Christopher Wellman argues that Joel Feinberg fails to grasp the distinction between will theories and interest theories. According to Wellman, this failure leads Feinberg to defend a will theory in “The Nature and Value of Rights” and an interest theory in “The Rights of Animals and Unborn Generations” and therefore to hold an inconsistent theory of rights. Wellman's interpretation of Feinberg is plausible. I will argue that there is another plausible interpretation and that the texts do not allow us to determine which of the two plausible interpretations is correct. I will also argue that this alternative interpretation allows Feinberg to avoid Wellman's charge of inconsistency but has problems of its own. Along the way, I will make some points about how to defend a theory of rights and about the strengths and weaknesses of the will and interest theories of rights. (Wellman holds that choice theories are a subset of will theories. I have argued elsewhere that it is better to see will theories as a subset of choice theories.¹

However, how one cuts up the field is irrelevant to the issue at hand. For the sake of argument, I will assume that Wellman's topology is correct.)

After an admirably clear presentation of Feinberg's view in the first two sections of his paper, Wellman begins his own arguments in the third section. First, he argues that Feinberg's argument for the view that animals can have rights is flawed. Feinberg's argument is a reductio. Feinberg holds that one cannot deny that animals have rights on pain of denying that human infants can have rights and that it is absurd to deny that human infants have rights. Wellman then presents Patricia Williams's argument for the view that denying that infants have rights is absurd. Williams argues that is it absurd to hold that infants do not have rights because this view implies that we have no moral reasons to refrain from mistreating infants. Wellman correctly responds that the view that infants do not have rights does not imply that we have no moral reasons to refrain from mistreating them. One could hold (and will theorists do hold) that there are non-rights-based moral reasons to treat infants well. Indeed, one could hold that the non-rights-based moral reasons for treating infants well are as strong (or even stronger) than the rights-based moral reasons for treating adults well.

Wellman has shown that Williams's argument is flawed. The problem for Wellman is that Feinberg never advances Williams's argument. Feinberg has a different argument for the view that it is absurd to hold that infants do not have rights. Feinberg's reason for thinking that infants have rights is that this is the settled law of all jurisdictions in the Anglo-Saxon world. Feinberg notes that “it is manifest that [human idiots and infants] have legal rights recognized and easily enforced by the courts.”²

Wellman quotes this argument in Section II of his paper but never attempts to show that it is flawed. Feinberg's argument is more powerful than Williams's because both interest theorists and will theorists have looked to the rights accorded by the law as models for rights in general. If the will theorist is free to ignore the settled law with regard to the rights of infants, the interest theorist may well argue that she is free to ignore the settled law that seems to conflict with her view. The will theorist who dismisses the settled law risks undercutting his own arguments.

There are at least two other reasons to think that the view that infants do not have rights is absurd. First, the view is radically at odds with the ordinary language of rights. The vast majority of competent speakers of English think that infants have rights. Second, the view implies that the moral reasons we have for not killing an infant are different from the moral reasons we have for not killing a typical adult. It is possible to hold both (1) that infants do not have rights; and (2) that there are non-rights-based moral reasons to treat infants well. However, one who takes this argumentative route must assert that while the reasons we have to refrain from killing an infant are just as strong as the reasons we have to refrain from killing a typical adult, the reasons are not the same. According to one who holds 1 and 2, we may cite the rights of an adult as a reason not to kill her but we cannot cite the rights of an infant as a reason not to kill him. That is implausible because most people think that the reasons one has not to kill an infant are the same as the reasons one has not to kill an adult.

Although Williams's argument for the absurdity of the denial of rights to infants is flawed, there are at least three independent arguments that Wellman has not considered. They seem to be sufficient to show the absurdity of the view. If the view that infants do not have rights is absurd, then it seems that any theory of rights that implies that view, such as the will theory of rights, must be rejected. Feinberg's assertion that “[i]t is simply not true…that the ability to understand what a right is and the ability to set legal machinery in motion by one's own initiative are necessary for the possession of rights” seems to be a decisive refutation of will theories of rights.

