Bryan Caplan  

Where Are the Pro-Life Utilitarians?

PRINT
What does it mean to suggest t... Bernanke and Poole on Secular ...
Pro-life utilitarians are very scarce.  A philosophy professor recently told me that he knows of zero pro-life utilitarians in the entire philosophy profession. 

This is deeply puzzling.  While I'm not a utilitarian, the utilitarian case against abortion seems very strong.  Consider: Even if a pregnant woman deeply resents her pregnancy, she is only pregnant for nine months.  How could this outweigh the lifetime's worth of utility the unwanted child gets to enjoy if he's carried to term? 

A bundle of empirical regularities reinforce this prima facie case.

1. Almost everyone is glad to be alive.  The unwanted infant may have a below-average quality of life, but below-average is usually excellent nonetheless.

2. There is a long waiting list - hence excess demand - to adopt healthy infants, so birth mothers need not raise their unwanted children.

3. Due to the endowment effect, unwanted children often become wanted by their birth mother once they're born - as many would-be adoptive parents discover to their sorrow.

4. Women who just miss the legal cutoff for abortion seem to quickly recover emotionally.  Pregnant women who think "A baby will ruin my life" are, on average, factually mistaken.

How could a utilitarian avoid the pro-life conclusion?  There are two tempting routes:

1. Argue that the utility of the unborn counts for nothing - at least until the fetus starts feeling pleasure and pain.  Convenient.  But once you accept the core utilitarian intuition - that the existence of pleasure is good, and the existence of pain is bad - the creation of creatures who will feel a lot more pleasure than pain seems like a great good.  Picture an uninhabited world capable of supporting happy lives.  How could a utilitarian not want to populate it?

2. Argue that each unwanted child has large negative social effects, even though people are eager to adopt.  Most obviously, utilitarians could embrace an extreme Malthusian story where the birth of one human statistically dooms another.  Once you accept this story, of course, saving any life becomes morally suspect.

When I present the utilitarian case against abortion, people normally reply, "But that implies a further moral duty to have tons of babies."  They're right.  From my perspective, that's yet another convincing argument against utilitarianism.  Creating life is a prime example of what utilitarians conceptually reject: actions that are morally good but not morally obligatory.  But given utilitarians' notorious willingness to bite bullets, why should they demur here?

P.S. If you do know of any pro-life utilitarians, please share URLs in the comments.


Comments and Sharing






COMMENTS (48 to date)
Thomas writes:

See Don Marquis' work.

Britonomist writes:

I think this is a common mistake when criticizing utilitarians: only considering the utility of one person, in this case the baby, or two people; just the baby and the mother. Utilitarians want to maximise utility of the entire system or society, so you need to think of the consequences of everyone, and all future people, when undertaking an action.

Regarding pro choice laws, there is more at stake here. Because the consequences of banning choice can affect millions of peoples utilities over decades to come, it can create a society where we are much worse off than otherwise, a net reduction in utility.

Also, what of negative utilitarianism, it's easy to justify abortion on those grounds.

roystgnr writes:

There was a pro-life utilitarian argument (of the effective altruism variety, naturally) on LessWrong a while back. Most people disagreed with the argument but thought it was very well presented, albeit not enough to promote to the main page. I was happily surprised by the discussion as well.

Ryan Murphy writes:

The alignment makes perfect sense from the standpoint of moral foundations theory. Pro-life arguments nearly all follow the sacredness foundation.

I am pro-life and I am a utilitarian if you broadly define utilitarianism.

type_erasure writes:

I am a utilitarian who believes that abortion is approximately morally equivalent to killing babies. I'm probably okay with making it legal anyway, because we need to draw a line somewhere and birth seems like a reasonable place for that. However, I also care about non-human animals, and so the effect of a marginal baby on our society seems overall negative to me, and so I endorse having fewer Americans and thus end up in favor of abortion. (Also, having a more well-off society is probably valuable for the far future, and fewer unwanted children helps with that.)

