My body and other objects: The internal limits of self-ownership

Eur J Philos. 2019;1–18. Abstract Common practices such as donating blood or selling hair assume rights of disposal over oneself that are similar to, if not indistinguishable from, property rights. However, a simple view of self‐ownership fails to capture relevant moral differences between parts of a person and other objects. In light of this, we require some account of the continuity in the form of ownership rights a person has over herself and other objects, which also acknowledges the normative differences between constitutive parts of a person, on the one hand, and external objects, on the other. This paper provides such an account by arguing that there are reasons internal to a general justification of property rights to limit the extent of powers included in ownership of different kinds of object, depending on how the person is situated in relation to them. Rejecting a typical Hohfeldian view of property as a univocal, gradable concept allows us to make space for a new approach to property and self‐ ownership: one which can make sense of various uses of the body as property without entailing that our relation to those parts is exhaustively characterised by an ordinary property right.


| INTRODUCTION
The standard view of property rights among legal scholars and philosophers alike is a Hohfeldian one. According to this view, property consists of rights, not things, and ownership consists in a bundle of claims, privileges, powers, and immunities held against others with respect to some thing. 1 This bundle theory of property is conducive to thinking of ownership as a gradable concept, whereby to have "full" ownership of a given object entails having the logically strongest set of possible incidents over that thing. According to this assumption, any bundle of property rights that holds less than this maximum set of incidents counts as something less than "full" ownership. This understanding of the structure of property rights has strong implications for theorising self-ownership as a principle of autonomy. For if the principle of self-ownership is supposed to be constitutive or explanatory of autonomy, then it would seem that anything less than "full" self-ownership would equate to something less than "full" autonomy. 2 Any denial or restriction of any logically possible stick in the bundle of property rights that could make up self-ownership would need special justification. This view treats the idea of property as a univocal concept which takes the same form no matter what kind of object property rights are taken to range over.
In the literature on self-ownership, this leads to a symmetry being assumed between the power of selfownership and the power of ownership over ordinary objects-that in both cases, the more incidents of property rights one holds over the thing that is owned, the "fuller" the right of ownership. 3 It is this assumption that underlies some of the most controversial conclusions about self-ownership, such as Nozick's permissive stance on voluntary slavery, or his argument that taxation amounts to a form of slavery. This jars with critics such as Alan Ryan (1992):

It is just because we take a relaxed view about people's rights over their cars, bicycles, books and the rest that Nozick can suggest that if these are my lungs, I can do what I like with them. Conversely, it is just because we don't take a relaxed view about people's rights over their bodies that Nozick can suggest
that we have no right to tax people against their will, just as we have no right to force them to marry against their will.
The crux of Ryan's point is that property rights, coming as they do with powers to alienate, are specifically applicable to ordinary, disposable objects precisely because they have relatively little moral importance. Our bodies, on the other hand, require a different set of rights to adequately reflect their different moral nature. To assert a symmetry between "full" property rights and "full" self-ownership as Nozick does is to try to play these conflicting intuitions both ways. Ryan and others take this as a point of departure from the idea of self-ownership, satisfied that this tension undermines the case for using the concept of property as a framework with which to express a fundamental principle of autonomy. This paper proposes an alternative response to Ryan's criticism, by suggesting that there is room within the Hohfeldian conception to resist the univocal view of property sketched above. Instead of thinking of "full selfownership" as consisting in the logically maximal set of Hohfeldian incidents each person could hold over herself, I argue that we can identify reasons internal to a general justification of property rights that justify limiting the extent of powers included in ownership of different kinds of object (including parts of one's body). This approach paves the way for an alternative view of ownership rights, one which provides the tools to explain differential limits to powers of ownership and alienation with respect to different kinds of object, depending on the way in which we relate to the object in question, and variable across different contexts. 4 The term "object" here is meant in a broad sense to include intangible goods which may become objects of intellectual property through trademarks, copyrights, or patents. Rejecting the homogeneity of the structure of property in this way allows us to make sense of various uses of the body as property in some contexts, without committing to the view that our relation to those parts is exhaustively characterised by a fixed "full" bundle of property rights.
The central proposal is as follows. Property rights are justified on the basis of protecting a basic normative power of control for persons and of enlarging the scope of objects over which a person is able to exert that control. 5 In order to determine the limits of property rights, we need to understand the way in which we relate to various kinds of object (including tangible and intangible objects, as well as parts of one's body) and how different uses of those objects impact that basic power. This pushes back against the assumption that a "full" property right in anything must entail having the maximum set of incidents of ownership that is logically conceivable 6 and that allowing any welfare-based considerations to bear on restricting that set is to invoke utilitarian trade-offs. Instead, it provides an alternative way of understanding what constitutes "full" ownership of some object, by explaining how object-and context-dependent limits on powers of ownership arise from reasons that are internal to a general justification of property rights. The internal limits in question can be determined by the extent to which a given framework of rights serves to protect or enhance this basic power when deployed as a way of regulating our use of different kinds of object.
Part of this project involves disentangling the notions of "property" and "ownership" by examining more closely the structure of relations of ownership-that is, the relations in which an owner stands to the objects she calls property, as well as to other people. I suggest that in order to understand property as a relational framework, it is necessary to explore further the nature of the relation between the person, the body, and other objects and to explain the significance of how the person is situated in this relation. And it is the very nature of these relations in different contexts which places internal limits on the way in which ownership can be construed. This allows us to conceive of the structure of ownership rights over oneself as continuous with the structure of ownership rights over ordinary objects, without assuming that the concept of ownership applies univocally in all cases. We can still recognise that there may be other values which come into play in the wider debate about how to shape and limit property rights. But understanding the way in which systems of property may be limited on their own terms allows us to get a grip on complex debates surrounding how to regulate uses of the body and other objects of personal importance, without having to draw on controversial essentialist claims about the value of certain things. It also provides an effective riposte to the libertarian who would staunchly oppose any limits on self-ownership on the basis of the primacy of the value of autonomy.

