I.

A natural strategy for opponents of redistributive policies to pursue is to emphasize people’s inviolability. Indeed, one way of trying to capture what animates views like libertarianism is to say that anti-redistributivists are particularly sensitive to violations of people that occur in economic contexts. Of course, they will acknowledge that there are many purported justifications for the state laying hold of our property in order to benefit others. Familiar accounts include various forms of contractualism and denials of the priority or even the reality of private property. But, says the anti-redistributivist, these accounts ultimately ignore something important about our inviolability. And a natural way to capture what is being ignored is to do so in terms of self-ownership. On the one hand, self-ownership might seem to confer on us a claim to goods the state proposes to remove and give to others; on the other hand, if the state proposes to make us work on other people’s behalf, then things are worse, since the state is in that case violating us even more directly.

We can begin with the claim about property in external goods. Locke develops this idea in a famous passage in the Second Treatise that runs as follows:

Robert Nozick, Murray Rothbard, and other libertarians have followed Locke in connecting self-ownership with ownership in external goods.^{Footnote 2} Rothbard, for instance, writes of a sculptor:

The Lockean chain thus proceeds from self-ownership, to ownership of our labor, to ownership of the products of our labor, which is taken to encompass the right to exclude others from what were initially common goods.

Private property is a notoriously complex topic, however my focus here isn’t the general theory of property, but the role that self-ownership plays within it. And in that context, it is quite unclear what contribution self-ownership is supposed to make to the crucial final element in the chain, namely, the right to exclude. We must bear in mind that what is to be explained here isn’t the right to make use of or otherwise benefit from some good—many people might share such a right—but rather the right to exclude others from exercising control.^{Footnote 4} The interesting thing about the money in your bank account isn’t so much that it’s yours but that it isn’t mine, so to speak—that you get to block me from accessing it, that you and not I decide what to spend it on. And the connection between self-ownership and the claim to exclude is, again, quite obscure. The problem isn’t in the first instance substantive doubts about self-ownership, or how the money in your account comes to be under your control. It’s that it is unclear how self-ownership in particular can contribute to the exclusionary claim. The question is: What is it about self-ownership that lets you exclude me from your money? The Lockean suggestion is supposed to be that self-ownership produces ownership over external goods via the ownership of our labor, but even setting aside the murky notion of owning an activity like labor, this appeal seems to involve a weird metaphysical relation: “mixing” our selves with the object so as to acquire ownership. But what we require at this point isn’t metaphysics, but a distinctively moral explanation: How is it that we come to be entitled to exclude others from some asset? Self-ownership is of no use in producing such an explanation, unless we take seriously the idea that the object of our labor literally becomes a part of us. But of course no one believes this. Whatever else we might make of self-ownership, the coconut tree that I work on does not become a part of my body.

This observation points to an obvious alternative. The significance of labor, on this alternative picture, is not that objects become a part of us through occult metaphysics, but that labor endows us with a straightforward moral claim to exclude others. This claim has nothing to do with self-ownership. To see this, just assume that critics of self-ownership are right. Kant, for example, argues that the notion of self-ownership is incoherent because it involves viewing people as both subjects who can be vested with ownership, and objects—mere things—that can be owned.^{Footnote 5} Others see it as a simple reductio that self-ownership implies we could do anything with ourselves we wanted, like sell ourselves into slavery or for other people’s amusement, which is supposed to be false. None of this would stand in the way of a moral claim to exclude. Self-ownership is not required in order to assert that our having worked hard on something can give us a claim—perhaps stronger or weaker, no doubt defeasible—against others attempting to exercise control over that thing.

Self-ownership thus points us precisely in the wrong direction: it makes us look toward metaphysics when we should be looking toward the kinds of moral considerations that strike us as compelling in the disposition of claims to control. Since my main goal is to suggest that invoking self-ownership is a mistake in light of better alternatives, I won’t develop a fleshed out theory of property here.^{Footnote 6} But I see no reason such a theory couldn’t be developed in a plausible way, without making mention of self-ownership. In our common sense picture of the world, we constantly negotiate ownership claims on the basis of complex moral considerations that have nothing to do with the self. Factors people routinely respond to include discovery, creation or invention, labor, fair exchange, and much else besides. As a small-scale example, consider a parent deciding whom to assign control over a picture Sandy has just painted (on her own paper, with her own markers, and so on) and which Arnold covets. What supports the judgment that Sandy should decide where to put the picture, or whether to destroy it or send it to grandma, is moral stuff all the way down—the fact that she drew the picture, that Arnold has no claim to the materials utilized, that there is plenty of paper left for him to make his own drawing, and so on. No need for fancy metaphysics or the self.