The word “seems” in the previous two sentences is important. Things are not always what they seem. I do not think that debates in rights theory (or philosophy in general) are best framed as debates over which views are absurd. Philosophy is not that simple. Perhaps it is possible in principle to use one argument to show that some complex theory on some complex philosophical topic is false. However, in practice it does not work that way. Rather, different, initially plausible, complex views have different strengths and weaknesses. In deciding which view is most plausible, one must review all strengths and weaknesses. I think that if a theory of rights implies that infants do not have rights, then that theory has a serious weakness. The theory may have sufficient strength to overcome this weakness, but the weakness remains. If Feinberg's theory of rights and the will theory of rights imply that infants do not have rights, that is a serious problem for those two theories.

In response to this line of argument, Wellman might well say that a defender of the will theory “is not making a moral claim about the status of young humans; instead, he is making a conceptual claim about the best way to conceive of rights for the special purposes of moral philosophy.”³

Wellman suggests that a theory of rights need not be too troubled about the ordinary language of rights because a theory is built for the special purposes of philosophy. Thus if a theory of rights departs from preanalytic use of a concept, if it implies that infants do not have rights, this is not a serious flaw. This response raises an important issue about how one defends a theory of rights.

There are two metrics by which to evaluate a theory of rights. The first is extensional adequacy. L.W. Sumner has presented this well:

The other way of evaluating a theory of rights is in terms of theoretical adequacy. One component of theoretical adequacy is nicely explained by Sumner: There is at least one other component of theoretical adequacy. In addition to drawing appropriate theoretical lines, one analysis of a concept is theoretically superior to another to the extent that the former solves or illuminates theoretical problems that the later does not. Suppose that theory of rights A provides an illuminating and satisfying solution to the problem of the rights of past and future generations but that theory of rights B does not. This would be a good theoretical reason to prefer theory A to theory B.

The quote above from Wellman suggests that theoretical adequacy is more important than extensional adequacy. I do not think that this is the case. I would not want to make the extremely strong claim that extensional adequacy is lexically prior to theoretical adequacy. Philosophy is not that simple, and there are some theoretical problems that are fatal to a theory. An inconsistent theory must be rejected. But I do believe that a theory of rights that is seriously inadequate when it comes to extension should not be accepted even if it has many virtues when it comes to theory. Theoretical matters provide good grounds to choose theory A over theory B only if the two theories are of roughly similar plausibility on the extensional side. If one were to take up Wellman's hint and decline to give extensional adequacy a significant priority over theoretical adequacy, one would no longer be defending a conception of the concept of rights. One would be creating a new concept “for the special purposes of moral philosophy.” Generally, and with some exceptions, the best theory of rights must cohere with the preanalytic extension of rights. There is no guarantee that rights as they are ordinarily understood will be of much help “for the special purposes of moral philosophy.” Philosophers can certainly wish and hope that rights as they are ordinarily understood do important theoretical work. But wishes do not make reality. Assuming that they seek a theory of rights and not a theory of smights, philosophers must remain open to the possibility that no extensionally adequate conception of rights has much theoretic use.

I argue above that Wellman has not successfully defended the view that a holder of a theory of rights need not be too upset if her preferred theory implies that infants do not have rights. Next, Wellman provides an interpretation of Feinberg's “Nature and Value of Rights” on which Feinberg is committed to the view that animals cannot have rights. Let us examine the textual evidence for this interpretation. The first textual evidence Wellman cites to support his interpretation is Feinberg's claim that “(t)he most conspicuous difference…between Nowheresvillians and ourselves has something to do with the activity of claiming.”⁶

One can see the plausibility of Wellman's interpretation. The “activity” of claiming is indicated as the difference between our world and Nowheresville, and this does provide reason to think that Feinberg must hold that a being that cannot engage in this “activity” cannot have rights.