Larry writes:

I don't think its fair to say Utilitarianism says something is a moral duty. Utilitarianism isn't interested in questions of duty vs supererogatory. Utilitarianism says having the child is better than not, and leaves it at that. If you choose not to have the child, Utilitarianism says you weren't as good as you could have been. If you want to know about duty go ask Deontology.

John Hamilton writes:

I'm a pro-life utilitarian! I also just finished my philosophy undergraduate degree, so no url yet... I'm curious how your fellow utilitarian blogger Scott Sumner would approach this question.

However, I also know someone (namely, my fiancée) who thinks life is pretty bad and, overall, not worth it (i.e, it would have been better to never have existed). She doesn't want kids. She argues she shouldn't have them, because life is mostly bad. Does utilitarianism entail an overall positive view on existence? I should also add she is open to adoption, since those kids exist anyway. Although now that I think about this, I suppose the logical thing for a pessimistic utilitarian to do would be suicide. Good thing she isn't a utilitarian...

blacktrance writes:

The obvious response to your argument against Route 1 is that the desirability of pleasure compared to pain is in the context of an existing being or fixed population, and that preferring pleasure to pain doesn't by itself commit you to anything when population isn't fixed.

Liam McDonald writes:

What a terrific argument, Bryan. Love it!

Thomas B writes:

To follow your logic, if a woman refuses an offer of sex when she is about to ovulate, that is wrong, or at least objectionable from a utilitarian standpoint?

To my mind, the utilitarian argument is the same.

Not that this makes it wrong on its own merits, but it may provide some insight into why pro-choice people would reject the original argument.

Separately, two other points. One, having a baby is not a mere matter of "gestate for 9 months, then forget it ever happened". Two, happiness is not a particularly useful metric for welfare: yes, you can be happy - "recover emotionally" - after having had your life ruined (it's commonly reported by accident victims who are left quadruplegic), but that doesn't mean you're not worse off.

Dale writes:

Hey, I'm the author of the Effective Altruism Abortion piece linked to above. I think it is highly likely that such considerations do pull strongly pro-life, though I go into some sub-arguments that pull in the opposite direction in the post (for example, utilitarians will reject many classic pro-life arguments. But they will also reject many classic pro-abortion arguments!)

There is also the argument from moral uncertainty. Even if you were 90% sure that unborn children were not morally significant, the 10% chance, multiplied by the huge moral cost if you are wrong, far outweighs other considerations. (I go into more detail about some example calculations in the post).

Also I know of one other guy who is an academic analytic philosopher and is probably pro-life, based on a few coded comments, but I am disinclined to 'out' him/her as I assume his discretion is out of fear of professional consequences.

Original post:

http://effectivereaction.wordpress.com/2014/12/31/blind-spots-compartmentalizing/

chris writes:

Isn't Derek Parfit's "repugnant conclusion" relevant here?
http://plato.stanford.edu/entries/repugnant-conclusion/

austrartsua writes:

The problem here, as pointed out above by Thomas B, is that your argument justifies rape. If a moral argument justifies rape, it must be flawed. This is the closest thing you will find to a proof by contradiction in philosophy.

Yes, this is a matter of rights. A women has the right to choose, or not to choose, to give birth to the baby. It is as simple as that. Utilitarianism is for people like Peter Singer.

[comment edited with commenter's permission--Econlib Ed.]

Tom West writes:

Every utilitarian I know has limits as to what they would accept happening to an individual in order to increase the happiness of others.

Many, including myself, would consider an involuntary pregnancy is far worse (and far more intimate) than slavery (at least for the duration).

Or do you think utilitarianism naturally implies full-on Matrix style "turn human beings into baby-making machines while feeding their brains signals to make them happy"?

Matt H writes:

[Comment removed for irrelevance.--Econlib Ed.]

Keith K. writes:

"The problem here, as pointed out above by Thomas B, is that your argument justifies rape."

