Self-ownership has played a prominent role in moral and political philosophy. It at least seems to play this role for some people: the fact that we are self-owners grounds a range of essential rights that we have, and duties that others owe to us, that are central to the appropriate limits of state power, and that have an important role in governing our interpersonal relationships with each other.
For example, John Locke thought it self-evident that all Men are self-owners, who are entitled, in the State of Nature, ‘to dispose of their Possessions, and Persons as they think fit, within the bounds of the Law of Nature, without asking leave, or depending upon the Will of any other Man.’Footnote 1 From this, he drew substantive conclusions in political philosophy, for example that justified political authority depends on consent.
More contemporary scholars, such as Robert Nozick, think that it is some kind of objection to certain distributive views that such views would make some people part owners of others, and that this involves a shift from classical liberal ideas of self-ownership to shared ownership of people.Footnote 2 And left-libertarians, such as Michael Otsuka,Footnote 3 Hillel Steiner,Footnote 4 and Peter Vallentyne hope to defend the view that roughly egalitarian conclusions are consistent with, and follow from, the best understanding of self-ownership. Vallentyne, for example, argues for the plausibility of full self-ownership as a fundamental moral principle that is consistent with some egalitarian social principles that he goes on to defend.Footnote 5 Furthermore, some liberal egalitarians, such as Phillipe van Parijs, rely on self-ownership to constrain egalitarian theories.Footnote 6
While these authors are not always clear about the role that self-ownership plays in their arguments, it is natural to read them as drawing substantive conclusions about the centrally important substantive questions in political philosophy from premises about self-ownership. Whether we are self-owners or full self-owners, then, seems to matter because it makes a difference to what we should believe about such substantive issues.
As we can see, the idea of self-ownership has seemed to many to be important in determining the justification, scope, and limits of the state on a wide range of topics. And there is something plausible about this idea, understood quite generally, aside from the views of any particular thinkers. For example, some might argue that because we are self-owners, others cannot own us, or at least they cannot do so unless we confer ownership on them. And self-ownership might then be thought to explain why slavery is wrong. But this more limited case might be extended. Because we are self-owners, it might be argued, no one can force us to work for the sake of others. And that also places limits on the right of the state to distribute the products of our labor to others, and on the taxes that can be secured from us for the sake of others.
It might also be thought that because we are self-owners, the state is not permitted to secure bodily resources from us for the sake of others. This prohibits compulsory organ donation, even where donation is not harmful, it restricts the right of the state to use our bodily resources to help make scientific advances without our consent, and it limits the right of the state to secure personal data about us for statistical analysis in order to shape government policy.
Finally, developing a version of Locke’s idea, it might be argued that the state must shape public policy and laws only through proper consultation with the citizens who are governed by it, because otherwise the state would wrongly take control of citizens, which is inconsistent with their owning themselves.
Furthermore, there is at least some intuitive power in the idea that these quite disparate features of justified state power are related to each other—that they all depend on some basic facts about the self, and the normative control that individuals appropriately have over themselves. At least at first, it seems quite natural to rely on the general idea of self-ownership to explain the impression that there is a unified foundation to a wide range of policies and practices of a justified state.
Thus, determining the proper role of self-ownership in political philosophy is extremely important. Knowing its proper role helps us to understand not only what policies should be adopted, but also the way in which such policies should be justified. For example, many different arguments might be offered to justify democracy. So we may be confident that democracy is the best form of governance. But it is nevertheless important to know why, both because this will determine the particular kind of democratic system that we should adopt, and because it is important to understand not only what we should do, but also why we should do it.
My main ambition in this essay is somewhat methodological: it is to show that references to self-ownership are normally redundant and misleading in moral and political philosophy, and philosophers are wrong to rely on either their truth or their denial in defending views in moral and political philosophy. I can’t prove that this is universally true, but I will argue that it is almost always true that we should not argue for some substantive conclusion in moral and political philosophy by claiming, or denying, that we are self-owners. The idea of self-ownership is normally best bypassed to allow us to focus on more fundamental questions about the significance of the person and the non-consequentialist principles that are grounded in respect for persons.
But I also have two substantive aims. First, I provide general reasons to deny that we are full self-owners. Second, I suggest an alternative idea that those attracted to self-ownership would do better to rely on—the idea that facts about selfhood ground rights and duties. A full account of this idea is beyond the scope of this essay, but I explore it to offer a promising avenue of research for those who have relied on self-ownership in moral and political philosophy, and to show just how controversial the idea is. Once this idea is clarified, friends of self-ownership will have a better understanding of how it needs to be defended, and its opponents have a better understanding of a view that is closely related to self-ownership that should be the target of their critiques.
Here is the argument in brief. As the idea of self-ownership relies on the idea of ownership, we should spell out what ownership involves. Sections I and II make some progress with this question: I argue that X’s ownership of some object O is constituted by a sufficient set of ownership facts about the relationship between X and O. Ownership facts are types of fact that can be constitutive parts of ownership—tokens of such types are constitutive parts of ownership where sufficient other ownership facts about X and O obtain. Ownership facts are primary rights and duties.