| PERSONS VERSUS PROPERTY
Some initial work is required to further motivate this approach. Looking back to Ryan's critique, one plausible way to understand the mistake attributed to Nozick is that it stems from a failure to recognise that there is a sharp distinction between persons and objects. And it is a mistake to treat the body as one among the many objects we can have property over, because when we are talking about rights, the body falls within, or perhaps constitutes, the boundary of the concept "person." Once we recognise this, it might seem obvious that the kinds of rights we attribute to persons differ substantially to the kinds of rights that persons can have over objects. Persons are the subjects of rights, whereas things are the objects of rights held by persons. In Ryan's words, we do not take a relaxed view about people's rights over themselves. As such, it would be appropriate to draw a sharp distinction between property (rights over objects) and rights over one's own person.
This interpretation ignores a crucial complicating factor, which is that people's bodies are objects too, and there are many respects in which we treat them as such. Notably, one thing that does distinguish one's body from other objects is that it is not something one must acquire in order to have rights in it. It is a constitutive part of one's person. This pretheoretic difference may be sufficient to establish natural rights of exclusive control for each individual over her own body. However, it leaves open the question of whether and to what extent those rights may be alienable, as property rights in external objects are ordinarily taken to be. Moreover, many of the ways in which we treat our bodies or body parts as objects do not involve any intellectual or moral mistake but rather arise from a reasonable interest in doing so.
Examples of such cases bring out the way in which the supposedly distinct spheres of person and object are much more closely interwoven than is often acknowledged. Although the link between the structure of rights over persons and those over objects may not be as continuous as assumed in the traditional literature on self-ownership, there is a need to be able to bridge the gap between the two spheres. I suggest that there is some basis for thinking of the structure of rights over the person as continuous with the structure of property rights over objects, while distinguishing differential limits to the scope of those rights as they pertain to different objects. That is, certain rights of the person are conceived as inalienable, some rights over objects (including aspects of the body) as potentially alienable, and others as straightforwardly so. We can do this without assuming that any limitation on the scope of an ownership right must arise from a trade-off between values external to the justification of property and the interests it centrally protects.
An example that illustrates the problem of assimilating the body under the abstract concept "person" and distinguishing this sharply from mere objects is the case of blood transfusions. Jean-Pierre Baud (2007) explains that this distinction came under real scrutiny in French law towards the middle of the 20th century, as technological advances changed how blood transfusions were carried out. Under French civil law, the body had been protected under the umbrella of the abstract legal notion of the person. As such, the body and its parts were not considered alienable, except in cases of slavery, where such alienation was just a consequence of the total alienation of the person. In cases where body parts had been severed, these were treated as minor cadavers, with funeral rites accordingly imposed for "notable parts of the body." When blood transfusions became possible at the beginning of the 20th century, they were initially carried out arm to arm and were therefore not construed as donations. Instead, they were characterised as a medical procedure carried out by a doctor and made possible by the act of a "donor." 6 By the 1950s, however, procedures in which blood was collected and stored outside the human body before being transfused into the patient had become commonplace.
Reluctance to recognise the blood as an object under the law meant that sale of human blood and blood products was not viewed as a transfer of property but instead labelled as "deliverance against payment" ("délivrance à titre onéreux"). Pharmacies did not buy blood in order to sell it on but rather accepted products to be "deposited in dispensaries." All this equivocating stemmed from an overarching concern to protect the dignity of the person and the sanctity of the body. The legal status of blood was thus largely set out in negative terms, on the insistence that blood was not a commodity and should not be considered to be a medicine-blood was not a thing. The result was that human blood and blood products were distributed for 40 odd years without having a determinate legal status.
As Baud explains, from a legal point of view, it was as though the blood did not exist. This was made evident in a most serious manner in the wake of the contaminated blood scandal, in which it was revealed that the Centre National de Transfusion Sanguine had knowingly provided transfusions of blood contaminated with HIV to haemophilia patients in 1984 and 1985.
Representatives of the transfusion centre argued on the basis of the legal non-existence of blood that the plaintiffs had no recourse against having been provided contaminated material. In a perverse twist of argument, they challenged as follows: How dare the court sanction such an affront to the dignity of the person by deigning to recognise blood as having the status of a dangerous product. To do so would entail considering the body to be a thing (Baud, 2007, p. 774). In response, the Paris Court of Appeal and a judgement of the High Court of Toulouse finally ruled in 1991 and 1992 that blood was a thing that could be bought and sold, categorising it somewhat inappropriately as a "substance hazardous to human health." 6 Donating parts of one's body amounts to the kind of act of disposal that we usually take to be the prerogative of a property owner. Baud explains that some theorists would like us to distinguish between "property ownership" and "belonging" ("la propriété" and "l'appartenance"), such that someone's hair can be said to belong to her, yet only becomes her property once cut from her head. Baud suggests that this is mere word play. If the rights entailed by my hair belonging to me include the right to cut it off and sell or donate it, then I must have had the right to alienate my hair as property all along. Any form of "belonging" which includes an exclusive right to dispose of the body part in question already includes those features of alienation which come with property ownership.
The critics' point is perhaps more clearly pressed as the suggestion that I could not transfer rights of ownership over my hair until it had become detached from my head. However, simply appealing to the fact that my hair, while still on my head, remains a constitutive part of me does not yet provide us with a principled reason to draw the distinction between belonging and property ownership in this way. Consider, for example, a person who promised to donate her hair to someone or contracted to do so. We can think of those as cases where the person alienates various claims, privileges, powers, and immunities with respect to her hair before it is cut from her head. This kind of alienation is functionally similar to transferring ownership before the hair is detached. And yet, simply appealing to the fact that the hair is still part of the person in this case, I suggest, provides no compelling reason to object to such practices. Some further explanation is required to account for the continuity between ownership/alienability of the detached body parts and ownership/alienability of those same parts while still integrated within the body.
