Wednesday, November 19, 2008

The pill

I will use the phrase "using oral contraception" (and cognates) to abbreviate the complex state of affairs of being a woman of normal fertility and using standard contemporary (not the older higher dose pill) oral contraception as the only form of contraception for the period of at least a year while being sexually active at an average sexual frequency. The following argument is sound when implicitly conditionalized on the present state of medical knowledge and technology:

  1. Someone who prevents the implantation of an embryo that he or she is a parent of causes the death of his or her innocent child. (Premise)
  2. It is wrong to act in a way that carries a significant risk of one's causing the death of one's innocent child without very grave reason. (Premise)
  3. Using oral contraception carries a significant risk of one's causing the death of an embryo that one is a parent of. (Premise)
  4. It is wrong to use oral contraception without very grave reason. (By (1)-(3))
  5. Very grave reason to use oral contraception is exceedingly rare if it ever occurs. (Premise)
  6. It is wrong to use oral contraception except perhaps in exceedingly rare cases. (By (4) and (5))
The same goes for the IUD. I do not know if the argument holds for implantable or injectable hormonal contraception, but at least unless one has very good reason to think that it does not, one has good reason to avoid that, too. I should add that I think marital direct contraception is always wrong, but the present conclusions are is controversial enough.

The argument has a formal feature that complicates things. The terms "significant risk" and "grave reason" are not defined explicitly. Rather, they must be taken to be partially interdefined by (2).[note 1] Thus, in defending (5), one needs to argue that what is exceedingly rare is the sort of reason that would fit into the exception in (2), where "significant risk" is the risk we get from the medical literature supporting (1).

I will not argue for (1) and (2). I think (2) is uncontroversial, and (1) will generally be accepted by pro-life folks.

We do not know exactly how often the use of oral contraception causes a failure to implant. A survey of data is given by Larimore and Stanford (2000). Hormonal contraception has three main modes of operation according to references (including product inserts in at least some cases): (a) prevention of ovulation, (b) modification of uterine environment that makes it inhospitable, and (c) modification of cervical mucus to make it harder for sperm to reach the ovum. The data in Wildt, et al. (1998) strongly suggests that sperm can travel through thick mucus. So (c) isn't a very effective method. Studies show that ovulation still occurs in 1.7-65% of cycles (see Larimore and Stanford, 2000; unfortunately, I don't know if this is for perfect use or typical use). Assume, conservatively, a not unrealistic fertilization rate of about 10% per cycle in which ovulation occurs.[note 2] Assuming independence (which isn't exactly right, admittedly), this gives us an annual fertilization rate of 2%-55% (assuming 12 cycles per year). But the established pregnancy rate for oral contraception is significantly lower than most of this range (the perfect use pregnancy rate is 1% or lower; the typical use pregnancy rates are higher), where an established pregnancy is one where implantation has occurred. So there is good reason to think that there are probably significantly more fertilizations than established pregnancies, and hence there is good reason to think that using oral contraception carries approximately a 1% to 50% chance (this will be a combined epistemic and nomic probability) per year of preventing a fertilization, and thereby causing the death of the embryo.

This will yield (3), assuming 1% to 50% counts as "significant risk" in the sense used in (2). But I think it clearly does count. A quick way to see this is to imagine that the risk is not to one's child, but to the user. The FDA would not approve a form of contraception which had a 1% to 50% annualized chance of resulting in the death of the user, and medication with that fatality rate would only be approved if the condition it treated was very grave indeed. If such a risk to the user would be unacceptable absent a very grave reason, a fortiori it would be unacceptable when the risk was to another party.[note 3] In fact, I doubt that we would approve of life-saving medication that had a 1% to 50% chance of causing the death of a bystander (imagine that it gives off noxious fumes or something).