On the other hand, there is another possible interpretation. One could look less at the word “activity” and more at the phrase “something to do with.” One could also note that at this point in the article, Feinberg is using the Nowheresville case to show merely that there is some sort of relationship between rights and claims. At this point in the article, he has not yet distinguished claiming that, making a claim, and having a claim. On this interpretation, the quotation above is merely a first, somewhat rough-and-ready pass at a view, and the “something to do with” indicates that he is does not require that a being actually be able to engage in the activity of claiming.

There are some texts that support this second interpretation: “one might have a claim without ever claiming that to which one is entitled, or without even knowing that one has the claim; for one might simply be ignorant of the fact that one is in a position to claim.”⁷

This text indicates that Feinberg thinks one can have a claim that one does not know about and that one never claims. This seems to weaken the link between having a claim and the activity of claiming. A right-holder cannot engage in the activity of claiming a right if she does not know that she has the right. Feinberg could see animals as beings that have claims without knowing that they have them.

The second part of the textual evidence for Wellman's interpretation of Feinberg begins with Feinberg's assertion that “[g]enerally speaking only the person who has title…or someone speaking in his name, can make claim to something as a matter of right. It is an important fact about rights (or claims), then, that they can be claimed only by those who have them.” Wellman then asserts that if Feinberg holds that animals can have rights because someone can make claims for them, then Feinberg must choose between the view that anyone can speak on behalf of the right-holder and the view that only someone the right-holder appoints can speak on behalf of the right holder. But Feinberg asserts: “Anyone can claim…that this umbrella is yours but only you or your representative can actually claim the umbrella.”⁸

This is plausibly read as the denial that anyone can speak on behalf of the right-holder. This implies that only someone the right-holder appoints can speak for the right holder, and since animals cannot appoint anyone, they cannot have rights.

There is a problem with this line of argument. It is that some rights theorists (interest theorists) deny that if one holds that animals can have rights, then one must choose between the view that anyone can speak on behalf of the right-holder or that only someone the right-holder appoints can speak on behalf of the right holder. One might hold (and interest theorists do hold) that there can be rights even if no one can speak on behalf of the right-holder. It is possible that this was Feinberg's view. Feinberg could consistently assert that “only you or your representative can actually claim” that to which you have a right (e.g., the umbrella) and that there are some cases in which no one can actually claim that to which a right-holder has a right.

The third argument Wellman makes in support of his interpretation of Feinberg is that “perhaps the best way to see that Feinberg cannot consistently posit animal rights is to return to what he says about the value of rights.”⁹

Wellman notes that Feinberg thinks the value of rights comes from the self-respect that one can have only if one thinks of oneself as a right-holder. Wellman then notes that animals would have the same concept of themselves whether they lived in our world or in Nowheresville. Wellman thinks that this argument is “decisive because if the nature and value of rights is thought to be most clearly revealed by contrasting our world with Nowheresville, then it seems to follow that no right-holder could be indifferent between these two worlds.”¹⁰

Wellman assumes that there is a tight relationship between one's theory of the nature of rights and one's theory of the value of rights. He assumes that what makes rights valuable largely determines the nature of rights. This is a reasonable assumption. But so is its denial. There is another plausible way to read Feinberg. He might have held that there is a looser connection between the nature of rights and the value of rights.¹¹

One might see the comments in the Nowheresville example about the value of rights as comments about why those beings that can value rights do value them. On this interpretation, animals and infants do not value their rights, but that does not mean that they do not have them.

Some of Feinberg's texts support this second interpretation. Consider the following from his earlier article, “Duties, Rights and Claims”:

This passage provides the most support for the second interpretation of Feinberg. Here he says that the verbal usage of “claim,” the activity of claiming, reveals why rights are valuable but does not say anything about the nature of rights. But substantive usage of “claim,” to have a claim, is what is crucial when one seeks to understand what rights are. On this view, animals can have a claim but cannot claim, can have rights but cannot claim them. However, this passage does not conclusively show that the second interpretation is correct. One could plausibly argue that this early article is not an accurate representation of Feinberg's later views.