Section III argues that in the light of this we do not have primary rights and duties because we are self-owners; we are self-owners because of the primary rights and duties we have. For this reason, self-ownership cannot ground primary moral rights and duties, and thus does not play an appropriate role in substantive moral argument. Section IV shows that while it remains a theoretical possibility that self-ownership grounds a further set of rights and duties—those that do not constitute self-ownership—there is no reason to think that this theoretical possibility is true in any but the most trivial sense. I show this by considering the most plausible example—secondary duties. Section V suggests that those who believe in self-ownership would be better understood as relying on claims about facts that constitute the self and our rights and duties. I briefly sketch an account of this kind and show that it has some plausibility, but also raise some questions about it.
I. Ownership and Self-Ownership
I am a self-owner if the idea of ownership applies to the case of the self. Thus, to understand self-ownership we must understand ownership.
A. Legal Ownership
Views in moral and political philosophy that rely on the idea of self-ownership treat it as a moral idea. This already creates a point of contention concerning self-ownership, since some might think that ownership is an essentially legal relationship—ownership is a product of the law, and we own property only because there is a law that says so.
This view comes in many variants. Some might think that a person, X, owns some object, O, iff there is a valid law that confers ownership of O on X. If legal positivists are right that there is no reason to think that valid laws are just, and if the validity of a law is insufficient for an obligation to obey it, the fact that X owns O tells us nothing about the moral entitlements that X, or anyone else has, with respect to O. If this is the right theory of ownership, then, the idea of self-ownership can play no role in a moral argument for a person’s entitlements over her mind and body.
Suppose, though, that there is an obligation to obey the law. It might similarly be argued that when the law confers ownership of O on X, it creates an obligation on others to treat X as owning O. So when X exercises the moral powers that come with ownership, these powers are morally effective. For example, if X decides that Y may not use O, X creates a moral obligation on Y not to use O because the law has conferred ownership on X.
However, even if this view is right, it does not justify the role that self-ownership has had in moral and political theory. As we saw in the introduction, the idea of self-ownership is supposed to play a role in determining what laws are just; it is supposed to explain why it is wrong to create a law that enslaves some for the sake of others, for example, or why certain redistributive policies are wrong. But the view under consideration would show only that if a valid law confers ownership on X, X acquires some normative powers that he would otherwise lack. The powers that he has would depend on the law. Thus self-ownership would not ground any restrictions on the law.
B. Moral ownership
Proponents of self-ownership are thus best off relying on a moral idea of self-ownership, and it is plausible that there is such a moral idea of ownership that corresponds to the legal idea.
Start with the content of legal ownership, by which I mean the set of legal facts that make it true that X owns O. When this set of legal facts obtains, X owns O because these facts obtain. These facts plausibly include a range of legal rights and duties that we can call “legal ownership facts.” Some standard candidates include: X’s liberty right to use O; X’s liberty right to destroy O; X’s claim right that others do not use, destroy, or otherwise affect O without X’s consent; X’s normative power to confer certain liberty rights on others, such as the right to use O; and X’s liberty right to transfer ownership on others with respect to O, with the normative powers required to do this.
X is a legal owner of O if the law recognizes that X has such a set of powers and rights that determine the legal relations between X and others and O. So, for example, X has the legal power to confer the legal liberty right on Y to use O if X can directly confer this right on Y as a matter of law, where if Y uses O, Y will have acted in a way that is legally permitted because of X’s act. This is a fact that can contribute to X being a legal owner of O. It is insufficient for X to be an owner, because non-owners can have this power—for example, trustees can have it. But it is a fact that can be a part of a larger set of rights and powers that together constitute legal ownership.
Now notice that there are moral equivalents of these legal relations. When these morally equivalent facts obtain, X is a moral owner of O. For example, X might have a moral liberty right to use O, which means that it is not morally wrong for her to do so; she might have a moral claim right that others not use, destroy, or otherwise affect O without her consent; she might have the normative power to give others the moral right to use O; and so on.
This leaves open the possibility that these moral facts obtain only when certain legal facts obtain. For example, Kantians think that there are no complete moral rights to property in the state of nature. But, of course, not everyone agrees that this is so, and it is plausible that it is not so.Footnote 7 Even some Kantians think that there are some personal rights in the state of nature—rights not to interfere with bodily integrity, for example. And they think that there are moral rights waiting in the wings that are ready to be completed by a state coming into existence.Footnote 8
Those who believe in moral self-ownership, then, are best understood as believing that people are moral self-owners because of the relevant range of moral ownership facts about the relationship between a person and herself. For example, a person is a moral self-owner if a sufficient number of facts of the following kind are true: X has the normative power to determine who is permitted to use X; X has the normative power to confer on others the normative power to use X; X has the liberty right to destroy X; and so on. This view depends on an understanding of what using and destroying X amounts to. I will briefly explore that question in the final Section. For now let us work with the standard assumption that X is constituted by her mind and body, and that using X involves using her mind and body; destroying X involves destroying X’s mind and body; and so on.
