1. Jean-Jacques Rousseau, Discourse on the Origin and Foundations of Inequality among Men, in The “Discourses” and Other Early Political Writings (Victor Gourevitch trans., 1997), at 167.

2. Id. Cf. Rousseau's characterization in the Emile of the independent individual as one who “has no need to put another's arms at the end of his own”; Rousseau, Jean-Jacques, Emile (Bloom, Allan trans., 1979), at 84Google Scholar.

3. See Immanuel Kant, Doctrine of Right, in Practical Philosophy 421–426 (Mary J. Gregor trans., 1996). For a helpful recent discussion of Kant's views on contract, see Arthur Ripstein, Force and Freedom: Kant's Legal and Political Philosophy (2009), ch. 5.

4. For the first kind of view, see, e.g., Charles Fried, Contract as Promise (1981); and the references in note 16infra.

5. Note that for present purposes I simply follow Rousseau in adopting that conception of freedom. For Rousseau's statement of his view, see Jean-Jacques Rousseau, 8 Lettres écrites de la montagne, in 3 Œuvres complètes (Bernard Gagnebin et al. ed., 1964), at 841. Kant holds a similar view; see Kant, supra note 3, at 393. For a modern view of freedom that also stresses the importance of independence but differs in important respects from those of Rousseau and Kant, see Philip Pettit, Republicanism: A Theory of Freedom and Government (1997). I discuss the conception of freedom as independence and the relation between the views of Kant and Pettit in Hodgson, Louis-Philippe, Kant on the Right to Freedom: A Defense, 120 Ethics 791–819 (2010)Google Scholar.

6. As I explain below, the problem here is not that others may not want to cooperate with me. That is just a consequence of our mutual independence and not a threat to my freedom as such. The problem arises when others agree to engage in collective action with me but remain free to do their part or not.

7. For examples of discussions aimed primarily at solving this problem, see Michael Bratman, Faces of Intention: Selected Essays on Intention and Agency (1999), chs. 5–8; J. David Velleman, How to Share an Intention, in The Possibility of Practical Reason 200–220 (2000). Even Margaret Gilbert seems to fit the description, despite her claim to be a holist rather than an individualist, since she insists that her theory “does not invoke any ontologically suspect kind of ‘social spirit’ or ‘group mind’”; Margaret Gilbert, Sociality and Responsibility: New Essays in Plural Subject Theory (2000), at 3.

8. To clarify, the thought here is not that for any given X, if there are countless things that I can accomplish only with the help of X, then X is essential to my freedom. As a referee for this journal pointed out, substituting “flying cars” for X plainly shows that this is an invalid form of reasoning. My claim is not that freedom demands that I have access to any specific action or object but rather that freedom demands that it be possible rightfully to acquire actions or objects, should any be available. The fact that I lack a flying car does not undermine my freedom, but the systematic impossibility of acquiring external objects would; likewise, the fact that you refuse to help me does not undermine my freedom, but the systematic impossibility of making the actions of others into my means would. What is at stake here, in other words, is the general possibility of making another's action into my means. So long as this possibility is compatible with the other's freedom (on which more below), freedom demands that it be unhindered. I am grateful to the referee for prompting me to be more explicit on this point. On the link between property rights and freedom, see Jeremy Waldron, Homelessness and the Issue of Freedom, in Liberal Rights: Collected Papers 1981–1991 309–338 (1993); and Louis-Philippe Hodgson, Kant on Property Rights and the State, 15 Kantian Rev. 57–87 (2010), at 58–63.

9. The assumption can be relaxed, as I explain at the end of the section.

10. To fix ideas, we can say that Larry and I share an intention in the sense that Michael Bratman has given to the expression. With some minor simplifications, this means that (1) each of us intends that we build cars together; (2) each of us intends that we do so in accordance with this common intention and by way of what Bratman calls “meshing subplans”; and (3) all of this is common knowledge between us; see Bratman, supra note 7, at 121. The notion invoked in the second clause is simple, if unusual: basically, we have meshing subplans to paint the house together if the subplans we have each formed to execute the project are compatible; see id. at 99 and 120. I invoke Bratman's view here partly because it is likely to be familiar to the reader but also because he explicity says that sharing an intention does not by itself create an obligation; forming a shared intention may normally coincide with the creation of an obligation (grounded in something like T.M. Scanlon's principle of fidelity), but that is not essential to shared intention—it is quite possible to rule out such obligations, which is what I want to do for now. For Bratman's view on this point, see id., ch. 7; for Scanlon's discussion of the principle of fidelity, see T.M. Scanlon, What We Owe to Each Other (1998), at 304.

