II. Libertarianism: Left, Right, and Far Right

One thing is clear: all libertarians as such are committed, at the very least, to the thesis that persons own themselves. What sets different camps of libertarians apart from each other has to do with how self-ownership rights license property in extrapersonal goods. We can state these differences as differences with respect to the so-called “Lockean proviso” on ownership—that one must leave “enough and as good” for others, in some sense—which on its broadest interpretation is a constraint on how much control of the extrapersonal world self-owners can rightfully claim. More specifically, I shall understand the proviso as, broadly, a “fair share” proviso.Footnote ⁵ On such a broad interpretation, the Lockean proviso can be seen either as a restriction on initial appropriation (as in Locke himself) and thus on what one can own, or as a restriction on the use of what one owns, or both (as in Robert Nozick's theory). Put another way, the proviso can either limit how much one can own or it can limit how extensive one's rights of ownership in a thing can be. (As we shall see in Section VI, my own concern is with the proviso understood as one on use.) In any case, the proviso rules out practices that would exclude others from a stake in the extrapersonal world, either by appropriating too much of that world or by using ownership claims to exclude others from it. The main differences between libertarians concern (1) whether the Lockean proviso has egalitarian implications and (2) whether there really is any place for the Lockean proviso after all.

One form of libertarianism, which I shall call simply “right-libertarianism,” holds that ownership of extrapersonal property (either its acquisition or its use) is subject to a Lockean proviso, but that the proviso has no egalitarian implications. Nozick is one example of such a right-libertarian, as is Eric Mack, who confines the proviso to use rather than acquisition. Right-libertarians are set apart from those even farther on the right—“no-proviso libertarians”—who deny that there is any Lockean proviso on either acquisition or use. Some representative no-proviso libertarians are Jan Narveson, Douglas Rasmussen and Douglas Den Uyl, and Murray Rothbard. Finally, “left-libertarians” hold that there is a Lockean proviso on the acquisition of property, and that this proviso, properly understood, has important egalitarian implications. Here one thinks of philosophers like Michael Otsuka, Hillel Steiner, Peter Vallentyne, and Philippe Van Parijs.

In what follows, I argue that both no-proviso libertarianism and left-libertarianism must reject the embodied conception of the self, albeit for different reasons. In particular, since no-proviso libertarians claim that the poverty of have-nots leaves their self-ownership untouched, they must hold that control of one's person is a separate matter from control of goods outside one's person, and thus they cannot countenance an embodied conception of the self in the case of have-nots. So too, left-libertarians must reject such a conception of the self in order to argue in the case of the haves that self-ownership is untouched despite the demand on them to make compensation for controlling parts of the extrapersonal world (and in this way to fund egalitarian schemes). If these conclusions are correct, then it seems clear that an embodied conception of the self, for haves and have-nots alike, means both that there must be a Lockean proviso and that the proviso cannot incorporate an egalitarian conception of justice. This conjunction of claims, of course, is exactly what right-libertarianism holds.

III. Embodiment

“Self” is said in many ways, so I begin by pointing out that in discussing the self I mean to discuss what some philosophers call “psychological identity.”Footnote ⁶ Psychological identity is to be distinguished from both metaphysical identity and personal identity, which concern the conditions under which one remains the same individual or the same person, respectively.Footnote ⁷ Psychological identity, rather, is an agent's first-personal sense of himself—a sense of what person one is, of one's nature or character, of one's options and abilities, and of one's place in the world. It is in this sense of the self that engaging in certain ongoing activities, projects, and relationships is an essential part of oneself.

The starkest alternative to the embodied conception of the self is what I have elsewhere called the “formalized” conception.Footnote ⁸ On the formalized conception, a self is fundamentally a particular, unique psychology. Here we can think of the self as the dimension of the person that engages in choice and practical reasoning. The actions of this self are performed in relation to the world around it, but the self and its actions remain always distinct from that world, strictly speaking. On such a view, ongoing activities and projects are seen as patterns of selection and response with respect to the world around one, and such patterns can continue unabated even when the particular details of one's worldly circumstances change (hence the description of such a self as “formalized”). Consequently, a formalized self is less vulnerable to serious interruption or upheaval brought about by external or even bodily changes: I may lose a leg or its use, say, but while this may be a serious and distressing loss, its impact is mainly on what particular projects I can engage in, not on who I am. It does not even impact what I can do, on a formalized understanding of doing: if what I do, really, is to direct my will in a certain way—so as to act well, say—in relation to the world around me, then I can continue to do the same in my new circumstances, too. If something stops me, it is not the loss per se.

Perhaps the clearest philosophical example of such a view comes from the ancient Stoics, who went so far as to define even the act of walking not in terms of what the legs do, but in terms of what the willing, selecting mind does.Footnote ⁹ The self, on this view, is one's will, and in this spirit, when Epictetus imagines someone threatening to fetter him, he replies, “What are you saying, man? Fetter me? You will fetter my leg; but not even Zeus himself can get the better of my choice.”Footnote ¹⁰ Notice that Epictetus does not reply merely, “Go ahead, fetter me—I can take it; I will still conduct myself well; you won't break my will.” His reply is more radical—that he cannot be fettered, because he is only his will. Such a conception of the self was also crucial to the Stoics' conception of virtue. Virtuous activity, on the formalized Stoic view, is the activity of selecting and reacting well with respect to one's circumstances, whatever they may be. Such a view also supported the Stoics' surprising position that the activities of the virtuous are invulnerable to circumstantial changes, and that the happiness of the virtuous person, which consists in such activities, is therefore beyond the reach of bad luck. But formalized conceptions of the self are found in many places even today, such as in the view, well attested in twentieth-century cognitive science and psychology, that a person's sense of identity and individuality is fundamentally psychological, a sense of a distinct personality related to the body as to something that the personality controls.Footnote ¹¹

However, twentieth-century studies of how people relate to their bodies also began to suggest that the earliest sense of individuality is a sense of physical rather than psychological distinctness, and that bodily motion is not the result of a person's controlling a body, but of a self that is aware of itself as a body.Footnote ¹² Such findings suggest that the self is not only a selector and controller, but indeed is partially constituted by its physical form and its particular repertoire of capacities for moving through and engaging the physical world.

Such an understanding of the self is also strongly suggested in cases of substantial bodily change. For instance, psychologists have observed a significant correlation between limb loss and depression.Footnote ¹³ This depression cannot in general be explained as resulting from the trauma of the violence that resulted in limb loss, since depression does not differ significantly between surgical and accidental amputations.Footnote ¹⁴ A better explanation is that the loss of a limb is often experienced as a kind of loss of self, a change that many patients find understandably difficult to endure. This hypothesis is bolstered by reports of the experience of loss of limb use, as in the case of degenerative diseases like multiple sclerosis. Such changes obviously affect one's sense of one's present possibilities for engaging the world and relating to other people, as well as the changing amounts of time required for acting. But they also affect one's sense of connection with one's past self, as it becomes increasingly difficult to identify oneself with a person who used to move in ways now impossible. Perhaps even more tragically, these changes have a significant impact on one's ability to project oneself into the future, since as degenerative diseases progress, one continues to lose various modes of being in the world.Footnote ¹⁵ Notice that in virtue of such changes, the entire structure of intentional agency changes as well.Footnote ¹⁶

