Many moons ago, I read the pre-print of a forthcoming essay by Perry Hendricks—whose work I always enjoy—titled “My body, not my choice: against legalised abortion”.
I disagreed with everything, and wrote a rip-roaring two page response. I decided against trying to publish it though, on the grounds that someone else would probably make my point for me.
I was right: Soon after Hendricks’ paper came out, Kyle van Oosterum [young man pictured below, 6,10”] wrote a two page response pressing the same objection I had, which you can read in paper and blog form.
[You’re on a date with THIS man, and you get a fax from Derek Parfit saying he’s solved the non-identity problem. What do you do?]
Why do van Oosterum and I reject Hendricks’s argument? First, here’s what his argument aims to establish. It aims to establish that bodily rights arguments for abortion—the kind defended by Judith Thomson and David Boonin, which in turn aim to show that abortion should be legal even if the foetus is a person—do not succeed.
In response to Boonin, who has the most well-developed version of the argument to-date, Hendricks (mistakenly) re-constructs Boonin’s version of the bodily rights argument as follows:
(1) Being a person does not give one the right to use another person’s body, even if it is needed for survival. [. . .]
(2) Therefore, even if the fetus is a person, it does not have the right to use its mother’s body, even if it is needed for survival.
(3) The state should only coerce A into letting B use her body if B has a right to A’s body.
(4) Therefore, the state should not coerce women into not having abortions, even if fetuses are persons.1
I say “mistakenly” because Boonin never commits to a claim as strong as (1), and has publicly said that he isn’t so committed. (1)—if true—would entail that parents have a right to let their babies die of thirst if hydrating them means walking to the tap to pour a glass of water, or starve their children by withholding breastmilk if providing it requires breastfeeding.
But Boonin has explicitly denied that he’s committed to (1), saying, with respect to the breastfeeding case: “If the burden of breastfeeding, as I understand it, is considerably less than the burden of undergoing an unwanted pregnancy, then no, it does follow from the claim that there’s not an obligation to carry an unwanted pregnancy to term that there wouldn’t be an obligation to breastfeed.”
Hendricks—writing under the misapprehension that Boonin does commit to a principle as strong as (1)—quite understandably hits him with the breastfeeding reductio:
CABIN: Sally is 9 months pregnant. Unfortunately—as occasionally happens—she doesn’t know that she’s pregnant. One day, while out hiking, a snowstorm unexpectedly hits, and she is forced to take shelter in a cabin. To make matters worse, she goes into labor while stuck in the cabin. The birth goes well, and her baby is healthy. Sally is stuck in her cabin for 7 days before she is finally dug out. Rescuers find her alive and well, but her infant is dead due to starvation—Sally did not feed her infant, despite having ample food for herself, and producing ample breastmilk (there was no baby formula available in the cabin).2
The intuition most everyone has about CABIN, myself included, is that Sally acted wrongly, and that, if you were in the cabin with Sally, it would be permissible—perhaps obligatory—for you to coerce Sally into breastfeeding her infant for seven days.
But all this shows is that in cases where providing bodily life-support to your child is comparatively undemanding (compared, that is, to pregnancy), you’re obligated to provide it, and others are allowed, and sometimes obliged, to force that provision. What it doesn’t show is that these things are true of cases where providing bodily life-support to your child is substantially more demanding, which is what Hendricks would need to show with regard to pregnancy.
In other words, the fact that pregnancy is significantly more demanding than breastfeeding might explain why it’s okay to refuse bodily support to a child in one case and not the other. But does this make a difference? We can test our intuitions with a modified version of the case. Consider this modified version of CABIN, and tell me if your intuitions change:
CABIN*: One day, while Sally is out hiking, a snowstorm unexpectedly hits, and she is forced to take shelter in a cabin. Sally is stuck in her cabin for 7 days before she is finally dug out. Rescuers find her alive and well. But they also find a dead infant. Sally explains that when she took refuge in the cabin, she found a baby, cryogenically frozen in a block of ice. Fortunately, when she put it by the stove, the ice melted and the baby sprung back to life. To go on living, however, it needed some milk. Unfortunately, due to its weakened condition, the only way for Sally to safely keep the baby alive was to strap him to her chest. And more unfortunately still, the only adhesive in the cabin with which to strap it was a roll of magic spell-o-tape (it’s a witch’s cabin—roll with me here). Along with the back pains that go along with strapping a baby to one’s body for nine months, spell-o-tape, which is imbued with all sorts of devilish properties, carries a number of magically-induced side effects: nausea without vomiting, nausea with vomiting, fatigue, bloating, mood swings, cramping, food aversions, and everything else on the What to Expect catalogue. To top it off, the spell-o-tape can only unstick after nine months, and, when it does, the peeling-off induce a pain that rivals the intensity and duration of human childbirth. Deciding she would rather not, Sally declined to strap the child to her chest. A few days later, he died of starvation3.
Intuitively, would it have been okay for you to coerce Sally into strapping the infant to her chest with spell-o-tape for nine months? If you’re inclined to say ‘no’, you should reject Hendricks’s argument outright. If you’re inclined to be agonistic, you should be unpersuaded by Hendricks’s argument, since he has the burden of proof. If you’re inclined to say ‘yes’, then you should think Hendricks is right, and van Oosterum and I dialectically screwed. But I doubt you will have that reaction, because it seems obvious—to me, at least—that it totally would not be okay to coerce Sally, at gunpoint, into strapping the child to her chest in CABIN*. If only pro-choice people have that intuition, so be it. But all that means is that Hendricks won’t persuade any pro-choice people.