It is important to traditional libertarians to develop a nonconventional theory of property, and we might worry that self-ownership plays an important role in this respect. The libertarian strategy, after all, is to insist that property rights must be adequately recognized in any legitimate social arrangement, so that we cannot simply announce that democratic majorities or constitutional conventions have decided to recognize laws that, say, ban or radically distort private property—any more than they may authorize attacks on the person. For this reason, libertarians tend to look toward Locke, and away from Hume and his successors, who typically view property as up for grabs until social institutions—which don’t typically themselves make essential reference to natural property rights—have been determined. ^{Footnote 7} An intriguing question I cannot do justice to here is what role self-ownership could play in a conventionalist theory, setting aside Hume’s notorious skepticism about the very notion of the self. Suppose that our recognition of your right to the money you earned working on a contract were merely conventional, subject to us changing our minds tomorrow. If ownership thus devolved into mere convention, presumably we couldn’t accept such an attitude toward ourselves, since it isn’t just a convention that I may not attack your body (the opposite “convention” would just be wrong). So there may be the interesting upshot that Lockeans are the ones interested in self-ownership because conventionalists have, so to speak, debased the currency of ownership to the point that applying the notion to the self seems misguided or even threatening.

But, returning to the objection, self-ownership doesn’t do anything to advance the goal of a nonconventional account of property by the earlier arguments, and appeals to homely moral notions like discovery or hard work can constrain appeals to convention. When Sandy tells Arnold that she worked hard for her salary, or that she created the painting, or that she traded for the car, there is no reason to take these statements as merely conventional, though of course much more needs to be said in any given case to settle exclusionary claims. So the Lockean program involves something like natural rights, but there is nothing about self-ownership that makes it uniquely suited to grounding that program.

All of this may sound uncharitable toward Locke, but one way of interpreting this criticism is as the suggestion that he misunderstood his own theory, in the same way that one might suppose he misunderstood other good ideas of his, such as the psychological theory of personal identity, which he (arguably) restricts to memory, instead of taking into account other mental states like intentions or dispositions. The important point there is the insight that sufficiently significant changes in our mental lives start to look like changes in personhood; the important point regarding property is that there are actions we can take that make a moral difference to our right to exclude others from control, and Locke gets at both points clearly enough. After all, he does correctly identify several morally significant features we might cite in assessing exclusionary claims. Labor is the obvious one, but of course he also emphasizes the value that labor adds, and that “enough and as good” of the commons must be left over if it is to be properly appropriated. Locke treats this as a proviso, but we can reinterpret it as just one more item on the long list of moral considerations that make a difference to our judgments about claims to control, on all fours with the rest.

Finally, one more modification to Locke bears mentioning here. Locke and philosophers following him tend to emphasize original acquisition in their discussions of property. Even egalitarians and left-libertarians agitating against strong private property claims tend to share this assumption—their discussions, too, take place against a backdrop of plots of land, harvesting coconuts, and so on. But of course very few of us do much original acquiring these days. Instead, members of advanced societies mostly participate in a service economy, engaged in professions like teaching, trucking, food services, or web design. It might be supposed that the value generated by services can be traced back to initial acquisition, but on reflection this picture is wrong as well. It is not the case that the service sector involves exchanging units of value that were initially mined from the earth, like pieces of gold. Japan is fifteen times richer than it was after the war, but not because of natural resource acquisition; getting richer in the modern sense just involves becoming more productive so that there are more goods and services available per person. Those of us interested in redistribution should not for the most part be spending our time arguing about land, oil, or gold; we should be arguing about who is entitled to value generated by a teacher spending time in the classroom, or a trucker spending time on the highway.^{Footnote 8}