I do not understand the leap in logic that equates consensual sex with rape in this line of reasoning. I do not absorb that a consensual activity is the same as a rights violation when I read "when a woman is already pregnant, the loss of her 9 months is less than the gain of the unborn child's likely decades of life". This is what Bryan's argument was based off the text here. I do not understand why anyone would consider Thomas B' argument strong. Can y'all elaborate why it justifies rape?

Bob Murphy writes:

[Comment removed for irrelevance.--Econlib Ed.]

Jameson writes:

How many utilitarians are there, pro-life or pro-choice? I mean because if there are a lot, that bothers me.

Sam W. writes:
I do not understand the leap in logic that equates consensual sex with rape in this line of reasoning. I do not absorb that a consensual activity is the same as a rights violation when I read "when a woman is already pregnant, the loss of her 9 months is less than the gain of the unborn child's likely decades of life". This is what Bryan's argument was based off the text here. I do not understand why anyone would consider Thomas B' argument strong. Can y'all elaborate why it justifies rape?

I think it puts it a little strongly but the simple form would be this.

Having babies is good. Really really really good.A women's opposition to having babies (either via not copulating or not bearing the child to term) is an amount of "pain" that is inconsequential compared to the good of having more babies. Therefore ignoring women's wishes with regard to childbirth is morally permissible/admirable because net good is higher.

Philo writes:

[Comment removed for irrelevance.--Econlib Ed.]

Tom West writes:

I think Sam W. has it right. Both rape and being forced to carry a fetus to term are both about as deep a violation of personal body integrity as it's possible to have.

At least that's how it looks from this male's perspective. If someone told me that I had a parasite that could save a life, and thus I was legally obligated to let it grow until it was 8 pounds, whereupon they could fairly safely cut it out of me, I'd not be very happy, even if I was told that statistically I'd be okay with it afterward.

I might *choose* to do it, but legally obligated? No.

And I doubt any utilitarian that I've met would disagree (with the male analog).

Floccina writes:

To austrartsua's argument could a utilitarian say the existing debate is not about rape but about abortion, rape must be considered separately, there is no utility in bringing up now?

Andrew Pearson writes:

This was something I thought about a while back. The more interesting scenario, and the stronger argument against utilitarianism, comes if you imagine a world with abortions but no contraception.

In such a world an abortion could argue that banning abortion would simply lead to lots of people abstaining from risky but enjoyable sexual behaviour, without significantly raising the birth rate. The exception, however, would be rape victims, who would not have chosen to engage in risky behaviour.

Thus, in a world without access to contraception, the utilitarian position on abortion would be "abortion should be available to everyone except rape victims". This is precisely the reverse of our ordinary intuitions, where even many pro-life people are willing to accept abortion in cases of rape.

Tom West writes:

the utilitarian position on abortion would be "abortion should be available to everyone except rape victims".

Which pretty much sums up why I don't have principles, only guidelines. It's just way too easy to game any set of principles into something ridiculous.

(Although admittedly, it *is* kind of a fun intellectual exercise to try.)

roystgnr writes:
Utilitarianism isn't interested in questions of duty vs supererogatory. Utilitarianism says having the child is better than not, and leaves it at that. If you choose not to have the child, Utilitarianism says you weren't as good as you could have been. If you want to know about duty go ask Deontology.

Utilitarianism is interested in questions of duty if you do it right. "Is doing X better for expected utility than not" is one kind of utilitarian question, but so is "Is treating X as a moral duty better for expected utility than not".

Treating the latter question as identical to the former would be a big mistake, but so would ignoring the latter question entirely. If (as is likely due to Schelling points, uncertainty, precommitments, etc) it turns out that optimal consequentialist actions include "create a limited set of rules and try to enforce them", then it follows that an optimal consequentialist action is not to instead say "that looks like deontology, ew!" and do something suboptimal instead.