II. The Relationship between Ownership and Ownership Facts
Earlier, I listed some legal and moral ownership facts. That way of putting things was intentionally vague, to leave open different views about the relationship between these facts and ownership.
A. Necessary and Sufficient Conditions
One possibility is that these facts are the necessary and sufficient conditions of ownership. A standard approach to developing our understanding of a concept or idea in philosophy is to try to find the necessary and sufficient conditions of that thing. However, this method is also notoriously difficult to execute, and often unpromising. We have not managed to find the necessary and sufficient conditions of many things that are of central concern in philosophy.
This shouldn’t be too surprising, because concepts do not normally develop by people specifying necessary and sufficient conditions for their use. Rather, they develop by using them to refer to instances of the thing in question. A concept is used to refer to a range of related particulars, where small variations across many dimensions are compatible with successful use. Thus, it tends to be appropriate to use a concept to refer to some instance of a thing even when some of the constitutive features of that thing are not present.
Take elephants. Very substantial constituent parts of elephants can be removed in nearby worlds, and we still have elephants. Thus, there are worlds with trunkless elephants, pink elephants, small elephants, cold-blooded elephants, brainless elephants, and so on. This is so even though their trunks, brains, warm-bloodedness, size, and color plausibly constitute elephants. Remove too many of these features, and there is no elephant. A small, pink, cold-blooded, trunkless, brainless creature is not an elephant.
The same thing seems true of ownership. For example, normally owners are permitted to destroy what they own, but this is not always true: owners of historically significant property, such as famous paintings, are not permitted to do this; neither are pet owners. Furthermore, if a pet is for life, and not just for Christmas, it may be wrong to sell a pet. A contract prohibiting selling concert tickets to others once they have been purchased does not prevent people from owning these tickets. And so on.
This does not imply that a reductive analysis of ownership is impossible in principle, only that a reductive analysis will be very complex because it will include many “or” clauses. X owns O if sufficient ownership facts obtain about the relationship between X and O, where there are many alternative ways of satisfying the demand for sufficiency. For this reason, it will almost certainly be impossible in practice to state the necessary and sufficient conditions of ownership.
B. Grounding and Constituting
Here are two relationships that might obtain between moral ownership facts and ownership that are more informative, and that might tell us more about self-ownership. On one view, moral ownership facts non-causally ground ownership without constituting it. On another, they constitute ownership.
Whichever relation holds, there is a dependence relationship between ownership facts and ownership. Dependence relations of this kind might be thought of as a family, where something more complex, or less fundamental, is built from something simpler or more fundamental.Footnote 9 We can see this from the fact that if f grounds O, f naturally features in an explanation of O, and the parts naturally feature in an explanation what they are parts of, but not vice versa.
The distinction between f grounding O without constituting O, and f constituting O is difficult fully to explicate, which is not surprising given that these relations are likely basic. Nothing ultimately turns on this distinction, but it is helpful to grasp it.Footnote 10 Consider the relationship between parts and a whole, such as the relationship between the legs and top of a table and the table. Its parts constitute the table when they are appropriately related (that is, attached to each other in the right way).
Contrast the relationship between a law and a decision of parliament. It is plausible that decisions of parliament ground the law (they may not be the full ground, but even if not they are a partial ground).Footnote 11 The law is not simply caused by the decision of parliament; the law comes into being directly through the decision of parliament, and the idea of grounding captures this relation. But the law is also not constituted by the decision of parliament. The decision makes a proposed law (in the UK, a Bill) into a law, but the law is constituted by facts about its content and its legal authority rather than by the decision of parliament.
Substantive moral arguments often make claims about the grounds of duties; indeed this is at the heart of moral argument. For example, consequentialists claim that the right is grounded in the good; many nonconsequentialists think otherwise— that there are facts other than the good that ground duties, such as facts about personhood.Footnote 12
Now compare two views about the relationship between ownership facts and ownership. On one view, ownership facts ground ownership without constituting it, in the same way that a decision of parliament to enact a law grounds a law, or the way that consequentialists claim that the good grounds the right. This view, though, does not seem right. It is more plausible that ownership facts constitute ownership. When sufficient ownership facts obtain about the relation between a person and an object, these facts together constitute ownership. More precisely, there is a relationship between a person and an object, where some of those properties are ownership facts, and when a sufficiency threshold of such properties is reached the person owns the object.
Admittedly, there is no argument for this claim about ownership. Because relations like grounding and constituting are primitive, we should expect the claim in the previous paragraph to be self-evident on reflection if it is true, and indeed it is. We can now also draw a contrast between facts that ground ownership in a non-constituting way and facts that constitute ownership. The former set of facts include, for example, gift and purchase—the fact that X gives Y a gift, or Y purchases a good from X, grounds Y’s ownership of the property gifted or purchased. These facts do not constitute ownership; they ground it in a noncausal way.