On this conception of the self, then, the self includes one's sense of practical and physical possibilities—that just is a part of the sense of what person one is. This is what it means to say that one's particular embodiment is part of the self. As one philosopher has put it, one's legs are not just objects in the world to which one is specially attached, but are indeed “the possibility of walking, running, playing tennis,” so that a loss of body function “represents a modification of the existential possibilities inherent in the lived body.”Footnote ¹⁷

The idea that the self includes one's sense of practical possibilities seems to be the best rationale, and perhaps the only rationale, for thinking that the self includes one's body. But notice that, for exactly the same reason, the self as a psychological identity would also include one's sense of participation in particular ongoing activities, projects, and relationships. After all, the sense of one's practical and physical possibilities obviously cannot be restricted to that part of the physical world within the boundary of one's skin. My sense of my place in the world is radically altered if I lose my legs or their use, but surely the loss of a partner, or a career, or a community, can have precisely the same effect. If I were to lose the people closest to me, for instance, the task of rebuilding my life would not be one of finding new people in relation to whom to act and select, but the much more radical task of redefining who I am going to be, of constructing a new self. Simply put, if my leg represents one set of possibilities of my embodied existence, then so too does my wife. Therefore, if psychological identity includes the body, then it also seems to include one's sense of participation in particular attachments—one's relationships, projects, and ongoing activities. This is what I shall mean, then, in saying that psychological identity and the activities in which it consists are “embodied”: the boundaries of the self include one's body as well as—and for the same reason—certain parts of the extrapersonal world.

On an embodied conception of the self, then, one's activities are not things that one does in relation to the world, but how one occupies and engages that world as a physical, worldly being. The self's ongoing activities, projects, and relationships can therefore be interrupted or terminated altogether by circumstantial changes: to lose a leg—or a spouse, or a career—on this view, is to lose forever a particular way of being in the world, leaving one to find a new self to be. As a result, a person's happiness and well-being are more vulnerable to circumstantial changes, since the person's identity and activities depend so crucially on the particular ways he or she inhabits the world.

It is this embodied conception of the self that I think is the more plausible one, for the reasons just given—although, again, limitations of space do not permit a more thorough demonstration of this claim. It is worth pointing out, though, that it is a commonplace of the libertarian conception of self-ownership that to own oneself includes at least the possession of control rights over one's physical person. It is, of course, possible for even a formalized self to acquire such control rights, for instance by bargaining for them within an original position.Footnote ¹⁸ For the libertarian, however, control rights over one's person are not rights for self-owners to acquire, but rights in which self-ownership consists. The libertarian is willing to demonstrate that, given self-ownership, one can also come to own parts of the world beyond one's person. What the libertarian does not do, however, is to argue from the ownership of the self to the ownership of one's person, as though these were different issues—as indeed they are taken to be on the formalized view.Footnote ¹⁹ The self that the libertarian takes to be owned, then, must at least include the whole physical person.

Nonetheless, an important disagreement among libertarians—as I intend to show in the rest of this essay—concerns whether these control rights end at the boundary of the skin. As I argue now, although no-proviso libertarians and left-libertarians alike accept that the self includes the body, nonetheless they cannot accept the embodiment of the self in parts of the extrapersonal world. These libertarians therefore adopt a third view of the self, which we can call the “quasi-embodied” conception. For their part, right-libertarians disagree with both no-proviso libertarians and left-libertarians in embracing the embodied conception of the self and rejecting the quasi-embodied conception. Furthermore, since I argued above that the rationale for thinking of the self as including a body is equally a rationale for thinking of the self as including parts of the extrapersonal world, the attempt to drive a wedge between embodiment in the body and embodiment in the extrapersonal world should seem at least prima facie implausible.

In what follows, I shall focus on how an embodied conception of the self bears on the libertarian's attitude toward the Lockean proviso, broadly construed, and thus on these different forms of libertarianism. I begin by arguing that no-proviso libertarianism, by rejecting the Lockean proviso, cannot accept the embodied conception for have-nots (Section IV). I then turn to left-libertarianism, whose egalitarian interpretation of the Lockean proviso is incompatible with the embodied conception for haves (Section V).

IV. No-Proviso Libertarianism

My aims in this section are twofold. First, I argue that given an embodied conception of the self, self-ownership requires that there be a Lockean proviso on ownership. I do so by arguing that no-proviso libertarianism can reject the Lockean proviso only by rejecting an embodied (as opposed to quasi-embodied) conception of the self, at least in the case of have-nots. More precisely, I aim to show that without a Lockean proviso, the self-ownership that remains intact for the very poor is one that must be separated from the ability to control parts of the external world. My other aim is to argue that the embodied conception of the self yields an independently more plausible account of the nature of the use of extrapersonal goods and what counts as infringing such use (when compared to the quasi-embodied conception).

No-proviso libertarians hold that while the welfare of the disadvantaged may be a matter of moral urgency, still it is not a matter of justice—that is, a matter that could license the introduction of coercively enforceable obligations. Jan Narveson has made the case for this claim as follows. To begin, he notes that if the welfare of the disadvantaged were a matter of justice, then it would be a demand of justice that such welfare be secured coercively if need be. His argument is then two-pronged: first, that such coercion would be unnecessary, and second, that in any case it would be unjust. As to the first, Narveson argues that there would be no need for such coercion in a capitalist society, since “the intelligent efforts of those who precede us create a world with far more for all of us than there would otherwise be.”Footnote ²⁰ I think that Narveson must surely be correct on this point. At any rate, he as well as John Locke and others are right to suggest that private ownership schemes actually increase opportunities for control of the extrapersonal world, rather than shrink them.Footnote ²¹ As Narveson puts the point, “The ‘secret’ of wealth is no secret at all: it lies in intelligent hard work plus respect for other people's productive activities.”Footnote ²² I shall return to this point in Section VI.

It is Narveson's second argument that concerns me at present, and it is also the deeper of the two. It has three key premises. First, for poverty to be a matter of justice it would have to threaten the self-ownership of the poor. To see this, suppose that poverty were a matter of justice, but did not threaten the self-ownership of the poor. The plight of the poor, being a matter of justice, would then potentially justify coercive restrictions on others. But such restrictions would violate the self-ownership of those persons, by putting them in a coercive plight in order to relieve the noncoercive plight of others, a relief that “is got at the expense of liberty for most of us,” and that liberty is, in fact, identical to self-ownership.Footnote ²³ And self-ownership is itself a matter of justice, if anything is.Footnote ²⁴ Therefore, since coercion not for the sake of self-ownership violates self-ownership, it follows that matters of justice—matters that justify coercion and delimit self-ownership—are only those in which infringements of self-ownership can arise.

Narveson's second premise is that the only infringements of self-ownership are invasive ones, namely, threats to one's person or one's property. This premise rests on Narveson's conception of the use of parts of the extrapersonal world. Those parts that one can be said to use, on Narveson's view, are “those elements of the environment over which one has exerted and continues intentionally to exert control.”Footnote ²⁵ This is because the use of an extrapersonal good is not “pie-in-the-sky,” imaginary or possible use, but actual use and the investment of effort; as Narveson says, “use is use.”Footnote ²⁶ Consequently, it is no infringement of the latecomer's self-ownership to deprive him of future use of that good, since he is not already using it. By contrast, interfering with a first-comer's use of some good is an infringement of his self-ownership, since this is to interfere with the first-comer's activity already under way.Footnote ²⁷ It follows that depriving someone of something counts as an infringement of his self-ownership just in case he is already using that thing. Since depriving someone of what he is already using is what we mean by an invasive infringement of his self-ownership, it follows for Narveson that all infringements of self-ownership are invasive. There is no such thing as a noninvasive infringement of a person's self-ownership.

