Suppose that while you’re driving home from work, you get into a car accident, and the other person is severely injured. He’ll die unless you give him a kidney. It would be morally admirable of you to donate your kidney to him, but many people find it doubtful that you should be legally required to donate your kidney. They would say that the government shouldn’t force you to provide this sort of bodily assistance, even though it’s necessary to save the other driver’s life.

Estimated reading time: 11 minutes

The Car Crash Analogy

Pro-choice advocates sometimes leverage the preceding thoughts to argue as follows:

P1. You shouldn’t be legally required to donate your kidney in the car crash case.
P2. If you shouldn’t be legally required to donate your kidney in the car crash case, then a pregnant woman shouldn’t be legally required to donate the use of her uterus to the fetus.
C. So, a pregnant woman shouldn’t be legally required to donate the use of her uterus to the fetus.

Call this the Car Crash Argument. Premise P1 is intuitively plausible to many people. Premise P2 is a parity premise: it’s supposed to be plausible because the two types of actions that it mentions—your donation of a kidney and the pregnant woman’s donation of the use of her uterus—are quite similar. In particular, both are acts that save the life of a person (the other driver in one case, the fetus in the other) whose life is in danger because of the potential helper’s voluntary act (the act of driving in one case, sexual intercourse in the other).¹ This similarity is supposed to make it plausible that if one of these acts shouldn’t be legally required, then neither should the other.

The Car Crash Argument, like Judith Thomson’s famous Violinist Argument, is a bodily rights argument. So most of the strategies typically used to rebut Thomson’s argument can also be used against the Car Crash Argument. For example, it might be argued that there is a special duty to assist one’s own biological children. If there is such a special duty, then there’s a relevant difference between the car crash case and the case of pregnancy: only in the latter case is the needy person the potential helper’s own biological child. As another example, it might be argued that it is worse to kill a person than to merely let her die. If killing is worse than letting die, then there’s another relevant difference between the car crash case and the case of pregnancy: only in the latter case is it true that the only way to refrain from assisting the needy person is to kill that person. Both of these potential responses target premise P2 of the Car Crash Argument, as they attempt to direct attention to a difference between the two actions mentioned in this premise that breaks the parity between them.

My preferred response to bodily rights arguments is the “responsibility objection,” so in what follows I’ll offer a response to the Car Crash Argument that draws on some considerations that pertain to the pregnant woman’s responsibility for the fetus’ needy state.

A Response to the Car Crash Argument

I’ll start by noting that there’s an ambiguity in the car crash scenario as I described it above: it’s not clear whether the car accident is your fault. I left this ambiguous on purpose because different pro-choice people who appeal to the car crash analogy fill in the details differently. Some invite us to imagine a version in which the accident is your fault (for example, see page 42 of this piece by a pro-choice philosopher), and others describe a version in which the accident isn’t your fault (for example, see here). I think the difference between these two versions of the case matters, so I’ll give them labels to keep them distinct:

Fault
You’re driving home from work, and because you’re drunk, you run a red light. As a result, you get into a car accident, and the other person is severely injured. He’ll die unless you give him a kidney (you’re the only match).

No Fault
You’re driving home from work. Everyone is driving carefully, but the traffic light malfunctions, giving everybody a green light. As a result, you and another car run into each other. The other driver is severely injured. He’ll die unless you give him a kidney (you’re the only match).

There are two ways of interpreting the Car Crash Argument. On one interpretation, the argument is appealing to Fault. On the other, it’s appealing to No Fault. On both interpretations, the argument faces problems.

First, suppose that the Car Crash Argument is appealing to No Fault. In that case, I argue that premise P2 of the argument is doubtful because there are two relevant differences between No Fault and the case of pregnancy. First, in the case of pregnancy, it’s a foreseeable result of the woman’s voluntary sex act that the fetus will end up needing her uterus in order to survive.² This is because it’s generally known that it’s not that unusual for pregnancy to follow sexual intercourse. In contrast, it’s not a foreseeable result of your act of driving that someone will end up needing your kidney in order to survive, for it’s incredibly rare for such a situation to arise as a result of driving. I’ve never even heard of this actually happening outside of philosophical thought experiments.

When a needy person’s neediness was a foreseeable result of a voluntary action of yours, this makes you more responsible for her neediness than you otherwise would be. And the more responsible you are for a needy person’s neediness, the more fair it is to require you to assist her, all else equal. So, there’s some reason to doubt premise P2. There’s reason to find it more plausible that a pregnant woman should be legally required to donate the use of her uterus than that you should be legally required to donate your kidney.

Here’s the second relevant difference. In the case of pregnancy, the pregnant woman bears a lot more responsibility for the fetus’ needy state than does the fetus himself. After all, the fetus bears no responsibility at all for his own needy state, for he isn’t yet capable of choosing to do things. In contrast, you bear no more responsibility for the other driver’s needy state than the driver himself bears. This is because all of the reasons for thinking that you’re responsible for the driver’s needy state suggest equally-compelling reasons for thinking that the driver is responsible for his own needy state. For example, it may be true that you bear at least a bit of responsibility for the driver’s needy state since you performed a voluntary act (the act of driving) that causally contributed to the driver becoming needy. But if this is so, then it’s also the case that the driver bears some responsibility for his own needy state, for he too performed a voluntary act (again, the act of driving) that causally contributed to his becoming needy.