11. For an eloquent defense of these and related ideas, see Pettit, supra note 5.

12. Ripstein stresses this point about Kant's conception of freedom in Ripstein, Arthur, Authority and Coercion, 32 Phil. & Pub. Aff. 2–35 (2004), at 8–11Google Scholar, and in Ripstein, supra note 3, ch. 2.

13. The clause “once he has agreed to help me” is meant to acknowledge that in some cases my depending on Larry's choices may be inevitable—for instance, if he is the only person who can help me and yet refuses to commit to doing so. In such a case, however, my depending on his choices is not contrary to freedom, since Larry's own freedom obviously demands that it be up to him to decide whether to help me or not. The problem arises once he has agreed to help me—once he has made his help available to me as a means—because there is then no reason for the decision to remain up to him.

14. See supra note 10 for references on how obligations may be thought to arise in the context of collective action.

15. It has been suggested to me that the moral obligation of others not to interfere with my choices provides all the independence to which I am entitled and that my status as an independent agent is not affected by whether they act on their obligation or not. But that cannot be right; if it were, a slave would have to count as independent in the relevant sense, since his master is undeniably morally obligated not to interfere with his choices. What the objection overlooks is that a person's status as an independent agent demands a certain kind of legal entrenchment—it demands that the person be granted what I call elsewhere full legal standing. Whatever his moral status might be, a slave is unfree because he lacks appropriate legal standing. As I explain below, the same thought holds (on a much smaller scale) for my collaboration with Larry: acting with him poses a threat to my freedom beause I lack legal standing with respect to his action. On the importance of the idea of full legal standing for freedom, see Hodgson, Kant on the Right, supra note 5, at 812–817.

16. I am assuming here that our having promised to do our share, without any explicit agreement about the use of force, cannot suffice to make the use of force consistent with our freedom. As Joseph Raz points out, to assume that promises can be enforced simply in virtue of the moral obligations they create is to endorse a discredited kind of legal moralism; see Raz, Joseph, Promises in Morality and Law, 95 Harv. L. Rev. 916–938 (1982)CrossRefGoogle Scholar. Moreover, the grounds in virtue of which promises are sometimes thought to be enforceable do not meet the demand of legitimacy. This is true of Raz's idea that the practice of promising should be enforced because of the social benefits it affords and also of Stephen Smith's suggestion that promises are enforceable because “promise making and promise keeping are constitutive elements of a close relationship”; Stephen A. Smith, Contract Theory (2004), at 74.

17. A referee for this journal has suggested to me that the point is convincing only if one assumes that each boxer has given the other revocable permission to use force against him. If one thinks that each has transferred a right to the other, as in a sale, then there is no reason to think that he can just change his mind about that. My view is that a person's right not to have force used against her body, unlike her right to a given external object, cannot be transferred away like that. The boxing example is meant to provide intuitive support for this view, but I concede that the matter would require further discussion. I am grateful to the referee for flagging the issue.

18. I take the idea of a contract right here in the broad sense of a right to another person's action. I do not attempt to account for all features commonly associated with contracts, although I do say more in the next section about one feature that strikes me as particularly important in this context and comment briefly on another feature in note 23infra.

19. Kant, supra note 3, at 388. Kant seems to hold that this is the only way in which coercion can be justified, although for present purposes we can remain agnostic as to whether that is correct; all we need is the idea that this is one way to reconcile the use of force with freedom.

20. Smith thus describes the requirement of freedom to contract as “arguably the most basic requirement for establishing a contractual obligation”; Smith, supra note 16, at 59.