Narveson's third and final premise is that the poor do not, just as such, face any invasive threat to their self-ownership, even if they are driven to accept low wages. “The poor or the sick are not unfree,” Narveson says; “that is not their problem. Nobody threatens to make them worse off than they already are” if they fail to take “very low-paying jobs…. [N]obody has a gun to their heads.”Footnote ²⁸ It follows from these three premises that, since poverty is not an invasive threat, it is no threat to self-ownership, and thus is not a matter of justice. Whatever standing welfare may have as a matter of moral concern, it is not one to be addressed politically, and thus not by a Lockean proviso on ownership.Footnote ²⁹

I agree with Narveson's first premise, and in the next section I shall consider and reject the rival view that there are matters of justice that both are independent of self-ownership and legitimately delimit self-ownership. Furthermore, I shall raise no objections here to Narveson's third premise, although I think that compelling objections have been raised against it.Footnote ³⁰ Instead, I want to focus now on Narveson's second premise, and argue that Narveson's narrow conception of infringements on self-ownership is incompatible with an embodied conception of the self, as well as implausible in its own right.

We can see this incompatibility in at least three ways. First, on an embodied conception of the self, the self consists in (among other things) its worldly interactions. The self is not strictly a psychological entity whose activities, though done with the world as their object, are nonetheless internal (as on the formalized view). Nor is the self strictly a psychological entity plus its physical embodiment, but separate from the extrapersonal world (as on the quasi-embodied view). Rather, one of the most important ways in which the self defines itself is through its sense of possibilities for interacting with the world. And it seems clear that such interaction is not limited to one's past or to what one is already doing, but is also forward-looking. As Eric Mack has observed, people “obviously do not live by past investments alone, but also by necessarily prospective interaction with extra-personal objects—interactions which, when viewed from some later point, may count as an investment of self.”Footnote ³¹ And as Gerald Cohen asks, if an act underway involves using a succession of worldly goods, why should we suppose it is no interference with the act to whisk away one of those goods, merely on the grounds that the agent has not touched it yet?Footnote ³²

On the embodied view, to control what one can do is to control who one can be. On such a view, then, we must broaden our understanding of infringements of self-ownership, to extend beyond obvious invasions of one's person—holding guns to heads, say—to include what Mack calls the nullification of world-interactive powers.

Moreover, this seems the more plausible view on independent grounds. As Mack has argued, there is no morally important difference between restricting a person's power to engage the world by encasing him in a cage you have made from materials you own, and encasing everything within his reach in such cages.Footnote ³³ Even though the latter case involves no invasion of his person, it is as great an infringement of his ability to engage the world as anything can be. On an embodied view of the self, then, there are such things as noninvasive infringements of self-ownership, and I find such a view to be independently more plausible than a quasi-embodied view.

That is one tension between Narveson's conception of use and the embodied conception of the self, and a second tension can be seen by noting that an embodied self also consists in (among other things) its particular attachments to things and persons in the world. These attachments by their very nature are extended through time; they include not only what attachments one has made in the past, but also how one's attachments extend into the future. Part of what it is to own such a self and to have control over it, therefore, is to have opportunities to take care for one's future.

Here it is worth noting with David Schmidtz that the inability to gain control over natural resources can prevent one from taking measures to protect and sustain those resources, and thus take care for one's future.Footnote ³⁴ Or, as Hillel Steiner has put the point, “Whether I have five other places I could permissibly be and you have none, … the adversity we would each experience from a lightning strike would be the same. However, the personal responsibility we would each have for that adversity would differ.”Footnote ³⁵ Taking responsibility for one's fate requires opportunities to control resources, such as finding a place to shelter from a storm. No one can guarantee his future, but differences in opportunities to do anything about one's future are real—and they are differences in control over the self. Again, then, on the embodied view the activities of self-owners extend well beyond what they have already touched, and thus on such a view not all infringements of self-ownership need be invasive ones. Moreover, this too seems an independently plausible way of thinking about infringement. Surely there is no morally relevant difference between (i) taking away a person's ability to take care for his future by physically restraining him and (ii) doing so by putting cages around all the things he might need to use in order to take care for his future.

Finally, the tension between Narveson's view and the embodied conception also becomes apparent through reflection on what is involved even in owning one's own body. Let me explain why. If a claim of justice regarding extrapersonal goods requires the original use of a first-comer, then does this requirement attach itself to one's person, as well? We cannot allow special pleading here, since we cannot take it for granted that one is the first-comer where one's own person is concerned; on the contrary, one's person is brought about and then helped along through the efforts of others—parents, family, and community. Where my person is concerned, I actually arrive fairly late on the scene; others were there well ahead of me. So is there any difference in principle between the claims of a latecomer in the case of extrapersonal goods and the claims of a latecomer in the case of his own person? Perhaps there is, on the grounds (say) that “first use” of a thing cannot generate ownership claims to it when the thing in question is a person—people cannot be owned by anyone else, including their procreators, but only by themselves. Alternatively, we might say that while the burden of proof falls to each of us to demonstrate that he should be entitled to exclude others from control of his person, nonetheless this is a burden of proof we each are able to meet.Footnote ³⁶ The upshot is the same either way: we own our bodies despite not having met any first-use requirement. But now the question is this: What rationale could there be for saying that we own our bodies, despite not meeting the first-use requirement, that does not also cast doubt on the appropriateness of the first-use requirement in at least some cases involving extrapersonal goods?

Narveson himself holds that we acquire rights over our bodies from each other, in virtue of the realization that “life for ourselves is bound to be worse” unless we grant one another “mutual recognition … of rights not to be molested by others.”Footnote ³⁷ The point at present, though, is not how one could acquire such a right, but the very fact that one could: we must set aside the narrow use restriction on ownership, in this case, because one's body is something over which one must have rights in order to have control over oneself. In the case of one's body, therefore, use cannot be a necessary condition of ownership if the things one can be said to use are restricted to things one has already got hold of. What matters here is not whether one got to it first, but whether one needs to control it in order to have control of oneself. Again, however, there can be no special pleading here: the reasons there are for thinking that one needs control rights over one's person seem equally to be reasons for thinking that one needs control over parts of the extrapersonal world. Depriving one of the opportunity to control parts of the extrapersonal world, therefore, is an infringement of self-ownership, even when that infringement is noninvasive, barring one from things one has not yet touched.

On the embodied view, then, use cannot be restricted to what one has already touched. Rather, use must be broad enough to include not only what one has already touched, but also what one needs to touch in order to have control of oneself—a view that also seems independently plausible. A self-owner must have control of his physical body, despite being a latecomer, but surely for the same reason he must also be able to control parts of the extrapersonal world. And of course it is precisely the opportunity to gain control over parts of the extrapersonal world that the Lockean proviso treats as a matter of justice. As I have argued, the prohibition of noninvasive infringements as a matter of justice cannot be in question as long as we think that the self that a poor or sick person owns is an embodied self.Footnote ³⁸ On an embodied conception of the self, therefore, there can be no question whether it is a matter of justice that a scheme of holdings should meet a Lockean proviso on ownership.

