20th WCP: Hypothetical Consent and Political Legitimacy

Hypothetical Consent and Justification

A commonly accepted criticism of the social contract approach to justifying political authority targets the idea of hypothetical consent. Since only actual agreements are binding, the argument goes, citizens are not bound to obey their governments on the ground that, under circumstances different from the ones in which they now find themselves, they would have agreed to submit to its authority. (1) The purpose of this paper is to rescue hypothetical consent from this objection. I begin by distinguishing political legitimacy from political obligation. (2) I argue that while hypothetical consent may not serve as an adequate ground for political obligation, it is capable of grounding political legitimacy.

I understand a theory of political legitimacy to give an account of the justice of political arrangements. (3) I understand a theory of political obligation to give an account of why and under what conditions, citizens are morally required to obey the rules constituting those arrangements. The social contract tradition offers us hypothetical consent theories of both political obligation and political legitimacy, frequently neglecting to distinguish the two ideas. Likewise, the common objection to hypothetical consent theories — that hypothetical contracts do not bind — is applied without sufficient attention to this distinction.

Consider Ronald Dworkin's account of this objection. He says, "hypothetical contracts do not supply an independent argument for the fairness of enforcing their terms. A hypothetical contract is not simply a pale form of an actual contract; it is no contract at all." (4) He draws an analogy with a game. While it may be true, he says, that one would have agreed to play by various rules had he been asked before the game began, it is not fair to enforce those rules against him if he had not actually agreed to them. Suppose you and I are about to play a game of Hearts and we agree that no hearts can be thrown on the first trick. If I then insist upon laying a heart on the first trick, I am acting unfairly. Suppose, however, we had not agreed that no hearts can be thrown on the first trick. If you were to insist, once the game began, that I cannot throw a heart on the first trick because I would have agreed before the game began that no hearts should be thrown on the first trick, then you are acting unfairly.

Now the fact that I would have agreed that no hearts should be thrown on the first trick, while it does not obligate me to refrain from laying a heart on the first trick, reveals something important about that practice. Assuming I am rational and have certain aims, the fact that I would have agreed to follow that rule shows that it is a reasonable rule. (5) In this situation my hypothetical consent demonstrates the reasonableness of the no-hearts-on-the-first-trick rule. If we maintain that fair rules are those that are reasonable, or, are those that card-players have reason to accept, then the fact that I have reason to accept the no-hearts-on-the-first-trick rule shows that it is a fair rule.

So, while hypothetical consent may not be sufficient to obligate a person to follow a rule, it can, under certain circumstances, establish the legitimacy of the rule. If we understand the hypothetical consent feature of social contract theories as creating legitimacy and not obligation, then we can see that the fact that hypothetical contracts are not binding does not pose a problem for those views. The fact that I would have agreed, in certain conditions, to submit myself to some political arrangement shows that that arrangement is reasonable and therefore legitimate. Further argument may be necessary to show that I ought to submit myself to that arrangement when the circumstances under which I would have agreed no longer obtain. (6)

But perhaps one should challenge the idea that in showing that persons have good reasons to accept a policy, one has shown that that policy is therefore legitimate. Why is a political system just simply because those who are subject to it have reason to abide by its rules? Perhaps it should be considered just on the ground that it produces the greatest overall happiness, or is an expression of God's will, or conforms to a hierarchy of authority discoverable in nature. Underlying hypothetical consent theories is the idea that, in Arthur Ripstein's words, "it is illegitimate to coerce people for the sake of other people's desires about what they should do." (7) It is legitimate to coerce people — that's what political authority does — but only if the coercion promotes citizens' legitimate aims as they define them. (8) Underlying this idea is the notion that all persons are equal in moral standing. If all are equal in moral standing, then no one person, or group, is entitled to determine how others define and pursue their ends.

So, hypothetical consent theories of political legitimacy presuppose a cluster of substantive moral principles. (9) They presuppose that persons are equal in moral status and hence, in Kant's terminology, cannot permissibly be used merely as means to another's ends. (10) Their own ends must be acknowledged as giving them a legitimate claim against those who would exercise power over them. Hypothetical consent theories, then, serve to justify political arrangements only if we are prepared to accept these precepts. If we agree that people should be subject only to policies that they have reason to accept, then hypothetical consent theories, insofar as they establish which policies people have reason to accept, are capable of determining which policies are just.