The picture cannot be as simple as always treating bodily materials as objects like any others, however. As a case in point, take a judgement under French criminal law which prosecuted cases of voluntary contamination of HIV through sex under the title "administration of a harmful substance." Something has gone awry if we start to define sexual interactions as a transaction of substances. To insist on this description of the wrongful action narrows the scope of the wrong in such a way as to separate the act of sexual intercourse from the "administration of the substance." This surely obscures something about the nature of the wrong, which is that the sex itself in such cases is a distinct kind of attack, regardless of whether the harmful pathogen in question is administered or not.
Baud offers a possible explanation for this, which is that in cases where parts of the body such as blood are characterised as objects, a legal fiction is required to confer that status on them. Unless something has happened to cause a certain part of the body to be recognised as an object under law (as separate from the person), exchanges of certain bodily fluids are simply biological processes which occur in the background and which fall outside the domain of legal regulation. This seems right: when we regulate interactions between persons, the proper objects of regulation are the actions of the people involved. The biological processes which occur within the body during those actions are most often incidental to the description of the action done by the person, though the intention to transmit a pathogen might factor in the assessment of distinguishing a sexual assault from a consensual encounter.
The point in question for my present purposes is that there are cases which compel us to recognise some aspect of the human body as an object under law-one which could be traded and exchanged under the same regulatory framework as property. And the reasons for doing so arise precisely from the importance of protecting something like the dignity of the person. But these reasons do not reach as far as recommending that we always treat those aspects of the body all simply as objects of property which are exchanged between people.
This brings out the puzzle that needs to be addressed. On the one hand, there are some uses of bodies and their parts which require us to treat them as transferable things, recognising their status as objects. In these cases, it is not clear that we can simply draw a sharp line to say that ownership only begins once the substance is separated from the body. Otherwise, we will be at a loss to explain why it is that I should be able to decide to donate some blood before it is extracted from me. On the other hand, there are many uses of the body which seem to belong squarely in the sphere of actions undertaken by persons and which it strikes us as a mistake (an intellectual, as much as a legal or moral mistake) to understand in terms of a transaction of materials akin to the gift or sale of a piece of property.
A question arises here about the nature of the link between persons and property. If it seems reasonable that a certain part of the body can be brought into the realm of property, we need an account of property rights that can explain when and why this is warranted. We need to be able to explain the continuity between the rights of disposal a person has over herself and the way in which she is able to alienate certain parts of herself as property.
But we also need to be able to explain where and why we ought to place limits on that kind of alienability.
The way in which rights of disposal over oneself and property rights become interwoven in the ways described above no doubt helps to explain why many find theories of self-ownership so appealing, as they purport to provide perfect continuity between property and the person. However, as Ryan points out, this supposedly neat link appears to play on conflicting sets of intuitions and leads us to question why the standard framework of object property as we know it is assumed as paradigmatic for understanding self-ownership. 9 A tempting alternative might be to opt for an approach which distinguishes more sharply between the normative status of persons and external objects. Kantian theory, for instance, draws a distinction between rights to your own person (the innate right of humanity) and acquired rights (rights to external objects of choice). Though the latter are not reducible to the rights to your own person, the normative basis of acquired rights is taken to depend on the right to one's own person. 10 However, this approach is not well placed to explain the continuity between a person's rights over various body parts and the right to alienate those parts as property as in the examples discussed above. For example, Pallikkathayil (2017) argues that there is some room in the Kantian account to explain how a detached body part could make the transition from part of a person to an object of property rights. However, the only legitimate way for this to happen would be for the body part in question to be in principle open to anybody to acquire as property after initial detachment from the originator's body. In this way, the Kantian approach is unable to account for the idea that I should have any privileged claim, above anybody else, to sell the hair that came from my own head. I would only do so if I happened to be the first person to perform the institutionally recognised act of acquisition to lay claim to the hair after it was cut.
Moreover, the Kantian approach still ends up drawing the normative link between property and the person too closely in its treatment of the significance of property violations, which on Arthur Ripstein's account end up amounting to coercion of the person. Ripstein (2010) explains how property rights depend on the innate right of humanity as follows: For Kant, property in an external thing-something other than your own person-is simply the right to have that thing at your disposal with which to set and pursue your own ends. Secure title in things is prerequisite to the capacity to use an object to set and pursue ends. (p. 67) Property rights, then, depend on purposiveness and provide a necessary framework within which people can rightfully enact their purposiveness in the world. The basic idea that the point of property is to enable us to make use of things for our own purposes, by creating and protecting a sphere of exclusive use, chimes with most conceptions of property. So does the claim that the structure of property rights parallels one's rights with respect to one's own person. But the extent to which Ripstein draws the link between purposiveness and property proves problematic. For Ripstein, the link between property and purposiveness is such that any unauthorised interference with my In a perhaps surprising streak of similarity with Nozickian theories of self-ownership, this Kantian line of thinking draws the parallel between the person and her property too strongly. If a colleague drinks out of my mug while I am out of the office, washes it up afterwards, and returns it to my desk so I am never any the wiser, she might have done something wrong insofar as she has violated some property right of mine but has she really coerced me? Just as Nozick's self-owner becomes partially enslaved by taxation, Ripstein's sovereign individual is made a slave to the mug thief. By presenting artificial property rights as the institutional solution that is required to give content to a fundamental natural right, the Kantian account imbues any legitimate system of property rights with the full normative significance of that fundamental right. I agree that we can think of institutional systems of property as mapping spheres of exclusive choice for persons with respect to their property. The point I am pressing is that it's a mistake to think that those institutional rights, once established, instantiate a sphere of freedom of such normative significance that any infringements on that sphere carry the same moral weight as interferences with my person. This is what is implied by bringing interferences with property under the umbrella of coercion.