Sam Haysom writes:

[Comment removed pending confirmation of email address. Email the webmaster@econlib.org to request restoring this comment. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Chip Smith writes:

David Benatar's pleasure/pain asymmetry (detailed in his book, Better Never to Have Been) provides a negative-utilitarian-friendly axis from which to derive an explicitly pro-abortion position. The important thing here is that the absolute absence of deprivation (of pleasure) of the unborn person/whatever needs to be weighed against the certainty of pain (including the pain of death) that will be experienced if the person/whatever comes into social existence. Notwithstanding arguments over fetal pain, pre-vital (or pre-natal) nonexistence entails no experiential deprivation (of good or bad things), so it is always better than being born, an event which guarantees suffering (even if slight) and death.

I know that Bryan disdains to engage with antinatalist arguments, but that doesn't mean they're not relevant!

Zachary Bartsch writes:

Bryan, Please see this quick analysis:

https://pacvae.wordpress.com/2014/08/19/kaldor-hicks-and-abortion/

Thomas B writes:

Just to revisit my earlier point, I did not say that the utilitarian position justifies rape. Rape involves a lack of consent; I argued that the utilitarian position would argue that it would be immoral for the woman to refuse consent. Thus, in the utilitarian morality it would not be rape, because the woman would consent.

Since virtually no-one argues that it is immoral for a woman to refuse consent to sex in our world, there is something wrong with the utilitarian position.

Separately, I thought it was well established that interpersonal utility comparisons were a no-no? This makes it impossible to compare the unborn person's utility in potentially living, with the living person's utility in not becoming a parent.

Rhys Southan writes:

It's possible for utilitarians to argue against abortion while arguing in favor of saving the lives of the already born. The focus would need to be on the feelings of the survivors in both cases. If someone who has already been alive for a while dies, this is devastating to the people who know and care about the person. Voluntary abortion doesn't usually cause harm to survivors in the same way.

There are people who get upset about abortion in an abstract sense, because they don't want it to exist or be legal, but even they can't easily get upset about the deaths of specific fetuses because they don't have any real personal connection with specific fetuses. If a woman has an abortion, and it's her decision, it's *possible* that she'll regret it later, but it's unlikely that she or anyone else will be as devastated about the death of the fetus as we tend to get over the deaths of people we know and care about.

Even so, you're right that the doctrine of utility maximization is a problem for utilitarians. Utilitarianism would work better if utilitarians did away with maximization, and with the concept of pleasure and pain aggregating across individuals. But I guess that wouldn't be utilitarianism anymore.

Rhys Southan writes:

In the first line, I meant, "argue for keeping abortion legal while..."

Richard writes:

Can't you make the same argument against population control? Obviously, if every women had 20 children each life would be pretty unbearable. But it seems sure that the optimal number is way more than 2-3, which simply keeps the population even.

Richard writes:

Oops, I posted the reply before getting to the end of the post. Sorry about that.

Richard writes:
Creating life is a prime example of what utilitarians conceptually reject: actions that are morally good but not morally obligatory. But given utilitarians' notorious willingness to bite bullets, why should they demur here?

I think the reason is that most utilitarians, like most educated people these days, hold their commitments to gender equality, gay equality, and anti-racism above anything else. They always have a post hoc justification of why their views do not violate the orthodoxy on certain topics, no matter how unconvincing.

Jameson writes:
A philosophy professor recently told me that he knows of zero pro-life utilitarians in the entire philosophy profession.

How many people does he know in the philosophy profession, utilitarian or not, who are pro-life?

Brian writes:

Bryan's argument is correct. It is not logically possible for a positive utilitarian to be anything but pro-life. On average, each person added to the world contributes a positive marginal utility, both for the person added and for those already existing. This same logic would also oppose population control.

The arguments made above that such logic would also justify rape is not correct. The timing, and not just the number, of births matters in the utilitarian calculus and it should be obvious that the ideal number for a given woman is finite, not infinite. Therefore, it is fine to say no or to use birth control if the timing of pregnancy is clearly suboptimal. Rape would be one such example.

So why are pro-life utilitarians so rare? Because utilitarians, by the nature of their position, reject traditional approaches to morality and tend to associate with others who do the same. Since their non-utilitarian associates reject the pro-life position, they reject it also as a matter of expediency. Their position is not logical but it is a rational response to their social environment.