It is in the case of have-nots that the no-proviso libertarian finds himself committed to rejecting the embodied conception of the self: poverty and sickness may leave one destitute, but even so one's self-ownership remains intact, since one can still reason, deliberate, and choose, and no one need be laying hands on one's body. To say that the Lockean proviso is otiose because self-ownership is untouched by one's material circumstances, is to deny that the self one owns is an embodied self. This is not only implausible but inconsistent, since it is surely the self-ownership of haves that necessitates their rights of control of the extrapersonal world in the first place, a necessity that arises only for embodied selves. For as Rothbard points out, trenching on one's person or holdings leaves the self untouched if the self is only considered as a “will”;Footnote ³⁹ self-ownership gives one the right to control one's person and extrapersonal property, then, only if the self one owns is embodied. Without the Lockean proviso, therefore, it is a matter of embodied selves for haves and other sorts of selves for have-nots.

I find the split between the self-ownership of have-nots and their ability to control property to be nowhere more palpable than in Douglas Rasmussen and Douglas Den Uyl's otherwise excellent book Norms of Liberty (2005). Now, I must point out that Rasmussen and Den Uyl would say that they reject the self-ownership thesis. But they understand the thesis they reject not in the broader way that I have understood it here, but in the narrower sense that the right of self-ownership would have to be a single right (rather than a cluster) whose existence is underived and a priori. Footnote ⁴⁰ However, they do espouse a right of “self-direction” by which one has a right against being directed to purposes to which one has not consented, a right that defines a moral space within which self-directed activities are protected against intrusion by others, and by which one has certain natural rights of property in the extrapersonal world.Footnote ⁴¹ As I said at the outset, self-ownership is often taken as a bundle of exactly these sorts of rights—to control oneself, to exclude others from controlling one, and so on—and it is on the basis of such rights that libertarians argue that one can have other property rights as well. So the right of “self-direction” can be treated as a self-ownership right after all, at least as I (and many others) have understood self-ownership rights.

Rasmussen and Den Uyl hold that persons have the right of self-direction because of the nature of the human self. The flourishing of a human being, they argue, is always the flourishing of a particular individual, and thus, far from being merely an individual instantiation of a single summum bonum, one's flourishing is in fact identical to the particular life of the one who is flourishing.Footnote ⁴² This suggests pretty clearly, I think, that Rasmussen and Den Uyl think of the flourishing self as a necessarily embodied self, in just the sense in which I have used that phrase here. Central to the flourishing of such a self, they argue, is self-direction, a right to lead a life of one's choosing, which is not only necessary for flourishing but is in fact a constituent of it.Footnote ⁴³ Moreover, for Rasmussen and Den Uyl there is no separating self-direction from the right to control what one creates; property rights are therefore rights to live according to one's choices, and thus are natural rights.Footnote ⁴⁴ “Taking control of another's property against that other's wishes,” they argue, “can now be seen as nothing less than taking control of one of the central relationships that constitute a human being's life.”Footnote ⁴⁵

With all of this I agree, but their conception of the self changes crucially when they turn to consider the case of have-nots. Rasmussen and Den Uyl consider the objection that being very poor can be an obstacle to being a self-directing person, and they reply that the very poor are self-directors, because even in great poverty one retains the powers of practical reasoning, deliberating, and choosing.Footnote ⁴⁶ But while these powers may demonstrate that poverty leaves intact the self-direction of a formalized or even a quasi-embodied self, that is not the conception of the self that was taken to require the right of self-direction in the first place—that self was an embodied one. Rasmussen and Den Uyl's argument, recall, was that control of extrapersonal property is crucial to self-direction, because self-direction is a right to lead an embodied life of one's choosing. But then we find that for some others (have-nots), self-direction is actually quite distinct from what they can control in the extrapersonal world. In a word, this means embodied selves for haves and other sorts of selves for have-nots.

It is clear, then, that an embodied conception of the self commits the libertarian to the existence of a Lockean proviso, that is, to treating one's opportunities to control parts of the extrapersonal world as a matter of justice. But to what proviso does it commit one, exactly? I go on now to argue that the interpretation of the Lockean proviso depends crucially on one's conception of the self, and in particular that the left-libertarian interpretation is incompatible with the embodied conception of the self. Then, in Section VI, I consider the proviso's interpretation according to right-libertarianism in light of the embodied conception.

V. Left-Libertarianism

Although libertarians and their critics have long held that self-ownership is incompatible with distributive equality, left-libertarians disagree. On the left-libertarian view, people do indeed have rights of ownership over themselves and parts of the extrapersonal world, but like any ownership right, the boundaries of self-ownership must be delineated by considerations of justice, including the justice of extrapersonal property. This is, after all, the traditional role of the Lockean proviso, and left-libertarians argue that the considerations of justice that enter the theory of extrapersonal property via the proviso are best understood in egalitarian terms. I now want to explore the conception of the self that is required by this alleged compatibility between egalitarian restrictions on extrapersonal property, on the one hand, and self-ownership, on the other. To do so, I shall discuss briefly the recent work of three representative left-libertarians, starting with (and spending the most time on) Michael Otsuka, before moving on to Hillel Steiner and Philippe Van Parijs.

Otsuka holds that self-ownership encompasses two main types of rights: rights of control and use of one's mind and body, and rights to all the income one can gain from controlling and using one's mind and body.Footnote ⁴⁷ Otsuka argues that the income right is subject to a Lockean proviso when, as in most cases, producing the income requires the acquisition of extrapersonal goods.Footnote ⁴⁸ In particular, Otsuka says, “You may acquire previously unowned worldly resources if and only if you leave enough so that everyone else can acquire an equally advantageous share of unowned worldly resources.”Footnote ⁴⁹ Otsuka's version of the Lockean proviso is both welfarist and egalitarian. It is welfarist because “enough and as good” is interpreted as leaving for others enough for the purpose of attaining a certain level of welfare, resulting from an individual's combined store of worldly and personal resources.Footnote ⁵⁰ And it is egalitarian because opportunities to derive welfare from unowned worldly goods are to be equal, despite inequalities in personal capacities.Footnote ⁵¹ Consequently, Otsuka holds that “[s]omeone else's share is as advantageous as yours if and only if it is such that she would be able (by producing, consuming, or trading) to better herself to the same degree as you.” Following Richard Arneson, he says that such “‘betterment’ is to be measured in terms of level of welfare understood as the ‘satisfaction of the self-interested preferences that the individual would have after ideal deliberation while thinking clearly with full pertinent information regarding those preferences’.”Footnote ⁵²

How is this version of the Lockean proviso—what Otsuka calls the “egalitarian proviso”—to be satisfied? Otsuka holds that the egalitarian proviso requires the members of each generation—starting with the generation of initial appropriators—to leave at least as great an opportunity to own extrapersonal resources for the next generation. He concludes that at one's death, resources at least as valuable as those one acquired in one's life must lapse back into a state of nonownership, that is, that such resources must return to the commons. The upshot of this requirement is that each person would have “only a lifetime leasehold on worldly resources,” including the ones that they improve. This requirement, Otsuka says, would not amount to a full prohibition on bequeathing, but it would prohibit all but very modest gifts that have little impact on the welfare of the recipients relative to other members of their generation.Footnote ⁵³

It is clear that Otsuka must show that the demands of the egalitarian proviso do not trench upon self-ownership: self-ownership, after all, is not a good to be balanced among others, but a boundary on what may be done in the name of any good. It is presumably for this very reason that Otsuka says that the egalitarian proviso does not conflict with or infringe upon self-ownership, but only restricts one from engaging in transfers that one has no right to engage in anyway.Footnote ⁵⁴ But how is this claim to be substantiated?

