In his essay "Do Animals Have a Right to Liberty?" James Rachels seeks to show that there are certain rights possessed by animals and that certain arguments purporting to show that it is not logically possible for animals to have rights are ineffective.[1] I shall primarily examine his constructive arguments designed to show that at least some animals have a right not to be tortured, a right to property, and a right to liberty. I do so in part I; in part II, I briefly consider his negative arguments concerning the possibility of animals having rights. It may well be the case that our common treatment of animals violates some important moral constraint; on this general point I am in agreement with Rachels. Still, it is philosophically desirable to correctly identify that constraint and provide something more than intuitive support of it. While Rachels's sentiments and general position are, I think, on the side of the angels, his arguments fail to show that animals have rights and that, for that reason, certain instances of imposing suffering or deprivation on them count as a violation of rights. Part of the difficulty he encounters, but insufficiently attends to, derives from the fact that we lack a clear grasp of basic notions such as interests and, more importantly, rights. Perhaps we can formulate a reasonable policy concerning the treatment of animals without possessing such perspicuity, but a certain caution seems to be in order.

I

Rachels's first application of his proposed method concerns the right not to be tortured. His reason for thinking that humans have such a right, quite aside from its postulation by the United Nations Declaration, seems to be that a "man has an interest in not being tortured because he has the capacity to suffer pain." More explicitly, then, we have the following argument:

1.  Human beings can suffer pain.

2.  Hence, human beings have an interest in not being tortured, (from  1)

3.  Hence, human beings have a right not to be tortured, (from 2)

His reconstructed parallel argument is, then:

a.  Animals (some) can suffer pain.

b.  Hence, animals have an interest in not being tortured, (from a)

c.  Hence, animals (some) have a right not to be tortured, (from b)

But consider a third parallel argument.

A.  Humans can suffer from sexual deprivation.

B.  Hence, humans have an interest in not being sexually deprived, (from A)

C.  Hence, humans have a right not to be sexually deprived, (from C)

It is not at all plausible to accept C, I submit, even though A and B are each plausible. For an equally dubious argument, substitute "be sick" (and cognates) in place of "sexual deprivation." For the sake of discussion let us treat the first inference in each of the arguments as unproblematic. There is undoubtedly a difficulty, however, in going from the second claim to the third. That is, it is not obviously correct to think a being's having an interest in something is a sufficient condition for concluding that it has a right to what is in its interest. So, I conclude that, in the absence of further argument, Rachels cannot justifiably infer that animals have a right not to be tortured (c) from his prior assumption (b). If there is any connection between a being's having interests (not, of course, the same as "being interested in")—and the concept of interest is one that needs fuller explication—and having rights, it may turn out that the former is only a necessary and not sufficient condition of the latter.

In Rachels's second attempt to show that animals have rights, he claims that if, as Locke urged, a man may acquire property by mixing his labor with something, so may the squirrel who gathers nuts to eat. Since, he alleges, "there is no relevant difference between the man and the squirrel," either both have a right or neither does. Rachels's argument may not be invalid; it is nonetheless inconclusive for the following reason. Locke's specification of the conditions in virtue of which a being (to be neutral) acquires a property right in something in the Second Treatise (see Sections 27 to 35) is more complicated and would read, more explicitly than Locke himself put it, something like this (where A is some agent and O is an object):

A owns (or comes to own) O if and only if

1. O was unowned (or, to avoid circularity, no one else has mixed his labor with O)

and 2. A mixed his labor with O

and 3. There is enough of O (or O-type stuff) left for others

and 4. What is left is as good as what A is taking.

Without broaching the difficulties with Locke's claims, it seems clear that "if Locke is right" quite a few conditions must be satisfied to acquire property, and it is not sure that squirrels could intentionally or wittingly satisfy these conditions. It is not clear, on the other hand, that one can unintentionally or unwittingly acquire property. Questions fly at us and our intuitions are diverse and shaky here. Does a squirrel's or a spider's energy expenditure count as Lockean labor? What of the energies of a Venus Fly Trap? Assuming that he (or she) mixed in the requisite labor, did the whale own Jonah? Perhaps Rachels is not wrong on this matter, but the argument is too sketchy to bear the weight he hopes to rest on it.