III. Are We Self-Owners?
The view of ownership outlined suggests that a person is a self-owner, if she is, because a sufficient set of ownership facts obtains with respect to the relationship between a person and her mind and body, and that these facts constitute her self-ownership.
A. Full self-owners
Libertarianism is sometimes identified not only as the view that people are self-owners, but that they are “full” self-owners. If self-ownership is constituted by sufficient ownership facts obtaining with respect to the relationship between a person and her mind and body, here is a natural way of understanding full self-ownership.
There are many ownership facts—facts that can partly constitute ownership when they are members of a sufficiently large set of such facts. Full ownership might then be understood to obtain because all possible members of such a set obtain. For example, suppose that the right to destroy a thing is an ownership fact—it is the type of thing that can partly constitute ownership, when other ownership facts obtain. If I own a Picasso, but it is wrong for me to destroy it, I don’t fully own the Picasso. A person is a full self-owner, on this understanding, if all ownership facts obtain with respect to the relationship between a person and her mind and body.
Some libertarians sometimes add a further condition to their understanding of full-self ownership: what we might call “maximal equality.” Maximal equality of full self-ownership, as libertarians defend it, holds if the strongest set of ownership rights obtains about the relationship between a person and herself that is compatible with an equal set of facts applying to others. This is so if the “rights-grounding” facts about ownership obtain in a way that gives rights their maximal strength, with an equality constraint.Footnote 13
This view is, though, not completely clear, because it is not clear what makes one set of rights stronger than another. Suppose that the right that we are interested in is a claim right. We might understand the strength of a claim right as a function of the stringency of the duties that others have not to perform the act that the right protects. Or suppose that the right that we are interested in is a liberty right. We might understand the strength of a liberty right as a function of the value that will be set back through the exercise of the right. But then we just have different scales of strength, and it is not clear how to develop an overall scale for strength.
B. Are we full self-owners?
We can set these difficulties aside because it is hard to believe that people are full self-owners for these two reasons.
First, libertarians agree that there are important moral differences between the relationship between a person and her mind and body and the relationship between a person and external objects. There is no reason to think, at the outset, that these differences are all in one direction—that they all favor more ownership facts obtaining about the relationship between a person and her mind and body than the relationship between a person and external objects. Indeed, given the special importance of the mind and body to the value of persons over time when compared with external objects, there is good reason to think that there are some restrictions on the ownership facts that apply to a person’s mind and body.
Second, it is hard to believe that the only morally relevant facts are ownership facts. It is also hard to believe that when self-ownership facts compete with other morally important facts, the former facts always win. Full self-ownership, though, implies otherwise.
These general arguments are borne out in examples. With respect to the first, there are things that owners are typically permitted to do with their property, and that seem clearly to be ownership facts, that do not obtain with respect to the relationship between me and my mind and body. Most importantly, I am permitted to transfer ownership of my property to another person for any reason at all. When I do, I give up ownership and confer it on another. But I do not have this normative power with respect to my mind and body—even if voluntary permanent enslavement is sometimes effective (something which many people will doubt) it is surely ineffective where there is no good reason for me voluntarily to enslave myself.
Several examples demonstrate that some ownership claims about the relationship between my mind and body are not true. Consider:
Low Risk: I go out driving very safely, creating a tiny risk of harming others.
Party: I hold a party for my kids in my garden, knowing that this will be a bit noisy for my neighbors.
I am permitted to do both these things, but it is not clear that they are compatible with full self-ownership.
Perhaps it might be argued that they are, because full self-ownership implies a liberty right to use my car and garden to secure my ends. But if this is right, it is hard to know what the implications of full self-ownership are—we need some kind of proportionality limit to determine the scope of my liberty rights, and the idea of full self-ownership then relies on independent considerations about the significance of harm, the Doctrine of Doing and Allowing, and so on.Footnote 14
Furthermore, consider:
Little Toe: A trolley is heading toward one billion people who it will kill if nothing is done. I can save the one billion only by diverting it onto a side-track, but if I do, Victim’s little toe will be bruised.
Obviously, I am required to save the one billion people. But full self-ownership, as I have defined it, implies otherwise. Here is why. Owners have rights not to have their property damaged, destroyed, or interfered with in other significant ways without their consent, and this is an ownership fact. If Victim is a full self-owner, Victim has a right not to have her little toe bruised on this basis. However, the one billion people do not have an ownership right to be saved from death. They have a right to be saved, but this right is not plausibly something that makes it true, in conjunction with other rights, that they own themselves.
To be sure, owners may have rights that others take steps to protect their property. But this right is not plausibly one of the things that constitute ownership. It is not a part of ownership, but rather an independent right that owners may enjoy over their property. Thus, the view that we are full-self owners implies that it is wrong for me to save the one billion, and this is obviously false.