Allowing for continuous transition between personal rights and property rights, on the one hand, while being able to distinguish an order of normative significance between personal interferences and property interferences, on the other, is key to making sense of transitional cases where we have reason to treat something straightforwardly as property in some contexts, but not in others. For example, consider prosthetic limbs-we have reason to allow a person to sell her prosthetic limb as a commodity, and yet also treat an attack on that object as assault, if it happens while she is wearing it. 11 A key feature of such cases is that certain kinds of interference with those objects constitute a wrong against the person which would not adequately be captured by describing it as a violation of a property right.
I suggest that a different kind of institutional account is better placed to deal with such cases. Rather than justifying systems of property on the basis of some fundamental natural right, we should think of property rights as justified on the basis that they serve some important interests of ours. The shape of any given framework of property rights will be determined by the extent it serves those interests and constrained by the need to protect people from being wronged in various serious ways. This approach requires some account of the structure of the underlying moral landscape, in order to anchor and constrain the justification of institutional rules of property. T. M. Scanlon has suggested a way of doing this using his contractualist principle of reasonable rejection. 12 The test of reasonable rejection is able to give us something like a concept of moral property rights, by allowing us to ask whether a principle allowing various kinds of interference with objects in which a person stands in some kind of relation could reasonably be rejected by the victim of that interference. This allows us to deal with typical "state of nature" scenarios, where, for example, a person has worked some land to cultivate crops, leaving plenty of land and resources around for others to do the same. If another person were to come and reap all her crops at harvest time for himself, the test of reasonable rejection could tell us she had been wronged by that interference.
When it comes to justifying general institutional frameworks for formalising property rights, the picture becomes more complex. On the one hand, we have to weigh the general reasons in favour of choosing certain forms of property rights against the reasons which count against them. But we also have to recognise that whether or not the rejection of a given proposal would be reasonable will be shaped to some extent by conventional systems already in place, including certain contingent sociological facts about the ways in which we value certain objects.
Moreover, the scope of a formal system of property rights will reach beyond the fundamental moral requirements of non-interference in the state of nature. This happens where some rule is needed to solve something like a coordination problem and where there is an acceptable range of forms that solution could take. 13 Though we can start with a moral basis of property using the test of reasonable rejection to furnish us with principles protecting against interference with some objects under certain circumstances, there will be other cases where the choice of principles is not so clear cut, and a number of different rules could reasonably be adopted. For example, if two people set upon working a piece of land at the same time, one of them trying to lay foundations for a building and the other clearing stones to plant crops, there would appear to be no principled moral way to determine which has a legitimate claim to the land. Such cases may be solved by the creation of conventions to determine which kinds of activities take precedence over others in terms of conferring ownership claims. There may not be a moral basis for choosing one particular convention over another, but enforcement of the convention could nevertheless be justified to the extent that it serves our interests in solving a coordination problem. Once a convention has arisen, it can then feature as a reason relevant to the judgement of what kinds of intrusions or interferences may be reasonably rejected in that domain, based on the expectations they engender.
So we can recognise that some problems in the state of nature might give rise to a range of justifiable institutional solutions, rather than one determinate blueprint for property rights. We can then notice that whichever solution is chosen and implemented from the range of possible policies for a given society, that chosen policy itself may have bearing on which interferences are those under that specific framework of rights which it would be reasonable to reject.
A proper understanding of the institutional justification of property thus has to take into account three distinct layers of analysis in the assessment of whether a given set of property rights serves our interests while protecting people from interferences which they could reasonably reject: (a) the function of the institutional structures which enforce rules for how we interact with one another; (b) the social conventions which provide informal regulation of our behaviours and influence cultural attitudes to certain objects and activities; and (c) the basic moral structure of our interactions with one another. Justifications for regulating certain activities through institutional structures such as property cannot therefore be made solely on the basis of a presocial understanding of rational beings in the state of nature. Rejecting the standard Kantian and self-ownership accounts of property in favour of the approach sketched above suggests that the ways in which we relate to various kinds of objects matters to the arguments we give in favour of or against uses of property rights in different contexts. Moreover, the way in which this relational aspect figures in those arguments will be internal to the general justification of property rights as an institutional framework, rather than relying on appeals to competing values.
A key aspect of this approach is the claim that institutional property rights do not map perfectly onto the underlying moral structure of our interactions. While there is a way of understanding ownership as a moral concept, the way in which general institutional rules of property are mapped out goes beyond a fundamental moral principle.
Moreover, the way in which the institutional framework is constrained by the basic moral structure of our interactions with one another gives us reason to posit that the extent of property rights may be more or less limited depending on the kind of object in question.
This approach involves rejecting the view that property is a univocal concept, whereby any limitation on the number of "sticks in the bundle" of a property right entails less than full ownership. An important question arises at this point, which is to what extent this would require a revision of the standard Hohfeldian conception of the structure of property rights. The next section shows how the institutional account can be developed within a Hohfeldian understanding of the structure property rights.

| PROPERTY VERSUS OWNERSHIP
Hohfeld famously argued that the correct conception of property is as rights, not things (Hohfeld, 1919). Thus, any interference with a person's property rights would constitute an interference with her property, regardless of whether any physical interference had occurred to the object over which she held those rights. A property right on the Hohfeldian view is not a single homogenous right, but rather consists of several distinct incidents, sorted into four basic components: (1) Privileges: A has a privilege to φ if and only if A has no duty not to φ.
(2) Claims: A has a claim that B φ if and only if B has a duty to A to φ.
(3) Powers: A has a power if and only if A has the ability to alter her own or another's Hohfeldian incidents.
(4) Immunities: A has an immunity if and only if B lacks the ability to alter A's Hohfeldian incidents. (Wenar, 2015) Though the four main incidents can be said to be determinate, the way in which they apply in different contexts to establish different elements that make up a given right is flexible or indeterminate.