The Agent-Dependent Nature of Reasonableness

If the real justification for a policy is that it is reasonable, one might ask, why do we need the notion of hypothetical consent? What does the notion of hypothetical consent add to the justification of political arrangements if the ultimate justifying ground is that the arrangements are reasonable? Why not simply provide an argument for the reasonableness, and hence justice, of the arrangements and forget about the idea of hypothetical consent? This objection gets the relation between reasonableness and hypothetical consent in social contract theories backwards. It supposes that something is worthy of consent because it is reasonable. But, in hypothetical consent theories, something is reasonable because it is worthy of consent.

Consider again the card game analogy. The point of the prohibition on throwing a heart on the first trick of a hand in Hearts is that it ensures that once the cards have been dealt, players' prospects will be determined largely by skill and not luck. In particular, it ensures that one will not, from the luck of the deal, lose the hand on the first trick. Therefore, it makes the game more interesting and more challenging. To say that the prohibition on throwing a heart on the first trick is reasonable is equivalent to saying that card-players who want the game to be interesting and challenging would adopt such a rule. The reasonableness of the rule consists in nothing more than this fact. This is not to say that every actual card-player would agree to this rule, but that the ideal card-player would. Reasonableness is defined by what people with certain desires, motivations and ends have reason to do, or, in other words, what they would consent to. Hence one could not argue directly — that is, without appeal to hypothetical consent — for the reasonableness of the rule against throwing a heart on the first trick. Such an approach would mistakenly suppose that a criterion of reasonableness, and hence fairness, for card game rules exists independent of what (idealized) card-players would choose.

Because it appears to do no justificatory work of its own in social contract theories, hypothetical consent is sometimes referred to as a "metaphorical device". (11) The notion of hypothetical consent, on this view, simply highlights or reveals what is reasonable and therefore just. As Dworkin states,

...you use the device of hypothetical agreement to make a point that might have been made without that device, which is that the solution recommended is so obviously fair and sensible that only someone with an immediate contrary interest could disagree. Your main argument is that your solution is fair and sensible, and the fact that I would have chosen it myself adds nothing of substance to that argument. If I am able to meet the main argument nothing remains, rising out of your claim that I would have agreed, to be answered or excused. (12)

Underlying this assertion is the idea that one could offer an argument for the reasonableness of political arrangements without appealing to what agents, endowed with specific aims and motives, have reason to accept. But there is no such agent-independent criterion of reasonableness, according to hypothetical contract theories. There is no independent order of value, on these accounts, to which one could appeal to establish the reasonableness, and hence justice, of a political system. So while it is true that the notion of hypothetical consent adds nothing to an argument for the reasonableness, and justice, of a political system, it is certainly not an extraneous apparatus that has nothing more than a metaphorical function. If what is reasonable just is what individuals with certain interests and motives would agree to, then the fact that a person, with certain interests and motives, would have agreed to a political system is the "main argument" for the justice of that political system.

It is important to distinguish the claim that hypothetical consent is a metaphorical notion from the claim that a hypothetical contract is a metaphorical notion, or in Rawls' terms, a "device of representation". (13) A hypothetical contract is indeed a metaphorical notion and Dworkin is right to say that it is no contract at all. (14) But it does not follow from the fact that hypothetical contracts are not real that hypothetical consent can do no justificatory work in arguments for certain kinds of political arrangements. We can see this if we examine, using the game analogy, the ways in which the logic of hypothetical consent differs from the logic of actual consent.

If you and I agree to follow a certain rule before the game begins, then we are bound to follow that rule. Our actual consent creates an actual verbal contract. By contrast, if you point out to me during play, that you and I would have agreed to a certain rule, then you are demonstrating the reasonableness of that rule. I may not be bound to follow that rule, though I might elect to if I am motivated to follow rules that are reasonable. Here our hypothetical consent is not anything we do. (15) It refers to a counterfactual claim about us made by you who is attempting to justify to me. Our hypothetical consent does not create a hypothetical contract that is somehow analogous to the actual contract created by actual consent. Nonetheless, the counterfactual claim that you and I would have agreed to follow a certain rule, if consulted, can be metaphorically represented as a hypothetical contract. But the fact that hypothetical consent does not create a contract — or, alternatively, that it creates a hypothetical contract that is merely metaphorical — does not entail that one has shown nothing of substance by establishing that someone would have made a certain promise.