There is a sense in which this shift in focus might make self-ownership seem more important. Services are a form of labor, and so the appeal to self-ownership can take a more direct form. What the teacher, graphic designer, and trucker are selling is their work, and it might seem especially compelling that we can exclude others from control over the value of the work we provide as a service to others. To put it a bit more grandly, the output of services is less dependent on external objects and therefore constitutes a purer reflection of our selves. A piece of code or a teaching session are relatively direct manifestations of our talents, ambitions, and personalities. Long haul trucking is of course less clearly connected to who we are, but even the trucker is selling something more directly connected to himself than is the harvester of wheat or the miner of gold. At the limit case are vendors of ideas—say someone getting paid as a script doctor in Hollywood, whose work may just consist of sitting in an armchair and fantasizing out loud about car chases and one-liners. Here, the chain from one’s self to what is sold is especially short, and so we may feel more tempted to frame ownership in terms of self-ownership.

However, on closer inspection, self-ownership remains unhelpful, even in the context of services. In the case of initial acquisition, what we are trying to explain is your claim to control a parcel of land or a chunk of gold. In the case of trade in services, what we are trying to explain is your claim to control the money that you receive in exchange. And in explaining why the money in the bank is yours to control and not mine, it remains unhelpful to invoke self-ownership. The money has not become a part of you. The metaphysics of the self and its mixings are once again irrelevant. The close connection between self and service is relevant to debates about property, but its importance lies in the fact that redistributing income from services is even harder to defend by appealing to a commons—the script doctor obviously isn’t appropriating a common resource. Left-libertarian maneuvers to acknowledge self-ownership while denying ownership over the commons in order to defend redistribution seem especially inapposite here. But none of this has anything to do with self-ownership.

II.

Next, set aside property claims and assume these are no longer under dispute. There is another way that self-ownership may seem relevant to redistribution. Writers on both the right and the left have assumed that the welfare state ultimately involves a claim to partial ownership in us. One version of this idea comes from libertarians like Nozick:

As the context makes clear, Nozick takes himself to be addressing a facet of the status quo, not some dystopian fiction. On his view, redistribution amounts to a form of compulsory labor open to objections from self-ownership. Perhaps more surprisingly, some egalitarians have agreed about this, and proceeded to conclude that we should deny self-ownership precisely to avoid Nozick’s criticisms. G. A. Cohen writes:

To his credit, Cohen takes the denial of self-ownership seriously to a fault, for example by examining what it means for an egalitarian to deny that you own your eyeballs, which he does deny. He accepts that it should give us pause, say, to view eyeballs as up for grabs in a world where some could benefit from an eyeball lottery, but in the end he suggests that the issue isn’t really the mere fact that eyeballs are a part of us that is relevant to our moral calculus, but more-general facts concerning, for instance, the “severe interference” in people’s lives involved in redistributing eyeballs (as opposed to mere income or wealth).^{Footnote 11} On this view, the important thing in considering redistribution isn’t the qualitative difference between self and other, but the quantitative difference redistribution makes to our interests. But both agree about the implications of self-ownership for redistribution.

The obvious doubt to raise about both of these accounts is whether redistributive taxation requires anything so strong as even a partial ownership claim. It is true that Cohen’s language is sometimes very strong, which encourages a somewhat alarming picture, as when he refers to the state enforcing a positive duty to “produce a surplus.” In fact, even Rawls originally referred to seeing the “distribution of natural talents as a common asset,” which likewise may suggest something like an ownership claim (though he tones this down in the second edition of A Theory of Justice).^{Footnote 12} Read in its least favorable light, enforcing the duty to produce a surplus suggests that the state could compel people to work who otherwise would not, that is, by throwing dilettante artists into labor camps. But let us assume a more charitable and weaker reading according to which the state merely has a claim to income and wealth we generate when we choose to work, in whichever way we do choose. (I set aside how this weaker reading harmonizes with what egalitarians like Cohen say elsewhere, namely, concerning our obligations of assistance and the supposed duty of the state to enforce these.)

