Erstwhile roomie Glen Whitman has a good, thoughtful post up on why most analogies used in abortion debates don’t really work. Now, I don’t often find myself reaching argument of this kind. I don’t think fetuses are persons or that they have any rights, and you don’t need elaborate thought experiments to explain why you’re not obligated to play host to a tapeworm under any circumstances. Still, plenty of interesting debate proceeds on the assumption that, if not from the instant of conception then at any rate sometime well before birth, the fetus is person-like enough to have either full blown rights or close enough to create a realmoral issue. For the sake of the rest of the post, I’ll assume that counterfactual. Suppose that at some point the fetus does become sufficiently personlike that we want to attribute rights to it, and we want to know how to regard abortion past that point: What then?
The key reason most analogies offered fail—and if you’ve spent any time following abortion threads on chat boards, they’re typically appallingly poorly thought out—is, I think, that (as Glen observes) there’s at least one salient way in which pregnancy is unlike any of the other cases Glen considers. That is: One of the parties involved in this case doesn’t exist at the outset (when one has sex), and then comes into being.
Now, to back up for a moment, we can imagine a principle for opposing abortion that would make this irrelevant. One might just say there’s a general obligation to provide lifesaving sustenance to anyone who needs it if you’re in a position to do so, even if this requires significant personal sacrifice. There are plenty of fine arguments for that position, but most abortion opponents don’t appear to believe it. It would, after all, require one to spend most of one’s income beyond the sustenance level on providing food and medicine to, say, starving children in the developing world. Plenty of abortion opponents will, of course, say one ought to (or perhaps must) do something to help the very needy—especially needy children whose condition is clearly not to any extent of their making. But most will not, I think, argue that anyone who fails to provide further aid—so long as there exist more children to help and another dollar the provider doesn’t need to live—commits a wrong equivalent to abortion.
So, returning to Glen’s analogies, the mother’s obligation must be regarded as being a specific obligation arising from her relation to that specific fetus. Now, recalling the argument of the last paragraph, it can’t only be that the mother is uniquely situated to provide nutrients. True, in the developing world case, many other people could potentially feed those same children, in a way that any other person could not (yet) house that particular fetus. But for practical purposes, given that there are starving or sick children that nobody else is helping, the situations are similar: If you don’t provide aid and nutrition for that child, it won’t be forthcoming from some other source.
So it’s got to be something else. This is where it’s relevant that, as many abortion opponents will somewhat confusedly say, the mother “accepted the risk” of pregnancy when she had sex. (Leaving aside, for the moment, cases of rape or incest, on which anti-abortionists split.)
Now, again, as Glen points out, on face this makes no sense. Even if you go out without a scarf in the cold, you’re not forbidden to use antibiotics because you “accepted the risk” of getting sick. Another, equally confused, locution has it that the mother “implicitly agreed” to the risk of pregnancy, though that won’t quite work either: Who is the agreement supposed to be with? Not the fetus, surely: It doesn’t exist yet. (One of Glen’s commenters implausibly suggests it’s the sexual partner; I’ll assume that one doesn’t need to be argued against.)
As Glen suggests, the only way it makes sense to introduce the mother’s voluntarily engaging in sex, knowing that pregnancy might result, is if we think of it as a kind of tort. And, still following Glen, that seems like a promising analogy. You’re not generally required (we’re supposing) to aid anyone who finds themselves in a dependent condition. But if you brought it about that the person was in that condition, then we tend to think you certainly are obligated.
Now, in tort law, we have different standards of liability—sometimes we have a negligence standard, where you’re only obligated if you failed to take reasonable precautions against your actions harming someone (in this case, birth control), whereas under strict liability, you’re responsible no matter what.
Interestingly, we don’t typically see anyone arguing for a negligence-standard type approach to abortion, wherein you’re free of obligation to the fetus provided you were taking precautions to prevent pregnancy. So let’s leave that aside and suppose a strict liability standard is applicable.
Here, as Glen points out, the prior non-existence makes things tricky. I’ve been talking loosely about “placing someone in a condition of dependence”, but in ordinary cases, what we mean is “relative to how they were before.” You’re normally not obligated if you leave someone in a dependent state because you’ve helped them from a still-worse condition, but not enough to leave them independent. Rather, it’s when you’ve left them in the lurch relative to their status quo ante—how they were before. That’s the baseline to which you’re obligated to return them, to “make them whole,” as the law puts it.
Glen suggests, then, that if the special obligation to the fetus is the product of an intentional tort—conception—then the remedy can only be to return the fetus to the status quo ante, non-existence. Which abortion accomplishes handily.
That is, of course, too clever by half. The problem, I think, is that it treats the relevant baseline as “non-existence” (to which one can be returned) and “never having existed” (which, absent a time machine, one can’t). But it does bring out how weird a case this presents, where our normal intuitions, which are bound up with baselines in assessing how we affect others, founder.
On the one hand, I think it’s coherent to say something like: “My life is so awful, it’d be better if I’d never existed.” And, similarly, I think it’s coherent (even plausible) to say that existing briefly, then being killed, is worse than not having existed in the first place. But there remains something very weird in comparing different imaginable states of being to non existence. The problem is that “never having existed” doesn’t fit in the ordinary scale of evaluation. It’s not like zero—like you get less and less happy until you get to the non-existence level, and then if you keep going you get to more and more suffering, so non-existence ends up being just like feeling sort of mezzo-mezzo.
Normally when we evaluate the badness of death, after all, it’s not really the “disutility of nonexistence” (which is incoherent) we’re actually thinking about. We don’t think: “Oh, yeah, being dead, that must suck.” You’re not there to mind, after all. Instead we look at it, as usual, from the perpsective of a baseline—the continuation of that person. But this is where it becomes hard to stick with that initial presumption of personhood, because it becomes hard to disentangle the value of life from people’s concrete desires, hopes, projects, and so on. The death of a young person, we usually say, is especially tragic because we see the erasure of an at least partially determinate trajectory. We’ll find it less sad when someone dies after a long, productive life having achieved many of their major aims—”now I can die happy,” someone who’s succeeded in one of their core goals might say.
Anyway, I think what this does is underscore, perhaps in a needlessly circuitous way, Glen’s point: Our ordinary analogies won’t work for abortion. Not just because a houseguest isn’t like a fetus in the womb, but because the ordinary concepts we fall back on when thinking about obligation—agreement (whether implicit or explicit) or harm (defined relative to some baseline)—aren’t applicable. Intuitively, creating a rights-bearing entity entails some kind of obligation to the creative being, but our intuitive categories of obligation don’t fit the case well.