One might grant that no agent-independent account of reasonableness exists and yet object to the idea that reasonableness is constituted by what agents, endowed with certain aims and motivations, would agree to. (16) Consider again the prohibition on laying a heart on the first trick in hand of Hearts. I maintained that the reasonableness of this rule is given by the fact that ideal card-players (i.e., those who wish the game to be interesting and challenging) would choose it. But they would choose it, of course, because it is a means to their ends. They want the game to be primarily skill-based, the no-hearts-on-the-first-trick rule helps ensure this, so, assuming they are rational, they would choose this rule. It seems that the reasonableness of the rule, then, consists in its being a means to certain ends. But if this is the case then the fact that ideal card-players would choose this rule is superfluous to the reasonableness of the rule, and it makes sense to say that they would choose the rule because it is reasonable. And if the rule is fair because it is reasonable, then it makes sense to say that they would choose it because it is fair.

My response to this objection depends on the distinction between a rule's being rational and its being reasonable. (17) Let us say that a rule is rational if it is a means to an agent's ends. The no-hearts-on-the-first-trick-rule is rational insofar as it is a means to ensuring that the game will be primarily skill-based once the cards have been dealt. But this rule is reasonable only if the end of creating a primarily skill-based game is itself reasonable. So let us say that a rule is reasonable if it is rational and its objectives are reasonable.

It follows from these remarks that the reasonableness of a rule, as per the objection above, cannot be given simply by its being a means to agents' ends. But we might say that the reasonableness of a rule is given by its being rational and its having reasonable objectives (where, of course, the notion of reasonable objectives is fleshed out.) It appears that the objection, modified in this way, stands since the fact that idealized agents would choose a certain rule does not, on this account, constitute its reasonableness. Once we flesh out the notion of reasonable objectives, however, we will see that the claim that a rule would be agreed to by idealized agents is equivalent to the claim that the rule is rational and has reasonable aims. Hence the fact that persons, characterized in a particular way, would agree to a rule establishes that that rule is reasonable and therefore just.

What authorizes the reasonableness of certain aims and objectives? What makes the aim of making a card game interesting and challenging a reasonable aim? Assuming there are no God-given objectives for card games, nor any objectives given by nature or by a mind-independent order of value, the answer is that reasonable objectives for card games are what good card-players would want the objectives to be. Since reasonable objectives for card games are those that ideal players would want, reasonable rules are those that achieve those objectives. Hence to say that reasonable rules for cards are those that are rational and have reasonable objectives is to say that reasonable rules for cards are those that ideal card-players would choose. And if fair rules are reasonable rules, then fair rules are those that ideal card-players would choose.

Naturally what counts as an ideal card-player is contestable. One might challenge the notion that ideal card-players wish the game to test their skills. Perhaps ideal card-players are better characterized as wanting winning to be based mostly on chance. As others have maintained, in particular with respect to Rawls, the heart of the issue in social contract approaches to political legitimacy concerns the characterization of the parties imagined to be choosing the rules that are thereby legitimated by their being chosen by those parties. Rules will be justified only if the aims and motivations of the choosing parties are reasonable. If, for example, we conceive of the parties as self-interested, acquisitive, glory-seeking adults who are roughly equal in ability, then the political arrangements they would choose will be justified to us. If we regard this description of the parties as unreasonable, then the arrangements they would choose will not be justified to us.

The fact that the description of the choosers contains normative elements should not provoke worries about the justificatory power of hypothetical consent theories, unless one is already worried about the moral presuppositions of hypothetical consent theories identified above. The normative features of the parties choosing the rules, along with whatever normative features are included in the description of the conditions of choice, largely embody those moral presuppositions. The choosers and their circumstances are designed, in hypothetical consent theories, so that they "model" (to use Rawls' term) the moral equality of persons and the sanctity of individuals' legitimate ends. Disagreement over the proper way to characterize choosers and their circumstances is rooted in disputes about what best models these principles. For example, social contract theorists generally subscribe to the notion that social cooperation should be for mutual advantage. (This principle is derived from the foundational ideal of moral equality.) Theorists disagree, however, on how this notion should be interpreted. (18) Their disagreement is reflected in their characterization of the parties in the imagined circumstances of choice. If the norms of cooperation are conceived of as a rational compromise necessary for the pursuit of one's non-moral interests, then it makes sense to characterize the hypothetical choosers as equipped with the capacity for instrumental rationality and as motivated exclusively by self-interest. If social cooperation is conceived of as a prerequisite for individuals' exercising their moral agency, then it makes sense to characterize the hypothetical choosers as having an interest in exercising that capacity and as representing agents who are motivated by reasons that do not refer directly to their non-moral interests.

