Self-ownership theorists argue that many of our most morally urgent and enforceable rights stem from the fact that we own ourselves. Even small infringements of a person’s ownership claim are morally impermissible. However, critics of self-ownership point out that the claim that people own their bodies commits the self-ownership theorist to several implausible conclusions. For example, self-ownership theory is based on several vague moral predicates, but any precisification of the required predicates is seemingly too permissive (because it allows people to impose deadly risks on innocent bystanders for no reason) or too restrictive (because it prohibits people from polluting or even interacting with others at all). Therefore, self-ownership theory is not justified because however the boundaries of self-ownership are defined, the theory is extensionally inadequate.
In this essay, I argue that this argument does not undermine the case for self-ownership theory because self-ownership theory does not require precisification of each moral concept that it is based on and, even if it did, the theory’s alleged extensional inadequacy does not undermine its justification. In Section I, I outline the argument in more detail. In Section II, I argue that vagueness within a moral theory doesn’t require a determinate precisification and that the presence of vague moral predicates in self-ownership theory can explain why certain cases related to the boundaries of self-ownership appear so problematic. Then, in Section III, I argue that the fact that the precisification of a theory makes it seem either too restrictive or too permissive in marginal cases does not on its own undermine the justification of the theory. In Section IV, I argue that self-ownership theory, or at least a version of it, may not even be that extensionally inadequate after all. Therefore, I conclude in Section V that self-ownership theory could still be justified, even though the theory is vague and it seems to have counterintuitive implications. Throughout, I argue that there are good reasons to think that self-ownership theory is justified.
My argument primarily addresses debates about self-ownership theory, but it has broader significance for theorizing in political philosophy as well. When theories appeal to concepts or empirical realities that are determined in a seemingly vague way, cases at conceptual and empirical boundaries potentially pose challenges to the theories. This is true narrowly for self-ownership theory and broadly for all moral theories. My arguments in response to this class of objections to self-ownership therefore serves as a more general argument against “vagueness objections” in moral theorizing. I argue that it is plausible that moral concepts can be vague in all the ways that vagueness can occur. Theories that are vague are not necessarily deficient. And even if moral concepts should not be vague, cases that arise at conceptual and empirical boundaries are aberrant, and theories should not be rejected solely on the basis of these kinds of cases. In this way, my arguments against prominent objections to self-ownership theory provide a blueprint for arguing against any class of objections to a moral theory that proceeds from an accusation of vagueness.
I. An Argument Against Self-Ownership Theory
Self-ownership theorists hold that even small infringements of a person’s right to their body, including exposures to small risks, are presumptively impermissible even when these infringements would provide significant gains to others. This thesis is often advanced in support of two conclusions. First, self-ownership protects people from having their bodies used without their consent, either to benefit themselves (paternalism) or to benefit others. Second, self-ownership provides a moral basis for people’s entitlement to keep the benefits of their bodily labor. This second conclusion supports broadly libertarian judgments about property rights and economic freedom. Most proponents of self-ownership agree that self-ownership supports very strong natural bodily rights. It is more controversial whether those rights extend beyond the body and if so, how far.
The second conclusion is controversial, even among self-ownership theorists. And the problems with self-ownership theory I address arise even if I limit the analysis of self-ownership to questions about bodily ownership. So I will focus on the view that self-ownership consists in a strong natural right that prohibits people from using other people’s bodies without their consent. Kasper Lippert-Rasmussen offers a clear characterization of the less controversial version of the thesis:
The Self-Ownership Thesis: Each person enjoys moral ownership of himself or herself (his/her body and mind).Footnote 1
This means that each person has the right to control her own body and mind and is entitled to compensation and apology when that right is violated. As self-owners, people have defensive rights against self-ownership rights. They also have the right to transfer (at least temporarily) control over themselves (or at least parts of themselves) to others.Footnote 2
The thesis captures the intuition that people are morally inviolable.Footnote 3 It explains why violence and threats of violence are presumptively wrong. Because self-ownership theorists hold rights of self-ownership to be uninfringeable, the theory can explain why slavery, rape, murder, unjustified imprisonment, and organ redistribution would be wrong, even if these sorts of violations would promote the greater good or bring about a more egalitarian distribution. There are many arguments for self-ownership, such as the Kantian claim that all people have intrinsic value in virtue of their autonomy or the more pluralist claim that self-ownership is entailed by a cluster of other natural rights.Footnote 4
Despite its intuitive appeal, critics often allege that the self-ownership thesis is false, because of its seemingly absolutist prohibition of rights-infringements. These critics observe that the boundaries of self-ownership are indeterminate, and once that indeterminacy is resolved, the theory is extensionally inadequate. For example, since the self-ownership thesis treats trivial infringements as equal in moral importance to serious infringements, if the thesis were true, then it would be wrong to violate a person’s right that her finger not be scratched in order to save the entire world. But intuitively, it would not be wrong to violate a person’s right that her finger not be scratched to save the world. So, the self-ownership thesis as stated must not be true.
David Sobel calls this the conflation problem.Footnote 5 Self-ownership theory allegedly conflates the wrongfulness of serious infringements with the wrongfulness of trivial infringements. Alongside this critique, Sobel and other critics allege other extensional inadequacies. For example, self-ownership theory seems to prohibit pollution and other small harms and risks in ways that rule out freedoms self-ownership theorists should support, such as freedom of movement and the right to use the commons. In this way, the view is potentially too restrictive. But any attempt to avoid ruling out freedom of movement and rights to use the commons seems too permissive. Rethinking the boundaries of self-ownership in a way that would allow people to impose some small harms and some risks of harms on each other would rule in too much. For example, it could permit people to play a game of Russian roulette with bystanders’ heads as long as they exposed bystanders to a sufficiently small risk of death.
These lines of critique are related to the precisification of concepts related to self-ownership. The charge is that any precisification of the concepts or terms that are important for self-ownership theory will be extensionally inadequate. This charge applies to the boundaries of several concepts that are central to self-ownership theory.
◦ Self: If A owns her skin, then it would be impermissible to pollute. If you do not own your skin, then it would be permissible, at least as far as self-ownership is concerned, to confiscate skin cells from people for the purpose of research.
◦ Acceptable Risk: If A has a right against x percent risk of death, then she has a right against a plane flying over her head and subjecting her to an x percent risk of death. If A does not have right against x percent risk of death, then it would be permissible, at least as far as self-ownership is concerned, to enter people into a death lottery that exposed them to x percent risk of death.