Hendricks knows about this worry, of course. He considers it in a follow up post on the blog of the APA. Here is his reply in full:
One problem with this sort of objection—which usually comes from childless men [🫣]—is that breastfeeding is a burden that’s additive: the longer a woman must breastfeed, the more burdensome it is (i.e. it’s more burdensome to breastfeed an infant for one month than to breastfeed an infant for one day, it’s more burdensome to breastfeed an infant for two months than one month, and so on). So, suppose that in The Cabin Case, it will be 18 months until Sarah is able to be dug out. Suppose, furthermore, that none of the solid food in the cabin can be made into baby food, and that the infant (for whatever reason) won’t be able to eat solid food until she’s 2 years old. This means that Sarah will have to breastfeed the infant for the whole 18 months in order for her (the infant) to survive. This is probably more burdensome than typical pregnancies. Nevertheless, it would be morally wrong for Sarah to let her infant starve—she must use her body to sustain the infant’s life. And this is true even though it’s probably more burdensome than typical pregnancies. So appeals to burdensomeness are on shaky ground.
Again, one might deny this. One might claim instead that the burdensomeness of breastfeeding an unwanted infant for 18 months is so great that it’s permissible for Sarah to let her infant die. This objection seems to me to rely on far too permissive a morality—morality requires us to sustain the lives of others even when doing so is very burdensome. For example, consider the fact that we must sustain the lives of severely disabled people who require assistance with eating, toileting, showering, and so on—it’s wrong to not provide them with care—even though doing so can be very burdensome. But despite its burdensome nature, providing care for disabled people is something we’re morally required to do—even if we don’t want to provide such care. Indeed, providing this care for the entire life of a severely disabled person is going to be more burdensome than a typical pregnancy. And so morality requires that we take on substantial burdens—even if it involves using our bodies to sustain the lives of others and even if we don’t want to.
I don’t think the childless men slight was fair, since I’m married to my blog, but in any case, two points are in order.
First, it’s not clear that 18 months of breastfeeding really is more demanding than 9 months of pregnancy. I doubt that, but whatever, I have no way of knowing, so judge that claim for yourself.
Second, and this is the key point, notice how the goalposts seemed to shift. The subtitle of the original paper was “against legalised abortion”. Yet in his reply to van Oosterum, Hendricks settles for arguing that it’s impermissible for Sarah to let her child die. But to show that abortion should be legal, not merely that it’s morally wrong, it needs to be intuitively plausible that it would be okay for you to coerce Sarah into committing to breastfeed for 18 months4. Consider the following case:
CABIN**: You—who are holding a shotgun—are in a cabin with Sarah and a baby. The only way to keep it alive is to strap it to Sarah’s chest for 18 months with 18 month-o-tape, which keeps the baby in strapped in position to breastfeed for 18 months, and won’t unstrap before then. Sarah violently refuses. Would it be permissible to force her, at gun point, to strap the baby to her chest?
It seems obvious to me that it wouldn’t be. But even if you aren’t sure what to say about CABIN**, you still shouldn’t be swung by Hendricks’s argument, since the burden is on his to give positive reason to think that it would be permissible to force Sarah, at gun point, to strap the baby to her chest for 18 months.
Third, and this is a point made in passing by van Oosterum, every cabin case we’ve considered so far—CABIN, CABIN*, and CABIN**—undercounts the opportunity costs of many pregnancies, meaning that even if you have the intuition, say, that Sarah in CABIN* can be permissibly forced to strap the baby to her chest with spell-o-tape, that still wouldn’t be enough to show that women in most real world can be permissibly forced to remain pregnant for 9 months.
This is because, in each version of the cabin case, Sarah is snowed in, so her decision of whether or not to breastfeed doesn’t adversely affect her career prospects, social life, or familial relationships. But these opportunity costs attend to many real world pregnancies, and add to their degree of demandingness. So even if you think Sarah can be permissibly coerced to breastfeed in all three cabin cases, that doesn’t get you the view abortion should be banned under real world conditions.
Hendricks, Perry. 2022. “My body, not my choice: against legalised abortion”. Journal of Medical Ethics 48: p. 457.
Ibid. p. 458.
You might object that in CABIN*, the child in question is not Sally’s son. If you think that’s important, we can just stipulate that he is.
Addendum: Since writing this, Hendricks has flagged to me that he didn’t intend to argue—in that blog post—that abortion should be illegal, but only that it’s morally wrong.
It seems like there's a difference between coercing somebody into doing something, and coercing them into *not* doing something. Consider the CABIN* case, with the following modification: Sarah has *already* strapped the child to her body, and now she wants to remove them. Let's also suppose (for the sake of making the case more analogous to abortion) that Sarah is the child's mother. It seems obvious to me that Sarah is doing something seriously wrong (i.e. morally equivalent to murder) by removing the child, and it seems plausible that society would be within its rights to stop her from doing so.
I think this intuition makes problems for Boonin's McFall v. Shimp example as well: there's a big difference between the state forcing McFall to undergo a painful medical procedure, and the state PREVENTING a pregnant woman from procuring an abortion. Consider: the state has the right to prevent people with body integrity disorder from amputating their healthy limbs, but it doesn't have the right to force people to undergo unwanted surgeries.
What do you think of Dr Calum Miller's arguments against abortion?