Anguish writes:

"How could this outweigh the lifetime's worth of utility the unwanted child gets to enjoy if he's carried to term?"

This is absurd. I strongly encourage reading David Benatar's "Why Coming into Existence is Always a Harm" I have included the link below.

It is quite obvious why.

article.sites.utexas.edu/upa/files/2014/02/Benatar-Better-Never-to-Have-Been-EXCERPT.pdf

Demosthenes writes:

I have long thought that an excellent use of charitable resources would be on the demand side of a market for unwanted babies. To put it crassly, pay women not to have abortions and use the charity to raise and place the children.

The marginal dollar spent on that charity would do a whole lot more good than the marginal dollar spent to elect pro-life politicians.

David Friedman writes:

Your post ignores the question of whether what the utilitarian wants to maximize is average utility, total utility, or some other measure. I have argued in the past (and Mead argued earlier) that maximizing average utility leads to implausible implications. But there are problems with total utility as well, at least if you define zero as the suicide point and utility along Von Neumann lines.

I have a chapter in one of Julian Simon's books that tries to construct an ordering of futures with differing populations along utilitarian lines, but it's only a partial ordering.

"What Does Optimum Population Mean?" Research in Population Economics, Vol. III (1981), Eds. Simon and Lindert.

Thomas B writes:

Brian at April 22 5:18pm writes,

"it should be obvious that the ideal number for a given woman is finite, not infinite. Therefore, it is fine to say no or to use birth control if the timing of pregnancy is clearly suboptimal."

I'm not sure this follows the utilitarian logic, but Brian's logic would then apply also to abortion: it is fine if the timing of pregnancy is clearly suboptimal.

Brian writes:

"Brian's logic would then apply also to abortion: it is fine if the timing of pregnancy is clearly suboptimal."

Thomas B,

Not so fast. Your claim was that by Bryan's utilitarian logic, a woman should not say no to rape if she is fertile. My logic definitely applies to your scenario but it does not necessarily apply to abortion itself. First, with regard to consent, given that the ideal number of births for a given woman is finite, a pregnancy imposes opportunity costs--you can't get pregnant by someone else while you are pregnant from the rapist. Why does that matter? Because the maximization of utility for everyone involved depends on mating with someone with whom you are compatible and with whom you are likely to share child raising duties. Since none of those conditions is likely to be met by mating with the rapist, the likely better outcome is to say no.

Can this logic be used to justify abortion itself, either as a result of rape or otherwise? Possibly, depending on what version of utilitarianism one espouses. However, utilitarianism is usually applied to those who already exist and have clearly identifiable future capacities for pain and pleasure. One does not try to maximize the utility of hypothetical human beings. In saying no to the rapist, the utilitarian has no obligation to maximize the utility of the imagined child that might result.

An abortion decision, however, necessarily effects the utility of an already existing human being. That's a different matter altogether. Under utilitarian reasoning, it is possible to justify abortion under certain "hard" cases, like the life or procreative health of the mother (if the birth of the child prevents many future births by the same woman, then the abortion might be morally preferred), but generally there would be a strong moral presumption against abortion.

Richard writes:

"One does not try to maximize the utility of hypothetical human beings...An abortion decision, however, necessarily effects [sic.] the utility of an already existing human being."

Brian,

This is begging the question. If utilitarianism does not try to maximize the utility of hypothetical people (I tend to disagree, but can take it as a premise), and fetuses are considered people, then the arguments for allowing them to survive vs. killing them follows the same path as for any other person. Occasional murder where expected utility is negative over the individual's life, and a "strong moral presumption against" murder in other cases.

In the first case, I don't understand how your model or utilitarianism deals with birth at all if it fails to consider hypothetical people. If I concieve a person, knowing full well its average utility will be negative over its life, is it not somehow wrong to actually conceive the child? Under your explanation, I see no reason it would be, given that I don't take into consideration the utility of hypothetical people, like the yet to be conceived child.