◦ Permissibility: If A’s right to her body is absolute, then it is always impermissible for someone to infringe on A’s right, no matter how great a good one can provide to others by doing so. If A’s right to her body is not absolute, then one can permissibly interfere with A’s self-ownership claim as long as it is for the greater good (even if it is for the greater good to A).
In each case, the strength and scope of self-ownership claims are indeterminate or vague due to vagueness in an underlying concept. This shows that self-ownership theory is vague. And in each case, making the concept more precise (for instance, specifying the boundaries of the self in a way that excludes only the top layer of skin cells) seems implausible as well.
Arguments against self-ownership theory then share a common form. They begin with the premise that self-ownership theory is based on several vague predicates, and then argue that any precisification of the predicates that are required to justify self-ownership theory, if known, would be either too permissive or too restrictive to justify self-ownership theory.Footnote 6 So self-ownership theory is not justified. This argument can be stated like this:
(1) Self-ownership theory is based on several vague moral predicates.
(2) Moral vagueness must have a determinate precisification.
(3) If the extensions of a precisification of a theory seem inadequate, then the theory is not justified.
(4) The extensions of any precisification of the moral predicates required to justify self-ownership theory are either too permissive or too restrictive.
(5) Therefore, self-ownership theory is not justified.
This argument appears in some form in several critiques of self-ownership. Richard Arneson writes,
The attraction of such appeals [to self-ownership theory] is at least partly a function of their vagueness. As we begin to clarify the self-ownership notion, its claim to represent the core of the liberal tradition becomes more dubious.Footnote 7
Sobel writes that “it may well seem that part of the problem [with self-ownership theory] stems from vagueness concerning exactly what rights I have in virtue of being a self-owner,” and then considers several attempts at precisification to show that the problems endure under several attempts at defending a determinate precisification of self-ownership.Footnote 8 G. A. Cohen cites Ronald Dworkin as pressing skepticism about the determinacy of self-ownership as well.Footnote 9 Andrew Kernohan echoes this sentiment in arguing that self-ownership cannot support the property rights required for capitalism.Footnote 10 As Michael Gorr notes, resolving the seeming indeterminacy of the self-ownership concept would reveal that the view is either trivial or false.Footnote 11
To avoid this problem, one may question the absolutism of the self-ownership thesis. If it were not impermissible to scratch another person’s finger to save the world in all cases, then the self-ownership theorist could avoid this kind of an objection. But part of the appeal of self-ownership theory is that it does not make people’s rights “hostage to empirical fortune.”Footnote 12 Proponents of the theory defend it as a way of protecting individual rights from the vagaries of the rights-holder’s circumstances. This problem is not unique to self-ownership. Any theory that specifies a range of deontological prohibitions encounters these sorts of objections. I explore ways of moderating the absolutism of self-ownership theory that avoid these objections in the Section IV.
Yet before we consider modifications and moderations, several other replies to this class of objections are available on behalf of self-ownership theory. Against (2), moral predicates may be vague in various ways, and it’s plausible that the moral predicates related to self-ownership are vague in these ways. Sobel’s concern seems to be that underlying vagueness in self-ownership produces vagueness in the resulting duties that people have in virtue of others’ self-ownership. But even if this were so, vagueness doesn’t necessarily undermine the validity of predicates involved in a theory or a theory itself. It is at least conceptually possible that the duties we have in virtue of self-ownership could themselves be vague.
Premise (3) is false because extensional inadequacy doesn’t undermine the justification of a theory. If (3) were true and extensional inadequacy did undermine the justification of self-ownership theory, it would undermine the justification of most other moral theories as well. We should not abandon all moral theory, so this argument does not give us reason to abandon self-ownership theory. Moreover, there are methodological reasons to avoid developing theories on the basis of marginal cases. Just as good cases make for bad law, I argue that marginal cases make for bad theories. If a theory did avoid problems of extensional inadequacy, it would be a bad theory for other reasons. For example, it would be ad hoc, and it would not be action-guiding.
But even if (2) and (3) were true, (4) is false. Even if moral predicates do require some determinate precisification and no precisification of the moral predicates related to self-ownership avoids the too-restrictive/too-permissive dilemma, these dilemmas are not as fatal to the theory of self-ownership as they appear. For example, self-ownership theorists might try to defend a precisification of the theory that defines the strength of self-ownership rights in a way that allows people to infringe rights for the sake of the greater good, at least in some cases, and defines the scope of self-ownership rights in a way that is not too restrictive or permissive.Footnote 13 Or self-ownership theorists can accept that implications that are presented as a reductio by critics are actually just implications of the view, and then try to make those extensions seem better than they do at first glance.Footnote 14
II. Self-Ownership without Precisification
The first premise of the argument against self-ownership theory is that it is based on several vague moral predicates, such as the vague concept of the self, the vague idea of “acceptable risk,” and the vague application of the term permissibility. The foregoing cases demonstrate that boundaries that divide the self from the other, the acceptable from the unacceptable, and the permissible infringement from the impermissible are indeterminate. The second premise is that the vagueness of these moral predicates must have a determinate precisification. Granting that self-ownership theory is based on several vague moral predicates means that it is difficult to develop a determinate precisification of self-ownership because the moral predicates on which it is based lack precisification.
The second premise is that self-ownership theory requires precisification. This premise is implicit in arguments against self-ownership theory that appeal to boundary problems. Boundary problems cast doubt on the theories’ ability to sustain a sharp boundary between selves and others, unacceptable and acceptable risk, or permissibility and impermissibility. Critics of self-ownership argue that any determinate precisification of self-ownership is extensionally inadequate. Implicit in the criticism is the view that the theory requires determinate precisification.
In this section, I respond by making the case that it’s not a problem if some of the moral predicates involved in self-ownership are vague.Footnote 15 As long as moral values can be vague more generally, which seems plausible, then the values involved in self-ownership theory could be semantically, epistemically, or ontically vague, just like other instances of vagueness.Footnote 16 For any of these ways in which predicates can be vague, the vagueness of aspects of self-ownership theory, such as permissibly, risk, and the self, does not undermine the moral force of these concepts within a theory or undermine the theory itself. What appear to be problems defining and defending sharp cutoffs stem not from problems with the theory, but from the unjustified expectation that proponents of a theory define and defend a sharp cutoff.
This argument shifts the burden of proof to the critics of self-ownership theory. I also argue that this is a more general problem for all moral theories. If people’s inability to define and defend all thresholds on which their moral theory relies were grounds for rejecting a theory, then far more moral theories than self-ownership would need to be rejected.
