My previous guest post on the argument from marginal cases got a number of provocative comments that I’d like to reply to; but I’m having some baffling and frustrating trouble with the comment form. So until I’ve gotten that sorted out, I’ll use this space to expand on one of the points that I wanted to make anyway: the use that my argument makes of Michael Thompson’s notion of “aristotelian categoricals,” such as “The domestic cat has four legs and a soft coat of fur” or “Humans walk on two legs” or (what’s important for the argument) “Humans are rational creatures.”
Here’s an attempted gloss of the argument that may express things more clearly than I did at first. I used Thompson’s notion to argue that appeals to species normality can do more than ethical vegetarians (such as James Rachels) seem to think that they can do, since they seem to take the species normality argument to amount to nothing more than an arbitrary appeal to the pure Cambridge relation of being a member of the same taxon as other members who are rational. Of course that has precious little ethical bearing by itself, but what I suggested is that there’s a much more charitable reading of the species-normality appeal on offer, in light of Thompson’s aristotelian categoricals. It’s not that “Humans are rational creatures” asserts that all humans are rational (as might have been said before “marginal cases” were introduced as a counterexample), or that it asserts “many (most) humans are rational creatures” (as Rachels and others seem to have misinterpreted it). Rather, “humans are rational creatures” asserts a teleological fact, about the form of life that is proper to creatures of the human kind, to the effect that each individual human is supposed to be able to exercise certain forms of rationality just as we are supposed to have two legs and cats are supposed to have four. I suggested that this makes an important difference, and the difference that it makes might help (1) explain the special moral standing that anti-vegetarians attribute to individual “marginal case” humans but not to individual normal animals, and (2) undermine Rachels’ and others’ attempt to offer hypothetical animals with freak intelligence as a counterexample. For the details, see the post.
Anyway, this brings me to a couple of comments by Alex. First:
Firstly, why does the fact that the paradigm cases of an object-type have a certain property commit us to saying that all objects of that type have that property? (e.g because paradigmatic humans are morally relevant, all humans are morally relevant) - won’t this commit you to (for example) the idea that all humans have the property of ‘having two arms’ because paradigmatic humans do?
The best response here is just a clarification of the argument. That you shouldn’t butcher and eat infants, or the severely mentally retarded, or the comatose, is one of the unargued premises of all the arguments, shared between by imaginary interlocutors. I don’t think it needs argument; it’s one of the background conditions of having a reasonable theory of ethics that you believe this, not something that a theory of ethics should have to convince you of.
What does need argument is the claim that there’s a difference between “marginal case” humans and normal animals that would explain extending that moral standing to the humans while denying it to the animals. The argument for that isn’t based on a general principle that “Paradigmatic humans have P” entails “All humans have P.” (I freely concede that for any reasonable candidate for the morally relevant sort of rationality, there will be many humans in abnormal or transitory circumstances who don’t have it.) The inference that I do want to endorse is that if “the adult human has P” is true as an aristotelian categorical, then “all humans have a natural capacity for P” is true as a universal generalization (for some important sense of the phrase “natural capacity”). Not all humans have two arms, but all humans do have the natural capacity (in some sense) for having two arms; that’s why armlessness is a tragedy for you but not for a trout, and that’s why a human with the intelligence of a cow is thought of as having a profound disability, but a cow with the intelligence of a cow is not: because the one is a case of having a rational faculty that’s damaged (perhaps irreparably), while the other is a case of not having any rational faculty at all.
Of course, at best this only explains more fully what the appeal to aristotelian categoricals and teleological talk about the human form of life is supposed to do. Which brings us to Alex’s second remark:
But the notion of “aristotelian categoricals”/”telos” conflicts wildly with a modern scientific worldview, and we can hardly justify redeeming it because it achieves some cuddly ethical conclusions. (Second key question) Don’t you have a give an independent reason to believe in these categoricals first, and then you may be entitled to refer to them to establish conclusions elsewhere.
… to which there are a couple of things to say.