Leif Wenar (2015) helpfully categorises these four basic components into first-and second-order rights. The firstorder rights (the privileges and claims) are those which hold directly over the object, while the second-order rights (the powers and immunities) are those which concern the alteration of the first-order rights. The second-order rights help to make sense of the idea that property is exclusionary while taking account of the fact that this exclusion of others is not static but that the owner can determine where, when, how, and in respect to whom that exclusion applies. The Hohfeldian analysis explains the complex normative structure and function of the concept of ownership, and the theoretical merits of this approach have led to it becoming the established theory of property within contemporary philosophy and legal theory.
Despite its merits, it is not without problems. One of the implications of the view of property as rights, not things, is that any interference with any of the incidents that make up a given property right counts as an interference with property. Among other things, this leads to problematically conservative interpretations of the Takings Clause of the Constitution of the United States of America (Wenar, 1997). Wenar points out that on the Hohfeldian view, any alteration or annulling of any of the incidents that make up a given property cluster would count as a taking of property, in the sense that this right would have been "taken" from the bundle held by the original owner. 14 In response to this problem, Wenar proposes that we can retain the complex Hohfeldian characterisation of property rights without abandoning completely the idea that property can be things, too. He suggests that we make the simple distinction between rights and the object of those rights-between property rights and property: private property is all those things over which private property rights are held. And private property rights are just those rights in that two-leveled structure of Hohfeldian rights (…) Property is what property rights are rights over. (Wenar, 1997(Wenar, p. 1944 The distinction between property rights and property allows us to stay firmly within the Hohfeldian framework of property rights, while retaining an interpretation of what constitutes a "taking" of property that is still grounded in the common-sense concept of removing some object from someone's possession. This separation of the ideas of ownership (or property rights) and property (things) allows us to make an important first step towards dismissing the univocal view of property. It makes it possible to challenge the notion that "full ownership" must entail the largest conceivable bundle of Hohfeldian incidents. To recap, the view that property is rights implies a univocal understanding of ownership. To compare two different sets of ownership rights in order to determine which is fuller, one need only compare the two bundles of incidents to see which contains the longest list of privileges, powers, claims, and immunities. The kind of object owned would not come into question, because the comparison of property just is a comparison of rights. The object only figures in a secondary sense-in that the right is held against other people, in reference to some object.
Once we have acknowledged that ownership rights may be altered without this entailing a taking of the property in question, we can also suggest that having less than the full abstract list of possible Hohfeldian incidents in a property right bundle need not entail that one has less than full ownership over a given piece of property. In particular, we can start to press the idea that different kinds of object by their nature impose different conceptual restrictions on what can be considered the maximum number of incidents in the bundle of ownership rights for that property. While it is in one sense conceivable to think of self-ownership in the maximising way described above, there is a separate question as to whether this is coherently conceivable within a given theory of the justification of property. It is this latter question which I pursue here. I have suggested above that an institutional approach to property based on a contractualist principle of reasonable rejection is best placed to tackle questions about the limits of ownership of different objects on a case-by-case basis by reference to the way in which enforcing property rights serves our interests in different contexts. Before addressing how it helps us with specific cases, it is necessary to provide an explanation of the general form of justification one might give for a system of private property to underpin the institutional account. That is, we need some explanation of which basic interests can be posited as justification for having a system of property in the first place and how those interests are taken to ground such a system.
The kind of general justification I have in mind is one which identifies a certain problem arising from some basic interests of individuals and posits institutional frameworks of property as a pragmatic solution to that problem. It starts from a familiar assumption that individuals have an interest in having secure use of some objects but that unless a person is physically holding a given object, her claim to possession of it is both indeterminate and precarious. 15 Having conventions to recognise certain rules for relations of possession can be seen as a solution to this problem. Hume suggests that we can view such conventions as arising in much the same way as languages develop to facilitate communication. They simply provide a pragmatic solution to a coordination problem.
So far, that sketch of the basic interest in secure use of objects and the problem of how to secure possession against others gives no determinate indication of how those interests may best be advanced. I suggested above, however, that there is a way of thinking about such conventions as having a moral basis, without entailing that justifiable property rights instantiate a boundary the crossing of which violates a basic moral principle. A contractualist approach to understanding the moral constraints on the possible sets of rules could narrow down a range of justifiable options, without entailing that any solution is morally required.
To understand how the institutional solutions may be justified by reference to some moral considerations, while not being directly reducible to natural rights, we can note that the way in which principles of possession of objects are established is closely aligned with principles of non-interference against the person. In particular, the cases where there will most clearly be a direct moral basis for claims to exclusive possession of some object will be those in which a person relies on that object to provide essential nourishment or shelter. 16 The reasons we have to protect ourselves against unwanted interference thus provide the strongest basis for justifying a system of private property in the first place. In this sense, we can think of the way in which we recognise artificial relations of possession over objects as both approximating and protecting the kind of natural control we have over our own bodies. 17 With that basis established, it can be extended by considering the reasons we have to value some system of rules for establishing secure possession of items beyond what is necessary for basic nourishment and shelter. Again, in the assessment of the way in which such general rules serve our interests, the bar for justification may be that they are ones which it is (all things considered) reasonable to adopt. However, it does not follow from the general justification of a rule in this way that any particular action which breaks that rule must be one which could itself be reasonably rejected.