Perhaps a look at the explicit conclusion of “The Nature and Value of Rights” will resolve this interpretational debate:

The material in roman typeface is Feinberg's conclusion about the nature of rights. The material in italics is his conclusion about the value of rights. There is some support for the second interpretation of Feinberg in this quote. Feinberg's explicit theory of the nature of rights asserts that a being has a right when the being has a claim against someone whose recognition as valid is called for by some set of governing rules or moral principles. A claim can be valid according to some set of rules or principles even if no one asserts that it is valid and, indeed, even if no one knows that it is valid. So it would seem that animals can have rights.

On the other hand, there is also support for Wellman's interpretation in this text. Notice Feinberg's assertion that to have a case meriting consideration is “to have reasons or grounds that put one in a position to engage in performative and propositional claiming.” The phrase “engage in performance” provides support for the view that Feinberg believes that to have a claim requires the ability to make a claim. On the other hand, one could hold that “to have reasons that put one in a position to engage” in an activity is not necessarily to be able to engage in it.

Just as there are issues in the interpretation of “The Nature and Value of Rights,” there are interpretational issues in “The Rights of Animals and Unborn Generations.” The conclusion of the second article is that:

This view is certainly consistent with an interest theory of rights and suggests that the point of rights is to protect interests. On the other hand, suggesting is not the same thing as saying, and the text is also consistent with a rejection of interest theories of rights. One could hold that the sorts of beings who have rights are those that can have interests but could decline to take the next step and hold that a being has a right just in case one of its interests is protected by another's duty or just in case one of its interests is the justification for another's duty.

The texts do not provide us with the resources to determine which of the two possible interpretations outlined above is correct. Feinberg himself never wrote an article that was a detailed attempt to determine what some famous scholar's view was. He was more interested in substantive matters and less interested in interpretational matters. I share Feinberg's interests. Therefore, let us turn from the debate over what Feinberg's view was and consider what really matters, the intellectual merits of the two possible interpretations of his view.

I think Wellman's critique of the first possible interpretation of Feinberg is correct. On this reading, Feinberg defends a will theory in “The Nature and Value of Rights” and an interest theory in “The Rights of Animals and Unborn Generations.” If the first possible interpretation is correct, then Feinberg's view is inconsistent, and we have noted that inconsistency is a fatal theoretical flaw.

On the second interpretation of Feinberg, matters are more complex. On this interpretation, “The Nature and Value of Rights” first defends a theory of the nature of rights. One has a right when one has a claim, and one can have a claim even if no one can make that claim. It then argues for a theory of the value of rights. The value of rights comes from the activity of claiming. In “The Rights of Animals and Unborn Generations,” Feinberg turned to the question of what sort of things can have rights and held that animals and infants have rights because they have interests and therefore have claims that those interests be respected. But they are not able to engage in the activity of claiming, so they do not value their rights. On the second interpretation, Feinberg's view is neither an instance of the will theory nor an instance of the interest theory.

What are we to say of Feinberg's theory of rights on this second interpretation? It has at least three flaws. First, it cannot account for immunity rights. No account of rights can be plausible unless it can account for the immunity rights found in contract law and the Bill of Rights. Cory owns the house at 436 Main Street in Decatur, Georgia. She has a right to it. A crucial part of her right to this house is an immunity to your transferring ownership of it to yourself. Your utterance of “The house at 436 Main Street belongs to me” has no effect. It is what lawyers call a legal nullity. A similar structure is found in my right to free speech. I have an immunity to George Bush's taking away my liberty to say that the Iraq war is unjust. His utterance of “It is henceforth illegal to say that the Iraq war is unjust” would have no legal effect.