Peter Vallentyne, Hillel Steiner, and Michael Otsuka respond to objections like this. In response to the problem of voluntary slavery, they claim that although this implication of full self-ownership is counterintuitive, we must balance that implication with the fact that the idea of full self-ownership itself has both intuitive appeal and theoretical simplicity. It is hard to see why full self-ownership is simpler than partial self-ownership—these are just different views about the scope of our bundle of rights and powers, and there is nothing simpler about a more extensive rather than a less extensive bundle.
In response to the second set of concerns, they distinguish full self-ownership in the strict sense and full self-ownership in a looser sense, where full self-ownership in the looser sense permits interfering with ownership rights when either:
1) There is only a very small probability that an act will result in an incursion on another;
2) If there is an incursion, the harm suffered will be trivial;
3) The harm was not reasonably foreseeable; or
4) The benefits to others of performing the action are enormous (for example, the avoidance of social catastrophe).
This second response is just a fig leaf, though. For these limits to be consistent with the idea of full self-ownership, we need a principled way to distinguish what makes self-ownership “looser” and what makes it “less than full.” Vallentyne, Steiner, and Otsuka provide no way of doing this. The idea of “loose full self-ownership” is just tantamount to acknowledging that we are not full self-owners, as Otsuka indeed used to acknowledge.Footnote 15
Once we see this, we have further reason to doubt the first response. For whatever the virtues of simplicity, and whatever the intuitive appeal of the general idea of full self-ownership—and I am skeptical of both things—Vallentyne, Steiner, and Otsuka do not believe in full self-ownership, and nor does anyone else who has moral views that are remotely plausible.
These objections do not count decisively against self-ownership because although most owners have the normative power to transfer ownership rights to others by simple declaration, and this power is a constitutive feature of ownership, this is not a necessary condition of a person being an owner.
Given that it is difficult to spell out exactly what facts are ownership facts, or to determine, even roughly, the sufficiency threshold for ownership, it is difficult to know whether people are self-owners. And this is consistent with my pre-theoretical intuitions about self-ownership—there is something similar about the relationship between me and myself and the relationship between me and the external objects that I own, but there are also differences. So it is not clear whether we are self-owners.
IV. Bypassing Self-Ownership
Fortunately, in order to make progress with the kinds of substantive issues in moral and political philosophy that are the subject of dispute between libertarians and their critics, we don’t need to make any progress with the question whether we are self-owners. I noted in the introduction that political philosophers seem to rely on self-ownership as a premise in an argument for substantive conclusions in political philosophy. Here I show that they ought not to do so.
The theory of ownership that I have defended claims that self-ownership is constituted from some deontic material, and that deontic material grounds further deontic truths. For example, I have the normative power to determine whether others are entitled to interfere with my body, and exercise of that power determines whether others are required not to interfere with my body.
Given this, some arguments in favor of deontic propositions that rest on self-ownership are viciously circular. Others are superficial; they are not strictly circular, but they are close enough to being circular. There are no arguments that rest on self-ownership that have the depth necessary to justify substantive conclusions about primary rights and duties in moral and political theory because they do not offer the grounds of deontic propositions. Self-ownership, in other words, is best understood as the conclusion of moral and political argument in favor of primary rights and duties rather than as a premise in such an argument.
Let me explain. Suppose that X is a self-owner because of some set of ownership facts {f} about the relationship between X and his mind and body. One such fact is that X has the normative power to determine whether others are permitted to interfere with his bodily integrity. But then, X’s normative power to determine whether others are permitted to interfere with his bodily integrity is not grounded in self-ownership. Self-ownership cannot, then, be a premise in an argument for the normative power. Either something else grounds the power, or it is morally basic. The fact that such a duty is a constituent part of self-ownership makes the idea that others have such duties because people are self-owners seem plausible. But it is false.
Compare the following. Suppose that I see a creature with a trunk, and I ask why it has one. You say: because it is an elephant. If this “because” is intended to offer some cause or ground of the elephant having a trunk, what you say is false. What you say may sound close to the truth: if this is an elephant, a sufficient set of elephant-constituting facts must obtain, and this is one of them. But you have not offered a reason why the creature has a trunk.
In response, some left libertarians might claim that self-ownership is a basic moral idea that grounds further moral ideas. Vallentyne, Steiner, and Otsuka, for example, claim that the idea of full self-ownership is morally appealing in itself. Others may believe that self-ownership, even if not full self-ownership, is appealing in this way. Perhaps those that find this appealing will think that self-ownership is morally basic, and can ground further moral claims, in a way that is familiar from intuitionism. But if my understanding of self-ownership is right, it cannot be morally basic in the strict sense—its parts constitute it. The parts may all be morally basic (though I doubt that they are), and that is the only sense in which self-ownership may be morally basic.