Libertarians can and do object to this weaker redistributive claim in just the same terms of self-ownership, of course. “Seizing the results of someone’s labor is equivalent to seizing hours from him and directing him to carry on various activities,” says Nozick.^{Footnote 13} We can render this vivid by imagining that you are a high powered consultant paid at the end of each day, that your redistributive tax burden is 20 percent, and that the way it works is that on Fridays everything you earn goes into the redistributive pot. On every other day, you are handed your check on the way out the door, but on Fridays when you are given your check for the day you are told that you must sign the check over to the state and drop it down the chute marked “For Egalitarian Redistribution.” The standard libertarian worry is that the situation I have just described captures the structure of ordinary redistributive taxation, and that it amounts to a kind of white collar labor camp.

It may even sound worse. Suppose our worker announces he doesn’t wish to show up on Fridays anymore. If the state announces that he must show up to work—or else—then we are back to our stronger, less charitable reading from before, in which the state asserts the right to compel you to work on its behalf, something close to the corvée labor of the Incas or of feudal France, except the labor is directed toward income, not construction. To avoid this implication, redistributivists should distinguish sharply between demanding that we work on behalf of others, and insisting that we owe part of our income to the state for redistribution. What they should claim is not that the state is enforcing a duty to work to benefit others, as Cohen sometimes seems to imply, but that we are not entitled to keep all of our income. One version of the story might go like this: our income emerges from within a system of social cooperation; the egalitarian insists that fair principles for social cooperation mean that we owe a fraction of our income to the state in support of institutions dictated by those principles, including redistributive welfarist policies. What the state is insisting on, then, isn’t that we must work if we don’t want to, but that when we do earn income we aren’t entitled to keep all of it. And what this means is that egalitarians should avoid the kind of setup I described earlier in which your work over a given interval is dedicated to redistribution. They can object that such a system can only reflect the principle that the worse off can compel the better-off to work on their behalf, whereas the principle they wish to invoke is rather that justice requires us to contribute some of what we earn to fair institutions, some of which will be redistributive.

Someone like Nozick might insist that this is a distinction without a difference and continue to pursue the self-ownership line. It remains the case, he might object, that we are compelled to work additional hours in order to attain the same income, no matter how important attaining that income might be, for instance to support a disabled child or unemployed spouse. And those who have more frugal desires or needs for income can accordingly work less and compel those around them to work on their behalf by invoking “just institutions.” (We can imagine a free-wheeling drifter who likes to travel, in effect compelling a factory worker to work longer hours so the worker can support his disabled child as well as the traveler who works less.) We can describe this as contributing to redistributive institutions, but the effect is the same as compelling others to work on our behalf.

Against this, I want to insist that the distinction really is meaningful. Consider the following comparison. You insist that I owe you money that I agreed in some contract to pay you under certain conditions that now obtain. I now raise an objection from self-ownership: I might as well deposit all the money I earn on Fridays into an envelope marked “For You.” “And don’t bother appealing to principles of keeping your promises or contracts,” I add. “These are but the functional equivalent of ownership claims, since either way I will need to work on your behalf, and all the more in view of my many obligations, including the orphans who depend on me.” Clearly something has gone wrong. The point is that we cannot generally infer from moral claims that have the implication that others owe us something that we are making an objectionable ownership claim over them, not even if they must work in order to meet their obligations. The nature of the claim we are making is a function of the moral principles behind it, not the actions required to meet that claim. In order to distinguish objectionable ownership claims, like those of a serf or slave owner, from unobjectionable moral principles that imply we owe something to others, we need only ask ourselves whether the claims assert control over other people, or merely dictate behavior as a consequence of principles that are in themselves directed toward moral ends that have nothing to do with such control.^{Footnote 14}

The upshot, then, is that no interpretation of egalitarian redistributive schemes is likely to be both charitable and turn on questions of self-ownership. Writers like Cohen do sometimes invite such a response, but our charity should extend beyond mere hermeneutics to broader dialectical charity, and at that level it makes little sense to argue over self-ownership. Likewise, mere self-ownership is never a defense against the demand that we fork over a part of our income, since such demands are never well construed as resting on ownership claims. However, just as in the case of property, I suggested abandoning self-ownership claims not because there was nothing to the relevant libertarian views but because there were far more plausible alternatives available; so something similar seems true to me in the present context. Libertarians should simply dispute the moral reasoning that produces the claim to our income without reference to self-ownership. The fundamental problem with redistributive taxation, libertarians should claim, isn’t that the state is becoming part-owner of us, but that it is demanding our property on the basis of false moral principles. Instead of debating self-ownership, libertarians should engage in straightforward debate over the moral principles governing the state.