Here we must turn to deeper questions regarding the kinds of considerations of justice that can delimit self-ownership, and there are two broad, mutually exclusive, and jointly exhaustive possibilities to consider. The first possibility is that we might defend the view (glimpsed in Section IV) that what any person may do under the banner of self-ownership can be limited only by the similar self-ownership rights of others—that the only limit on self-ownership is self-ownership.Footnote ⁵⁵ It seems clear, though, that the egalitarian proviso could not be derived solely from the self-ownership rights of have-nots, even given the embodied conception of the self. To be sure, from the fact that one owns oneself, and that the self so owned is embodied in projects involving the extrapersonal world, it follows that one is entitled to a fair opportunity to participate within a fair scheme by which people can acquire parts of the extrapersonal world.Footnote ⁵⁶ But that comes far short of the egalitarian proviso, which entitles one not only to the opportunity of acquiring extrapersonal property, but indeed to the opportunity of acquiring extrapersonal property sufficient to give one as good a chance to attain welfare as anyone, whatever differences there may be in personal capacities to improve resources. It is no surprise, then, that neither Otsuka nor left-libertarians generally take their various egalitarian renderings of the Lockean proviso to follow from the thesis of self-ownership; rather, those renderings are taken to arise from considerations of justice independent of self-ownership.

And that is the second possibility, namely, that it is false that the only limit on self-ownership is self-ownership; on this view, there are considerations of justice that both are independent of self-ownership and serve to delimit self-ownership. Here we should notice that Otsuka does not think that there are considerations of justice demanding that one give others access to one's person, which would of course conflict with the self-ownership right of control of one's person. However, since Otsuka thinks that considerations of justice demand that one give others access to one's income, and that this is no infringement of the self-ownership right to income from one's person, he must of course hold that where self-ownership is concerned, there is a crucial difference between control of one's person and control of the extrapersonal world. Indeed, this is exactly what he does hold: the Lockean proviso, he says, holds in cases of producing income by using extrapersonal goods, but not when one uses parts of one's person, as for instance when one weaves a textile from one's own fallen hairs.Footnote ⁵⁷ Thus, the same proviso that would trench upon self-ownership rights to income, in the case of income derived strictly from one's person, does not trench upon that right in the case of income derived from extrapersonal property.Footnote ⁵⁸ What, then, is this crucial difference between personal and extrapersonal goods?

That there is an important difference is obvious: the world is here for all of us, but your person is just for you. So far, however, that idea is fairly vague: in what sense is the world “for all of us”? One possibility is that the world is “for all of us” in the sense that every one of us needs the world not only as a necessary container and sustainer of our physical presence in it, but also, and just as importantly, as a place in which to construct embodied human lives and embodied human selves. But notice that on this rationale, there is no longer the difference that Otsuka needs between personal and extrapersonal goods. On this rationale, the right to control the self is a right to control one's person and parts of the extrapersonal world, for the very same reason—that the self is embodied in both of these. Because we are self-owners, each of us has a right to control all of himself and some of the world. Therefore, on the present rationale—which is, of course, the embodied conception of the self—the world's being here for all of us is a fact of self-ownership after all, so the Lockean proviso would no longer be independent of self-ownership. The only limits on self-ownership suggested by this view, then, are themselves considerations of self-ownership. Yet, as we have seen, considerations of self-ownership alone do not generate the egalitarian proviso. Thus, Otsuka must hold that while it is self-ownership that gives one rights of control of one's person, independent considerations of justice must come in to generate the Lockean proviso, if it is an egalitarian proviso.

A better rationale for the left-libertarian is that the difference between extrapersonal property and property in one's person is that the extrapersonal world is commonly owned, as indeed left-libertarians generally hold.Footnote ⁵⁹ Common ownership, unlike joint ownership, does not require one to secure the permission of all in order to use or appropriate parts of the extrapersonal world, but it does require one to make compensation to others for doing so. Notice that this rationale for thinking that the world is “for all of us” does not follow from the thesis of self-ownership: from that thesis, it follows that one is entitled to an opportunity to access the extrapersonal world, not that one is entitled to compensation from others should they avail themselves of such an opportunity, just as on Otsuka's view. So if common ownership is a demand of justice, then it is a demand independent of self-ownership that can come to bear on extrapersonal ownership by way of a Lockean proviso, with egalitarian implications.

The real question for Otsuka, then, is whether common world ownership is a matter of justice and therefore consistent with self-ownership, which is a matter of justice if anything is. And the answer is: It depends on how we conceive of the self that is owned. According to Otsuka, common world ownership requires persons to make compensation to others for what they take from the commons, and this requirement must not infringe, but only delimit, self-ownership. However, if the self that is owned is an embodied self, then such demands for compensation necessarily infringe self-ownership. Let me explain why.

If persons are self-owners, then at the very least no person need justify his existence to any other person, and no person can owe compensation to others for the fact that he exists—that he is a human being, in the world, leading a human life. Now, if the self that is owned is an embodied self, then necessarily that self exists and leads a human life only by controlling parts of the extrapersonal world. Removing goods from the commons, then, is the price of leading a human life. However, given the egalitarian proviso, removing the sorts of goods from the commons that can underwrite the leading of a human life entails an enforceable claim by others to compensation. Therefore, since (ex hypothesi) the owned self is an embodied self, the egalitarian proviso requires that one compensate others for one's leading an embodied existence. And that is to say that the egalitarian proviso requires one to compensate others for the very fact that one is a self-owner. But of course, to own oneself is for there to be no one from whom one need lease oneself. The egalitarian proviso is therefore incompatible with an embodied conception of the self that is owned.

Note that the argument here is not that the compensating act is incompatible with ownership of an embodied self, although I think it is. On Otsuka's view, that compensation consists in being compelled to appropriate only on condition that one's appropriation be temporary. More precisely, the compensation is that one forgo embodying oneself in any world-interactive projects, such as saving for the future of others, that would involve more than a “lifetime leasehold” on extrapersonal property. Whatever we say about the sort of compensation required, though, the real issue is that the very fact that any compensation is called for violates self-ownership, if the self is an embodied one.Footnote ⁶⁰

In order to ground the egalitarian proviso in the demands of common ownership, then, Otsuka must either embrace an embodied conception of the self and reject the thesis of self-ownership, or he must reject an embodied conception of the self so that the proviso no longer infringes self-ownership. It is clear that left-libertarianism must reject a formalized conception of the self, given the general libertarian thesis that self-ownership as such includes rights to one's person. The left-libertarian must draw the boundaries of the self so as to include one's body but exclude the extrapersonal world. In other words, left-libertarianism requires a quasi-embodied conception of the self.

Consequently, there is a crucial disagreement over the boundaries of the self between left-libertarians (who accept common ownership) and right-libertarians (who reject it). Moreover, the left-libertarian's conception of a quasi-embodied self seems seriously problematic, on at least two grounds. One is that, as I have noted already, the quasi-embodied conception of the self is independently implausible. The quasi-embodied view would drive a wedge between the body and the extrapersonal world where the self is concerned, but as we have seen, the reasons for drawing the boundaries of the self so as to include the body are the same as the reasons for drawing the boundaries so as to include parts of the extrapersonal world. This is because both are crucial to the nature of the self as a world-engaging being.

The other problem is that for the left-libertarian the quasi-embodied conception of the self is a double-edged sword. First, the left-libertarian must draw the boundaries of the self well short of the extrapersonal world in order to preserve consistency between self-ownership on the one hand and compensatory demands for using commonly owned goods on the other. But second, the left-libertarian must also motivate the very idea of a natural right of common world ownership in the first place; yet a self with such a right would surely have to be an embodied one. How could self-owners be anything short of embodied selves and also have some sort of natural claim over things in the extrapersonal world? Indeed, common ownership itself seems to be incoherent in this respect, drawing the boundaries of the self in one place in making persons common world owners and in another place in making persons incur liabilities to others when they use parts of the extrapersonal world.

