the attorney. The key to seeing why and how a contractarian can make use of a strategy like this for evading Cohen’s criticism is remembering that contractarianism, as we are now understanding it, coexists alongside some (non-contractarian) theory of interpersonal morality. So the contractarian can, and I propose should, say that the entire enterprise of political morality is an outgrowth of, and is ultimately grounded in, some principle of interpersonal morality. This allows contractarianism to be foundational in the most robust sense that anyone could want from a political morality having acknowledged the distinction between the state’s obligations-quaperson and its obligations-qua-state. It makes no sense to demand any kind of fundamentality from a political morality other than its containing the most fundamental truths regarding the state’s obligations-qua-state.
20
benjamin sachs
contractarianism as political morality
draft paper
Contractarianism can be foundational in this sense while being factsensitive. As I said earlier, my view is that a contractarian theory should begin with a state of nature thought experiment from which it is inferred that the state has such-and-such a purpose. From such a claim, plus the interpersonal moral principle ‘agents should do what it is their role or purpose to do’ and the supplemental claim that the state is an agent, the contractarian can infer certain obligations incumbent upon the state-quastate—that is, the contractarian can infer certain foundational principles of political morality. And all of this will be fact-sensitive because it will be based on a claim about the role of the state, which will itself be supported by the state of nature thought experiment, which includes facts about what the state of nature and the contractors are like. The third objection is that whereas interpersonal morality forbids getting one’s way through the use of threats, the state is permitted, indeed positively expected, to do just that, which shows that (contrary to what I have been arguing) the constraints of interpersonal morality do not apply to the state. By way of response, the first thing I want to point out is that it is not per se immoral, or even per se pro tanto immoral, to use threats to get what one wants. If a woman threatens to break up with her boyfriend unless he agrees to get married, she does no wrong. So there has to be something special about the state’s use of threats. Plausibly, what makes the state’s threats morally distinctive is that a lot of the time what is threatened is the use of force—throwing someone in prison, e.g. This is important, because while it isn’t generally wrong to issue threats, it is generally wrong to issues a threat such that the threatened action would itself be wrong.31 And of course it is wrong, usually, to use force against others. Thus we arrive at a more nuanced identification of the obligation of interpersonal morality from which states are (apparently) exempt: the obligation to not issue threats of immoral action. I agree that states are exempt from this moral obligation, at least if we’re talking about legitimate states insofar as they use their coercive power in the service of enforcing just laws. But I do not concede that this shows that interpersonal morality doesn’t apply to states. Consider an analogy: If I sign a mortgage agreement with a bank, that agreement is likely to contain a provision whereby the bank is licensed to forcibly remove me from my home if I stop making payments on that mortgage. Forcibly removing people from their home is, I assume we all agree, barred As I argue in “Why Coercion is Wrong when it’s Wrong”, Australasian Journal of Philosophy 91 (2013):63-82.
31
21
benjamin sachs
contractarianism as political morality
draft paper
by interpersonal morality, yet it is not wrong for the bank to remove me from my home if I stop making payments on my mortgage. But of course this doesn’t show that the bank isn’t constrained by interpersonal morality. Rather, what’s happened in this case is that the bank and I have each employed our moral powers to override some of the interpersonal obligations and permissions that would otherwise hold between us. Having done this, what can be said (at most) is that one element of interpersonal morality no longer applies to the bank. Similarly, we could say that because of some special relationship that holds between a legitimate state and its people, the part of morality that would ordinarily bar the state from using threats of imprisonment to get its people to obey its laws does not apply. This is the more parsimonious thing to say (though it does leave us needing some account of the special relationship just mentioned and how it acquires its moral overriding power). Conversely, if we go the other way—if we say that interpersonal morality does not apply to states—our job is considerably more challenging: we have to explain why the state is morally barred from lying, stealing, killing, etc. (as it surely is), without alluding to interpersonal morality. The fourth objection takes the form of the following argument: A political morality tells us some of the facts about what the state is obligated to do, and presumably the state is permitted to do everything it is obligated to do. Therefore, a political morality tells us, indirectly, something about what the state is permitted to do. Furthermore, since (presumably) the people are obligated to allow the state to do anything that it is permitted to do, a political morality thereby tells us (indirectly, again) something about the people’s obligation to obey the state. And this is the job of a theory of legitimacy, by my earlier definition of ‘legitimacy’. So it seems a political morality must do double duty as at least a partial theory of legitimacy. This argument is problematic for me, as I have been proposing that contractarianism is compelling as a political morality while insisting that it is not compelling as a theory of legitimacy. The problem with this argument is its first inference. I don’t accept that, in general, agents are permitted to do whatever they’re obligated to do.32 Suppose, however, that that inference is valid. Still, I can construct I have an argument for this; see my Explaining Right and Wrong (unpublished book manuscript), Ch. 3. Moreover, if there are cases in which someone is obligated-but-notpermitted to do something, we would expect to find those cases wherever we find role moralities (which is of course what I’ve said contractarianism plausibly is—it’s the true role morality of the state). This is because role moralities always exist against the backdrop of interpersonal morality, and the presence of two moralities creates the possibility of conflict. Consider, for instance, how the role morality of being an attorney obligated the attorney to act in the best interest of her client. In some cases this seems to require
32
22
benjamin sachs
contractarianism as political morality
draft paper
