I am not a utilitarian. I can respect utilitarianism, I think most utilitarians are fine people, I have no problem with Effective Altruism (which tends to be a movement full of utilitarians), and I kinda agree with the argument that all ethical systems are utilitarian, just optimizing for different things1. So while I’m not one, I understand where utilitarians are coming from.
What I don’t understand is why so many of them are pro-choice.
I can understand the basic pro-choice argument from a non-utilitarian perspective. If you don’t see a human fetus as a person with rights and moral value, then it would make sense to allow the mother (who is definitely a person with rights and moral value) to do whatever she thinks is best, including killing her fetus. I don’t agree with this argument, but I understand it. It makes sense from a general deontological point of view: normally the rule is you don’t kill people, but a fetus isn’t a person so the rule doesn’t apply. If it doesn’t apply then go with the other rule, which is people get to decide what to do with their bodies. If a fetus isn’t a person, then you can justify abortion from a deontological perspective.
But how could you justify abortion as maximizing utility in any way?
Let’s not even argue about whether the fetus is a person with rights and such yet. For the sake of this argument, I’ll concede that point. However, person or no, a fetus is a human who, left alone, will probably survive to birth, and then have about 75 years of life ahead of her on average. Unless you believe that most human lives are a net negative in terms of utility (which doesn’t seem to be a common utilitarian stance) then an abortion wipes out a whole heck of a lot of expected utility. Certainly carrying an unwanted baby to term comes with negative utility, but could it really come with so much negative utility that it wipes out the gains from an entire human lifetime?
Now you might say that the woman’s suffering is happening right now, and the future utility of living a life is only possible utility in the future, and that utilitarians should maximize current utility, not future utility. That would certainly be an argument you could make, but if you actually believed it you’d have a very strange kind of utilitarianism. For one thing, if only present suffering and pleasure matter then killing someone via heroin overdose is morally fine. Possibly even praiseworthy, you’ll be increasing his present utility quite a bit until he stops breathing.
Clint Eastwood has a great line in the movie Unforgiven. “It's a hell of a thing killing a man. You take away all he's got, and all he's ever gonna' have.” I think most utilitarians recognize this, which is why they put so much value on saving lives and aren’t, you know, totally chill with murder. On the typical utilitarian view killing someone is bad not only because of their pain and suffering in the present, but also that you cut off their ability to experience any utility in this life again. No more soul stirring sunsets to watch, no more hugging their children, no more delicious meals, no more accomplishments, nothing. That’s a lot of lost utility, and most forms of utilitarianism recognize that.
Yet somehow when it comes to a human fetus all that goes out the window: it’s just a “potential person”, so you can do whatever you want to it. Why? I mean a full grown adult that you plan to murder just has “potential” future utility; maybe his life will totally suck from here on out, and killing him is actually doing a favor. I can see a utilitarian justifying that in the case of mercy killing, for a person who has no real chance of surviving a painful illness for instance. But there is no reason to suspect that a fetus is likely to experience a life that is net-negative for them in terms of utility; indeed, we have every reason to think the opposite. Killing a 50 year old wipes out about 28 years of expected life, killing a fetus wipes out 78. Now depending on the age of the fetus that number is going to switch around a bit: at week 5 there’s about a 20% chance of the fetus miscarrying, so aborting a 5 week old fetus arguably only wipes out an expected 62 years of life. AT 8 weeks the fetus has a 4% chance of dying before birth, and at 14 weeks its down to 1%. All of which still have a much larger expected loss of years of life than killing a 50 year old; so if, as a utilitarian, you think it’s wrong to murder a 50 year old then you should also think it’s wrong to kill a fetus.
Some might object that the kind of fetuses that are aborted are disproportionately likely to have lives that are less happy than average, coming from poor families, or families with higher risk of abuse and neglect. That’s a fine objection, but the utilitarian answer would be to put the kid up for adoption after birth (where he will almost certainly be snatched up by one of the 2 million happy and functional couples who are waiting to adopt), not to kill them! Do utilitarians recommend that children who are currently being neglected or abused be killed in order to raise their utility? No? Then why would you be fine with killing a fetus that only has the possibility of being abused or neglected?
I’m sure there must be some utilitarian arguments in favor of abortion, but I can’t see it.
As a virtue ethics guy myself, I’d be optimizing for maximum virtue.
It's because nobody is a utilitarian, although yes, lots of EA-ish people emphasize consequentialist arguments in an unusually wide range of cases. This social class is deeply embedded in a fundamentally liberal-individualist culture, much more deeply than in the purely intellectual commitments of consequentialism, and is thus extremely loath to justify coercion in the name of (for the sake of argument, let's assume) better outcomes. It is indeed, all in all, hypocritical to be willing to advocate for the rights of, say, shrimp but not for those of human fetuses, though as a pro-choice non-consequentialist myself I personally am glad that nobody seems to be biting that bullet.
Utilitarians don’t agree on whether there’s a moral imperative to create more people if it reduces average happiness - Google the “repugnant conclusion” to see more. But the claim can also shift to assuming the mother in question has already decided how many children she will have in her life, and will use contraceptives later if she had a child now. Then the future child would often be happier than the current child.