Summary

I have argued that while hypothetical consent may not be capable of establishing political obligation, it has the power to justify political arrangements, if one accepts two principles. One principle is ethical, the other metaethical. The ethical principle states that persons are equal in moral worth and are, consequently, not justified in coercing one another for the sake of their own ends. The metaethical principle denies the existence of an independent order of value, given by God or by nature, to which we could appeal to determine what is just. If there are just political principles, they are the principles that persons have reason to accept.

Notes

(1) Ronald Dworkin, "The Original Position," Reading Rawls, Norman Daniels, ed. (New York: Basic Books, 1975), p. 17-21.

(2) This distinction is discussed by Jeremy Waldron in "Theoretical Foundations of Liberalism," Liberal Rights: Collected Papers 1981-1991 (Cambridge: Cambridge University Press, 1993): 45-50. My way of drawing the distinction differs from his.

(3) "Political legitimacy", in this context, is a normative concept and not to be confused with the descriptive notion of "political legitimacy" that political scientists sometimes use to indicate that a political system us regarded as legitimate by its citizens.

(4) Ronald Dworkin, "The Original Position," in Norman Daniels, ed. Reading Rawls (New York: Basic Books, 1975), p. 17.

(5) Ibid.

(6) See Ronald Milo, "Contractarian Constructivism," The Journal of Philosophy 92, 4 (April 1995): 181-205, p. 185.

(7) Arthur Ripstein, "Foundationalism in Political Theory," Philosophy & Public Affairs 16 (Spring 1987): 115-37, p. 133.

(8) Hobbes, for example, held that political authority is justified by the fact that people's interest in self-preservation is best served under a sovereign's reign. Rawls, by contrast, holds that people's moral interests in pursuing a conception of the good and exercising their capacity for justice can be met only in political society. For a discussion of the differences between Hobbesian and Rawlsian contract theories see Samuel Freeman, "Reason and Agreement in Social Contract Views," Philosophy & Public Affairs 19 (Spring 1990): 122-57. See also, Thomas Nagel, "Moral Conflict and Political Legitimacy," Philosophy & Public Affairs 16 (Summer 1987): 215-40.

(9) For an argument that Hobbes's contract theory, despite its appearance of avoiding any moral presuppositions, relies on the notion that persons should be free to pursue their own conceptions of the good, see Ripstein.

(10) Immanuel Kant, Groundwork of the Metaphysics of Morals, trans. H.J. Paton (New York: Harper Torchbooks, 1964), p. 96.

(11)Thomas Scanlon, "Nozick on Rights, Liberty and Property," Philosophy & Public Affairs 6 (Fall 1976): 3-25, p. 17. See also Dworkin, p. 37.

(12) Dworkin, p. 18.

(13) John Rawls, Political Liberalism (New York: Columbia University Press, 1993), p. 24. He refers to the "original position" as a "device of representation".

(14) For an argument that no contractual agreement on the two principles of justice occurs in the original position and that therefore the two principles are not justified by a contract, see Jean Hampton, "Contracts and Choices: Does Rawls Have a Social Contract Theory?" The Journal of Philosophy 77, 6 (June 1980): 315-38.

(15) As Jeremy Waldron maintains, "When we move from asking what people actually accept to asking what they would accept under certain conditions, we shift our emphasis away from the will and focus on the reasons that people might have for exercising their will in one way rather than another." Waldron, p. 55.

(16) This objection is due to Bruce Landesman.

(17) For a discussion of this distinction and the relation between the reasonable and the rational, see Rawls, PL, pp. 48-54.

(18) See Freeman, pp. 123-31.