First, I didn’t intend my post to provide compelling reasons to reject the ethical vegetarian position (indeed, I accept a form of it, so a fortiori I don’t think there are compelling reasons to reject all forms of it). But I do think that the arguments in favor of the position aren’t as good as they could be, and that the devotion that nearly all ethical vegetarians show to the argument from marginal cases is misguided. One of my reasons for thinking that is that I think there are alternatives on offer that start from philosophical premises that most ethical vegetarians simply haven’t (yet) demonstrated an understanding of. Maybe those premises are wrong (I don’t think they are, but I don’t think they entail the anti-vegetarian conclusion, either), but ethical vegetarians will have to recognize the premises and give reasons to doubt them, or to doubt the inference drawn from them, before any progress will have been made. A full explanation and defense of the independent reasons for Thompson’s claims is better found in Thompson’s essay (there’s also a good discussion in Philippa Foot’s book Natural Goodness), and was in any case beyond the scope of the post.
Second, though, I do hope to use the rest of this post to take up the subject anew, and offer at least a sketch for the defense of Thompson’s claims about “aristotelian categoricals” and about the sort of teleological talk that go along with them. Partly because I think that something like Thompson’s picture is necessary for any kind of reasonable account of the nature of goodness (and thus for the foundations of any reasonable ethical theory). But the work that aristotelian categoricals will do for you in ethical theory is just a side benefit, not the primary reason for accepting them into our philosophical picture. The primary reason is that statements such as these:
The domestic cat has four legs and a soft coat of fur.
Coyotes hunt small game.
During the mating season, the male emperor penguin warms the egg while the female returns to the sea to feed.
Humans are rational animals.
Humans are the only known animals that use language, but not the only animals that use tools.
… are all both meaningful and true, and commonplace bits of what has sometimes been called “natural history.” But it’s hard — indeed, I think, impossible — to give a good account of what they truly say by squeezing them into any the familiar set of logical quantifiers, whether existential, universal, or statistical. A good analysis of (1)-(5) needs to do at least two things: (i) it needs to be materially adequate (i.e., it needs to provide a gloss that’s true when they are true and false when they are false); and (ii) it needs to be semantically serious (i.e., it needs to provide a gloss of the statements, not some other statements that are considered easier to deal with).
But if you interpret (1) to mean “all domestic cats have four legs and a soft coat of fur,” then your interpretation is not even materially adequate; after all, that’s not true (just ask poor Tibbles, who has been shaved and maimed in a tragic accident). If you interpret it as “some domestic cats,” “many domestic cats,” “most domestic cats,” or even “the overwhelming majority of domestic cats have four legs and a soft coat of fur,” then that will be something true, but it won’t be semantically serious. It doesn’t capture all of what is meant by the categorical statement, and you can see this by considering some parallel cases: some cats are blind, many cats are tabbies, either most cats are male or most cats are female — I don’t know which — and the overwhelming majority of cats are vaccinated against common diseases. But “The domestic cat is blind,” “the domestic cat is a tabby,” “the domestic cat is male,” “the domestic cat is female,” “the domestic cat is vaccinated against common diseases” are all quite obviously false. Material adequacy breaks down in the parallel cases because semantic seriousness wasn’t maintained in the original case.
