Abortion insurance

Dan Fabulich (daniel.fabulich@yale.edu)
Mon, 01 Mar 1999 05:19:49 -0500

Timothy Bates wrote:
>What is the extropian position on these two? We no doubt all agree about
>free crypto, and certainly I am glad the Nuremberg decision happened this
>way. They are perfect example of first-use force hiding behind free speech.
>These anti own-body control memes retard the rate at which we get to
>transcend ourselves

Abortion is one of those problems for which an honest libertarian will have to claim that libertarianism does not have a simple answer. If the aborted fetus was a citizen who bears rights, then abortion is probably force against a non-aggressor, unless we conclude that the fetus itself would rather risk abortion than pay for abortion laws. Then again, if the aborted fetus is not a citizen but rather a part of the mother, then abortion is not force at all; outlawing abortion in this case is force against a non-aggressor.

If libertarianism is defined purely in terms of the prohibition of force against non-aggressors, then we can clearly see that libertarianism will agree with both sides of the abortion debate, depending in this case on the definition of a person and what qualifies consent.

Normally I like to fall back on Marshall efficiency to help answer these questions, since it more or less maps directly on to expected utility, and thus leaves us, more or less, with utilitarianism. (Friedman provides a great intro to Marshall efficiency in his "Expanded Postcript for Libertarians" in his _Machinery of Freedom_.)

In an ideal situation we imagine that a "rational fetus" would be willing to pay quite a bit for its own protection if it could do so; perhaps more than it would cost to protect it, even when adjusted for the accrual of interest. On the other hand, it seems like trying to effectively enforce abortion laws would require an extremely intrusive gov't, and therefore an extremely costly gov't; since abortions can and are performed as back-alley affairs where abortions are illegal, we might not expect such laws to be effective no matter how expensive. The cost of enforcing them may be so high that even the imagined "rational fetus" wouldn't be willing to pay for it; that the fetus would be willing to accept a higher risk of abortion rather than pay the enormously high cost of abortion insurance.

If, in fact, abortion is so costly that the fetus would be unwilling to pay for its enforcement, then we may think of the freedom to choose abortion in terms of assumed consent; in the same way we make decisions for people who are comatose and cannot directly provide their consent. Assumed consent is not entirely unreasonable in this context: consider a starving man who steals food when its owner is away, and provides money to pay for the damages. In this case, we allow the fetus to suffer a risk to its own life, and as a result it doesn't have to pay the costs of insurance when it gets older, in the form of taxes paid to an intrusive gov't.

If it doesn't seem plausible that a "rational fetus" would make this choice, consider also the fact that many unwanted children go on to live exceptionally poor lives, so the "insurance" we provide may save the fetus's life, but that fetus may grow up to live rather unhappily in abject poverty, which is only made worse by the presence of an intrusive gov't with increased taxes. We have heard this argument before in the argument of abortion in a somewhat different context. Formerly, this argument may be seen as pretty absurd: would the fetus's live be so bad that the fetus would rather commit suicide than live it? Obviously not. So this consideration alone is not a good reason for us to allow abortions to go on.

Here, however, we might imagine a "rational fetus" considering the problem thus: "If I pay for abortion insurance, I can greatly reduce the risk of my being aborted before I am brought to term. However, if we consider the cases where I would have been aborted but wasn't, many of those leave me with very little well-being; since the cost of insuring against abortion is so high to begin with, I'd want to expect that I'd be getting something really good for my money; in most of the cases where the insurance would be useful, I wouldn't."

I think I have outlined a correct analysis of the problem of abortion in terms of Marshall efficiency; I think most of those those who disagree with me in this case would also disagree with using Marshall efficiency as a moral indicator, and perhaps with utilitarianism in general. Certainly, if you believe that abortion is simply wrong, not as a derivative from some other foundational rule but as a simple moral prohibition which cannot be derived nor rejected, you will not find my argument very compelling. Most people do not think this way, however: most people who think that abortion is wrong think that it is wrong because we have a prohibition against doing harm to begin with. This claim is problematic, and one which shouldn't be accepted by a libertarian: we should, at the very least, presume that we can harm an aggressor.