My mother is a professor of English. As a child, when I would ask, “Can I go outside and play?” she would respond, “I don't know if you can but you may.” Claims (and their correlative duties) concern what one may or may not do. Cory has a claim that you not hit her, and you have a duty not to hit her. You may not hit her. Immunities (and their correlative disabilities) concern what one can or cannot do within a particular set of rules. Cory has immunity to your transferring ownership of the house at 436 Main Street. You have a disability. You cannot transfer ownership of the house.

On either interpretation, Feinberg's theory is entirely focused on claims. Feinberg holds that “a right is a kind of claim.” Feinberg has overlooked the fact that in addition to actions we may not do, there are actions we cannot do. Some rights are correlative to actions we cannot do. Because of this flaw, Feinberg's theory cannot provide an adequate account of such paradigm rights as Cory's right to her house and my right to free speech.

The second problem with the second interpretation of Feinberg's view is that it has no theory of the relational nature of rights. (On the first interpretation, it has two theories of the relational nature of rights.) Feinberg is aware that some duties are to others, are relational, while others are not. If you and I agree that I will give you ten dollars in exchange for your copy of Feinberg's Doing and Deserving, and you send me the book, then I have a duty to give you ten dollars. This is an example of a relational duty, a duty to someone. Relational duties are duties whose performance is owed to the correlative right-holder. Feinberg's example of a nonrelational duty is the duty to stop a red light.

Another standard example of a nonrelational duty is the duty to give to charity. I think that because I am financially well-off I have a duty to give to charity. But this duty is not to anyone.

Feinberg's discussion of making claim to, claiming that, and having a claim does not tell us why the duties correlative to rights are relational. On the second interpretation, Feinberg does not answer the question: Why is the duty correlative to a right a duty to the right-holder? In this respect and on the second interpretation, Feinberg's theory is inferior to the interest and will theories of rights.

The final problem with Feinberg's view is that on either interpretation, it provides an insufficiently detailed analysis of the link between rights and other moral concepts. There is clearly some link between one's rights and what others are forbidden to do. Rights clearly have some link to obligations. Feinberg analyzes rights in terms of claims and then distinguishes three kinds of claims and claiming. This provides a link between rights and duties because duties are the correlatives of claims. But Feinberg does not attempt to go further and explicate the link between rights and such concepts as what is forbidden to do, what one has an obligation to do, and what one can do. He sets such questions aside with some suggestive but incomplete remarks. Feinberg's theory is incomplete. This is a criticism in only some minimal sense. No one has a right that Feinberg provide a complete theory of rights.

Feinberg was justly famous for noting important distinctions and for writing with great clarity. Wellman's shocking headline is “Feinberg Conflates Fundamental Distinction!” Wellman thinks that Feinberg's theory of rights has the fatal theoretical flaw of inconsistency. On the other hand, Wellman correctly notes that on his interpretation, Feinberg's view has the merit of providing insights to both will theorists and interest theorists. An inconsistent view can have the merit of providing resources to both sides in a debate, and in Feinberg's case the resources provided were considerable.

My shocking headline is “Feinberg Fails to Write Clearly!” I do not believe that there is enough evidence in the texts to determine conclusively what Feinberg's view was. What insights can be gained from the second interpretation of Feinberg's view? For at least the last thirty years, the debate over rights has been dominated by a debate between interest theories and will theories. Elsewhere I have argued that defenders of the interest theory have made very powerful points against the will theory and that defenders of the will theory have made very powerful points against the interest theory.¹⁶

Wellman sees this argumentative standoff; he concludes that it cannot be resolved and that both views are reasonable. I see the standoff and conclude that neither view is reasonable. We should consider the possibility that both the interest and the will theories are false and so we should seek a theory of rights that is neither an interest theory nor a will theory. The insight I take from the second interpretation of Feinberg's view is that it is possible to develop a plausible “third” theory of rights, a theory of rights that is neither an interest theory nor a will theory. I think this insight is of considerable importance.