Thus, in taking seriously the have-nots' alleged right of common ownership, Otsuka must employ an embodied conception of the self, and in arguing that that right does not trench on the self-ownership of the haves, he must employ some other conception. Ironically, this form of left-libertarianism turns out to suffer from the same basic problem as no-proviso libertarianism: an embodied self for some, another sort of self for others.

However, while Otsuka's view requires compensation for the use of resources full stop, we might instead restrict compensation only to the use of resources beyond a fair per-capita share. This is the view of Hillel Steiner, who holds that a Lockean proviso sets a limit on what one may come to own in virtue of being a self-owner, and in particular that such a proviso grants one unencumbered title only to what one produces either from self-owned resources or from one's per-capita share of unowned, raw natural resources.Footnote ⁶¹

For Steiner, this proviso is especially important when considering the use of a rather special sort of natural resource, namely, “germ-line genetic information.” Germ-line genetic information, he argues, is an important resource, since it determines one's native endowments and thus the costs involved in bringing one to a certain level of welfare. And since it is an unowned natural resource, the Lockean proviso requires its division into per-capita shares.Footnote ⁶² Perhaps, then, we could generate an egalitarian proviso just by focusing on the ownership of bodies, without bringing the extrapersonal world into the picture.

I find Steiner less than fully explicit on this point, but I think it is clear that we must distinguish between the germ-line itself—a genetic heritage spanning generations—and the genetic information that stems from that line. This is so, because the genetic information itself takes the form of a particular physical person, and such persons, so far from being unowned natural resources, are self-owners on Steiner's view. What no such person can claim as rightfully his, in contrast, is the fact that a certain natural heritage brought it about that he has the particular endowments he has. To focus on the genetic information, then, is to focus on the particular individual, whereas to focus on the germ-line is to focus on the biological provenance of that individual. Every individual owns himself, but no individual “owns” his biological provenance; the latter, then, must be regarded as a natural resource. The Lockean proviso therefore extends to the use of the germ-line.

Now, as I have just noted, the use of the germ-line results in a new individual with a particular set of personal endowments. However, biological parents have no germ-lines to use but their own, of course, and the use of different germ-lines results in persons with unequal bundles of endowments, an inequality notably expressed in the differences in costs associated with bringing different offspring to a similar level of ability-value.Footnote ⁶³ In that case, the values of our naturally distributed shares of the natural resource of germ-lines are not divided on a per-capita basis, so the Lockean proviso requires compensation for the differences. Consequently, on Steiner's view, persons are all subject to the following enforceable rule: if you procreate, then you must participate in a scheme whereby the costs of developing ability-value in offspring are roughly equalized between sets of parents. This restriction, for Steiner, is no infringement of self-ownership, since it is only a demand of justice that enters via the Lockean proviso to set limits on what a self-owner can be said to own in the first place.

The question for Steiner is essentially the same as that for Otsuka: Given that persons are self-owners, if the owned self is embodied, can such a restriction in fact be a requirement of justice? I do not think that it can, because such a restriction demands compensation from a self-owner for using an owned self to engage in a project that is, on an embodied view of the self, paradigmatic of self-ownership.

My argument for this has two premises. The first is that on an embodied view of the self, having offspring is a paradigmatic project of a self-owner if anything is. Recall that on an embodied view, the reason that the self extends to the body is the same reason that it also extends beyond the body, to include other persons, relationships, and projects. The self clearly extends to one's offspring, and even to one's potential for offspring, since these determine the kinds of relationships one can engage in, what kinds of values one can bring about in the world, what sorts of roles one can play and virtues one can develop, and how one will relate to the human world and its future. Likewise, the self also extends to one's relationship with one's partner, and, here too, questions of procreation determine what sort of couple a couple can be, what joint projects they can share and what joint abilities they can develop, and how they as a couple will relate to the human world and its future. It is for these sorts of reasons, of course, that procreation can play such a profound role in human life.Footnote ⁶⁴ Having offspring is among the projects by which humans typically give their lives and their identities a particular definition; thus, on an embodied view, having offspring is the project of a self-owner, if anything is.

The second premise is that all that is involved in procreation is a pair of selves, each of which is a self-owner. Return to the distinction between genetic information and lines of such information. In particular, when I procreate, do I use the information or the line? Well, it is clear enough that I use the information; that information, after all, exists as the particular physical individual that I am, including its procreative powers and genetic package. But is there any sense in which I use the line? Only metaphorically. After all, focusing on the germ-line, as opposed to the particular bundle of genetic information arising from it, is really just a way of making the historical point that I am not the originator of myself. It is therefore misleading to think of the germ-line as some reified thing which I literally use in bearing offspring. All that I “use” is my particular embodiment—myself—which I own.

From these two points, it follows that the restriction imposed by Steiner's version of the Lockean proviso is a requirement on self-owners that they make compensation for using selves that they own to engage in projects that are, on the embodied view, paradigmatic of self-ownership. Once again, then, to bring in considerations of common ownership as demands of justice via the Lockean proviso is to require self-owners to make compensation in the event that they do the things that self-owners do, if the self is embodied. In short, it requires self-owners to make compensation for being self-owners; and that, as I said above, is just to deny their self-ownership.

Thus, Steiner must reject at least one of the two premises I defended above. On the one hand, if he were to deny that all that is involved in procreation is a pair of owned selves (the second premise), then he would have to hold that, even if some particular genetic information could be in some sense mine, still it could in no sense be me. On the other hand, if Steiner were to sever the link between procreation and self-ownership (the first premise), then he must hold that the self is untouched by restrictions on how it may relate to the extrapersonal world. Either way, then, he would have to reject an embodied conception of the self.

Lastly, this separation of self from world is nowhere more explicit, I think, than in Philippe Van Parijs's discussion of “real freedom.” According to Van Parijs, a society with what he calls real freedom would have three characteristics: (1) it would provide for the security of its members; (2) it would respect the self-ownership of its members; and (3) it would otherwise create the best possible opportunities for each of its members to do whatever he might want to do.Footnote ⁶⁵ Quite rightly, Van Parijs holds that the maximization of such opportunities must stop short of imposing forced labor upon those who pay the tax bill, given that members of a free society are self-owners. However, where considerations of self-ownership place the ceiling on taxation depends crucially on how self-ownership is connected to ownership of extrapersonal property. On Van Parijs's view, self-ownership is a separate matter from extrapersonal property. For instance, he holds that even a socialist state that forbids wage labor and owning capital still respects self-ownership, on the grounds that such a scheme restricts only what may be done with external objects, and not what may be done with oneself.Footnote ⁶⁶ Although an analysis of socialism falls outside the scope of this essay, it should be clear that socialism can combine self-ownership and a ban on private ownership of material capital only by treating what one may do with oneself as a separate question from what one may do with external objects.

By keeping questions of self-ownership and extrapersonal property ownership so distinct, Van Parijs can set the ceiling on maximizing opportunities at the point where taxes “generate the highest yield … that can be durably generated” under the type of socioeconomic regime in place.Footnote ⁶⁷ However, this conclusion requires rejecting an embodied conception of the self, since on that conception the ownership of one's person and the ownership of extrapersonal property are anything but distinct issues.

