The Challenge of Marginal Cases

This is the opening post in a series on the challenge of marginal cases and several prominent responses to that challenge.

Most people believe that human beings have full moral significance, whereas nonhuman animals have only partial moral significance.  For example, most people believe that the use of nonhuman animals in vivisection may be morally justified, provided the practice yields clear and significant benefits and provided the animal subjects do not suffer ‘gratuitously.’  Most people would also claim that morality categorically prohibits the use of human beings in vivisection, no matter how clear and significant the benefits or how ‘humanely’ the human subjects were treated.  Let’s call this view about the unequal moral significance of human beings and nonhuman animals the conventional view.

Accompanying the conventional view is a conventional justification–viz., human beings and nonhuman animals have unequal moral significance because human beings have capacities that nonhuman animals lack.  For example, human beings, unlike nonhuman animals, are capable of understanding morality and governing their conduct in accordance with its norms.  The form of justification here pays tribute to an intuitively compelling higher-order moral principle, which we may call the relevant difference principle, and state as follows: If two individuals, A and B, have unequal moral significance, there must be a morally relevant difference between A and B.

Suppose we assume that nonhuman animals lack the capacity for moral agency, and that lacking that capacity is what justifies our treatment of them, or what explains why they have less than full moral significance. There are human beings who lack this capacity as well. Indeed, there are human beings whose cognitive and emotional endowment is roughly equivalent, and in many cases inferior, to that of certain nonhuman animals.  Just to fix ideas, picture a healthy adult chimpanzee alongside a profoundly mentally retarded human being.  What should we say about the moral significance of the severely disabled human being?

Perhaps one is prepared to say:  ‘Ah, yes–there are human beings, too, who have only partial moral significance.  Other things equal, we are as morally justified in using them in vivisection as we would be their nonhuman animal counterparts. That we don’t in fact use them in this way has more to do with extra- or non-moral considerations than moral ones.’

Most people, I suspect, are not prepared to say this: the profoundly mentally retarded human being has full moral significance.  Enter the challenge of marginal cases, which confronts us with what seems like impeccable reasoning:

1. Relevant difference principle: If two individuals, A and B, have unequal moral significance, there must be a morally relevant difference between A and B.

2. There is no morally relevant difference between (A) a profoundly mentally retarded human being and (B) a nonhuman animal whose cognitive and emotional endowment is roughly equivalent to (A)’s.

3. (A) has full moral significance.

4.  Therefore, (B) has full moral significance.

The relevant differences principle seems to capture something essential to moral reasoning; it appears pretty invulnerable.  Since we’re assuming the truth of (3),  that leaves (2) as the only viable target.  We either have to explain why (2) is false, or, assuming the argument’s validity, give up the conventional view.

Not surprisingly, what we find in the literature on the challenge of marginal cases are attempts to refute (2)–that is, attempts to identify a morally relevant difference between (A) and (B).  In the next post I’ll review one such attempt and explain why I think it fails.




This entry was posted in Uncategorized. Bookmark the permalink.

7 Responses to The Challenge of Marginal Cases

  1. Nates says:

    Of course, what we also find in the literature are attempts to embrace (2) in the name of animal rights. This strikes me as being a disastrous approach–even if you are all for animal rights–but there you go.

    It would be great if someone would come up with a novel way of refuting (2). Maybe even two people, co-writing a paper.

    • David says:

      –“It would be great if someone would come up with a novel way of refuting (2). Maybe even two people, co-writing a paper.”

      This series of posts will wrap up with our response!

  2. juan says:

    David, I don’t know if this is the theory you are going to talk about, but one could reject (2) because type (A) individuals have relations to us which animals don’t. We care for them, so their moral worth is based on our caring for them, maybe just like, for a parent, her child is more important than a stranger’s.
    Of course, you could say that the parent’s caring for her child is based on the further relations of having given birth to it, nurturing it and so on. And there might be type (A) creatures so wretched that nobody ever cared for them, so the response would be that this relation of caring and all that it comes with doesn’t apply to them.

    But what if we say that type (A) individuals are more morally significant for us than type (B) because they have the property ‘being somebody’s child’ for example? We can surely relate to that property as parents, or as people who know what a child is and how important children are.

    I’m not denying there may be a response to this. If you push the ‘why is that property morally important?’ line far enough, and for every property I throw at you, then you will probably conclude that (2) is indeed true. You can apply the same sort of reasoning to the parent case: from an objective point of view, why is it morally significant for the parent that she gave birth to the child, or that it is her child? You can point to the exact sameness of intrinsic properties of this child to another random child. And that may be true. Still, you will not be able to convince the parent that her child is not worth caring for more than another child. It’s also possible even that the parent acknowledges everything you say and still go on caring, just as a brute attitude. The same may be true of (A) type beings. We can end up with a brute attitude of caring for these individuals in a way we don’t for animals.

    • David says:

      Hi Juan,

      Absolutely! Several philosophers have offered ‘relational’ accounts of why the challenge of marginal cases fails. But others are not so sure–Norcross, I think, says that this would make the moral status of the marginal cases unacceptably contingent on our continuing to value or care about the human relationships into which marginal cases figure. Of course Norcross’s worry here may be lame–it seems unlikely given everything we know about human beings (much less other mammals) that the parent-child relationship will cease to loom large. And as for the obvious rejoinder, ‘what about the hypothetical orphan who has no relations,’ we can still have thoughts like, ‘That’s somebody’s child,” and thoughts like that affect us profoundly in that ‘brute’ way you mentioned.

  3. juan says:

    By the way, I’m teaching intro to ethics this fall, and Norcross’ paper is the first thing I’m doing. I had never even heard of the guy before you taught it in intro. But I liked it a lot, so I’m teaching it too.

  4. Josh says:

    Okay – without having done any of the reading, I’ve got another proposal (which may or may not have an analogue in the actual literature about this question):

    I’m thinking about some sort of class-based account. Something like: a deeply mentally retarded individual must be accorded the same rights as a “normally” functioning human being, because even though the former lacks many capacities that the latter has, the former belongs to the same class – i.e., “humankind” – that another animal does not.

    I know that sounds question-begging. But – we regard humans as all having something in common – just the fact that they *are* humans, and we accord all members of that class with the same respect, regardless of whether they are an “ideal type” of that class or not. But that ideal type is still important – since the ideal (or fully functional human being) has a set of traits that command it respect – abilities to feel pleasure and pain, ability to reason, etc., then even a human being without all those qualities ought to be treated as though it has them, because we can recognize such a human being as a not-fully-functional member of the class.

    This is different from the relational account- it’s not that those one of those beings gave birth to another, it’s that they share in some sort of species-being, even if they are both not identical realizations of that species.

    Again I know this is a naive take but I thought I’d throw it in.

  5. Nates says:

    Yes, this is called the argument from kinds. It’s widely held (including by Rawls, so you’re in good company!), but there are worries. For one thing, it looks disturbingly analogous to arguments designed to defend the special rights of certain races or genders over others, and so it often gets tagged as a form of speciesism. Even many of the people who ultimately endorse it feel uneasy about the argument. (Again, this includes Rawls, who hems and haws about the limits of his theory at this point.)

    Still, the argument has prospects, if only one could find a way to properly ground the appeal to fellow membership in the human species. Yes, that would be just the ticket…

Leave a Reply to Josh Cancel reply

Your email address will not be published. Required fields are marked *