A. Vague language and self-ownership
The boundaries of self-ownership may seem vague because the way we talk about the self or conventional understandings of ownership are vague. If so, then seeming dilemmas related to restrictiveness or permissiveness around the boundaries of self-ownership may arise because we talk about the boundaries in a vague way.
Surely at least some terms that are relevant to self-ownership theory are semantically vague. For example, some people understand “body” to include genetic material and body parts that are no longer attached to the body while others have a more limited understanding of what “body” includes.Footnote 17 From cases of seeming semantic vagueness, we might conclude just that our natural language about bodies doesn’t always “carve at the joints” in ways that draws a natural, uncontroversial boundary for people’s bodies. If so, then semantic indeterminacy about a descriptive term (such as “body,” or “self”) may cause a kind of indeterminacy about moral predicates, like self-ownership, to the extent that the moral claims people have over their bodies depend on what we call a “body” or a “self.”
It is less clear whether moral predicates related to acceptable risk or permissibility can be vague in this way. Whether they can be would depend partly on metaethical views about what moral predicates are in the first place, and about what they should do. If moral predicates can be semantically vague, then so can moral predicates related to self-ownership theory. But even semantic vagueness is not the best explanation for the seeming indeterminacy of moral predicates related to acceptable risk and permissibility. The vagueness of other terms related to self-ownership theory could go some way in explaining the seeming indeterminacy of the theory.
Tom Dougherty surveys the extent to which semantic accounts of moral vagueness are compatible with different metaethical views.Footnote 18 For those who think that the correct application of moral predicates depends in some way on people’s conventional use of a term or on idealized attitudes about a term, it could be that linguistic conventions for terms like “permissible” or “owned” are indeterminate in their application.Footnote 19 One worry about this view is that, if it were true, it would be hard to explain how people who are correctly complying with the convention could ever be wrong about their use of a moral predicate.
More plausibly, the application of some moral predicates related to ownership could partly depend on usage or conventions. If so, then just as people may have only partial beliefs about descriptive terms like “self” and “body,” perhaps people can have partial beliefs about moral predicates, such as “acceptable risk” or “permissibility” or “ownership.” If this view were true, then the moral facts, which are partly based on conventions, would be vague. One problem with this view is that no set of extensions for a moral predicate will stand out with what Dougherty calls a “special ethical glow.” The range of acceptable conventions could only mark out a range of acceptable uses for any particular moral predicate.
Alternatively, one may hold that the content of moral predicates does not depend on people’s attitudes or conventions. For example, if some kind of attitude-independent moral realism is true, it would be strange if the content of moral predicates were determined by their use. Miriam Schoenfeild develops this argument against the view that seemingly vague moral predicates are explained by accounts of semantic vagueness.Footnote 20 She notes that if moral predicates like “permissible” were semantically vague, then it would imply that collecting data about the uses of the term or asking an expert to offer more precise conditions for a moral term would offer us a precise answer to whether it was true that harming skin cells is permissible or not. But it doesn’t seem that researching a conventional use of a moral term would settle facts about the permissibility of damaging peoples’ skin cells. So, the indeterminacy of permissibility on which self-ownership accounts are based is not best explained as a kind of semantic vagueness.
Schoenfeild’s analysis suggests that if people have natural ownership rights that are not defined by conventions, then settling ownership conventions in a particular way (for example, allowing some people’s skin cells to be owned by others) would not settle facts about whether a person’s self-ownership rights were violated. Similarly, it would be surprising if the facts about whether it is acceptably risky for a person to drive in a busy city, or whether bumping a person on the subway were impermissible, were settled by taking a poll. It would also be surprising if the facts about ownership, acceptable risk, and permissibility could be settled by asking how an ideal agent would apply the term.
But even if the seeming indeterminacy of the moral predicates involved in self-ownership theory is not fully explained by the vague use of terms, the precisification of the language of the self and of terms like “acceptable risk” and “permissibility” through a better understanding of conventions or laws may clarify the boundaries of self-ownership to an extent. If so, then further precisification of relevant terms could potentially answer seeming dilemmas related to self-ownership theory being too permissive or too restrictive.
B. Unknowable self-ownership boundaries
Alternatively, it could be that even if we had a perfect language when discussing self-ownership theory and the moral predicates involved in it, we still could not know how to apply the terms in particular cases. This epistemic vagueness may also explain why self-ownership theory appears to encounter dilemmas related to being too permissive or too restrictive. For example, there could be a sharp cutoff between acceptable and unacceptable risk, but that cutoff could be unknowable, which would make it difficult to apply the predicate “x is acceptably risky” in cases where people believe that they are within a margin of error around the cutoff. It may not be that the theory has nothing to say at these borderline cases, but rather we don’t know what it has to say.
Timothy Williamson defends this epistemic account of vagueness more generally. He argues that in some cases, a term seems vague because the term is applied in cases where there is a risk of error in its application. For example, “bald” and “hairy” needn’t be understood as semantically vague for it to be difficult or even impossible to know whether a person is bald or hairy. This is because it’s difficult to know how many hairs a person has on his head. Perhaps our knowledge of the moral predicates involved in self-ownership is like Williamson’s account of baldness. We know that people own themselves and that it is impermissible to infringe on those ownership rights or to subject them to unacceptable risks, but we’re uncertain in some cases whether we can correctly say that a particular small infringement was an impermissible violation of an ownership claim.
This strikes me as a plausible view. Sometimes it’s hard or impossible to know if an act violates a person’s ownership rights by imposing risks on her or by interfering with the boundaries of the body in a certain way. Some moral predicates are knowable in core cases but unknowable in marginal cases. Even if we had a perfect language to describe the nature of permissibility or acceptable risk or the self, these predicates’ precise application in borderline cases could still be impossible to know.
Some people will be skeptical of this solution because they doubt that moral predicates can be determinate while their boundaries are unknowable. For example, Dougherty and Schoenfeild are skeptical that moral predicates can be epistemically vague. Dougherty worries that this account would imply that morality is arbitrary since the application of moral predicates would depend on sharp and unknowable cutoffs. Schoenfeild worries that on this view morality is not action-guiding. I argue that Dougherty’s objection is not that bad for any moral theory and that Schoenfeild’s objection is false.