an adequate defense of my position. The most plausible reason why the first inference would be valid—that is, why it would be the case that role morality obligates the state to do only things that interpersonal morality permits the state to do—is that interpersonal morality constrains role morality: in other words, we cannot obligate an agent qua role-bearer to do anything that she is prohibited qua agent from doing. If this were the case, then from the fact that a political morality obligates the state to φ we could infer that interpersonal morality permits the state to φ and therefore the correct theory of legitimacy requires the people to allow the state’s φ-ing. But in this scenario the correct political morality, whatever it is, still wouldn’t ground any of the facts as to what the people are obligated to allow the state to do. And surely a theory doesn’t count as a theory of legitimacy if it doesn’t tell us how such facts are grounded. So if the first inference is valid then contractarianism could nevertheless be the true political morality while being a false theory of legitimacy. The fifth objection is that the viability of the form of contractarianism I’m describing here would amount to no more than a Pyrrhic victory for contractarianism, since it has so many of the critical moral issues vis à vis the state being settled by something other than the contractarian theory. But what this objection describes as a weakness of my version of contractarianism I consider a virtue. This goes back to what I said in the introduction about the historical development of contractarianism. Contractarianism burst on to the philosophical scene at a moment when everything of moral importance seemed up for grabs. But this is no longer the world in which we live. Much has been settled, and the contractarian— and indeed any political theorist, for that matter—should be happy to take on board anything that has found a solid philosophical footing. This is the major payoff of building a version of contractarianism that leaves room for interpersonal morality. The contractarian can take on board the widely accepted moral-equality-of-persons thesis and use it to explain certain constraints on the contracting situation (e.g. that the contract must be unanimously agreed to). Similarly, the contractarian can make use of human rights theory and say that no matter what the outcome of the contracting situation, the state is morally prohibited from throwing people in prison on account of their political beliefs. The contractarian can, in addition, simply bring on board an independently developed theory of legitimacy, such as the idea that the individual is obligated to obey the law just when, and because, it has come into force through a democratic prothe attorney to seek the acquittal of someone she knows is guilty. And, plausibly, interpersonal morality prohibits all of us qua agents from knowingly acting so as to make it less likely that substantive justice is done. This would be a case in which the attorney is obligated-qua-attorney to do something that she is prohibited-qua-agent from doing.
23
benjamin sachs
contractarianism as political morality
draft paper
cess that meets certain requirements of representativeness, openness, etc. Even granting all this, there are still plenty of questions to be answered regarding what the state morally ought to do. Some of them, I contend, should be answered by the contractarian theory, though not all of them; there is still more whittling to be done before we find contractarianism’s subject matter. For presumably one outcome of the state of nature thought experiment would be a decision in favor of democracy, and democracy of course means leaving some decisions in the hands of the people through their representatives. In other words, believing in democracy means believing that many questions of the form, “What morally ought the state to do about X” have the following answer: “The will of the people morally ought to determine how the state acts with respect to X”. Thus we have found yet more moral questions about the state that are not to be answered (except indirectly) by the contractarian theory, even if contractarianism is correct. However, there are many questions about what the state morally ought to do that aren’t settled by interpersonal morality and that democrats themselves would concede are not to be settled by the will of the people—I have in mind, roughly, what are often called constitutional matters. One of them is the matter of which questions are to be settled by the will of the people. Should we plump for a separation of powers in which a constitution and a court with the power of judicial review limit the will of the people, exercised through the legislature? This question has to be answered before the legislature itself is set up. Another such question is how to structure the system of voting and districting. Another two examples: (1) many people believe it is the job of the state to fund the arts and scientific inquiry, and (2) many people believe that it is the job of the state to enforce interpersonal morality per se. But, in each case, others disagree. This is not an issue to be resolved by appealing to the democratic will of the people, since what’s at issue is whether, and if so when, the state should ever refuse to engage in some activity that its people want it to engage in. And despite the fact that the state is constrained by interpersonal morality, these questions cannot be settled by identifying the true interpersonal morality. These disputes seem to be distinctly about the state, and as such it takes a political morality to answer them. Contractarianism is well suited to answer them, not just for the obvious reason that contractarianism is a political morality, but furthermore because contractarianism begins by asking what the role of the state is, which seems like an issue that it would be helpful to have a grip on if we’re to determine whether the state should be funding the arts and scientific inquiry and enforcing interpersonal morality. Furthermore,
24
benjamin sachs
contractarianism as political morality
draft paper
these are very important issues—issues of life and death, in some cases. So even conceding all I have here, there is still plenty of importance for contractarianism to accomplish, even if not quite the most important things or the things that historically contractarianism has been taken as providing guidance on.
25
the aristotelian society
president: Susan James (Birkbeck) president-elect: Tim Crane (Cambridge) honorary director: Rory Madden (UCL) editor: Matthew Soteriou (Warwick) lines of thought series editor: Scott Sturgeon (Oxford) executive committee: Corine Besson (Sussex) / Kimberley Brownlee (Warwick) Rowan Cruft (Stirling) / Alison Hills (Oxford) / Adrian Moore (Oxford) / Samir Okasha (Bristol) / Robert Stern (Sheffield) managing editor: Hannah Carnegy (UCL) assistant editor: David Harris graphic & web designer: Mark Cortes Favis administrator: Josephine Salverda (UCL)
w w w. a r i s t o t e l i a n s o c i e t y. o r g . u k