Consider Joel Feinberg's example of a backpacker who is caught in a life-threatening blizzard and breaks into a nearby cabin to weather the storm, helping himself to food from the cupboards and burning furniture for warmth. 18 Although the owner's property right might feature in the assessment of the permissibility of the backpacker's act, that reason, I take it, would not be sufficient to outweigh the reason he had to save his life. Despite the owner's property right in the furniture, it would not be reasonable for him to reject a moral principle allowing the backpacker to burn it to save his life. Feinberg explains this case using Judith Jarvis Thomson's distinction between infringing and violating a right: while there may be cases in which it is justifiable to infringe a right, one violates someone's right if one infringes it without justification. 19 On Feinberg and Thomson's view, although rights infringements are justified, they leave a moral residue which requires the rights infringer to compensate the person whose rights were infringed. However, one might well question why compensation in this case would be morally required. If our backpacker was destitute, and the furniture he burned a priceless antique, one might think it reasonable for him to reject a principle requiring him to take on the significant debt of replacing the antique in order to save his life. Conversely, one might well think it unreasonable for the property owner to reject a principle requiring him to sacrifice his antique chair in order to save someone's life unless that principle included a compensation clause. A contractualist could thus make the case that the cabin owner has a moral duty of aid to the backpacker that includes no claim to compensation. However, it is compatible to hold that the owner has a moral duty to sacrifice his property in aid of the backpacker's life, while questioning whether this duty is enforceable. Thomson puts the question this way: why should the property owner be forced to pay the cost of saving the backpacker's life? After all, the point of property rights is to secure a sphere of exclusive control so that I can choose to exclude or include at my own discretion, whether or not my choices are perfectly morally motivated.
The proposed approach to property rights allows the possibility of an institutional justification for the owner's claim to compensation, even if compensation is not required by a basic moral principle. In order to meet people's interests in stable and secure possession of things like holiday cabins and antique furniture, it may be justifiable for a system of property rights to include compensation claims for owners in situations like the one described above.
There could, however, be a range of justifiable ways of structuring these claims. Instead of requiring the backpacker to compensate the cabin owner directly, a system could collect tax to support a compensation fund for such eventualities. Another option might be to require all individuals to carry public liability insurance to ensure that they be able to pay compensation without incurring significant amounts of debt. These would be ways of building in compensation claims for property infringements, without imposing unreasonable burdens on those who are morally justified in infringing someone's property rights. On this view, we could say that the backpacker infringed the cabin owner's property right but did not wrong him in doing so.
The examples explored in Section 2 of this paper suggested that there may be cases where it makes sense to treat the body under the framework of property, given our interest in having access to things like blood banks.

Clearly, the relation between a person and her body is not a relation between two distinct objects, insofar as my body
is (a constitutive part of) me. And yet, the link is a somewhat complicated one-bodily stuff can become a separate "thing," and ordinary "things" can be amalgamated to the body. So there is a certain amount of physical detachability that makes at least parts of our bodies more akin to mere objects, and thus perhaps appropriately subject to the framework of property, on the model described above. Beyond this, a parallel problem to the precariousness of the possession of objects arises in the realm of persons and their actions, too. Namely, it is useful to be able to treat certain undetachable aspects of our person as alienable-to create an artificial relation of alienability, where no actual detachability exists. The structural frameworks which allow us to securely trade with others stretch beyond the realm of mere objects, so a similar story can be told about how conventions arise to secure individuals' claims against each other for the provision of services. The way in which these arise can be construed as addressing a parallel problem to that of securing possession over objects. Namely, that the control a person has over her own body or actions is too secure and that we need to be able to construct artificial relations of alienation in order to be able to give others secure and determinate claims over one's actions or services.
The mechanisms for this can be fleshed out using the Hohfeldian rights framework. Take the example of a person, Alice, who agrees to work on Bert's farm for a fee. Alice has agreed to work under Bert's orders for the day on general farmyard tasks. The Hohfeldian framework allows us to track the changes in the normative relation between Alice and Bert with respect to some of Alice's actions. Namely, before Alice agreed to work for Bert that day, Alice would have had a privilege to go fishing if she chose to do so. Now, however, Alice has effectively undertaken a duty to Bert not to go fishing and so has waived that privilege. Alice has also transferred to Bert a claim that Alice labour on the farm that day, because Alice has undertaken a duty to do so. In entering the agreement, Alice was utilising her powers to alter her Hohfeldian incidents with respect to Bert. In doing so, we can see that Alice has also transferred certain powers to Bert and thereby lost certain corresponding immunities. For example, Bert now has the power to order Alice to muck out the stables, or to harvest the corn, or to shear the sheep. At the moment that Bert gives any of these orders (providing they fall under the remit of the original agreement), this imposes on Alice a duty to perform the action that has been asked of her and gives Bert a new claim against her. Alice thereby lacks the immunity from Bert altering her Hohfeldian incidents within these parameters. Again, we could conceive of the object of the agreement in this case as being Alice's consent to transfer certain normative relations holding between her and her actions to Bert and the agreement not to renege on the transference of these incidents.
The Hohfeldian structure thus allows us to conceive of a certain artificial relation holding between Alice and her mental and physical capacities. This relation allows us to make sense of exchanges in services and actions by conceiving of Alice engaging in a transfer of the relation of possession of her actions along the same lines as she can transfer her property in external objects. This mirrors the way in which the artificial relation of possession was justified as serving our interests in securing exclusive use of external objects. Just as the relation between person and object can be thought of as modelled on the fixed and constant control a person has over her mind and body, so in turn can an artificial relation of alienability emerge between a person and her mental and physical capacities, that is modelled on the artificial relation of possession. The way in which these two relations mirror each other might be put in simple terms as follows: external objects are too easily transferable, so our control of them needs to be made as stable as possible by approximating the stability of our control over our bodies and minds. The exclusivity of a person's control over her body and mind, on the other hand, is too stable, so we need a way of giving other people claims over our actions and services in a similar way to the way in which we can easily hand over control of external objects to them. This allows us to square the circle as to whether the shape of property over external objects is derived from the structure of bodily rights, or vice versa. But it allows us to do so in a way that takes account of the fact that such property relations arise as an artificial feature within the account of the justification of formal regulative structures.
We construct these relations because they are a useful way to formalise our interactions with others in a way that can be publically recognised and enforced by a third party to provide the security and stability to protect our interests. It formalises our interactions as transactions.