Dougherty argues that if moral predicates like acceptable risk and permissible are determined by very small changes, then seemingly insignificant shifts could make a big difference morally. For example, a change from a 38 percent risk of harm to a 39 percent risk of harm could determine whether a risk is acceptable or unacceptable. A change between an intervention costing its victim one year of life versus one year and one day of life could determine whether it is in fact a permissible infringement. But if it turned out that moral predicates, like acceptable risk and permissibility, were determined by such small differences, then perhaps the distinction between unacceptability and acceptability or impermissibility and permissibly would not matter so much morally after all, since they would ultimately track seemingly arbitrary small differences.
Sobel echoes something like Dougherty’s argument, specifically with reference to self-ownership cases. Sobel writes:
“If the [self-ownership theory] is developed in the threshold manner, then there will implausibly be cases where risks just below the threshold are no problem as far as my rights are concerned but just over it is a full rights violation. Such views will be forced to maintain that arbitrarily small additional impositions of risks make a very great moral difference—a greater difference than a larger amount of risk that took us near to the threshold.”Footnote 21
The objection is that if self-ownership theory specifies some sharp, unknowable cutoff between acceptable and unacceptable risk, or what is permissible and impermissible, then the theory seems arbitrary because small and unknowable distinctions would have disproportional moral significance.
The intuition behind this objection seems to be that if moral facts supervene on nonmoral facts, then small differences in the world should align with small moral differences as well. But this intuition begs the question against any moral theory that appeals to nonmoral concepts or properties with thresholds. For example, if moral status begins at conception, then the difference between an egg and a fertilized embryo is small but greatly morally significant. If agency is the basis of moral status, then the difference in capacities between an agent and a non-agent may be small but greatly morally significant. If hedonistic utilitarianism is true, then the difference between perceptible and imperceptible pain is small but morally transformative.
So too, if self-ownership theory is true, then the difference between a permissible and impermissible infringement may in fact be determined by a small, unknowable distinction. For example, it could be that a person’s entitlement against exposure to risk is determined by whether she would be rationally required to account for the risk in her deliberation if she knew about it, and that the boundaries of this requirement are very fine-tuned such that small changes in risk exposure could be morally transformative.
A related objection is that it is problematic if unknowable distinctions have great moral significance because morality must be knowable in order to be action-guiding.Footnote 22 The objection asserts that there cannot be moral facts whose application to particular cases is in principle unknowable. A concept is “in principle unknowable” if it is unknowable even if it were defined in a perfectly precise way.
The first response is to deny that morality must be knowable as a general matter by denying that a person’s perceptions of her circumstances matter for the moral status of her actions.Footnote 23 People who think that the moral status of an act is determined by its consequences sometimes say things like this.Footnote 24 If that’s true, then it could be that whether an act trespasses the boundaries of the body, is unacceptably risky, or is impermissible, is unproblematically unknowable.
The second response is to deny that the particular kind of unknowability involved in cases of vague moral predicates is a problem by denying that the unknowability associated with vagueness prevents morality from being action-guiding. A moral predicate related to self-ownership, such as acceptable risk or permissible interference, could be in principle unknowable as applied to borderline cases and yet still action-guiding, if it is action-guiding in clear cases and if in borderline cases people adopt meta-norms that inform their actions. For example, say the threshold between unacceptable risk and acceptable risk is unknowable. And assume further that the unknowability of the threshold is known. In response to the known unknowability of the margins of acceptable risk, people may judge that they have a moral duty to adopt a meta-norm of caution as they approach the margins.Footnote 25 Such a principle could be action-guiding in the face of unavoidable moral ignorance.Footnote 26
So even if it were impossible to know whether an act violated a person’s claim to self-ownership, all that implies is that the act may be permissible if the actor is morally lucky but impermissible if he is morally unlucky, and he cannot know at the outset whether he will be lucky or unlucky. But these cases are not unfamiliar. Just as a would-be drunk driver should adopt a meta-norm of caution, and refrain from drunk driving on the grounds that it is better to take a taxi home than to potentially harm a pedestrian while driving drunk, meta-norms can inform behavior in this same manner when the application of other moral predicates related to self-ownership is unknowable.
For example, let’s stipulate that a person cannot know whether it is permissible to pollute the air with pollen by cutting down a tree or if it is permissible to reduce someone’s life expectancy by one year in a particular case. A person can still know that the application of “permissibility” is unknowable in these cases. And in light of her knowledge of her lack of knowledge, she may adopt a meta-norm of responding to this uncertainty. She may decide that it is best to refrain from exposing people to seemingly minor harms when she cannot know whether doing so would be permissible. Or she may decide that as long as she truly cannot know whether an act would be permissible, she is off the hook and therefore at liberty to do it. Either way, her judgment that the facts about permissibility in these cases are unknowable is still action-guiding.
Deploying meta-norms to make decisions in boundary cases can also enable the self-ownership theorist to address the objection that self-ownership theory would be arbitrary if people’s claims depended on small unknowable distinctions. After all, it could be that people should adopt a meta-norm to apply a higher justificatory standard to actions whose permissibility is known to be closer to the unknowable threshold of a violation of a person’s self-ownership claim.Footnote 27
C. Ontic moral vagueness
So far, I have argued that self-ownership may seem extensionally inadequate in boundary cases because the language of self-ownership is indeterminate or the nature of people’s claims are unknowable around certain thresholds. A third option is that even if we had a perfect language when discussing self-ownership theory and the moral predicates involved in it and had perfect knowledge about how to apply the terms in particular cases, self-ownership theory would still encounter dilemmas related to being too permissive or too restrictive in these boundary cases. If so, perhaps some aspects of the theory are ontically vague, meaning that the underlying nature of the self, of acceptable risk, or of the theory’s claims about permissibility, are indeterminate.
Ontic vagueness is controversial as a general matter, and the question whether moral predicates can be ontically vague is probably more controversial. But I think that it is plausible that moral predicates can be ontically vague. And if they can, then this explanation may also resolve some apparent paradoxes in self-ownership theory. Maybe cases in which self-ownership theory seems too permissive or restrictive are actually cases in which the theory is indeterminate.
One possibility is that self-ownership claims appear indeterminate because they rely on descriptive predicates that are ontically vague. For example, it could be that the boundaries of the body are not fully determinate. After all, people modify their bodies by using prostheses and pacemakers, by adding cosmetic enhancements like breast implants and horns, and by removing gall bladders and tumors. We can imagine a world where people modified their bodies even further or a world where people connected their body parts to virtual avatars that extended their sphere of bodily use. These practices and technologies could blur the line between the body and external objects. If so, then even if the language we use to talk about bodies were perfectly precise and even if we had full information about the physical composition of a person’s body, it would still be indeterminate whether a recently removed skin cell was a part of someone’s body. This indeterminacy should not, however, undermine our confidence that we broadly understand what bodies are.
