“Why Clement’s Argument Is, In Fact, Individualistic Rather Than Relational, And Why It’s a Good Thing That It Is, Too”
 
         This comment is supposed to be brief and brevity does not come naturally to me.  So I decided that if I couldn’t give a long paper, I would at least give a short paper with a long title.  The title is: “Why Clement’s Argument Is, In Fact, Individualistic Rather Than Relational, And Why It’s a Good Thing That It Is, Too”.  In the interest of brevity, my reading the title just now will also serve as the introduction.
        So on to the body of the paper: I want to begin by making sure that we are clear about what, on Clement’s account, makes a moral theory individualistic or relational.  An individualistic approach, according to Clement, is one that focuses on the “characteristics of individual beings and their moral implications,” while a relational approach is one that instead focuses on “relationships between beings and their moral implications” (1-2).  In considering what duties we have toward so-called marginal humans, for example, the individualist asks “What characteristic do even the most impaired humans have that makes them morally considerable?,” while the relationalist insists that our responsibilities toward them “depend not on their individual capacities but on our relationships to them” (5).  The question to which the individualist gives one answer and the relationalist gives another, then, is not the surface-level question, “which properties are morally relevant?” (since different individualists may answer this question differently as may different relationalists), but rather the more theoretically fundamental question: “which subjects are the bearers of morally relevant properties?”  To this question, the individualist answers that it is individual beings who are the bearers of morally relevant properties, and the relationalist answers that it is instead the relationships between individual beings and the agents who have moral obligations toward them.
        Now it is clear enough, on this account, why the argument from marginal cases counts as an individualistic argument: it grounds its normative claim that (at least some) nonhuman animals and marginal humans must be treated comparably on the descriptive claim that (at least some) nonhuman animals have the same characteristics as do marginal humans.  But it is far less clear why the argument that Clement proposes in its place should be viewed as any less individualistic in this sense.  The argument begins as follows: “We treat dogs as valued companions, while we treat pigs as mere objects which are useful to us.  But dogs and pigs have similar capacities for intelligence, sociability, etc. . . .  Consistency requires that we treat dogs and pigs with similar levels of moral respect” (pp. 7-8).  It then continues by attempting to overcome the objection that Clement believes limits the force of the argument from marginal cases: that it leaves open the question of whether we should therefore treat pigs like more like dogs or treat dogs more like pigs.  Clement argues that relational considerations yield an answer to this question, and one favorable to the animal welfare cause: “our relationships with companion animals allow us to know that dogs are deserving of moral respect.  On the other hand, our lack of relationships with other animals of similar capacities, like pigs, allow us to rationalize their use and abuse.  So pigs should be treated more like dogs, rather than vice versa” (p. 8).
        It is at this point that Clement considers the response I wish to pursue: that the appeal to this consideration still leaves the argument as a whole an individualistic one, since it still focuses on the claim that individual dogs and individual pigs have comparable individual characteristics, and still urges that it is in virtue of this fact about them as individuals that morality demands that they be accorded comparable moral status.  Clement’s reply to this concern runs as follows: “I believe that the relational aspect of this argument is significant.  For this approach calls attention to the fact that it is our relationships with animals that allow us to discover what those animals are like, and thus what our moral responsibilities toward them are” (p. 8).  Or, as she puts the point a bit later, “Since we learn about animals in themselves through our relationships with them, we need a moral approach which focuses on such relations” (9).  Clement’s response to the claim that her argument is individualistic rather than relational, then, seems to turn on two claims: (1) that we learn about what animals are like through our relationships with them, and (2) that a position that appeals to this claim is relational rather than individualistic.  I want now to raise some doubts about both claims.
        Let me begin with the first claim, the claim that we learn about what animals are like through our relationships with them.  One reason for denying this claim is that there are ways of learning about animals other than by forming relationships with them of the sort that Clement describes.  Indeed, Clement’s own argument presupposes that this is the case: her argument assumes that (a) we know that pigs have dog-like capacities for intelligence and sociability and that (b) we do not have relationships with pigs.  But if both of these claims are true, then it must be the case that we have ways of knowing what pigs are like that do not involve our having relationships with them.  If this were not the case, after all, then we couldn’t know what we do know about pigs.