For this reason, the idea of self-ownership cannot play the role that the good plays for consequentialists, or that personhood plays for many nonconsequentialists—consequentialists claim that the good grounds duties; nonconsequentialists that personhood does so. They may be wrong about these things, but these are the kinds of things that can be grounds of duties. Self-ownership, in contrast, is not.
V. Self-Ownership and Secondary Duties
It does not follow that no deontic claims can be defended by appealing to self-ownership, only that those claims that partially constitute self-ownership cannot be defended in that way. A person may have a bundle of rights and duties that constitute self-ownership, and she or others may have a further set of rights and duties because she has this bundle.
This follows from the more general idea that some set of facts {f} may constitute some entity, E, and E may ground some further fact. The fact that this is a table partially grounds my knowledge that it is a table, for example. My knowledge that it is a table is not grounded simply in the constituent parts of the table; the fact that it is a table cannot be eliminated in a complete grounding explanation of my knowledge that it is a table.
Similarly, some might claim that self-ownership is constituted by certain rights and duties, but that once it is constituted thus, self-ownership itself, and not simply its constituent parts, grounds further rights and duties. If so, self-ownership plays a role in moral and political philosophy that cannot be eliminated. Although this is a theoretical possibility, I doubt that it is true; or rather, I doubt that the way that it is true, if it is, gives it much importance.
First, let us acknowledge that ownership can itself sometimes ground certain rights and duties, rather than just being constituted by those rights and duties. For example, suppose that a law decrees that only those who own a certain amount of property have a right to settle in a country, and that law is valid. The right to settle in a country is not constitutive of ownership, and it arises as a result of ownership. So this is a case where the fact that a person is an owner grounds some independent right or duty.
This case is superficial, though, in this sense: the right to settle arises only because ownership has been incorporated into the law as a trigger of the right to settle. So although some rights and duties may arise as a result of ownership in this case, ownership does not provide a deep explanation of those rights and duties. It is the authority of the law, rather than ownership, that deeply explains the right to settle.
Some might think that ownership, and also self-ownership, can provide a deeper explanation of certain secondary rights and duties. A secondary duty is a duty that exists in virtue of the violation of a primary duty. As ownership, including self-ownership, is constituted by a set of rights, and thus a set of correlative duties on others, violations of the rights that constitute ownership or self-ownership ground a set of secondary duties.
For example, suppose that X interferes with Y’s body, harming Y. The fact that Y is a self-owner, it may seem, can ground X’s duty to compensate Y for the harm Y suffers. Even in this case, though, the best explanation of X’s duty to compensate Y does not refer to self-ownership. It is better to refer directly to the source of the particular duty that X has not to interfere without Y’s consent. For example, suppose that Y has the power to determine whether X is required not to interfere with Y’s body because of the importance of Y having autonomous control over what happens to her body. X’s duty to compensate Y is then grounded in the fact that X has failed to respect Y’s autonomy by failing to respond to a duty that is grounded in Y’s autonomy. This explanation is more direct and more illuminating than an explanation that appeals to self-ownership. Appeals to self-ownership, in this case, are at best misleading, because they are put forward as though it is the whole of self-ownership, rather than one of its constitutive parts, that grounds X’s secondary duty.
Self-ownership could play a role in explaining secondary duties only if the fact that a person has sufficient rights and powers to meet some threshold that implies she is a self-owner has normative significance in itself. But there is no reason to think this. Like ownership, self-ownership can be a useful way of referring to a set of rights and powers that a person has, but it does not have independent normative significance.
VI. The Self in Parts
Even if self-ownership is the conclusion of a moral argument, rather than a premise in a moral argument, there may nevertheless be some general premise in a moral argument that partially grounds a range of rights and duties that together constitute self-ownership. Those who think that we are self-owners should look to facts about the self to play this role.
If this idea could be vindicated, it would give friends of self-ownership much of what they want. It would show that the rights and duties that together constitute self-ownership have a common source in the moral significance of the self. It would show that the rights and duties that constitute self-ownership are related to each other because they are grounded in this common source. And that would explain why this collection of rights and duties seems to stand and fall together. It would thus help to show that the collection of facts that make self-ownership true are not simply related to each other in an ad hoc way, but rather have a common source.
Furthermore, this would vindicate the intuition that I noted at the outset—that a range of policies that might otherwise seem unrelated seem justified because they appropriately respond to personhood. If this is true, there is a deep response to the charge that Lockean views that distinguish significantly between our entitlements to our mind and body, and our entitlements to natural resources, lack coherence.Footnote 16 The distinction between these entitlements might be explained in a quite general way, even if there is room for disagreement about what, exactly, our entitlements over either thing are. Lockean views of a more moderate or more extreme kind will have coherence because of the common source of the relevant rights and duties, while leaving open the range of competing considerations that fully determine our rights and duties both over our minds and bodies and over external resources.