Some people deny the conceptual possibility of ontic vagueness on the grounds that saying it is sometimes indeterminate whether something is true or false seems to violate the law of excluded middle. Against this denial, Elizabeth Barnes defends the claim that ontic vagueness is possible. On her view, something like a body could have more than one precisification, and for all precisifications of “body,” it is determinately true or false whether something is part of the body or not, but indeterminate whether we are in the world where the precisification of “body” is actualized in a way that includes that body part or not.Footnote 28 Or as Barnes puts it, it could be “perfectly determinate that things are precise but . . . indeterminate which precise way things are.” I’m not committed to Barnes’s specific defense of ontic vagueness here, only that metaphysical indeterminacy is at least possible.
If we grant that some things can be ontically vague, then perhaps ontic moral vagueness is possible too. If so, then perhaps self-ownership claims seem indeterminate because they appeal to moral predicates that are ontically vague. For example, even if we had full information about the risk of death that a choice would impose on people and a perfect language that precisely specified what “acceptable risk” meant, whether a choice that imposed a 0.3 percent risk of death on a person was acceptably risky could still be indeterminate. Or, if the basis of self-ownership rights is an indeterminate convention or practice, then self-ownership rights could be indeterminate as well. Or if self-ownership rights are independent of conventional understandings, it could still be the case that the permissibility of an infringement or the acceptability of exposing someone to risk could be indeterminate within certain parameters.
If it were the case that the boundaries of bodies, acceptable risk, and permissibly were ontically vague then, as in the epistemic case, a person may reasonably adopt a meta-norm to inform her actions in response to the indeterminacy of these boundaries. This characterization of self-ownership theory could also explain why marginal cases strike us as such a challenge to the theory. One aspect of Sobel’s objection to self-ownership theory was that any characterization of a sharp cutoff seems arbitrary, ad hoc, and extensionally inadequate. But if some of the predicates on which the theory is based are ontically vague, then this would explain why attempts at a precisification make the theory appear arbitrary, ad hoc, and inadequate. These attempts would be misguided if they amount to attempts to specify something that is itself indeterminate.
D. Partners in indeterminacy
Self-ownership can be vague at various points in the theory (related to the boundaries of the self, acceptable risk, and permissibility). The theory can be vague in three ways. The language of self-ownership can be imprecise. The truth about self-ownership can be unknowable as it applies in boundary cases. Or the truth about the boundaries of the self, acceptable risk, and permissibility can themselves be indeterminate.
Perhaps some readers are skeptical about some of these accounts of moral vagueness, but those who allow that there can be moral vagueness should also allow that the vagueness of self-ownership may not undermine the theory any more than vagueness undermines other moral theories. And all moral theories appeal to values or moral predicates that are vague and that either do or do not require determinate prequalification. Arneson writes that self-ownership theory is less determinate than rival theories, such as act utilitarianism or Rawls’s liberal egalitarian view. But both theories also appeal to vague descriptive terms, such as “perceptible pain” or “person,” which rely on a threshold-concept to ground a person’s moral status. For example, definitions of “persons” that aim to distinguish persons from animals encounter marginal cases of persons, such as children, people with cognitive disabilities, or cognitively advanced animals, like gorillas.
Like self-ownership theory, both theories also must offer an account of acceptable risk. For example, act utilitarians must defend thresholds beyond which existential risks are so great that the moral imperative to avoid them outweighs the moral value of other projects. Liberal egalitarians must defend a theory of whether people can reasonably reject a policy that exposes them to some level of risk, or whether exposing a particular group to risk is compatible with treating them as equals.
The result of this analysis is that self-ownership theory may be vague in a way that does not require a determinate precisification, in the same way that other moral theories can be vague in ways that do not require determinate precisification. To close this section, I also want to note one potential broader implication of the foregoing arguments. It could be that, as a general matter, seeming moral dilemmas are best explained by cases in which the moral principles at stake are either indeterminate and unknowable or ontically vague. If so, then it may be that some cases that appear to be cases in which moral principles fall short are actually cases in which the borderline between permission and obligation or permission and prohibition are themselves indeterminate.Footnote 29
III. Why Marginal Cases Make Bad Theory
In the previous section, I developed a series of arguments to defuse the claim that boundary cases were problems for self-ownership theory. Seeming problems associated with attempting to draw sharp cutoffs may stem not from the theory but from the desire for a sharp cutoff. I then argued that this is a more general problem for all moral theories. Many people are skeptical that moral vagueness is intelligible, though. And some are skeptical that any epistemic or ontic vagueness is intelligible.Footnote 30 If these critics are right, then would borderline cases that have implausible extensions under any precisification then count as objections to self-ownership theory?
In this section, I argue that even if vague moral predicates must have some determinate precisification, moral theories nevertheless should not generally be rejected solely on the grounds that the extensions of a precisification of a theory seem inadequate. I argue that theories that explain complex phenomena either have counterexamples or they are ad hoc. If one must tradeoff in theory selection between an ad hoc model and a model with counterexamples, one should choose a model with counterexamples. Therefore, the fact that there are counterexamples to a theory is not a sufficient reason to reject the theory. Even if self-ownership theory encounters some strange counterexamples that seem to imply it is either too permissive or too restrictive, that’s not a reason to reject self-ownership theory.
I develop two versions of this argument. In the first, I argue that extensional adequacy must be balanced against other values and that theories should not be held hostage to their implications in particular cases if the theory has other merits. In the second, I argue that a theory’s extensional adequacy for marginal cases should be given even less weight in assessing the merits of the theory.
A. Discounting cases
Self-ownership can have surprising implications in particular cases. In addition to Sobel’s arguments against the theory, other critics of self-ownership theory, such as Steven Wall, reject self-ownership theory on the grounds that it seemingly prohibits even borderline cases of paternalism, such as soft-paternalism in cases where a person’s agential capacities are compromised.Footnote 31 And G. A. Cohen rejects self-ownership theory on the grounds that the theory prohibits seemingly permissible forms of redistribution while it allows seemingly impermissible forms of exploitation.Footnote 32
These critics assume that the counterintuitive implications, implausible extensions, and counterexamples to self-ownership theory discredit the theory. Thinking about cases may be unavoidable in thinking about the merits of a theory.Footnote 33 But the fact that philosophical arguments are generally evaluated on the basis of intuition does not imply that philosophers should discount philosophical arguments merely because they have counterintuitive implications for particular cases, even if this judgment is widely shared.Footnote 34 On its own, widespread disagreement with a judgment about a particular case is insufficient to establish that a general theory is false on the grounds that it yields that case-based judgment.