When person A bears a lot more responsibility for person B’s neediness than does person B, this makes it fairer to require person A to assist person B than it otherwise would be. So we again see that there are some grounds for thinking that it is fairer to require the pregnant woman to provide bodily assistance to her fetus than it is to require you to provide bodily assistance to the car crash victim in the case of No Fault.

I conclude that if we take the Car Crash Argument to be appealing to No Fault, it’s reasonable to doubt premise P2, for there are two relevant differences between the actions mentioned in this premise.

Now suppose that the Car Crash Argument is appealing to Fault. In that case, I think we should reject premise P1. If you’re drunk driving and run into someone, and if you thereby bring it about that she’ll die unless you donate a kidney to her, then you should be legally required to donate your kidney. It seems unfair for you to be allowed to just walk away and allow the victim of your drunk driving to die. To my knowledge, nowhere in the US does the law require you to donate your kidney in this sort of scenario, but it seems reasonable that the law should be changed.

It’s worth noting that even David Boonin, one of the best defenders of the pro-choice bodily rights argument, is inclined to agree that people like the drunk driver should be forced to provide bodily assistance to his victim. Boonin doesn’t consider the car crash case specifically, but he does consider a relevantly similar scenario. Paraphrasing Boonin, his scenario is as follows:

Poisoning
David Shimp knows that if he puts a certain toxin in Robert McFall’s drink, it might cause McFall to get aplastic anemia. Shimp also knows that if McFall gets aplastic anemia, he’ll die unless Shimp donates some of his bone marrow to him. Knowing all of this, Shimp puts the toxin in McFall’s drink anyway, just for the fun of it. Sure enough, McFall gets aplastic anemia.³

Boonin observes that it does seem right to say that Shimp owes McFall some bone marrow, and he suggests (correctly, in my view) that this is so because Shimp wrongfully harmed McFall. And this case is relevantly similar to the case of Fault, in which you’ve wrongfully harmed your victim. So it should also seem right that you owe your spare kidney to the victim of your drunk driving in Fault. And if this is so, there’s reason to think that you should be legally required to give your kidney to your victim.

Here’s a summary of my response to the Car Crash Argument. Either it appeals to Fault or it appeals to No Fault. If it appeals to No Fault, then there are good reasons to reject premise P2. And if it appeals to Fault, then premise P1 can be doubted. Either way, there are good grounds for finding the Car Crash Argument unconvincing.

How the Car Crash Analogy Can Help Pro-Lifers Refine Their Views

The car crash analogy comes up frequently on the Abortiondebate subreddit. Sometimes, the analogy is used to develop something along the lines of the Car Crash Argument. I’ve argued that this use of the analogy is unsuccessful. But there’s a another way that pro-choice advocates sometimes use the analogy that I think is much better. I’ve seen the following sort of discussion unfold many times:

Pro-life: The woman had sex knowing that she might become pregnant as a result. So, she consented to being pregnant. That’s why the fetus has a right to be in her uterus.

Pro-choice: Consent to sex is not consent to pregnancy. If you think otherwise, it’s because you’re assuming that it’s in general the case that if you consented to do something, then you’ve thereby consented to all of the possible consequences of doing that thing. And this general principle is false. We know this because it’s surely not true that someone who consents to driving has thereby consented to being in a car accident, and this general principle implies otherwise.⁴

I think the car crash analogy succeeds in showing that it’s not generally true that if you’ve consented to do something, then you’ve thereby consented to all of the possible consequences of doing that thing. So pro-lifers who want to defend the pro-life position by way of showing that the woman has consented to being pregnant should appeal to a general principle that’s different from the one that the pro-choice person attributes to the pro-lifer in the above exchange. Perhaps pro-choice advocates on social media like using the car crash analogy because a lot of pro-lifers have a hard time articulating a general principle that both (i) implies that consenting to sex is sufficient for consenting to pregnancy, and (ii) does not imply something implausible in the car crash case.

A good way to approach developing such a principle is to consider what potentially relevant differences there are between the case of pregnancy and the car crash case. In this article, I’ve directed attention to two such differences that have to do with the pregnant woman’s degree of responsibility for the fetus’ needy state. These differences suggest something like the following principle:

If you perform a voluntary action that has as a foreseeable result that someone will die unless you provide him with bodily assistance, and if you bear a lot more responsibility for his being in his needy state than he does, then your voluntary action counts as (tacit) consent to provide him with that assistance.

This general principle is much better than the one that pro-choice people often attribute to pro-lifers, for it is not easily refuted by either of the versions of the car crash analogy (Fault and No Fault). And this general principle does have the pro-life-friendly implication that the pregnant woman has tacitly consented to provide the fetus with the use of her uterus. For as I have observed, it’s true both that (i) the pregnant woman performed a voluntary act that had as a foreseeable result that the fetal person will die unless she provides him with this bodily assistance, and (ii) the woman bears a lot more responsibility for the fetal person’s being in his needy state than he does.

To sum up, the car crash analogy is used by pro-choice advocates for multiple purposes. One of the main ways this analogy is used is to press the Car Crash Argument, and I’ve explained why I don’t find this argument persuasive. A second way the car crash analogy is used is to challenge a common strategy for establishing that consent to sex is sufficient for consent to pregnancy. I’ve suggested that this use of the car crash analogy helps pro-life advocates to avoid relying on an overly simple principle about when consent to one thing amounts to consent to another.

Endnotes