To see that this approach has at least some initial promise, let us explore in a little more detail the fact that various rights that we have over interference with, and control of, our minds and bodies do have a common source. Consider our right not to be enslaved and manipulated, our right not to have our bodily organs removed without our consent, our right not to have our private thoughts made public, or to have those thoughts interfered with, our right not to have our genetic material stored and explored without good reason, and our right to determine what will be done with our remains after we die. These rights are all concerned with certain kinds of intrusive interference or control. But they seem related to a wider set of rights, such as the right to have a say over the laws that govern us, the right to be free to pursue our own ends, and the right to form cooperative relationships with others to pursue mutually valuable goals. Although these rights are in some ways quite diverse, they do seem to have a common source.
Furthermore, it seems that the relationship between these rights is not explained simply by the fact that they promote our interest in welfare or autonomy. I have these rights even when performing the acts that violate them does more to protect the interest that others have in welfare and autonomy than mine is set back. It is wrong to remove one of my kidneys, for example, even if doing so will enable another person to be protected from a significantly graver harm. The right to live in a democratic state is not completely dependent on democracy securing our independent interests more effectively than other political arrangements, and paternalistic interference is wrong even when others know better than I do what will satisfy my interests. These rights seem stringent, quite independently of considerations of welfare, because they are concerned with deep and intimate properties that I have as a person.
Although self-ownership cannot explain this group of rights, for reasons I have offered, they might nevertheless be explained by the fact that they concern what is mine in another sense. When I refer to a hand, or an arm, or a thought as mine, I need not mean that I own the hand or the arm or the thought. I may rather mean that the hand, the arm, and the thought are constituent parts of me. To see this, note that I can describe a leg as belonging to a table, or a trunk as belonging to an elephant, without implying that the table owns the leg or the elephant owns the trunk. What I mean, rather, is that the leg is a constituent part of the table and the trunk is a constituent part of the elephant.
It might then be argued that the fact that my hand, arm, and thoughts are constituent parts of me grounds the rights and duties that are ownership facts about these things, and that may ground self-ownership. In other words, our minds and bodies constitute us, and the fact that we are constituted by these things is itself morally significant. Its significance grounds a set of rights and duties that are parts of a whole that makes us self-owners; these rights and duties thus have a common source.
Both the claim that our minds and bodies constitute us, and the claim that this fact is morally significant, are controversial. The first claim is metaphysically controversial: many people think that our bodies do not constitute us. The second claim is morally controversial: it is not immediately transparent that there is an important moral difference between constituent parts of a person and external objects. Obviously, I cannot explore either of these questions with the depth they deserve, but here are some remarks.
A standard view in the metaphysics of personal identity is that identity does not go with the body, but rather goes with the mind, or with the mind when the same brain supports it. Those who are drawn to this view of personal identity, the view that rejects the significance of the body to personal identity, typically do so because of thought experiments like this:
Brain Transplant: my brain is transplanted from my skull into your skull, replacing your brain.
I have acquired a new body; you have not acquired a new mind. Although some find cases like this hard to imagine and evaluate, this conclusion seems sufficiently uncontroversial that a good theory of personal identity should either explain it, or at least be consistent with it.
However, the truth of the view that I go with my brain in Brain Transplant does not support the view that the body is not a constituent part of me. It only supports the view that it is not a necessary constituent part of me. Recall my discussion of elephants earlier. Suppose that I cut off an elephant’s trunk and fix it onto a trunkless body. The elephant has lost its trunk; it has not gained a trunkless body. Identity goes with the remainder of the elephant’s body, not with the trunk. But this does demonstrate that the trunk is not a constituent part of the elephant—it obviously is. It just demonstrates that it is not a necessary constituent part of the elephant.
Similarly, some parts of my body are constituent parts of me, even if they are not necessary constituent parts of me. Consider, my leg, my hand, my eye, my fingers, and so on. These are parts of me. The fact that I persist when any of these parts is removed, or even when all of these parts is removed, does not demonstrate otherwise. Some parts of me are sufficiently large to be necessary constituent parts of me. The whole of the brain, for example, is both part of me, and a necessary constituent part of me. But parts of my brain are parts of me, but not necessary constituent parts of me—we can see this from the fact that part of my brain can be removed, leaving me without that part. The whole of the brain is not the only constituent part of me. It is not even the smallest necessary constituent part of me—if a large part of my brain were transplanted into another person’s body, identity would go with the large part of the brain, rather than with the remainder of my brain and my body.
In contrast, external objects, even those that I am intimately related to such as my house or my clothes, are not parts of me. And there are some borderline cases, where it is not clear whether something is part of me, such as my hair or my fingernails. There may be no fact of the matter whether they are part of me.
Now, as is familiar, relationships of rational concern do not go with identity. We know this from fission examples—if the two halves of my brain are each transplanted into two bodies, I am not identical with either of the two people who then exist. Yet it may be rational to be concerned about both in just the same way as it is rational to be concerned about my future self—what is sometimes called “egoistic concern.”Footnote 17
It also seems rational to be concerned about these future individuals in a different way than it would be rational to be concerned about a mere replica of me. I have reason to be concerned about what happens to me in the future, and the products of fission, in an egoistic way, where I have reason to be concerned about what happens to my replica in a way that I have reason to be concerned about others.
