Against the Impairment Argument: A Reply to Hendricks

In an article of this journal, Perry Hendricks makes a novel argument for the immorality of abortion. According to his impairment argument, abortion is immoral because: (a) it is wrong to impair a fetus to the nth degree, such as causing the fetus to have fetal alcohol syndrome (FAS); (b) it is wrong to impair a fetus to the n+1 degree (to cause the fetus to be more impaired than to have FAS); (c) killing the fetus impairs the fetus to the n+1 degree (causes it to be more impaired than to have FAS); (d) abortion kills the fetus; (e) therefore, abortion is immoral. The impairment argument is a promising account for the wrongness of abortion because it does not rely on the controversial metaphysical premise that a fetus is a person. This article aims to show, that despite some immediate advantages over the rival theories of the immorality of abortion there is a reason to believe that the impairment argument is untenable. That is because there are goods that can be achieved by abortion but that cannot be achieved by impairing the fetus.


| 863
RÄSÄNEN someone reading (or writing) this paper. 4 It simply relies on a quite common assumption that if it is wrong to cause something bad to happen, it is even worse, morally, to cause something worse to happen. 5 There are several ways one could contest the impairment argument. For example, Bruce Blackshaw argues that killing does not impair the fetus the same way as impairing does. 6 Dustin Crummet argues that the ceteris paribus clause of the impairment principle is not met in cases of pregnancy because carrying an unwanted pregnancy to term is much more burdensome than is refraining from excessive drinking for 9 months. 7 Here, I will show that the ceteris paribus clause in the second premise of the argument is not met. Put another way: all things are not equal, as there are certain goods that cannot be achieved by impairing the fetus, but which can be achieved by killing it. 8 To see the problem of the impairment principle, we can apply the argument to non-human animals as well. Many people believe it is wrong to impair animals. For example, it is wrong to blind a horse.
If it is wrong to blind a horse, according to the argument, it is also wrong to kill a horse. 9 But horses are often killed, for example, for food. If the impairment argument works, not only is it wrong to kill fetuses-it is also wrong to kill many non-human animals. 10 Advocates of the impairment argument could contest my claim and argue the ceteris paribus clause is not met in the case of an animal, but that it is met in the case of a fetus. Hendricks explains the ceteris paribus clause as follows: If, for example, the impairment referred to in the consequent produces a particularly valuable good, but the impairment referred to in the antecedent does not produce such a good (or the good it produces is significantly less valuable), then the ceteris paribus clause is not met, and

TIP [the impairment principle] does not kick in. 11
For instance, one could claim that, by killing the aforementioned horse, there is a valuable good produced (its edible meat) that is not produced by blinding it, while there are no valuable goods that are achieved by killing the fetus that would not also be achieved by impairing it.
However, there are two problems with this reasoning. First, if one would eat the fetus after killing it, then according to the argument, abortion would not be immoral, as there would then be an achievable good (the fetus' edible meat) that would not be possible to achieve if the fetus were merely blinded. But surely, eating the fetus after killing it does not make abortion less wrong; on the contrary, it makes it worse.
Second, some goods can be achieved through abortion that cannot be achieved by impairing the fetus, thus the impairment principle fails.

| G OODS OF ABORTI ON
What goods can be achieved by abortion, but not by impairing the fetus with FAS? Hendricks briefly considers some possible goods, and finds them lacking. These are: the good of the woman's autonomy, the good of not having to be an unwilling parent, and the good of not having a child.
I believe the most obvious good that comes from abortion is that the child that would otherwise exist does not exist. Some might be skeptical whether there is anything good about "non-existing." Nonexistence does, however, provide several goods. Non-existence is good for (a) the non-existent entity itself, (b) its prospective parents, and (c) others (including animals, the environment, and fellow humans).
First, non-existence is good for the one who does not exist because she would avoid the pains she would otherwise inevitably suffer in life if she were to exist. 12 Abortion prevents someone from coming into existence (or kills them, depending on one's view of when a fetus becomes a person), and that someone would inevitably suffer during her life. When a woman has an abortion, she prevents the pains and discomforts of life from befalling on the fetus (or future person). Of course, abortion also prevents the happiness and many pleasures of life that the fetus (or future person) would have experienced if not aborted, but that is not relevant here. What is relevant is that the lack of pain and discomfort of the future person cannot be achieved by impairing the fetus.
Hendricks does not claim that the only option for the ceteris paribus clause not to be met is that the bad outweigh the good; he simply claims that it is enough to refute the argument that abortion can achieve some goods that impairment cannot.
Second, the death of the fetus is good for the prospective parents because they would avoid the burdens of parenthood, such as losing sleep due to the infant crying at night. These cannot be achieved by impairing the fetus, for example, by giving it FAS. Again, surely there are goods of having a child that cannot be achieved by not having one, and in at least some (perhaps most) cases, the goods of being a parent outweigh the goods of not being a parent.
Again, however, the point is not to compare the outcomes of being

RÄSÄNEN
vs. not being a parent, but simply to show that there are goods that can only be achieved by killing the fetus-not by impairing it.
One could object here that the goods of not producing a child can be achieved by adopting one instead, so abortion is not necessary. Hendricks suggests this. While it is true that some of the goods of not having a child can be achieved by adoption (such as not losing sleep because of a crying infant), not all goods of not having a child can be achieved by adoption.
For example, in some rare cases, the genetic parents may well be required, morally, to do something demanding and harmful for themselves because of the well-being of the child. According to McMahan, many people intuitively believe that this biological relationship gives the sperm donor a special reason to provide bone marrow for the child, although it might not mean that the sperm donor is morally obligated to provide the bone marrow. It is only to say that he has a moral reason to provide it that someone completely unrelated to the child does not have. While the sperm donor might not have a moral obligation to donate the bone marrow, I believe it is safe to assume that he is at least obligated to consider the choice more seriously than would be expected from a total stranger. This moral reason to consider the donation does not disappear after the child is adopted away, but it does disappear when the child is killed as a fetus. Therefore, there is a good of not having a child that can only be achieved by an abortion, so the impairment argument fails.
One might object that in the above case, there is nothing that harms the parents, and avoiding the harms of parenting is the main good that is avoided when there is no child. But in the above example there simply is a risk of harm-because it is not certain that the adopted child will ever need bone marrow-and that is a very different thing from actual harm. But is that so? If a dentist tells me that one of my teeth has suffered internal resorption, slowly destroying the tooth from inside, and that every week there is a 10% chance that I will suffer terrible pain and discomfort, I would not be satisfied to hear that it is just a risk of pain and suffering, even though I might never experience any pain related to the tooth. I would not be satisfied because the risk itself, at least once I am aware of it, causes me harm, since I could not enjoy my life normally. Similarly, someone putting a child up for adoption might be harmed (for instance psychologically) because she might constantly be aware that the child could someday find her and ask for, say, a bone marrow transplant. 14 Last, but not least, having yet another human on the planet would be bad for the environment, other humans, and non-human animals, because humans tend to over-consume limited natural resources at the cost of other living things. 15 The harm that every human inflicts on others is something that can be reduced by abortion, but not by impairing the fetus. Therefore, the impairment argument fails.

| CON CLUS ION
I have argued here that the impairment principle, although initially promising, cannot show that abortion is wrong. This is because there are goods that one can achieve by killing the fetus that one cannot achieve by impairing the fetus. Thus, the ceteris paribus clause in the second premise of the argument is not met. Therefore, I think, we can add the impairment principle to the list of pro-life arguments that are not convincing. 16

CO N FLI C T O F I NTE R E S T
The author declares no conflict of interest.