Extreme Cases and the Moral Limitations of 'Capacity-Criteria'
Allow me to begin with a rather uncontroversial fact: However little we may know about the neurobiological basis of such capacities, certain non-human animals at the very least seem to have some psychological capacities (cognitive, emotional, etc.) which certain humans seem to lack. Compare, for instance, the cognitive capacities of a highly intelligent chimpanzee and those of a human child with a severe intellectual disability. In certain cognitive tasks (puzzles, games, etc.) a psychologist might ask subjects to perform, the former might markedly outperform the latter.
Many appeal to this fact about capacities in condemning the countless inconsistencies between, or biases in, our treatment of animals and humans. Some even go so far as to contend that particular capacities are the solely relevant or, at least, main criterion for determining how we ought to treat living creatures. Perhaps the most vocal apostle of this view is the utilitarian philosopher Peter Singer, who condemns the human exploitation of other animals (as food, in experimentation, in industrial uses, and so on) by appealing to these animals’ capacities for pain, or sentience. For various domains of human activity, this moral appeal may, indeed, be sufficient. But the problem of animal experimentation reveals a chink in the armor of these capacity-centered criteria.
Such facts about capacities surely compel us to consider seriously the interests of non-human animals. However, in our general treatment of both humans and non-humans, capacities (whether for pain, explicit consent, etc.) are morally relevant only to a rather limited extent. That is, they are useful only in establishing a certain base criterion for relevantly differentiating the recipients of our treatment—beyond which they remain, by themselves, inconclusive. These limitations can be demonstrated most vividly by examining extreme cases of biomedical experimentation in which various capacity-criteria that have otherwise proven dependable guides suddenly lose their moral relevance because they fail to be relevant distinguishers. In addressing such extreme cases, we must, instead, use an alternative approach, one that draws on capacity-criteria when relevant but does not depend on them as the sole moral appeal. And, in such cases, not only can a form of ‘weak speciesism’ be justified, but also this alleged bias would in fact be more morally permissible than a kind of species-blind treatment.
Consider the following thought experiment. A group of medical researchers have discovered a drug that they believe has a very good chance of treating some pandemic neurological disease. In order to test the drug’s efficacy and potentially debilitating side-effects, which include severe neural degeneration, they must administer the drug to a living subject. Furthermore, due to various neurobiological factors, suppose that the drug is only effective on humans and a certain species of highly intelligent macaques. The researchers now face the dilemma of who to use as the subject of their experiment. Suppose they have four subjects to choose from: (1) a macaque that was bred for experimentation; (2) a medically braindead elderly human with no prospect of awakening nor any loved ones attending his bedside; (3) a severely developmentally-challenged orphan whom the macaque markedly outperforms in virtually all cognitive tests; and (4) a self-sacrificing, consenting adult.
Let’s construct three subcases out of these four subjects and examine how the subjects’ possessing various capacities might bear on the morality of testing on them. In Case 1, the experimenters must choose between the macaque and braindead human. Here, clearly, the fact that the macaque demonstrates certain psychological capacities which the human patient does not is morally relevant. The macaque meets a certain base standard for sentience, pain, and various rudimentary cognitive functions, whereas the braindead patient does not, lacking even the capacity for pain. So, given this fact, the choice here is unambiguous: the braindead human. To choose otherwise would be to preferentially treat the braindead patient on the sole basis of his membership to the species group Homo sapiens—an act of pernicious, or ‘strong’ speciesism.
In Case 2, the experimenters must choose between using the macaque and an unusually altruistic human who, fully aware of the risks involved, has provided explicit consent to being administered the drug. In this case, both potential subjects meet the base standard for sentience, so that particular capacity-criterion can no longer serve as a relevant deciding factor. Both subjects will experience pain and the degeneration of their mental faculties; ranking their relative suffering would be futile and scientifically unfounded. On account of this capacity-criterion’s indeterminacy, we must, instead, appeal to another one—namely, the capacity for consent. This criterion, unlike the criterion in Case 1, serves here to qualify subjects for participation, not exempt them from it; additionally, this criterion was not morally relevant in Case 1 because neither subject could give consent. So the experimenters must choose the consenting adult.