This whole discussion of personal identity is consistent with the view that my relationship of concern is grounded in the relationship between my constituent parts and me. Even in the case of fission, where personal identity does not hold, the relationship between the two people I am divided into and my current self is significant because my constituent parts persist in these future individuals.Footnote 18
Now consider some basic rights, such as the right not to be used that nonconsequentialists defend. To take a very familiar case, suppose that I am thrown in front of a trolley in order that my body prevents the trolley from killing five. This seems wrong because I am used to save the five. But it might be objected that I am not used at all. My body is used, but some people think that my body is not me. My embodied mind, they claim, is identical to me, but my embodied mind is not situated in the parts of my body that are used to stop the trolley.
In response, my body is a constituent part of me, and that is used to stop the trolley. This plausibly makes it especially wrong to use my body to stop the trolley. And intuition tells us that the principle that it is wrong to use people extends not only to facts that are necessary constituent parts of a person, such as the brain, but constituent parts that are not necessary in this way.
Because they are parts of me, parts of my body and brain are not proper objects of redistribution on standard egalitarian or prioritarian grounds. In this way, my mind and body have distinct normative importance compared with external resources. When it comes to external resources, it is plausible that who gets to use what is determined by some familiar distributive principles, such as egalitarian or prioritarian principles. But this is not true of my mind and body because they are constituent parts of me.
For this reason, contra G. A. Cohen, it might be argued that the significance of the self does seem to make it especially wrong to transplant an eye from a baby with two eyes into a baby with none simply because doing so satisfies egalitarian principles. Footnote 19 Unlike the case of external resources, an argument in favor of doing this cannot simply be that this gives each person equal opportunities or resources. The fact that the eyes are a constituent part of the baby with two of them makes it harder to justify the transplant than it would be to justify redistributing external resources that happen to be available to one person rather than another.
Cohen claims that we would be just as troubled in the case where some people get two eyes by chance as a result of their standing under ocular trees where eyes fall into their sockets as we are in the case where they come to have these eyes naturally.Footnote 20 This case does seem to me less clear, though, partly because it is plausible that the eyes that fall into a person’s sockets become a part of the person once they do so. Where the eyes have fallen into a person’s hand, and are not yet part of the person, it seems to me very clear that the eyes should be distributed fairly, so that each person gets one.
Now, I should admit that it is hard to argue further for the view that I have sketched here. It is hard to see why the physical integrity of a person that is constitutive of her identity has the moral significance that it seems to have. A more revisionist view, where we redistribute physical resources that people are born with to ensure equality of welfare, or equality of opportunity for welfare, may seem more plausible to some. All I will say at this point is that the idea that physical integrity has fundamental moral importance is highly intuitive, and whether we should accept or reject it depends to a degree on the extent to which our intuitions should be taken as guides to moral truth. That is a difficult methodological question that I cannot answer here.
So far, we have focused on the use of bodily parts. We can now see, at least in outline, though, how to extend this idea to a wider set of rights. Consider the right not to have my mind and body controlled by others, even where this maximizes my welfare. This right might be justified by the fact that it is appropriate that I have control over what happens to my body, and this is so because the self is an integrated whole. The appropriate way to respond to a self, which is an integrated organism with a mind and body, is to respect the decisions that this self makes for itself.
If this idea can be defended, and it has at least some initial plausibility, it would help to vindicate a wide range of moral and political rights, such as the right to have a say in the laws that govern a person, and the right to be free to form associative relationships with others to pursue common goals. These rights all seem grounded in the fact that it is appropriate that we govern our minds and bodies, where we have the relevant set of rational capacities, because our minds and bodies constitute the self.
This, of course, is no more than a sketch. But it is a sketch of something that has the kind of initial appeal that, I think, draws people to believe that self-ownership is an important moral idea. While it is very hard to develop and defend, it is also more plausible as a foundation of the justification and limits of the liberal state than self-ownership.
VII. Conclusion
Disputes about self-ownership, including full self-ownership, have been at the heart of political philosophy. This debate, I have suggested, has not been well conducted. I propose replacing this debate with a debate about the moral significance of the self. I have suggested some reasons to favor the view that the self has moral significance, and underpins a range of rights that have led some to believe in self-ownership. I have also argued that full self-ownership, at least as I have understood it here, is not a plausible view. Furthermore, I have offered some reasons why the idea that the self is morally significant in the way I have suggested is a bit mysterious.
A deeper exploration of the moral significance of the self promises to have implications in a wide range of serious ethical issues, many of which have been topics of debate between friends and opponents of self-ownership. A clearer grasp of the controversial underlying issues in morality and metaphysics that I have identified will not immediately resolve these issues, but I hope that it will at least encourage us better to understand them and the kinds of argument that are needed to resolve them.