Rachels's effort to support the claim that animals have a third right concerns liberty. After considering the possibility that beings have a right to any intrinsic good they are capable of enjoying and liberty is an intrinsic good (to later conclude that animals have a right to liberty) Rachels, thinking few philosophers would agree that liberty is an intrinsic good, adopts another strategy and argues that limitations on their liberty, nonetheless, harm the interests of animals. The claim that animals are frequently and significantly harmed in serious ways by captivity, as in his zoo examples and in the case of factory-farming, is, I think, incontrovertible. Similarly, there is little doubt that many experiments on animals inflict extraordinary suffering, and are often done with little attempt to distinguish between cases where the research promises trivial or, alternatively, quite beneficial results. In the midst of documenting these points he proposes a principle similar to one previously considered, "Any creature that has interests has at least a prima facie right not to have those interests needlessly harmed." Let it be conceded that many animals have interests, that many are harmed and, in that sense, their "interests are harmed." Let it also be conceded that this harm often occurs needlessly. We still need an argument to show what Rachels asserts: that creatures with interests thereby have correlative prima facie rights of a sort.[2]

There is an evident alternative view to take here, and it is not merely a callous skepticism. It is simply that we have a duty not to impose suffering or serious deprivation on other sentient creatures and, hence, not on many animals, and we may have a strict obligation to so refrain in the absence of any countervailing moral considerations. Now it may be that the said claim of duty would be entailed by the claim that animals have rights or a prima facie right, say, not to be caused to suffer by rational beings, but the duty may exist without a correlative right. Rachels's arguments do support the claim that rational beings have certain duties toward animals of a certain sort, but he fails to argue the further assumption (if he indeed is implicitly making it) that such duties entail the mentioned rights. Alternatively, if he believes that the fact that a creature has interests entails that that being has rights of a sort, this needs to be argued, I think, and the type of objection to such a claim which I posed early on needs to be subverted. Perhaps it can be; my only point is that the task remains.

II

Partly for the reason of the difficulty just mentioned, I shall not remark at length on Rachels's substantial remarks on "moral freedom." His argument concerning such freedom and the related claim that some animals show compassion is, I submit, inessential to his basic contention about rights. He wishes to maintain that animals have interests and, therefore, prima facie rights. It is reasonable to assume that the set of animals exhibiting compassion constitutes a subset of the animals possessing interests; hence, if his prior argument based on interests were successful there would be no need to appeal to the particular further trait of exhibition of compassion as a basis for rights. Any animal showing a compassionate disposition which lost just that disposition (due, say, to alteration in glandular balance) would presumably still have interests and, hence, on Rachels's view, rights as well. Further, if he were to suggest that only animals exhibiting compassion have rights, the scope of animals possessing rights would surely be restricted in a way that he would disavow; we may assume, then, that he would reject such a move.

In making final remarks, I shall bypass the opportunity to fully explore Rachels's counterarguments in his Appendix, ones designed to show that it is possible for animals to have rights. However, I shall press the general theme; his objections to McCloskey and Ritchie are reasonable ones, but there are, I think, more tempting ways of arguing that animals are not the sort of entities that can have rights.[3] Rachels focuses on similarities between humans and animals and tends to deemphasize certain differences. This fact may be relevant, for one might argue, roughly along the lines of Michael Tooley or S. I. Benn (in another context), that it is analytic that only persons have rights, that only entities that can function as agents and, perhaps, can conceive of forwarding their projects by an exercising of rights could, in fact, have any rights at all; that animals lack this capacity, and hence, regardless of notable similarities to humans that have rights, animals are not the sorts of entities that can have rights.[4] I am not persuaded that this line of thought is compelling, but it strikes me as far more tempting than the objections which Rachels, in fact, considers. He cannot be faulted, of course, for not having explored all possible objections, and I offer the point only to advance the inquiry.

So far then I have tried to show that each of Rachels's attempts to show that animals have a right not to be tortured, a right to property if certain conditions are met, or a right to liberty is beset with one or more difficulties. More generally the supposition that whatever creatures have interests have correlative rights needs defense, though many of the facts to which he legitimately and usefully calls our attention support a claim that we have certain specific duties to animals. Whether such duties correspond to rights needs investigation. That we have much neglected duties toward animals I have no doubt; the precise nature and extent of those duties is the more burdensome question. I have further suggested that on certain views about the grounds for having rights, even "natural rights," it may be that some humans (e.g., neonatal ones) as well as all animals do not possess rights; such an objection deserves examination.

As a final demurrer I note that at the outset Rachels states that one of his motives is to "cast doubt on the importance of human rights," but the body of his paper does not tend to do this at all. In fact, Rachels seems to seriously weaken the mentioned claim very quickly by defining "human rights" to mean "distinctively human rights" (rights possessed only by human beings if at all). Now even if many rights generally called human rights are not human rights in the sense of distinctively human rights (supposing that some are shared with animals), with the result that the number of distinctively human rights is fewer than previously thought, it is most difficult to see how this goes any distance toward the possible task of casting doubt on the importance of human rights, distinctively human ones or not. Nor is it clear why Rachels maintains "that human rights are not nearly so interesting or important as philosophers and politicians have thought."