21. I say “in the relevant sense” to acknowledge that the decision does not have to be truly B's in any deeper sense—that is, it does not have to spring from B's deep rational nature or anything to that effect. The point is that the decision must be B's as opposed to someone else's.

22. Contract law does not require that a person intend to be legally bound for a contractual obligation to arise; see Restatement (Second) of Contracts §21 (1979). For an illuminating discussion of the role of consent and of the idea of voluntariness in contract, see T.M. Scanlon, Promises and Contracts, in The Difficulty of Tolerance 234–269 (2003).

23. The condition I outline in the text could be seen as underlying a puzzling feature of Anglo-American legal systems: the notorious doctrine of consideration, which states that a promise is not enforceable unless something has been given in exchange for it; see Restatement (Second) of Contracts §§17(1) and 71 (1979). Nothing I say here entails that a system of contract rights must include such a requirement, but the doctrine of consideration can be seen as one way (among others) of articulating a requirement of control: it ensures that one can avoid becoming contractually bound if one so wishes. That said, there is nothing wrong with a system in which gratuitous promises are deemed enforceable, so long as the parties have appropriate control over how such promises modify their legal situations. The argument of Section III suggests a different way to account for the consideration doctrine. I said at the end of that section that your help need not be essential to my project for the problem of dependence to arise but that it has to be material. One might view the requirement of consideration as providing a rough measure of whether that is the case. The thought would be that if I gave you something in return for your promise, then I must take your action to be material to my project in some way. There are undoubtedly other ways to go about this, but such a measure does have its advantages—for one thing, it does not require courts to make substantive judgments as to what constitutes material help. For a more ambitious (and, in my view, misguided) argument claiming that the doctrine of consideration is necessary to ensure that contractual obligations are suitably reciprocal, see Benson, Peter, The Unity of Contract Law, in The Theory of Contract Law: New Essays 153–184 (Benson, Peter ed., 2001)CrossRefGoogle Scholar. For an argument that the doctrine is both internally inconsistent and unfair in many of its applications, see Fried, supra note 4, ch. 3.

24. For a more thorough discussion of the tension between the private enforcement of rights and freedom, with special attention to the case of property rights, see Hodgson, Kant on Property Rights, supra note 8, at 66–78. The additional complication in the present case is that we are assuming that the two parties have antecedently entered into an agreement about the use of force, which is not normally the case for property rights. As I argue in the text, however, this ends up making no difference.

25. See id. at 76–78.

26. The point I am making here can be seen as one part of the Kantian argument for the state. I discuss that argument at greater length in id.; see especially the remarks, id. at 79, on the extent of the Kantian state's legitimate power.

27. Note that the point does not have the same force in the case of corporations. Given the kind of agent they are, compelling specific performance does not pose the same threat to their freedom as it does for individuals and is consequently easier to justify. The idea that compelling specific performance “poses a special risk to personal liberty” is noted in Smith, supra note 16, at 155; see also id. at 400–402, where he notes that the point does not hold for corporations.

28. See Fuller, Lon and Perdue, William, The Reliance Interest in Contract Damages, 46 Yale L.J. 52–96 and 373–420 (1936), at 52–66CrossRefGoogle Scholar. Although the article was coauthored, the theoretical framework it presents is universally attributed to Fuller.

29. On Fuller's view, the award of expectation damages is justified only because these are easier to ascertain than reliance damages and because reliance and expectation damages tend to converge in a market economy. On the award of expectation damages as the standard remedy, see Restatement (Second) of Contracts §347 (1979).

30. The specification “as far as pursuing her end is concerned” is important. The question is not what will make the wronged party feel as happy or as satisfied as the defaulting party's performance would have. In this respect, the standard is strictly objective.

31. In other words, expectation damages should be awarded even for wholly executory contracts because a party's entitlement to damages does not depend on any reliance interest but rather on the fact that the aim of contract rights is to make it possible for one person to have the action of another as her means. Through a contract, an agent acquires an entitlement to another's performance and hence to adequate compensation should the other fail to perform. On the importance for contract theory of the idea that a party is entitled to expectation damages for the breach of a wholly executory contract, see Benson, supra note 23, at 119–120.