Consider the strange case of knowing that conceiving a child would bring a small (net) happiness to existing people, while the child itself would (net) suffer greatly. You wouldn't consider the child's suffering (they're hypothetical, existing people are not), conceive the child, then decide the child's suffering will be too great and abort the fetus (Their suffering now can be considered, and they now exist). Then, you would conceive another child, which would suffer the same fate, over and over again. In short, I don't see how not accepting the utility of hypothetical people doesn't make you temporally inconsistent when a possible action is "Make hypothetical people".

Secondarily, even if we accept the somewhat strange claim that we should only maximize utility of existing people, we still need to assign personhood. You never once use the word person in your comment, instead using "human being". I can't speak for you, but I place little to no value on the life of a cell line in a lab (other than its value to thinking humans), but might value an alien species or intelligent machine. This is why the word "person" is used rather than "Human being" in these conversations. But if "person" and "human being" need not be identical, we need to identify when a human being is or is not a person. This is a common talking point so there's no need to further belabor it, but one you seem to gloss over entirely, simply stating utilitarianism values existing human beings and fetuses are existing human beings. As I said before, this is begging the question. The questions are, should utilitarians value hypothetical people, and if not, are fetuses people.

Henry writes:

[Comment removed for supplying false email address. Email the webmaster@econlib.org to request restoring your comment privileges. We'd be happy to publish your comment. A valid email address is nevertheless required to post comments on EconLog and EconTalk.--Econlib Ed.]

Brian writes:

Richard,

Thanks for your thoughtful response. You make many reasonable points.
With regard to hypothetical people, some utilitarians may count them, but others surely do not. From my perspective, if the goal is maximizing the sum of everyone's utility functions, it's hard to see how a hypothetical person would have a utility function to contribute.
You say "In the first case, I don't understand how your model or utilitarianism deals with birth at all if it fails to consider hypothetical people." As I think you understand, the decision whether to have a child would be based on its effect on the utility of those already alive. If you knew ahead of time that your child was going to become Adolph Hitler, or destroy the World Trade Center, surely you would favor not having the child? Utilitarian logic would say not having the child would be the moral thing to do.
With regard to your hypothetical of the negative personal utility child, the utilitarian response depends somewhat on what utility you choose to maximize. But here's why in practice the decision really can't be made based on the hypothetical child's utility. Let's assume that the child's utility IS included in the calculation. Even if the child's utility is expected to be negative, the actual utility is highly uncertain. There's no way to predict ahead of time whether the child will actually have positive or negative utility. But the impact of a single person on everyone else's utility is very well known, since it is calculated from the average of all previous examples. The small but certain payoff must be weighted much more strongly than the highly negative but highly uncertain individual utility.
I will comment on some of your other points later, if I am able.

Peter Gerdes writes:

As a utilitarian my short answer is this: murder isn't essentially wrong. Murder is wrong because of the externalities, i.e., people mourning the victim, feeling the need for vengence, the loss of societal investment in the victim.


Longer answer: I suspect that in general producing more humans right now is overall negative in terms of utility unless you have reason to believe you have particularly good genes or non-transferable parenting advantages.

Brian writes:

" I suspect that in general producing more humans right now is overall negative in terms of utility"

Peter,

If I may ask, why do you think this? I would say it is certainly not true that the marginal utility of humans is negative, but I'd like to know why you think otherwise.

Jonatan writes:

I see no reason to distinguish categorically between a potential human that has reached the "cells in a womb" state and a potential human on a prior stage such as the "deciding whether to try and have a child stage".

Thus there is no utilitarian reason to be more focused on the pro-life question specifically, than the overall question of whether we should aim for more living people.

And that is a question that all of us have to consider, not just utilitarians: How do we think about unborn potential human lives?

Also, I would argue that we are all utilitarians. I have never heard a single argument that's not about in the end improving lives for people ( / other creatures.) Principles that give disadvantages in the short run / in specific circumstances, are always argued with the implicit evaluation that they are good for people overall.

Alex Nicolin writes:

[Comment removed pending confirmation of email address. Email the webmaster@econlib.org to request restoring this comment. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Comments for this entry have been closed
Return to top