For this reason, critics of self-ownership theory who cite the theory’s counterintuitive implications are committed to a controversial methodological claim that theories ought to be evaluated on the basis of whether they have intuitive implications for particular cases. This methodological approach of weighing intuitions about principles or theories against intuitions about cases is sometimes called reflective equilibrium. On closer inspection, this methodological assumption is either a bad method or the method does not provide grounds for rejecting the theory. Reflective equilibrium is a bad method if it holds that extensional adequacy for cases is a necessary condition for endorsing a theory. Reflective equilibrium is irrelevant to whether self-ownership theory is true if it only holds that intuitions about cases should play a role but that these intuitions ought to weigh against other theoretical and methodological desiderata.
If we understand reflective equilibrium as a method of evaluating principles and theories primarily in terms of their adequacy for cases, then it is a poor method for three reasons. First, people’s case-based intuitions are potentially very unreliable. Second, the method gives no guidance for determining which case-based intuitions should have greater weight in theory evaluation. Third, because the method holds moral theories hostage to people’s case-based intuitions, it inhibits progress in moral theorizing.
The first reason to discount the relevance of case-based intuitions for evaluating a theory is that people’s intuitions about cases are unreliable, especially in areas of philosophy that deal with complex social phenomena with many moving parts or areas where people’s intuitions are especially likely to be influenced by normatively extraneous factors, such as people’s social identities or desirability biases.
Debates in political philosophy, including those that address self-ownership, are clearly debates about complex social phenomena. In these circumstances, Williamson writes,
The more adjustable parts a model has, the more opportunities it offers the model-builder to rig the results, to gerrymander the model by setting parameters and arranging structure in ad hoc ways to fit preconceived prejudices. Simplicity, elegance, symmetry, naturalness, and similar virtues are indications that the results have not been so rigged. Such virtues may thus ease us into making unexpected discoveries and alert us to our errors.Footnote 35
If ever there were an area of philosophy where theorists examining complex social phenomena faced the temptation to rig the results to fit their preconceived prejudices, it must be political philosophy. Not only is politics complex in this way, but theorists’ intuitions are also especially vulnerable to influence from ideological and other identity-based biases.Footnote 36
For these reasons, those engaged in theory selection in political philosophy should consider the model-building approach. Proponents of a model-building approach to philosophy hold that extensional adequacy is only one among several desiderata that a theory could have, and not a necessary condition for a model to be informative and useful. And model-builders should be especially wary of models that seem extensionally adequate but have many moving parts or trade simplicity for nuance.Footnote 37 No sufficiently coherent moral theory can accommodate the bulk of people’s case-specific intuitions.Footnote 38 Self-ownership theory, like other theories, is a simplification of the normative domain. Proponents of any general moral theory should adopt skepticism about case-specific intuitions and be open to revision of case-specific judgments.
If we think of theorizing as a process of simplification or modeling judgments of right action, the fact that self-ownership theory is comparatively parsimonious should count in its favor. I say that the theory is comparatively parsimonious because like other moral theories, it appeals to some notion of rights over the self, requires a theory of permissibility, and in some cases, requires judgments of acceptable risk. All rival political principles that appeal to rights also require these theoretical judgments, but must add additional theoretical apparatus in order to secure more intuitive conclusions.
The second reason to discount the relevance of case-based intuitions for evaluating a theory is that, even if one rejects the model-building approach, it’s unclear which case-based intuitions one should reject. Self-ownership has very plausible implications in a range of cases. It explains why assault is morally impermissible, why a kidney tax is wrong, and, more generally, it affirms a widely held commitment to inviolability. Many people intuitively think of themselves as self-owners. But when we take this view seriously, it has some surprising implications. Within the method of reflective equilibrium, there is no reason to give priority to one class of cases over the other. Some people will feel the force of one set of intuitions more strongly than others, and from these different intuitive starting points, the method cannot adjudicate disputes between them.Footnote 39
The third reason to discount the relevance of case-based intuitions for evaluating a theory is that doing so potentially deters moral progress. While the fact that a theory runs counter to a consensus judgment may indicate that the theory is false, it may also indicate that people should revise their judgments or that there is a widely shared reason for people’s counter-intuitions.Footnote 40 The method of reflective equilibrium cannot account for these different possibilities. There is nothing internal to the method that can tell people when to reject a theory on the basis of an intuitive judgment about cases and when to revise their judgments. In this way, a strong methodological commitment to extensional adequacy holds theories hostage to the conventional wisdom of their time.
A proponent of reflective equilibrium may respond to this line of criticism by arguing that the method does not require that theories have plausible implications in all cases, only that the balance of implausible implications does not outweigh the independent merits of a theory. One may argue, for example, that a good theory is sufficiently universal and elegant to produce generalizable judgments or hypotheses and sufficiently defined and nuanced to capture the phenomenon under consideration. So too, a good moral theory is sufficiently parsimonious to produce simple rules that inform right action, yet sufficiently intuitive to serve as a guide to action in actual cases. These two desiderata trade-off against each other. A model that is parsimonious has fewer parts and variables. A model that is intuitive in a range of cases has enough parts to generate a conclusion that seems plausible in a range of cases.
Self-ownership theory is no exception to this tradeoff. A theory of self-ownership that is very simple and general will seem implausible in some cases while a theory of self-ownership that does not generate seemingly implausible conclusions will look less like a theory of self-ownership. For this reason, arguments against self-ownership often focus on marginal cases where either the theory commits its proponents to seeming reductios of the view or requires proponents of self-ownership to abandon core aspects of the theory.
If this is how we understand reflective equilibrium, as a method of weighing tradeoffs between theoretical virtues like parsimony and robustness and extensional adequacy in particular cases, then I have no objection to the approach, and it is similar to what I am calling model-building. On this interpretation, the method of reflective equilibrium still provides no independent guidance for adjudicating between theories, though, and so such a method would not favor rejecting self-ownership theory primarily on the grounds that it has implausible implications, no matter how counterintuitive those implications were.