That leaves (5) to be argued for. But that's easy. Given that the "very grave reason" would have to be one that would justify taking an annualized 1% to 50% risk of being the cause of one's innocent child's death, it seems clearly that only extreme circumstances will yield such a reason. If using oral contraception is needed to save someone's life, a case might be made (though note the caution at the end of the previous paragraph). But remember that I defined "using oral contraception" as including sex at a normal sexual frequency and with no other contraception being used. So that would have to be a case where having sex at a normal sexual frequency was needed to save a life and no other contraception was possible. Maybe if a woman had to have sex with a dictator for a year or he else he would kill her (or someone else), and if pregnancy would result in the woman's death, and if the use of non-abortifacient contraception were impossible, this could be argued to be a case like that (in the end I deny it, because it is wrong to commit adultery even to save a life). But such cases are, indeed, exceedingly rare. Perhaps saving someone from serious disability would qualify. Could "saving a marriage" qualify? I doubt it. Would it be permissible for a couple to undergo a "marriage saving treatment" that had a 1% to 50% chance of killing one of their children? And the case would have to be such that the consequences of pregnancy would be very grave, and that no other contraception was possible. (And even then it would be wrong if, as I think, marital direct contraception is always wrong—but that requires a different argument.) So I think (5) is very plausible.


Gordon Knight said...

This is a little off the point, but I wonder about implications of # 2
From what I understand about 60% of naturally conceived embryos fail to implant for natural reasons.

If I am a woman (I am not, but this is philosophy, so lets be hypohetical),and i accept # then does it not seem that i am obliged not only to not engage in an activity that increases the chance of my embryos not being implanted, but the stronger thesis that i should take pains to prevent te non-implantation of embryos? And of course I can do this, if I become sterilized.

Sterilization has a cost, buts hard to see how that cost comes even close to the death of one's children.

jawats said...

See here -

An excerpt:

PROF. SANDEL: Thank you. I have two questions about the rate of natural embryo loss in human beings. The first is what percent of fertilized eggs fail to implant or are otherwise lost? And the second question is is it the case that all of these lost embryos contain genetic defects that would have prevented their normal development and birth?

DR. OPITZ: The answer to your first question is that it is enormous. Estimates range all the way from 60 percent to 80 percent of the very earliest stages, cleavage stages, for example, that are lost.

Gordon Knight said...

sorry, it should have occured to me that abstinence and other forms of birth control may also be a way of avoiding the death of one's babies--so its a disjunctive conclusion:

If I should take reasonable efforts to prevent my embryos from not implanting, I should abstain, become sterilized, or perhaps use another form of birth control (e.g.Condoms), or if AP is right that natural family planning works well enough, that would be an option. One think I could not do, it seems, is try to conceive a child.

Alexander R Pruss said...

Everybody (perhaps with a few exceptions) dies. We do not say that their parents are their causes of death (at least not in any morally relevant sense) simply because they have conceived them. Furthermore, the mere length of time elapsed between conception and death should not be relevant. One is no more and no less a cause of someone's death because that death takes five days instead of 80 years. What matters is the causality and activity, not the temporal distance.

The problem I am highlighting with the pill is not merely that the child will die quickly. There is nothing intrinsically immoral about conceiving a child that one knows will die soon (though if one does so casually, that bespeaks a serious lack of moral sensitivity). Rather, the problem is that in the case I am discussing, the cause of the child's death will be the mother's use of the pill.

(It's worth noting parenthetically that killing children is still wrong in regions which have high child mortality.)

If I am right that length of time does not matter and that the embryo is a subject of moral rights just as much as the adult, modifying the case as follows should change nothing: The mother takes a contraceptive pill which a significant proportion of the time allows conception to occur but with the child receiving a growth abnormality that causes him or her to die on the 20th birthday.

Now, let us suppose that Joe's mother knowingly used such a pill when Joe was conceived. Joe is approaching is 20th birthday. It seems that although he ought to be grateful to his mother for the curtailed life that he had managed to have, he has a genuine moral claim against his mother for doing something that she knew would cause his death (and, one might wish to add, untimely death).