How might this sort of exchange go? Just to make clear that it need not involve self-ownership, here is a rough sketch. Recall the earlier suggestion that Lockeans about property support their claims by making appeals to the sort of humdrum considerations that we invoke all the time to exclude others from control. (“I earned that money!” “I invented that contraption!” “I paid for that thing!”) If the target is supposed to be an egalitarian theory of justice that targets us for expropriation and redistribution, the right objection is simply that moral norms governing property aren’t adequately reflected in such a theory. Of course, the truth only emerges in the details of this dialectic, but in any case there’s no reason to think that things will turn on the self. The sort of doubts that will be raised will most likely include such things as whether our theory of justice adequately reflects the fact that there are nonconventional Lockean facts about who is entitled to what virtue of various activities, such as creation, discovery, trade, hard work, and so on.

III.

Even when self-ownership is clearly relevant to redistribution, any plausible delineation of the concept will take a form with stark limits that must be respected. To illustrate this point, it will be helpful to describe a specific account in a little more detail. Robert Taylor has argued that Kantians aren’t committed to the denial of self-ownership; in fact, they should see themselves as its most natural defenders. He makes three central claims. The first is that we have an obligation of physical noninterference toward one another, “that is, a duty to respect the bodily integrity of other people.” This obligation is a “universal perfect duty: all rational beings owe it to all other rational beings,” and thus is not subject to discretionary implementation.^{Footnote 15} The second claim is that these duties of noninterference are equivalent to a bundle of correlative rights that amount to self-ownership. The idea is that at least one plausible conception of self-ownership consists simply in a series of rights against others, especially to exclude them from trespassing upon us, or from using us in various ways. And these rights are precisely what we get in virtue of our reciprocal duties of noninterference toward one another.^{Footnote 16} And finally, Taylor points out that there is no reason for Kantians to object that such a conception of self-ownership involves seeing oneself as an object that could be treated as one pleases, since rights do not imply liberties:

So according to this Kantian conception, we have all kinds of self-regarding duties that significantly constrain our liberties or Hohfeldian privileges, but that is consistent with our possessing Hohfeldian claim-rights against others that amount to self-ownership. Self-ownership, in other words, is an entirely outward-facing notion, which prevents it from coming into tension with self-regarding Kantian doctrines.

The example of suicide is perhaps unfortunate, since it isn’t uncontroversial that we do in fact have a right against others not to interfere with that particular Kantian wrong, but the more general point about rights and liberties nevertheless holds. You certainly have a right against me interfering in minor private wrongs that you perform in your home, like telling lies to those you love, while not enjoying the moral liberty to do so. But as cases like suicide bring out, there are important questions about the strength and scope of this duty of noninterference. In particular, notice that this duty, while often strong, is hardly absolute. There are all kinds of sufficiently good reasons for you to trespass upon my person in addition to (arguably) preventing certain kinds of self-harm, for example to shove me aside so you can save your child from walking into traffic, to brush up against me to catch a baseball for your kid, or to pluck a hair from my head for some medically significant purpose. There will always be some disagreement about particular cases, but it seems clear enough that we do not have a perfect Kantian duty of noninterference. The key observation here is not the cliché end-of-the-world scenario that outweighs some supposed absolute right, but rather that rights come in different shapes and sizes.^{Footnote 18} Under certain circumstances—temporary suicidal despair, say—my right to noninterference is weakened; my general right against small-scale interferences like being jostled in public is always quite weak; and more importantly for our purposes, our right against interference in general is subject to balancing considerations, or what I prefer to think of as threshold reasoning, whereby those who would trespass upon our person must reach some threshold over and above a marginal utility gain, but usually not one that is insurmountable. Clearly our claim against those who would lay their hands on our organs or eyeballs is very strong—perhaps the threshold here is asymptotic—but not in virtue of a perfect duty of noninterference. If there were such a duty, we would face powerful moral constraints on shoving people or plucking their hairs or rustling up against them, whereas it is easy to think of cases in which such interferences are permissible once we breach some threshold of importance.