        A second reason for denying the first claim needed to sustain Clement’s response is that having a relationship with an animal can distort our understanding of its nature as easily as it can reveal it.  Because we care very much about dogs and not very much about pigs, for example, we may be inclined, out of sentiment, wishful thinking, and so forth, to think more highly of a dog’s abilities than the facts truly merit.  In the same way that we would not trust a child psychologist’s estimate of her own child’s mental abilities as much as we would trust her estimate of the abilities of a stranger’s child, so we should not trust our own estimates of the abilities of the animals we love as much as we should trust our estimates of the abilities of those animals about whom we are more indifferent (and we should trust these, in turn, more than our estimates of the abilities of those animals we find positively repellent).  Forming special relationships with nonhuman animals, then, seems neither necessary, nor particularly desirable, for helping us discover what they are really like.
        But let us now suppose that I am mistaken about this and assume that the best or only way for us to discover what an animal is like requires us to have a special relationship with it.  Does this suffice to render Clement’s position relational rather than individualistic?  I believe that the answer is no.  The claim would suffice if the following argument were sound: (a) the kind of relationship a person has with an animal is relevant to how accurately a person can determine the individual properties of an animal (Clement’s claim); (b) accurately determining an animal’s individual properties is relevant to accurately determining what a person’s moral obligations to the animal are (the individualist’s claim); therefore (c) the kind of relationship a person has with an animal is relevant to what a person’s moral obligations to the animal are (the relationalist’s claim).  But this argument is unsound: it does not follow from (a) and (b) that the existence of the relation between the person and the animal is relevant to what the person’s moral obligations toward the animal are; it follows only that it plays a causal role in his uncovering the existence and content of those obligations.  After all, a person’s studying an animal with an electroencephelograph might also be instrumentally valuable in determining the animal’s individual properties, which might in turn be relevant to determining the person’s moral obligations to the animal, but this would not show that the fact that the person was studying the animal with an electroencephelograph was itself relevant to what his moral obligations were.  What is paradigmatic of the individualistic approach, it is important to remember, is that it asks of possible subjects of moral concern: “What characteristic do they have that makes them morally considerable?”  This is still the question that Clement is asking about animals, even when she appeals to facts about them as individuals that we may uncover through our relationships with them when she attempts to answer it.  And so on Clement’s account, just as surely as on the account presented by the argument from marginal cases, it is individual beings, rather than the relations between them, that are the bearers of morally relevant properties.
        I have argued thus far that Clement’s argument is, at bottom, individualistic rather than relational.  It might be thought that this is merely a terminological quibble.  What difference does it make?  But I think this fact about the argument has at least two important substantive implications.  The first is that it shows Clement’s argument to be neither stronger nor weaker than the argument from marginal cases.  Clement argues that “the individualism of the argument from marginal cases is in fact its main shortcoming” (2).  But if I am correct that her argument is just as individualistic as the argument from marginal cases, then it follows that either (a) her argument, too, has a serious shortcoming, or that (b) the argument from marginal cases does not.  Clement, of course, attempts to show that her argument does not have such a shortcoming by maintaining that our relationships with dogs enable us to know that we should treat pigs like dogs rather than the other way around.  I have argued that her so arguing is consistent with her argument’s being an individualistic one.  And if I am correct about this, then there is also nothing inconsistent about a defender of the argument from marginal cases offering a parallel response to the parallel challenge: yes, it is true, she can say, the argument from marginal cases is equally consistent with treating marginal humans like animals and treating animals like marginal humans, but our relationships with the senile, the retarded, the severely deformed, enable us to see that they are worthy of our respect.  It is due to our lack of comparable interactions with comparably endowed nonhuman animals that we fail to perceive this in their case.  Therefore, we conclude that we should treat such animals as we now treat marginal humans, rather than treat marginal humans as we currently treat such animals.  My point, I should emphasize, is not that this is a decisive response to the objection to the argument from marginal cases, but rather that it is just as effective or as ineffective here as is the parallel response to the parallel objection to Clement’s argument, and that nothing about the individualism of the argument from marginal cases prevents its advocate from appealing to it.
        This first implication of my recharacterization of Clement’s argument is therefore partly favorable and partly unfavorable to her position: it suggests that her position is at least as strong as the argument from marginal cases, but it also suggests that it is no stronger than it.  The second implication is more straightforwardly favorable.  For there is, I believe, a powerful objection to any attempt to offer a relational argument in defense of the moral status of animals.  It thus turns out to be a good thing that, on my account (although not on hers), Clement has not offered such an argument in the first place.