Here is a different line of thought. Let us suppose that there was a variant of the pill, call it the pill*, which prevents fertilization just as much as the pill does, but which does not prevent implantation. But, if we take numbers somewhere in the middle of the 2-55% annual fertilization range, the pill* would be an extremely ineffective contraceptive, indeed so ineffective that it would not make much prudential sense to use it. Therefore, if, given all the relevant data and having thought the matter through, one is choosing to use the pill, one is doing so in part because of the ways in which it differs from the pill*. But the pill does not differ from the pill* in respect of its prevention of fertilization. Therefore, if one uses the pill given all the relevant data and having thought the matter through, one is using it because of the abortifacient effect. And that is clearly wrong, assuming of course that abortion is clearly wrong.

Gordon Knight said...

"There is nothing intrinsically immoral about conceiving a child one knows will die soon'

But there is something immoral in conceiving a child one knows may die soon, if the death was in some way dependent on your choice to use contraception (even if the purpose of the contraception was to prevent fertilization?)

There are some differences here, in both philosophical intuition and empirical data. I personally find it hard to accept that a bein without a brain should be considered to have the same value as person with a brain--not that the brain in itself counts, but consciousness counts, and we have reason to think that the sort of consciousness that human beings and higher animals have is associated with brains. this does not support a strong pro-choice positoin, but your example had to do with implantation.. a pre-brain event if there ever was one.

But let us assume:

It is a seriously bad thing for an embryo to be destroyed.

It is not a seriously bad thing for no embryo to be conceived

AP: do you agree with these two claims? it seems that if you do, then, if you accept

one ought to act in such a way as to produce good things and avoid bad things.

there is a prima facie duty to do what you can not to conceive.

the argument i have in mind does not pressupose consequentialism.. only that consequences matter.

Do you want to have a view according to which causation by inaction has no moral signifance at all?

Alexander R Pruss said...


The level of consciousness that a child of, say, 4 months (after birth, i.e., 13 months after conception) has is probably below the level of a normal adult horse. Yet it is permissible to kill and eat the horse if this is necessary to our survival, but not so in the case of the child. This suggests that whatever feature is central to distinguishing the bearers of rights from those who do not bear rights, it is not consciousness.

Anyway, yes, one should avoid the bad and pursue the good. But sometimes the good brings the bad along with it. Thus it is in the case of human life--it comes with death at the end of it. In cases where the good brings the bad along with it, it can still be permissible to pursue the good, as long as the bad is not a means to the end, the bad is not disproportionate to the good, etc. The evil of death is not disproportionate to the good of life--it's worth being alive even if the price is death. In procreating, the couple do not intend the death of the child, either as an end or as a means. Moreover, while the couple's procreative action starts a causal chain leading to the death of the child, the procreative action is only very indirectly the cause of the death of the child.

But in the contraceptive case, the couple's action of contracepting is directly the cause of the death of the child. Moreover, while the procreating couple can say that they are acting for the sake of the value of life, a value to which the evil of death is not disproportionate, the orally contracepting couple is not typically acting for the sake of the value of life, but a much lesser value. Even if the woman's life would be endangered by pregnancy, there are other highly reliable ways of preventing pregnancy, such as total abstinence, radical NFP regimes (e.g., ones where the couple has sex only once per cycle, on a day chosen precisely to minimize the chance of conception--eight or nine days after ovulation seems about right intuitively).

Causation by inaction, if one can even call that "causation", has moral significance, but a lot less than active causation. Suppose you were told by someone with certain knowledge of this that if a certain baby grows up, he'll become another Hitler. It would be wrong to kill the baby. But at the same time, there might be no moral duty to save the baby from attack by another. (In fact, the standard conditions in the doctrine of double effect would perhaps prohibit one from saving the baby. Why? Because the action of saving the baby would have a good effect--the saving of the baby--and a bad effect--all the bad stuff that the dictator would do. But one may only do an action that has a good and a bad effect when the bad effect is proportionate to the good. And in this case, the bad effect seems disproportionate--the deaths of millions.)