This means that any plausible conception of self-ownership will need to be structured in such a way as to allow for occasional violations. Perhaps some will wish to interpret this as a denial of self-ownership altogether, since they view anything worth the name as absolutely prohibiting trespasses upon our person. (This seems to be how Cohen viewed matters.) But as I have just argued, no such conception will turn out to be defensible, and a more porous version seems to me a worthy enough inheritor of the mantle. And if we do accept a conception of self-ownership that is subject to threshold-reasoning, then we must also accept that invoking self-ownership to settle disputes about redistribution won’t always be straightforward. This is because we won’t be able to rule out redistributive demands that infringe on self-ownership without substantive engagement with whatever moral considerations are at stake, without assessing how important these are relative to the infringements on our person, determining whether these breach the relevant thresholds, and so on. That is, we cannot simply invoke self-ownership in order to sweep away the relevant demands.

Let me work through an example. Suppose the egalitarian demands that the able-bodied work on behalf of the worse off—the sort of claim I was suggesting we avoid earlier on the grounds of charity. There is no pretense, in other words, that we owe a fraction of our income to the state on grounds of a fair division of the commons or the like; the claim is just that we are required to work for others against our will, perhaps according to an aggressive interpretation of the maxim, “From each according to his ability, to each according to his needs.” According to the account we are now working with, we cannot simply announce that trespasses upon our person are ruled out automatically. So we must consider and weigh up the extent to which our interests are infringed by the demand that we go work on behalf of others, as well as the interests of those whose needs will go unmet if we don’t work. This may look like victory to someone like Cohen, substantially similar to the simple denial of self-ownership.

Before taking stock, it is important to emphasize that our weighing-up should not take a utilitarian form, or at least that nothing I have said so far supports the view that we simply have to do whatever the marginal utility considerations favor. The fact that considerations of self-ownership aren’t automatically decisive does not mean that self-ownership, like other forms of ownership, doesn’t have deontic contours that must be respected. To illustrate the point, there are circumstances under which you are entitled to break down my door and take my property, even when I specifically deny you permission to do so, as when you are saving a child going into anaphylactic shock. But it would be a mistake to infer from this sort of case that there is nothing to the disposition of property beyond a calculus of net-benefit. There are at least two reasons for this. One is that non-utilitarians will of course deny that the mere fact of net-benefit is a sufficient reason to take someone’s property; they will insist either that there is some high threshold that must be met first, or similarly that we can attach extra weight to our own interests, or something of the sort. The mere fact that you would get more out of driving my car than I would does not mean you get to take it, short of fairly extreme circumstances. Second, there is a deontic structure that should inform our deliberations about these matters. For example, if you break into my house, you should try to make things right with me afterward. It is not the case that the net-benefits imply you did nothing wrong and therefore have nothing to discuss with me afterward. Either you or the third-party beneficiary have a duty to apologize, offer repayment, avoid such situations if at all possible, and so on. Likewise, backward-looking deontic features like desert make a difference. If you need to break down my door because you have been negligent in some way, that will alter what you are allowed to do, and what I may deny you. So when thinking about non-absolute ownership claims, including self-ownership, we must not collapse into simple-minded utilitarian reasoning about net-benefit—or at least libertarians need not do so.

With that clarification in place, we can return to the egalitarian’s demand that we work on other people’s behalf, and the fact that we must consider and weigh up the interests of those involved, though we are free to do so in characteristically deontic terms. In general, those around us must reach a high threshold before preempting our authority over what we do with ourselves, just as is the case with home-invasion. But there are sometimes sufficient reasons for invading people’s homes, and so we must contemplate the possibility of sufficient reasons for putting us to work. Of course, when it comes down to assessing these considerations, we might not be at all impressed. Suppose it turns out that without compelling people to work, there will be those who are unable to provide for themselves, who will consequently be dependent on the charity of others, and some people will receive no such charity and consequently suffer. Presumably it is something like this that is in the background of Cohen’s doubts about self-ownership. Some libertarians will view this as simply insufficiently weighty to breach the threshold at which we may exercise control over other people. Others might try something more subtle: they might concede that at least in some extreme cases the threshold is breached, but then insist that the afore-mentioned deontic structures kick in: perhaps those who compel others to work or confiscate their property owe them repayment (when possible) and apology, and have stringent duties to avoid burden-shifting in advance when that is in their power. A position of this sort—which I myself find plausible—concedes a fair amount to even Cohen’s egalitarianism, but maintains that questions of redistribution are much broader than the narrow focus on the permissibility of some act of forced labor or redistributive taxation would suggest. Conceding to the egalitarian view on these narrow questions does not mean conceding the broader point about repayment, duties to avoid burdening others, apology, and so on. Whether the resulting picture would look more egalitarian or libertarian (as I suspect) is not something we can glean from the narrow question of whether anyone can ever be required to make a transfer payment or to work on behalf of others.