        The objection arises from the distinction moral philosophers standardly draw between negative duties, or duties of noninterference, and positive duties, or duties of assistance.  It seems quite plausible to suppose that the nature of the relationship between A and B is relevant to the content of the positive duties that A has toward B.  But it seems far less plausible to claim that this is so of negative duties.  It is easy to imagine circumstances in which, for example, whether or not it would be permissible for you to decline to give food, or medicine, or clothing to a needy person depends, at least in part, on whether that person is a stranger to you, a rival, a fellow countryman, a friend, an enemy, your parent, your teacher, your student, your lover, or your child.  But it is extremely difficult to imagine circumstances in which whether or not it would be permissible for you to kill, or rape, or injure, or steal from a person would depend, even in part, on such factors.  We may well think worse of a person who kills her own child than of one who kills a stranger, but surely this is not because we think that you have a stronger right not to be killed by your mother than you have not to be killed by a stranger.  The strength of one’s right not to be harmed does not vary from potential harmer to potential harmer, but the stregth of one’s right to assistance may well vary from potential benefactor to potential benefactor.
        Moreover, this intuition about the difference between what is relevant to positive and negative duties makes good theoretical sense.  A’s positive duties to B concern the ways in which A is obligated to postively interact with B, so it makes sense that facts both about B and about A’s relation to B would be relavent to this.  But A’s negative duties to B concern only the respects in which B has the right to be left alone, and here it makes sense that facts about B alone can be relevant to establishing them.  In addition, not only does this distinction seem intuitively plausible and theoretically sound, but it also seems to be implicit in Clement’s own account as well.  When she speaks of the sorts of duties that relational considerations can ground, she speaks only of “special responsibilities” that we can have to others in virtue of the relationships in which we stand to them.  It seems natural to suppose that such special responsibilities are meant to include the duties we have to care for some individuals and not for others, but it seems equally unlikely that they are intended to include, implausibly, special duties we have to refrain from harming some and not others.  And in the book in which she develops her account of vulnerability in more detail, Clement is explicit about this, referring to her position as the “vulnerability model of our care obligations,” rather than as the vulnerability model of our obligations not to harm, or of our obligations generally.
        Now I do not want to suggest that the question of what positive duties we might have to care for animals is a trivial one.  But I do want to suggest that the primary concern of those concerned to defend animal welfare has been to establish the negative duties that we have not to harm them.  And this is as it should be.  Such questions as whether it is morally permissible for us to kill animals for food and clothing, to hunt them for sport, to confine them for entertainment, or to perform experiments on them for our benefit, are both more theoretically fundamental and more practically urgent than are such questions as whether we are obligated to care for stray animals, to prevent deer populations from starving, to rescue animals that accidentally become trapped, and so on.  An approach to determining what our moral responsibilities toward animals are that can answer the second kind of question but not the first cannot be acceptable as a basis for grounding a full-blown defense of the moral status of animals.  And since, as I have suggested, the relational approach to our moral responsibilities is just such an approach, it proves to be a good thing that Clement’s argument proves to be an individualistic one after all.
        Let me conclude by considering one response that Clement might offer to this last claim.  Yes, she might agree, on some relational accounts, it turns out that we can only justify the special duties we have to some, not the basic duties we have to all.  But on the particular version of relationalism that appeals to the notion of vulnerability, she might argue, this is not so.  Since, as Clement suggests, all animals are such that we can harm them, it follows that all animals are vulnerable to us and are, therefore, subjects of a basic duty on our part not to exploit that vulnerability (p. 7).  It would therefore follow that this sort of relational account can provide what the defender of animal welfare is most centrally looking for, a reason to believe that all animals have a basic right not to be harmed.
        I am willing to agree that this sort of account may, in fact, produce what the defender of animal welfare is looking for.  But it seems to me that it can do so only because, once again, it is at bottom, an individualistic argument.  Clement maintains that the notion of vulnerability is relational rather than individualistic, since one is not simply vulnerable simpliciter, but rather is vulnerable to some particular individual (p. 7).  But if the concept of vulnerability is truly to be stretched to apply to all animals, and to the acts of all people who could conceivably harm them, then the superficially relational claim that an animal is vulnerable to some particular person ultimately reduces to the more fundamentally individualistic claim that the animal is an individual that is capable of being harmed, that the animal, that is, has interests.  Now I have no substantive objection to a position that attempts to ground our duties to animals in the claim that animals have interests.  But the fact that Clement’s position seems to come to this does seem, again, to suggest that her position is a powerful one just to the extent that it is an individualistic one and that to the extent that it is a relational one, it is also a problematic one.

David Boonin
Dept. of Philosophy
University of Colorado