Anonymous said...

I believe (3) is in dispute in the scientific literature though. I'm not that familiar with all the studies, but they have a good discussion going over at the Life Training Institute Blog.

What do you think of their arguments?

Alexander R Pruss said...

Well, as long as it's under serious dispute, (3) is still true--i.e., there is a significant risk that there are abortifacient effects.

Here is a thought. Suppose we had a pharmacological contraceptive and some data that inconclusively suggested that a not insignificant percentage of women using the contraceptive--say, 1% or more--died each year. Moreover, nobody had done any good, large-scale studies to ensure that the contraceptive was safe for women. In such a case, (a) the FDA shouldn't approve the contraceptive, and (b) if the FDA by some fluke approved it, no doctor should prescribe it, and (c) if a doctor were to prescribe it, nobody should agree to take it.

Drugs are guilty until proven innocent--they cannot be administered (except in experimental settings, to subjects fully apprised of the risks, and so on) without safety studies.

Gordon Knight said...

We have very different moral intuitions. I think that if I knew a baby would become a hitler, and the only way to prevent the results of Hittler's reign would be to kill the baby, i ought to kill the baby. It is hard thing, but it seems to me that anyone who holds that there can be a just war should also allow this. I am not willing to let the negative value of the death of millions of people to be outweighed by the differnce between killing and letting die one individual.

I don't know what to say aboutthe horse case, though I do think the case for animal rights is not ridiculous for the reason you mention. I would be more inclined to say its the capacity for consciousness, rather than level, which is a necessary condition for having rights (even if the horse does not have a right to life, I do think it has a right not to be tortured). My "capacity" I mean an active capacity, not merely being a causal antecedent state. A "person" who is brain dead does not have that capacity, but I do when I am in a dreamless sleep. Such a capacity is a necessary, not sufficient, condition for having rights, including the right to life.

But consider this case

Suppose me and my wife know that 6 out of ten of our children will suffer and die when they are six months old-- a short life followed by a two weeks of agony. Would you agree that in this circumstance procreation would be a serious moral wrong?

Now if death by itself (independent of suffering) is seriously bad, then we do not need to add the business about suffering. It would be equally wrong for such a coule to conceive a child knowing 6 out of ten will die young. Of course throughout much of human history infant mortality was very high, but in those circumstances, the high risk of infant death was a necessary risk for the preservation of the species.

Of course I am not arguing against making babies, but rather than the moral significance of failure to implant is much less than death of a child. At least I think this is what we ordinarily assume when as a couple we choose to have children knowing that many fertilized eggs will likely perish.

Alexander R Pruss said...

On the capacity vs. active capacity business, see this old post of mine.

I do not think a couple wrongs a child in procreating him even though he will die an early death. A short life is better than no life. The case where the death is painful is harder, but in the end I'd say the same thing, and anyway the deaths of embryos who do not implant are probably not painful.

But there is a clear difference between procreating a child even though he will die an early death, and taking a drug that will cause one's next child to die an early death.

Alexander R Pruss said...

I should add, though, that there probably will be other reasons to refrain from procreation under the circumstances you describe:
- the medical costs of intensive care
- the emotional costs of infant death (embryonic death is not difficult emotionally in cases where the parents do not know about it)
- the fact that even though the action is morally permissible in itself, it might reflect and contribute to an insensitivity to death and suffering

Madeleine said...

I wonder if you can help me either understand the force of Don Marquis' continual identity problem that he argues the embryo faces pre-segmentation or, preferably, give me an argument to overcome it.

My husband, a Christian Philosopher, otherwise very pro-life is persuaded by Marquis' stance and it is causing divisions in our home and on our blog. I need to either understand it or offer him a counter argument. Are you able to assist?

Alexander R Pruss said...

Your wish is my command. There is a post in queue, hopefully for tomorrow.