Moreover, I think we can also say that we can harm someone who explicitly consents to be harmed. This is why libertarianism is not defined in terms of harm, but rather in terms of force, which is, by definition, some action committed against the will of the person(s) being forced. In light of that, it also seems reasonable that libertarians should accept some kind of assumed consent; here I adopt a theory of consent which assumes consent in terms of what a "rational" agent might do, but this is not the only such theory of assumed consent. Some may argue that assumed consent in the case of abortion is not meaningful: a fetus, though alive, cannot formulate its own consent, and therefore any harm committed against it is necessarily forced harm. This objection overlooks the fact that although the fetus cannot give its consent, neither can it deny consent. And since it can do neither, the argument that it WOULD give its own consent, if it could, should convince us (presuming that my argument is correct!) that it is right for us to do some amount of harm onto the fetus (in the form of the risk of abortion) by the assumption of assumed consent.

These objections have been raised against the claim that assumed consent even makes a difference here: that abortion is simply wrong by assumption, or that a fetus which cannot conceive of consent cannot be assumed to give it to us. Further objections might agree that it matters whether or not the fetus would give its consent, and go on to claim that my argument does not show that the fetus would actually do so. For example, the fetus may not be "rational;" if not, we may have to provide abortion insurance for the fetus, even though the fetus is making an irrational decision for itself. We may be morally obliged to do so if it would ask us to. I do not have the right to forcibly prevent my irrational friend from committing suicide if he decides to do so; therefore I certainly have no right to withhold abortion insurance from an irrational fetus when that insurance may save its life.

However, am I REALLY forbidden from preventing my friend from committing suicide? What if I know that this bout of suicidal tendancies is temporary, and that if I prevent him from committing suicide now, he will thank me later and praise my decision? While a theory of assumed consent could certainly be abused to justify all kinds of governmental atrocities, (and has, and will be again!) it seems like in the case of a suicide it would also result in an ethical wrongdoing if we do not accept assumed consent. Libertarianism does not tell us anything about assumed consent, since in order to formulate the idea of libertarianism one must already know the definition of force and consent. However, utilitarianism in this case can provide us with a clear answer: assumed consent results in optimal net utility as evinced by my friend's praise of me after the fact, so we should accept it. (This only applies when the theory is correctly formulated: clearly a theory of consent which assumed that you agreed to let me rape you simply because you were wearing a skimpy dress would be an ill-formed theory!)

This still does not show that the fetus would give its consent; after all, I may be absuing the theory of assumed consent in this very case! There are a host of different ways this objection might be raised against me, the most plausible of which is perhaps the argument that by my reasoning I have assumed that the fetus would be willing to accept its own death in exchange for the cost of abortion insurance. We do not accept the argument that "abortion is okay because most would-be abortions live unhappy lives," because unhappy life is usually better than no life at all. However, this objection is based on a mistake about my argument and the position from which we must consider the fetus: we are not imagining a fetus considering whether it should be aborted or whether it should not be aborted (and have to pay for the cost of preventing its abortion), since obviously in this case the fetus would accept the latter; instead, the fetus must consider the possibility of getting NOTHING from abortion insurance and still having to pay for it. From this perspective, it may well be willing to suffer an increase in the RISK of abortion in exchange for not paying for the abortion insurance, which may well be useless.

Some may still claim that this argument leads to unethical conclusions, and therefore we should reject it out of hand. However, the main reason I chose Marshall efficiency is because it maps so well onto utilitarianism; in this case, if my argument is valid, then not only is it an argument within libertarianism for rejecting abortion laws, but a utilitarian argument as well. Not everyone would be better off without abortion laws: clearly, many fetuses who are aborted will be much worse off than they would have been in a state which banned abortion. However, those who were worse off would not be MORE worse off than others would be made better off by the freedom to choose; on average, most people would be better off than the aborted fetuses were worse off, if my Marshall analysis is correct.

While utilitarianism is not anywhere near universally accepted as an ethical theory, those who believe that utilitarianism is right will believe this argument from Marshall efficiency, presuming the analysis itself is valid. Even if you don't accept utilitarianism, most of us do believe that improving the happiness of all people, even at the expense of some, is a worthwhile moral consideration. If my argument is ethically wrong, then it must be that both utilitarianism and libertarianism must be wrong here as moral theories [qualified by the assumption that assumed consent is a valid libertarian consideration]. Perhaps they are, but if you disagree with both utilitarianism and libertarianism, you really aren't the sort of person to whom I have intended this argument. If you are of this camp, I am sorry to have wasted your time: you will not find my argument convincing, and I have nothing further to say to you.

For everyone else, however, I hope I've given you some food for thought on this debate, and hopefully allowed you to view abortion from a direction which you're not used to. I know I've never heard an abortion argument presented in these terms before, so at the very least the argument is new on me; hopefully you'll appreciate the argument's novelty even if you reject its content.

Good luck!


               -THEN WHAT DOES-