This is instructive: it would seem that the only way to bring in considerations of justice, independent of self-ownership, via the Lockean proviso is to treat self-ownership and other forms of ownership as separate issues. But on an embodied conception of the self, self-ownership and other forms of ownership are not separate issues at all. On that conception, it is because we are embodied self-owners that one must be able to control both one's own person and parts of the extrapersonal world. Therefore, given an embodied conception of the self, the only restrictions that could enter via the Lockean proviso are matters of self-ownership.Footnote ⁶⁸

VI. Right-Libertarianism and the Lockean Proviso

We can now see that conjoining the thesis that persons are self-owners with the embodied conception of the self yields the view that I have called right-libertarianism. As I argued in Section IV, the embodied conception requires a Lockean proviso on either the acquisition or the use of extrapersonal property (or both), on the grounds that intrusions against a self-owner include both invasions of one's physical person and certain noninvasive infringements of one's ability to engage the extrapersonal world. Right-libertarianism therefore commits to the embodied conception in the case of have-nots, who as self-owners must as a matter of justice have free access to a fair scheme within which private property can be acquired and exchanged. Moreover, I argued in Section V that the only demands of justice that can enter via the Lockean proviso are demands arising from self-ownership itself. This is because demands entering via the proviso must not be in tension with self-ownership; thus, if demands of justice independent of self-ownership were to enter via the proviso, then we would have to show that those demands do not trench upon self-ownership even though they place significant restrictions on the ability to control the extrapersonal world. But in order to show that, we would have to separate the question of self-ownership from the question of control of the extrapersonal world. Yet it is precisely this separation that an embodied conception of the self rules out; so the only demands of justice that can enter via the Lockean proviso are other demands of self-ownership. In that case, right-libertarianism commits to the embodied conception of the self also in the case of haves: because they are self-owners, the only constraints of justice on their acquisition or use of property are those stemming from self-ownership. And these constraints have no egalitarian implications.

Therefore, different forms of libertarianism require different conceptions of the self. As I said earlier, this important difference has not received sufficient attention among libertarians. Even so, it has not been altogether ignored. Perhaps the most vocal right-libertarian on this point is Eric Mack, who has insisted on the importance of what I call the embodied conception of the self for libertarian self-ownership. Human beings, Mack writes,

Embodiment, Mack notes, is crucial to the idea that self-ownership gives one rights not only against invasions of one's person, but also against threats to one's “world-interactive powers,” such as the ability to use, transform, and appropriate parts of the extrapersonal world.Footnote ⁷⁰

Because right-libertarianism embraces the embodied conception of the self, it denies that self-ownership is a separate issue from ownership of extrapersonal goods, and therefore insists that ownership of extrapersonal goods is subject to a proviso forbidding property owners from worsening the situation of others, through either the acquisition or the use of property. I now want to develop and clarify that proviso by addressing four major questions about it. First, does the Lockean proviso, as I have defended it here, set limits on the acquisition of property, or its use, or both? Second, the Lockean proviso protects against certain kinds of worsening of the situation of others, but what kinds of worsening are these, and what liberties are protected by the proviso? Third, what sorts of objects are such that their disposal is subject to the Lockean proviso? Finally, does the Lockean proviso make any “real” difference between right-libertarians and no-proviso libertarians, or (as some have suggested) does it turn out merely to be a fifth wheel, so to speak? I take up these four questions in turn.

Many discussions of the Lockean proviso focus on it as a limit on how much a person may rightly acquire, and, as such, these discussions typically focus on thought experiments involving first-comers to virgin land, with the aim of devising a proviso that will prevent first-comers from leaving fewer or inferior goods for latecomers. However, I find such a focus seriously problematic for at least two reasons. One problem is that this approach implicitly assumes that acquisition is zero-sum—that somehow if one person removes something for the commons, there is thereby necessarily that much less for everyone else, leaving latecomers only the scraps. Yet, over time anyway, this is just the opposite of what happens. I recall attending a meeting of philosophers a few years ago in which one participant undertook to remind the others, with an air of perfect self-evidence, that the first-comers to the area where we were staying had made us all worse off by grabbing everything up centuries before we could get there. My thoughts immediately ran back over my morning before the meeting, and as I remembered how easy it was to find soap and hot running water, how easy breakfast had been to come by, and indeed how I was able to pay for all these things by means of work in my own specialized field, I wondered how any comparison between myself and the region's first settlers could possibly show me to be the less fortunate one. On the contrary, acquisition and use are parts of a productive cycle that generates not just enough and as good, but more and superior. As a result, latecomers usually come off far better than first-comers. Acquisition is generally positive-sum, so it seems pointless to devise a Lockean proviso on acquisition as if it were necessarily zero-sum.Footnote ⁷¹

The other problem is that I find the notion of leaving “enough and as good” of raw, unowned, natural resources to be conceptually problematic. Even if I can count how much stuff is in a given area, I can count the number of resources there only if I know rather a lot about people's plans, abilities, skills, and needs, and about what sorts of interactions are possible between persons controlling various parts of the world. Without knowing that, we cannot even know which materials count as valuable resources to be used and which count as rubbish that would be better cleared away.Footnote ⁷² Thus, “enough and as good” presupposes a world more like our own anyway, in which much is already controlled by various persons.

I do not find it useful, therefore, to think of the Lockean proviso as a proviso on the acquisition of property.Footnote ⁷³ That leaves the question of use: the relevant question is not “how much” one may acquire, but what sorts of rights of control one can assert over a thing in virtue of becoming its owner. So understood, the proviso forbids an owner from so using what he owns that others are worsened with respect to their opportunities to become owners, as when one uses what one owns to close off another's interactions with the world. But what counts as “worsening” here?

This brings us to our second question: The Lockean proviso protects persons from uses of property by others that would deprive them of certain opportunities, but what opportunities are these? To put this another way, the right-libertarian holds that people have rights of control over themselves and that such rights also require rights of control over the extrapersonal world; but what kinds of rights are these latter? It should be clear that such rights cannot be rights to own some particular piece of property or even some particular amount of private property; nor can they be rights to a given level of opportunity to use and appropriate extrapersonal goods. This is because property rights so construed would be rights to appropriate for oneself the efforts of others, if need be—to provide one with these goods, or this level of access to welfare—and this would clearly conflict with the self-ownership of those whose efforts are thereby appropriated. Rather, the proviso protects one against being deprived of an opportunity to use and appropriate extrapersonal goods. More specifically, the proviso protects one's right to participate in a practice by which private property can be acquired and exchanged.Footnote ⁷⁴ This does not mean that one has a right to a formal practice of private property even if there currently is none, since that again would be a right to the efforts of others in creating a certain social order. The right in question is, rather, a right not to be excluded from a practice of private property, if there should be any. Violations of such a right include the sorts of noninvasive infringements of self-ownership that we discussed in Section IV, such as when one person uses what he owns to keep another from interacting with the world so as to acquire parts of it.

Furthermore, we saw above that forbidding the earning of wages or the private owning of capital can be made consistent with self-ownership only by rejecting an embodied conception of the self. By contrast, on my view, to own an embodied self is also to have a right not to be excluded from a practice of private property, within any political order with a system of property. This is because the world-interaction of an embodied self involves the opportunity to acquire durable control over parts of the extrapersonal world, including the opportunity to do so through exchange of one's own efforts.