This picture allows us to establish context-dependent limits to property rights. These emerge from the way that the account provides a justification of property based on the assumption of certain fundamental interests in noninterference and the way in which formal structures attributing relations of possession protect this interest. The Hohfeldian framework is helpful in this regard. If we take another look at the Hohfeldian incidents, we can make a further distinction over and above the one between first-and second-order rights. Namely, incidents (1), (2), and (4) can be considered static ones. 20 That is not to say that there can be no change in these incidents, rather, these incidents only change when somebody has wielded a power to change them. We might think, then, that the flexibility of an ownership right is really characterised by the powers included in that ownership right (incident 3). These are what allow you to make various uses of your property in cooperation or trade with other individuals, by altering the specific Hohfeldian incidents that hold against others. I suggest that differential limits on powers of ownership can be grounded in a certain understanding of the importance of Hohfeldian powers from the point of view of an institutional approach to property rights.
One of the questions that Ryan prompts us to ask is: can I really do whatever I want with my lungs? This a question of alienation: ought I have the right to destroy my lungs, or to give or sell them to somebody else? Before addressing this question about the legitimacy of alienating specific parts of one's body, it will be easier to address a more general question: if we conceive of a person as having ownership rights over herself, ought she be able to alienate those rights entirely? We should distinguish here between alienating incidents within specific parameters (e.g., Alice transferring to Bert the power to give Alice work orders for the day) and alienating all one's incidents by alienating one's powers over them all.
For ordinary objects, full alienation of the second sort usually involves actual transference of the object. This is where Wenar's distinction between property rights and property becomes helpful. With this distinction, we can say that alienation of property rights usually also involves alienation of the property itself. In simple terms, if I transfer to you all property rights over a bottle of wine, you will most probably also take that bottle of wine home with you (and away from me). So while transference of the relation of property might happen in theoretical terms while I am still holding the wine bottle in my hands, the object then can be, and usually is, physically removed from me.
In contrast, while it is conceivable to think that a person could alienate all her Hohfeldian incidents to someone else, there is clearly no corresponding physical detachment of the person from her body or mind that can take place. 21 In that case, we might think that alienation of the ownership rights over the person would be possible, but not alienation of the property.
It might seem trivial to labour this point but viewed in the context of the account of property we have been exploring, it helps us to make an important distinction. Some theorists have suggested that what distinguishes property rights from personal rights is that the former are in principle alienable, while the latter are not. 22 Philosophical disagreement over the question of voluntary slavery, however, shows that personal rights are at least conceptually alienable, in principle. 23 The challenge is to provide an answer as to why there might be limits to the ways in which we can alienate certain rights over ourselves. The account of property rights I have proposed provides a way of doing this by drawing on reasons that are internal to the general justification of systems of ownership rights. Certain ways of alienating control of parts of a person that cannot be separated from herself, although conceptually and practically possible, will undermine the basic interest which a system of ownership rights is predicated upon protecting.
At the heart of the account is the assumption that individuals have a basic interest in having secure use of certain external objects, modelled on the kind of security of control we enjoy over our own physical and mental capacities. This is no coincidence, because secure use of our physical and mental capacities is also the precondition of us being able to establish secure use of external objects. 24 Another way to characterise this basic interest, then, is as an interest in having the basic power to exclusively control various things (starting with our own bodies/minds and extending out to external objects).
This can be seen as the basic power that it is assumed must be attributed to the individual as the political atom in this account. It is by attributing this power to individuals that frameworks of rights provide the security from unwanted interference which is taken in many liberal accounts as the starting point for providing a general justification of the enforcement of regulative structures by the state. The fact that external objects are physically detachable from the person means that when one alienates a piece of property to another person, this has no impact on that basic power. It straightforwardly serves our interests to be able to discard some items of property completely. This lack of impact on the basic power is precisely because there is no real connection between the person and the external object.
But now we can contrast this to thinking what it would be to completely alienate all incidents over one's body in the same way. This would be to demand of the state that it no longer attribute to that person the very same power that was posited as requiring institutional protection in order to secure a fundamental interest of the individual. In order to understand exactly what kind of mistake this involves, we can turn to an approach suggested by Scanlon on how to assess the concept of self-ownership.  examines the argument that self-ownership entails that taxation is akin to slavery because an individual is entitled to the full amount of what another is willing to pay her for her labour: To assess this argument we need to ask what makes the idea of self-ownership appealing, and whether the reasons that lie behind its appeal support the idea that individuals are entitled to the full amount of what others are willing to pay for their services. (p. 114) This is a method which Scanlon applies to various ideas, including the ideas of equality, liberty, and coercion. His approach is "to try to identify the reasons that give these concepts their importance and to ask when these reasons apply." The considerations which give legitimacy to the idea of self-ownership in one's labour are "the reasons people have to choose their occupation, to be able to quit a job if they wish, and so on." 25 Taking these to be the reasons which support the appeal of the concept of self-ownership in one's labour, Scanlon suggests that these very same reasons support the regulation of markets and some level of redistribution through taxation. This is because unfettered markets and lack of taxation will lead to high levels of inequality, which in turn create negative externalities.
The crucial point is that the negative externalities created are ones which threaten people's abilities to choose their occupation and to be able to quit a job if they wish to-the very considerations which Scanlon takes to be central to the appeal of the concept of ownership of one's labour. He concludes from this that the reasons that support the concept of self-ownership support a system which has inbuilt limits on the claims which certain people can make on the total pot of resources. Taxes are not viewed as deductions from a person's income, in the sense that some of his money is taken away from him, "Rather, these taxes reflect the limits on the claims to resources that he can come to have within a legitimate system of property and market exchange." 26 In other words, the reasons that support the basic concept of self-ownership serve to shape the structure of institutional frameworks which are justified on the basis of that concept. And those reasons might place certain internal limits on the claims which an individual can reasonably make within that system.