And the indeterminacy of the criterion here is not simply a function of our epistemic uncertainty (namely, about the physiology of pain). As I hinted at above, it is tempting to compare the respective pain experienced by the macaque and the human: If there were a well-justified ordering of pain, then we could, it seems, resolve this capacity-criterion’s indeterminacy and preserve the criterion’s status as a ‘trump-card’ appeal in the moral discussion. However, this is actually not the case. Suppose we came to some kind of scientifically well-founded conclusion that, beyond any reasonable doubt, the pain of the consenting adult was quantitatively greater and, so, more ‘profound’ (in terms of its emotional, intellectual, or even somatic content) than the macaque’s, thus compelling us to choose the macaque for the experiment. This knowledge would result in a devastating clash between the capacity-criteria for consent and for pain—a problem which might, perhaps, serve as the coup de grâce for this very notion of a capacity-criterion. For the following question is raised: Is it justified to ignore the consenting human’s unbending commitment to self-sacrifice here in the name of some pain-centered, utilitarian calculus? In practice, it would seem rather draconian for an institution to so callously ignore the interests and deeply-held desires of this consenting human in the enforcement of such a principle. Who are we to say that her pain is more morally relevant to the situation than and, thus, trumps her own moral convictions? This would be a clear violation of her individual liberty, limiting her pursuit of the ‘good life’ as she conceives it. So, even if we were to glean reliable information about the relative magnitudes of these pain-experiences, we would still likely have to choose the consenting adult. Nevertheless, it is probably best to avoid this rather spurious line of reasoning altogether.
Lastly, in Case 3, the experimenters choose between the macaque and a developmentally-challenged orphan who has the capacity for pain but not consent. Here, clearly, the two previous criteria cannot be morally relevant, since both subjects meet the base standard for sentience and yet neither can give consent. So, as before, we might proceed in looking for some other relevantly differentiating capacity-criterion to solve this quandary. But this is surely a fool’s errand. The macaque, by any reasonable measure, possesses psychological capacities much greater than that of the orphan. The only conceivable capacity-criterion here would be a cognition-based one. Yet this would lead us to either an inconsistency with our stance in Case 2 (the consenting adult clearly has greater cognitive capacities than the macaque), or a slippery slope (cognitive capacity, it follows, would be an equally valid tiebreaker in choosing between an infant and a non-consenting adult). Capacity-criteria, then, do not suffice here, and in their place, I propose an alternative approach.
Part of the problem with capacity-criteria is that they are awfully simplistic; no experiment exists in a vacuum, but, rather, all take place in a particular social context. So, in addition to the consequences for the subjects, there are other morally relevant factors to consider in making good distinctions between the two choices in each case. In the previous cases, these other factors seemed, at least, to be overshadowed by the salient differences in consent and sentience. But, in Case 3, to fill in this gap of indeterminacy left by the capacity-criteria, we must look elsewhere. One such factor to consider is our obligation in a society to caring for children and the disabled—a commitment which is undeniably stronger than the one we have to nurturing non-human animals. Additionally, experimenting on certain disabled humans might have the consequence of threatening other members of the society —perhaps ones who are developmentally challenged but not as severely as the orphan in Case 3, or, say, relatives of the orphan—in a way that would not be a concern if the macaque were selected. In a more circumspect approach, one which seeks to examine all relevant criteria for differentiating, not just the ones related to capacities, the macaque would have to be chosen over the orphan.*
So, here, where capacity-criteria are no longer relevant, we are obligated to act on the grounds of ‘weak’ speciesism—in which the fact that the subject is human becomes relevant but no longer can be if there are more salient discrepancies in terms of capacities, such as consent or sentience—because to ignore these other morally relevant facts would be to commit a serious blunder in our deliberation. This is not to deny, however, the power of capacity-criteria in denouncing stronger manifestations of speciesism. In this regard, capacity criteria are useful as the sufficient conditions for an animal’s inclusion in our moral community: for determining, in our moral discussions, whose interests we ought to consider in our decisions. But these criteria compose merely one star in the constellation we must use to navigate these rather difficult problems.
* Admittedly, this judgment is grounded in an assumption about the static attitudes of our current society. If in some ideal society, certain members succeeded at mitigating these threats perceived by the disabled members (or their family), then perhaps it would be permissible for the orphan to be chosen as a subject. But, regardless, this misalignment of the actual society and a more ideal one is precisely such a nuance that the capacity-criteria approach neglects.