At this point—and this is our third issue—one might worry that, by expanding the boundaries of the self, right-libertarianism gives one either too many control rights or too few. On the one hand, by blurring the line between the self and the extrapersonal world, perhaps right-libertarianism extends the sorts of rights of control one has over one's body to all sorts of things in the extrapersonal world, even including other persons and their property. Simply put, if the self can include things like loving relationships and careers, then does the owner of such a self thereby acquire rights to control the persons and resources on which those relationships and careers depend? And on the other hand, by blurring the distinction between self and world and then placing a Lockean proviso on things in the world, does the right-libertarian thereby open the door to placing a Lockean proviso on one's person as well? If extrapersonal goods are subject to a Lockean proviso, then why aren't personal goods—such as body parts and substances—subject to such a proviso as well?Footnote ⁷⁵ Each of these queries concerns the same basic question: Over what sorts of things does the Lockean proviso range? If one person cannot use what he owns to deprive another person of a place to stand, then why can he nonetheless deprive the other of a job in his firm, or of his blood and other tissues, or of his hand in marriage? After all, the point of the proviso is to protect persons' efforts to acquire control over things they may bring within the boundaries of the self, but on the embodied view these boundaries are broad enough to include others and their possessions.

However, this oversimplifies what the proviso actually does, which is to protect efforts to acquire not full stop, but in a way that is constrained by the fact that others are self-owners as well. Since the proviso protects opportunities to gain rights of use over things by appropriating them, the proviso protects opportunities only with respect to appropriable things, things that are not already owned. The thesis that persons own themselves, interpreted via the embodied conception of the self, automatically removes persons and what persons already own from the set of things that are appropriable and ripe for the picking. In that case, any claims one would stake to a person or what that person owns will exist only by the agreement of that person. One may make other persons or their possessions crucial parts of one's plans, but one does so always at one's own risk. Simply put, the Lockean proviso attaches only to appropriable things, and given the prior ownership of embodied selves, these things do not include persons or their property. If another person ends an intimate relationship with me, or closes the place where I work, or withholds an organ I need, then the outcome may be undesirable for me, but the Lockean proviso surely promised me no protection against such an outcome.

This brings us, finally, to a point where we are ready to consider the difference that the Lockean proviso makes for right-libertarianism. I have argued that right-libertarianism is committed to a conception of the self which in turn entails a Lockean proviso, but does this latter commitment actually make any difference worth mentioning between right-libertarians and no-proviso libertarians? Narveson, for one, is highly skeptical about this, claiming that what I have called right-libertarianism probably would not have “a real-world-divergence” from his own no-proviso libertarianism.Footnote ⁷⁶ Does the Lockean proviso make any significant difference?

Here a thought experiment will be useful. Imagine that a person—call him “Landlocked”—finds his land hemmed in by his neighbor's land and that the neighbor refuses to grant Landlocked an easement allowing him to bring things onto his property. Right-libertarianism treats cases like this one as cases in which one person's use of his property compromises another person's world-interactive powers, and thus as problematic from the point of view of self-ownership. But here there are a couple of concerns. One concern is that characterizing the case in terms of self-ownership makes no difference to the process of resolving the dispute. The neighbor has an ownership right that turns out, on this occasion, to be inconvenient for Landlocked, and that is how the impasse must be addressed; casting things in terms of self-ownership gets us nowhere. It would perhaps be better, then, to say that while Landlocked's plight is unenviable, still the situation is not a question of justice or injustice, but should be settled by other sorts of considerations. So even though right-libertarianism need not cast all conflicts in terms of the Lockean proviso, it does seem committed to treating this conflict in those terms. But doing so gains us nothing—and that, of course, is exactly what the no-proviso libertarian would have predicted.Footnote ⁷⁷

But should we agree that casting such a conflict in terms of the Lockean proviso really makes no difference to the resolution of the conflict? I do not think so. Part of the point of marking off questions of justice from other sorts of considerations, such as inconvenience, is surely to establish the lexical priority of the former questions to the latter questions. It seems clear that, with or without the Lockean proviso, the neighbor's situation will be regarded as a matter of justice: the question, after all, is what Landlocked and the neighbor may rightfully do with land that the neighbor indisputably owns. Without the proviso, then, as a matter of justice an arbiter must consider the claims of the neighbor as lexically prior to those of Landlocked. But now it is clear that framing Landlocked's plight in terms of the Lockean proviso will make a significant difference in the arbiter's deliberations about the conflict after all, because with the proviso in place, Landlocked's plight must now enter the arbiter's deliberations right alongside the neighbor's plight. Resolving the dispute, then, would not be a matter of balancing a claim of justice with a matter of mere inconvenience, but of balancing two claims of justice. My aim is not to say that such a conflict must therefore be decided in one way rather than another, but only to say that framing the issue in terms of the Lockean proviso does make a difference—and, I think, a most appropriate difference—in how one must deliberate about the resolution of such a conflict. Since this observation generalizes, I conclude that there are important “real-world” differences between no-proviso libertarians and right-libertarians. Their different conceptions of the self entail importantly different views about the line between injustice and inconvenience.

However, this way of addressing the first concern now raises another: precisely because we can frame such a conflict as a conflict between the claims of two self-owners as self-owners, we must now appeal to something other than considerations of self-ownership to resolve such a conflict.Footnote ⁷⁸ In that case, the worry goes, the conflict must be resolved by appeal to some independent consideration, but then we would have done just as well to have appealed to that consideration in the first place, leaving the Lockean proviso out of it. The Lockean proviso may entail that such a conflict is a matter of conflicting claims of self-owners, but that just means that something other than considerations of such claims must come in to adjudicate the dispute. The Lockean proviso still seems redundant.

But such a conclusion would be unwarranted. For the right-libertarian, the fundamental rationale behind the Lockean proviso is the conviction that the boundaries of the self are broad enough to include those parts of the extrapersonal world that are crucial to the life one constructs. And this same rationale can be a valuable guide in thinking about conflicting claims, such as the ones that arise between Landlocked and his neighbor. Given that the rationale behind the Lockean proviso lies in the value of opportunities to engage the extrapersonal world, the task for the arbiter is now to identify what sorts of rulings would be consistent with a larger private property scheme in which such opportunities are best made available. Again, this is not to say that the arbiter must therefore decide the case this way or that way; a rationale for a resolution is not itself a resolution. Nonetheless, it is to say that the considerations whereby such conflicts are to be resolved are themselves prompted by the same rationale that prompts the Lockean proviso. They are therefore not independent of the proviso after all, but are unified with it by a larger conception of human nature and well-being.

Of course, it may well be that the right-libertarian and the no-proviso libertarian would recommend the very same tie-breaking principle to our imaginary arbiter, so that these two ways of framing the dispute could even lead to identical resolutions. But even so, this does not show that the Lockean proviso makes no important difference between right-libertarians and no-proviso libertarians. This is because right-libertarians and no-proviso libertarians set their political principles within different larger conceptions of the nature of the self, and these conceptions represent different ways of understanding our nature, our well-being, and how we relate to one another. The difference between right-libertarians and no-proviso libertarians is therefore one that obtains at the level of an overall conception of human life. For the right-libertarian, then, the Lockean proviso is anything but idle. It is the thing that tethers him to a distinctive outlook on the nature of human existence. And that is a difference that crops up everywhere. What is it to lose a loved one? What is it to lose property? What is it to lose a homeland? How independent are identity and well-being from things that happen in the extrapersonal world? Since such questions go to the heart of our self-understanding, they should matter to philosophers if anything does. And one can answer such questions only by taking a position on the boundaries of the self.