We can take the same approach here, with the question framed in a slightly different way: what makes the existence of frameworks of property appealing, and what are the reasons behind attributing powers of ownership to individuals? The first reasons identified above were based in a fundamental interest of individuals in establishing security of control over themselves and other objects. I suggested that systems of property can be thought of as creating artificial relations of possession over external objects to approximate the security of a person's control of her mental and physical capacities. Furthermore, the powers of alienation that come with property rights can be seen as responding to reasons that emerge after security of possession has been established, namely, the interest in trading with others. It is only once there is some system to protect secure, exclusive possession that the notion of being able to trade possessions makes sense. The reasons that support the possibility of alienation are in this regard dependent on the reasons that support establishing security of possession.
This hierarchical way of conceiving of the reasons supporting systems of property allows us to think more clearly about the kind of structure of property rights which could legitimately be supported on the basis of these reasons. In particular, we can suggest that these reasons support internal limits on a system of property rights to ensure that powers of alienation do not go so far as to undermine the basic security of the individual's possession of her mental and physical capacities on which the whole thing was predicated. From this point of view, we can say that there is nothing in the reasons that make the existence of property frameworks appealing that would support enabling the kind of wholesale alienation of a person's body explored above. This gives us reason to posit a sliding scale of differing limits for the ownership powers that we can conceive of the individual exerting over her own mind and body and the kind that we can conceive of her holding over external objects. Namely, on the terms of this kind of justification of property rights, there is no reason to think that if we allow individuals to subject aspects of their body to the framework of property, this would have to include attributing to the individual the full power to alienate in the same way that ownership rights over external objects include this power.
Nevertheless, the analysis showed a mirroring between the power held over one's body and mind and that held over objects. The way that these are interlinked, I suggest, paired with the challenges posed by the examples discussed in Section 2, motivate the proposal that it could be useful to make provision for individuals to subject their bodies to the framework of property in certain cases, even if the structure of the power of ownership that is attributed to an individual over her body is not symmetrical to ownership of other objects. Furthermore, this kind of provision would be compatible with the Scanlonian analysis of the basic reasons which support the existence of frameworks of property discussed above.
One might object that one could well accept the Hohfeldian characterisation of artificial relations of alienability being established in the way described above but still resist bringing this all under the umbrella of property. Instead, we could simply accept that there are different clusters of Hohfeldian rights for different objects, and for persons and their actions, perhaps even for different parts of people in different circumstances, but we needn't talk in terms of property rights in order to understand this. In response to this point, I needn't commit to the strong claim that these rights over our person are property rights in some fundamental sense. What the account above brings out is the fact that this basic Hohfeldian structure allows one to treat one's body as though it is property, by engaging in these artificial relations of alienation. Or, in the cases where parts of the body are physically detached and transferred from one person to another, by engaging in actual alienation of the thing as property. Once we are engaging in these kinds of transactions, we can say that we are operating within the framework of property, broadly construed. This allows for a transition between bodily rights and property rights via the notion of ownership, without committing to a direct symmetry between the two.

| CONCLUSION
It is perhaps a curious upshot of this picture that it leaves us with little reason to think that there is any particular moral significance to labelling something as "property." 27 This, I suggest, is a strength of the approach. We saw above that the reluctance to recognise certain bodily materials as property was based in a concern that doing so would be an affront to the dignity of the human body as constitutive of the person. This kind of worry, I take it, has often led to a framing of debates around the treatment of various bodily materials in terms of what ought to count as property or not. These debates are often conducted as though to label something as "property" in any context is to confer on it a specific set of values which must be exhaustive of its identity. This leads to a kind of essentialism about parts of the body which makes it very difficult, if not impossible, to make sense of the changing ways we make use of various parts of ourselves in different contexts. That approach gets things precisely the wrong way around. We should instead think of Hohfeldian rights as offering a certain transactional framework, the limits of which can be understood as context dependent in the way explained above. We start first with an understanding of the way in which we value parts of ourselves in different contexts and how that figures in the assessment of reasons in favour of various regulative frameworks. The limits to the way we transact with various things will vary with respect to different objects and contexts, but the basic form of rights as understood with respect to both people and objects is continuous. 28 By avoiding the kind of essentialism described above, this paves the way for a pragmatic approach to clarifying debates around how far persons ought to be able to subject themselves and their bodies to various markets or contracts. We might ask, for instance, how far do certain activities such as prostitution or contract surrogacy involve an alienation of powers that comes close to threatening the basic capacity for control that was identified as a fundamental basis of the justification of ownership? Rather than providing a yes/no answer to the question of whether such activities could be legitimate, the approach would allow an examination of the conditions under which it might be possible for such contracts to be regulated in specific ways which would protect, rather than threaten, that power.
It also provides a way of explaining why certain transactions may be more problematic than others. For example, if I sell my kidney to somebody else while it is still in me, this will have implications not only for my powers over my kidney but also for my normative power of control over the rest of my body. If the transaction gives the new owner a claim to take possession of my kidney, that affects claims I hold over the rest of my body against physical interference. The importance of protecting those claims against threats from this kind of alienation will be proportionate to the danger posed to the basic power of control which we established was inalienable. This approach requires us to take into consideration structural aspects of certain markets and the power relations within them. For example, following Scanlon, if we think it important that a person have the exclusive right to decide what happens to her body, this may give us reason to think that nobody else, including the state, may interfere with her decision to sell her kidney. On the other hand, there may be structural features of such markets that make us think again. We may think that the high price offered for a kidney exerts undue force on the decisions of those who are very poor, effectively threatening their power to control their own bodies. 29 These are suggestion of how the account provides a more nuanced approach to important practical questions.
These are of course complex matters and will need more work to fine-tune the details. The purpose of this paper has been to lay the groundwork for proceeding with a view of ownership based on an institutional account of property that can begin to make sense of the continuity between the structure of rights of disposal over one's person and rights over objects, without ignoring the normatively significant points of differentiation between the two. 5 The normative power of control I have in mind is similar to that proposed by Christopher Essert (2016). Essert argues that we ought to think of systems of property rights as justified on the basis of their necessity for protecting a person's normative control, lack of which would leave her open to domination.