The dialectical point, in any case, is that even when self-ownership really is relevant to redistribution, that fact won’t always settle the issue since no defensible conception of self-ownership implies correlative duties that are absolute. To make progress, we will rather need to discuss the complex structures of need, desert, liability, apology, and repayment.

IV.

Let me conclude by addressing more directly the fundamental question of whether we should think about debates about redistribution in terms of the self and attacks on our right to self-determination. The general strategy behind this approach is to root objections to redistribution in ideals concerning the separateness and moral independence of persons, and to draw attention to the ways in which mundane-seeming matters of taxation and government spending ultimately undermine those ideals—ideals now sometimes associated with Rawls, but really core ideals of the enlightenment. The strategy, as conceded earlier, is a natural one. If I insist that you must work on my behalf, or turn over a fraction of your wages to me, it is tempting to frame your objection in terms of an attack on you. To the extent that my demand interferes with your freedom to chart your own destiny, my demand may seem incompatible with a genuine recognition of your full and separate moral independence—the fact that you are not just a resource for me. To make a state-analogy, if we issue demands that another country pay us tribute or otherwise benefit us for reasons that aren’t independently justified—say, as compensation for wrongs they have committed—then it is natural to interpret this as an attack on their sovereignty, and ultimately the country itself.

We can think of examples in which this is just the right approach. Corvée labor on behalf of an aristocracy that views us as a resource to build its pleasure domes would be an apt case. Here, our independence obviously is under threat, and self-ownership or something close to it would be the right notion to invoke. But now consider two versions of contemporary egalitarianism, highbrow and lowbrow. For the highbrow version, we can entertain some variation on Rawlsian contractarianism, which instructs us to consider a hypothetical choice scenario in which we occupy the perspective of the worse off, or perhaps some form of luck egalitarianism, according to which being worse off through no fault of your own is an injustice. For a lowbrow egalitarianism, we can conjure a campaign speech calling for redistributive policies, of the sort one might hear nowadays:

Earlier, I suggested that it was uncharitable to interpret redistributive taxation as a form of compulsory labor. Here, let me point out that doing so misdiagnoses the deep roots of the disagreements in this area.

The deep disagreements at the highbrow level are likely to turn on such questions as whether there is a nonconventional moral basis for property that contractarian devices must respect, or whether we think we are morally entitled to shift our misfortunes onto others when we suffer from bad luck. At the lowbrow level, the disagreement is more likely to take the form of rival economic appraisals and varying pictures of what fair social cooperation looks like. These disagreements really are fundamental. It matters deeply whether we think that it doesn’t make sense to talk about your claim to your salary until we have fixed the background social institutions, or whether my bad luck is ipso facto your bad luck, or whether it is always unfair for some to gain more from a social setup than others. But none of these disagreements, it seems to me, is best interpreted in terms of a trespass on your person, and so none of them is best interpreted in terms of self-ownership. For the libertarian to make his or her case, the best approach is to try to develop a plausible Lockean theory of property in the context of a service economy, in order, for example, to rebut claims about the “myth of ownership”; or to contest claims about bad luck giving us permission to shift our burdens onto others; or to argue that we cannot infer the justice of a social setup from a cui bono principle, the way a detective assesses motive in a crime drama. None of these arguments is likely to be easy to make or uncontroversial, whereas the principle of self-ownership may seem simpler and harder to contest. And yet it points us in the wrong direction, away from the true sources of disagreement about justice.