This is not to say that extensional adequacy is entirely irrelevant, only that it isn’t always necessary and other model-theoretic desiderata can be equally important in evaluating a theory, depending on one’s purpose.Footnote 41 For example, a successful scientific or game-theoretic model of human behavior would be tractable and get broadly consistent results across a range of simplifications, even if it did not accurately describe all observed phenomena. Philosophical models may be evaluated on similar grounds.Footnote 42 If self-ownership theory were generally tractable and got consistent results across various specifications, it would be an advantage of the theory, even if it did not affirm all of our pre-theoretic intuitions.
These methodological claims may strike some readers as partisan to the view I am defending. But it is worth noting that I am not arguing for discounting intuitive judgments about particular cases entirely and also that the general practice of discounting case-based intuitions does not necessarily favor self-ownership. Self-ownership theory has some advantages over more complex or nuanced theories, but it is not as powerful a theory as rival views, such as utilitarianism. On the other hand, self-ownership theory is more extensionally adequate in core case than rival views. The tradeoff, then, against parsimony, robustness, and tractability may be worth it.
The point of these methodological arguments is not to establish that political philosophy done right will support a commitment to full self-ownership. Rather, my aim has been to defuse attempts to reject self-ownership theory on the grounds that it is extensionally inadequate. Critics of the theory must make the case for a methodological balance that highly weighs extensional adequacy over other model-theoretic desiderata in theory selection and then show that whatever they are arguing in favor of fares better than self-ownership theory by those criteria.
B. Discounting marginal cases
Even if extensional adequacy were very important for the success of a theory in a broad sense, my second line of response to arguments that dismiss self-ownership theory for its implications in particular cases is that extensional adequacy should not be required in borderline cases. On this view, if self-ownership theory is powerful and intuitive in a wide range of cases, then failure at boundary cases is less damaging to the theory than failure in core cases would be. This line of response builds on arguments that some kinds of case-based intuitions are better than others.Footnote 43
To illustrate this argumentative strategy, consider an analogy to utilitarianism. The theory may appear to break down in cases of imperceptible harms. If the theory were made to include imperceptible harms that approached the margin of perceptibility, then a great many acts that caused no experienced harm of any kind would seemingly be wrong according to utilitarianism. But if imperceptible harms are excluded from the analysis, then sets of acts where many people inflict imperceptible harms that together add up to a perceptible harm would be permissible according to utilitarianism. Should we infer from this dilemma that utilitarianism is incoherent if it is extensionally inadequate at the thresholds of perceptible harm? The theory should not be rejected on these grounds if in core cases—those of perceptible harm and actions that are imperceptible and not harmful—the theory gave clear guidance that was broadly plausible in a range of other cases.Footnote 44
Similarly, if the theory of self-ownership is simple, strong, coherent, powerful across a range of cases, tractable, and broadly extensionally adequate, then the presence of marginal cases related to photons of light or sound waves or pollen hitting people’s skin and eardrums do not qualify as grounds for rejecting the theory any more than the Liar’s paradox is grounds to reject principles in logic. So perhaps, even if self-ownership theory does entail that it is either permissible to enter a person into a death lottery or that it is wrong to fly airplanes overhead, that observation would not on its own discredit the theory itself. Even if the theory implied that killing one person to avert a nuclear disaster was impermissible, if it got plausible judgments about the permissibility of killing in all the core cases, then perhaps the theory should not be thrown out entirely on the basis of a hypothetical mass killing.
Joshua Gert develops a more general argument against focusing on exotic and marginal cases when evaluating moral theories.Footnote 45 Drawing an analogy to color, Gert notes that the boundaries between colors may be vague or indeterminate, but the difference between yellow and purple is clear. Gert acknowledges that in marginal cases there may be no correct answers or the correct answers may be vague, as I suggested above. If this were the case, it would not undermine the judgment of the theory in clearer cases of self-ownership. Or, Gert suggests, some marginal cases are mistakenly seen as challenges to a moral theory, but these judgments are artifacts of psychological biases, cultural norms, or evolutionary considerations.
For these reasons, seemingly borderline cases of bodily rights violations are not reliable guides to the ethics of self-ownership. Freiman and Lerner find that people are more condemnatory and resistant to invasive but minor violations of self-ownership (painlessly taking a body part or fluid) compared to noninvasive violations of a similar order.Footnote 46 On their view, these cases show that core judgments of self-ownership violations are really motivated by evolved disgust reactions. But another interpretation of these data is that people’s intuitive reactions to marginal cases, such as painless and minor violations that are never perceptible to the victim, are a poor guide to the more general subject of self-ownership theory because these judgments are artifacts of evolution and psychological biases.
As above, this is not to say that theorists should not take marginal cases seriously and attempt to overcome biases, norms, and evolved responses. It is worth trying to refine a theory in ways that avoid dilemmas and paradoxes. Marginal cases can also be instructive in learning about the true nature of a theory and learning about its implications in limiting circumstances. There is value in considering marginal cases because they can illustrate the nature of a theory. But when we do consider these cases, we should resist the temptation to reject the theory on the grounds that it has counterintuitive implications in them.
Perhaps we expect theories to do too much. All theories that function well as theories function especially poorly in rare circumstances where the theory does not fit well with the world. And to the extent that theories change to fit the data or our intuitions, they look less like theories and more like mere descriptions.
IV. Blasphemy and Bullet-Biting
The foregoing sections focused on undermining the premise that the constituent principles of self-ownership theory required determinate precisification and rejecting the use of the premise that any precisification of the predicates required to justify self-ownership theory is either too permissive or too restrictive to justify self-ownership theory. These responses to the argument against self-ownership theory are essentially meta-philosophical. They reject the premises on methodological grounds rather than substantive grounds.
Another response to the argument accepts the claim that any precisification of the predicates required to justify self-ownership theory will appear at first to be either too permissive or too restrictive but denies that it is in fact too permissive or too restrictive. In other words, one might deny the fourth premise of the argument and defend a counterintuitive interpretation of self-ownership or accept that the theory has some strange and counterintuitive conclusions in marginal cases on the grounds that the theory is true. The first response entails revising self-ownership theory in ways that may seem blasphemous to committed proponents of self-ownership theory but nevertheless preserve the main contours of the view. The second response is an exercise in bullet biting for the sake of preserving self-ownership theory.
A. Small revisions to self-ownership theory
Consider first an exercise in blasphemy against the theory of self-ownership. Maybe the metaphysics of the body is determinate and knowable, but if our language perfectly described the knowable and sharp boundaries of a person’s body, we would be surprised to find that skin cells are no longer a part of a person’s body if they have been detached for more than 30 seconds or if they are more than 30 feet away. Or maybe people do not have exclusive rights to control the entire top layer of their epidermis. That is, maybe self-ownership theory as it is currently understood includes more extensive protections for bodily integrity than it should.