So how can we understand categoricals like (1)-(5)? Well, you could decide that we can’t, and toss them out as not precise enough for the uses of a logical or scentific language. That, though, would seem to me to be a gross error about the nature of logic and its relation to everyday language, of the sort exposed by Wittgenstein when he said:
F. P. Ramsey once emphasized in conversation with me that logic was a ‘normative science’. I do not know exactly what he had in mind, but it was doubtless closely related to what only dawned on me later: namely, that in philosophy we often compare the use of words with games and calculi which have fixed rules, but cannot say that someone who is using language must be playing such a game. —But if you say that our languages only approximate to such calculi you are standing on the very brink of a misunderstanding. For then it may look as if what we were talking about were an ideal language. As if our logic were, so to speak, a logic for a vacuum. —Whereas logic does not treat of language—or of thought—in the sense in which a natural science treats of a natural phenomenon, and the most that can be said is that we construct ideal languages. But here the word “ideal” is liable to mislead, for it sounds as if these languages were better, more perfect, than our everyday language; and as if it took the logician to shew people at last what a proper sentence looked like. (Philosophical Investigations, § 81)
If we take seriously our responsibility to get everyday language right, rather than discarding it in favor of the will-o’-the-wisp of an “ideal” language, what sorts of understandings might be on offer? Well, Thompson’s suggestion is that we can understand the sorts of natural-historical statements such as (1)-(5) by reference to the role that those traits play in the natural life cycle of that kind of organism, that is, to facts about the form of life that that sort of creature enjoys and the way that that life is supposed to go. We say that the domestic cat has four legs and a soft coat of fur because that’s part of the way that domestic cats live under normal circumstances; having four legs is how they walk and having a soft coat of fur is how they stay warm. There are domestic cats out there that have two or three legs, or are bald; but that is something abnormal about them, due to abnormal circumstances (whether hereditary, or congenital, or acquired in the course of their life). In this case, it’s something that would generally be considered a defect, something wrong with the poor creature; but you can imagine cases (cats that can talk, or cats that can leap four stories) where the extraordinary trait would be preternatural, or better than cats are expected, in the normal course of things, to have.
What’s important is to see how these terms — “natural,” “form of life,” “supposed to go,” “normal,” “abnormal,” etc. — involve us in teleological talk. The fact that conditions are normal or abnormal, that the natural course of events is disrupted or allowed to proceed, are involved with teleological notions, notions of the natural ends that certain sorts of creatures have and the functions that their various distinctive traits serve in realizing those ends. And it’s not immediately obvious whether there is any way that this kind of teleological talk could be reduced to non-teleological talk. (It’s certainly not merely statistical: if all the cats in the world lost their hair through a mysterious virus, that wouldn’t make “domestic cats have a soft coat of fur,” or make baldness normal for the domestic cat. It would mean only that for the time being all cats are abnormal, due to abnormal conditions. Nor will appeals to evolutionary history do, since making sense of evolutionary history already requires you to talk about terms such as species and life-cycle functions such as eating, keeping warm, reproducing, etc., all of which involve you in teleological talk about the roles that each activity serves in the organism’s form of life. And, to ascend up a level of abstraction, if “fitness” isn’t a teleological notion, then what in the world is?)
That brings me to the last of the objections: isn’t Aristotelian teleology just the sort of thing that was rightly expelled from proper natural science in the 16th century? I don’t think, actually, that teleological talk does conflict with a modern scientific worldview — if “a modern scientific worldview” means the worldview presupposed by actual working science. I do recognize that it conflicts with any number of explicit philosophies of science, from early modern mechanism to high logical positivism to the modern day; but, well, so what? There’s good reason to think that it conflicts with them not because teleological language is, in and of itself, anti-scientific, but because the expulsion of teleological language accompanied the remarkable success of two specific branches of science — mechanics and chemistry — that for the past five centuries scientists and philosophers have repeatedly tried to “reform” all the special sciences by imposing standards of language specific to mechanics and chemistry on them. But there’s precious little reason to think the methods or forms of language appropriate to mechanics and chemistry are also what will work best for biology, geology, ecology, paleontology, epidemiology, metereology, tidology, psychology, economics, sociology, anthropology, etc. etc. etc. In fact, as I think Thompson convincingly shows, these methodological constraints are distinctly inappropriate for biology (and “natural history” broadly), because teleological talk about life cycles and natural functions is both irreducible to non-teleological terms, and essential to understanding commonplace biological statements about species, their traits, their lives, the organization of their bodies and the operation of their organs or tissues, and indeed the foundational concept of “life” itself.
Of course, I’ve only sketched out the arguments here, and referred to further discussions. I don’t expect this to be a knock-down argument in favor of aristotelian categoricals and their employment in inferring teleological facts. But if you think that you’ve got a way to reduce them to terms without teleological import, or to understand the categoricals without mentioning teleological facts at all, or to do biology while dispensing with both the categoricals and the teleology, well, try me.