Turning to questions of risk, concerns about the precisification of the acceptable boundaries of risk may not highlight a deep problem with self-ownership theory but rather reveal that the theory is in need of revision or an accompanying deontological decision theory. Seth Lazar, for example, argues that deontological theorists can adopt a decision theory for risk that resembles a more consequentialist decision-rule without committing deontology to a principle that treats people as means to the maximization of some other, agent-neutral value.Footnote 47 Here again, the slight revision to the theory as it is currently understood may enable self-ownership theorists to avoid seeming extensional inadequacy.
Even for questions of permissibility, proponents of self-ownership theory may consider revising parts of the view to an extent to allow for more forms of interference and to avoid objections related to the view being too permissive. For example, in Section II, I argued that people should respond to cases of indeterminacy related to self-ownership theory by adopting meta-norms, such as a norm of caution. There are other metanormative principles that may inform how people respond to self-ownership too, such as a norm of minimizing harm in response to indeterminacy. If these principles informed people’s action in the face of indeterminate self-ownership claims, then even if it would be permissible to infringe on some rights that looked like bodily rights once a sharp boundary was specified, people would have good reason to play it safe and refrain unless it was for the sake of the greater good. On this view, one could revise self-ownership theory to allow for some infringements for the sake of the greater good by noting that when there is a much greater good to be achieved and it is seemingly indeterminate whether one’s self-ownership claim is violated at all, then some interference for the sake of the greater good could be justified.
An alternative way of revising the theory is to limit the ambitions of the theory. Perhaps the theory itself does not define the boundaries of what is permissible and impermissible, acceptable and unacceptable risk, or what counts as a body or not. Instead, perhaps self-ownership theory is one component of a broader set of theoretical commitments that public officials should consider when deciding what to do. In this way, self-ownership theorists could insulate the theory from a charge of extensional inadequacy by claiming instead that the theory only describes one corner of the moral domain and is not meant to inform what people ought to do all things considered.Footnote 48
One worry about these strategies for revising self-ownership theory to accommodate or address too permissive/too restrictive dilemmas is that the theory could encounter the aforementioned worry that once specified in a plausible way, the theory is too general to offer any distinctive guidance in particular cases. For any of these cases, the self-ownership theory would need a story for drawing the lines at a particular point that appeals to values internal to the theory. If not, then a revised version of self-ownership theory encounters the aforementioned objections related to being ad hoc. I have not given such a story in this section. Instead I illustrated how a theory could inform action in core and borderline cases even if it is specified in a general and indeterminate way. Revising the theory in these ways would not make it so that the theory allowed for coercive organ redistribution or paternalism or forced labor, but it would enable the theory to inform conduct in a broader range of cases and avoid seeming dilemmas related to the precisification of vague aspects of the theory.
B. Revisions to case-based intuitions
A second line of defense for self-ownership theory is to bite bullets. Assume there is a sharp boundary between permissibility and impermissibility. If we knew the nature of that boundary in particular cases, then it would be wrong to interfere with a person’s bodily rights, no matter how substantial the gains to others. The same can be said for imposing small risks on people. Wherever that boundary of acceptability lies, it would then be permissible by the lights of self-ownership theory to pollute or to play Russian roulette with people’s heads when the risk of death associated with each act was below the relevant threshold. There may be other moral reasons, of course, not to play Russian roulette with a person’s head. But self-ownership theory on its own could permit it. And whenever the threshold of risk would deem it unacceptable to play Russian roulette with other people’s heads, pollution that exposed people to a similar level of risk would be unacceptable as well.
Perhaps we should therefore reconsider whether it counts against a theory that it finds that driving is morally impermissible because it exposes people to an unacceptable risk of injury or death and consider whether driving is permissible instead. The fact that self-ownership theory implies that nonvoluntary slavery and marital rape are morally impermissible would have been counted against the theory two hundred years ago. It could be that self-ownership theory is true and in two hundred years our descendants will view our acceptance of driving, polluting, inflicting risks on one another, and living in cities where we bump into people’s bodies and assault them with noise and light as morally unacceptable.
Here again, my argument in defense of self-ownership theory in this section may be read as an argument for evaluating self-ownership theory in the same way we evaluate other theories. All theories that are not ad hoc in the way I described in the third section will yield some surprising conclusions. But in moral theory, it may be that surprising conclusions are the ones we should accept.
V. Conclusion
In 1956 Elizabeth Anscombe petitioned and protested against Oxford’s decision to award an honorary degree to Harry Truman. Anscombe objected to honoring Truman on the grounds that he had committed mass murder. Later, she published a pamphlet explaining her disapproval (Mr. Truman’s Degree) and further defended her position in War and Murder, where she addressed objections to her position, including an objection she termed “Where do you draw the line?” Responding to this question, Anscombe writes,
The fact of twilight does not mean you cannot tell day from night. There are borderline cases, where it is difficult to distinguish, in what is done, between means and what is incidental to, yet in the circumstances inseparable from, those means. The obliteration bombing of a city is not a borderline case.Footnote 49
The same can be said for twilight cases related to other moral principles, such as the principle that it is impermissible to knowingly enslave or assault someone, that people should respect others’ entitlements to control their own lives, that paternalism is wrong, and that bodies are inviolable.
Self-ownership theory prohibits paternalistic interference, forced labor, violence, and at least some instances of redistribution. The difference between this theory and a theory that permits paternalism, forced labor, violence, and more extensive redistribution, including the redistribution of body parts, is night and day. There is more at stake in these debates than a methodological dispute. Theorizing on the basis of borderline cases can lead us astray and cause people to reject a promising theory that describes an important feature of the moral domain.
If it seems that self-ownership theory, or any moral theory, doesn’t get the right answers, it may not be because the theory is deficient but because people are asking the wrong questions. While borderline cases seemingly challenge a range of moral theories, such cases may reflect genuine empirical or conceptual indeterminacy. Even if borderline cases require precisification, and precisification seems extensionally inadequate, cases like these are an unreliable guide to the truth of a theory. Or if they are a reliable guide to the truth of a theory, it may only reveal that self-ownership theory was slightly different than we thought or had more radical implications than it seemed.
Beyond self-ownership theory, this argument has more general implications for evaluating all moral theories in light of indeterminate boundaries and borderline cases. Whatever we make of twilight cases, it’s a mistake to sacrifice daylight to avoid the dawn and dusk.
