Saturday, April 17, 2010

Marquis on Contraception and Identifying Victims

Marquis' (1989) 'Why Abortion is Immoral' famously argues that abortion deprives the fetus of a "future like ours". Towards the end of the paper, he argues that "nothing at all is denied such a future by contraception" (p.201). I think this view is defensible, but not for the reasons he suggests.

In response to the claim that contraception "deprives both the sperm and the ovum separately of a valuable future like ours", Marquis writes:
On this alternative, too many futures are lost. Contraception was supposed to be wrong, because it deprived us of one future of value, not two.

This is too quick. We can imagine "fusion" cases where two people (A and B) merge together to share a single future C. (It helps if A and B are intrinsic duplicates, or at least extremely similar.) As in fission cases, we might not want to describe this as a case of strict "identity" across time, but it's clearly compatible with what matters in survival. That is, even if the logic of identity prevents us from saying both that C is A and that C is B (because A isn't B, and identity is transitive), we should still hold that C's life constitutes a "future of value" for each of A and B, such that it'd be harmful to each to deprive them of this future by killing them before they merge.

Since it's possible to wrongly deprive two people of a single shared future, Marquis' response here is insufficient: more work is required to explain why contraception doesn't deprive both the sperm and egg of a valuable future. (I think there are two respectable answers here. Pro-lifers should appeal to a kind of biological essentialism, and say that gametes are just a fundamentally different kind of entity from human beings, and hence do not develop into the conceptus, but rather are replaced by it. But my own view is that only persons have futures in the relevant sense, so I don't think fetuses are harmed by death any more than sperm are.)

Marquis continues (pp.201-2):
One might attempt to avoid this problem by holding that contraception deprives the combination of sperm and ovum of a valuable future like ours. But here the definite article misleads. At the time of conception, there are hundreds of possible combinations... there is no nonarbitrarily identifiable subject of the loss in the case of contraception.

The notion that victims need to be "identifiable" yields atrocious results. Suppose a large roomful of people are about to be gassed. A teleportation machine is set to randomly choose one of these people and teleport them to safety. Would Marquis have us believe that it'd be okay to turn off the machine and let everyone die, just because "there are hundreds of possible" beneficiaries, and no particular person we can pick out as the one who would have been rescued?

Again, we need a different explanation of why it is that mereological sums of sperms and eggs aren't harmed by contraception. In particular, we should say that these gerrymandered "objects" aren't the right kind of entity to be the subjects of future experiences. (Only persons, or organisms, or some such, exhibit the requisite internal unity.)


  1. I always wondered if he wrote that paper just because there was such a dearth of pro-life arguments from philosophers (at least, non-Christian ones), so he figured it was guaranteed to be extensively reprinted and cited. I suppose I'm probably being too cynical.

  2. I had similar thoughts. My example involved the claim that I have no obligation not to spin a cannon at random and fire, since there is no person such that I determinately would have killed him.

  3. About Marquis' first argument. I cannot shake the view that identity matters. If due to fusion, C stops existing (because after the fusion there is no entity that is C), then how does C get something that is just as good as literal survival? He no longer exists. So he gets nothing. Can you awake me from this dogmatic slumber?

    Relatedly, you write, "it'd be harmful to each [of A and B] to deprive them of this future by killing them before they merge." But you can't kill them before they merge, for the simple reason that they don't exist then, and you can't kill things that don't exist.

    About Marquis' second argument. I agree that this is a bad argument, for the reasons you say (I think Norcross made a similar point in a reply piece to Marquis in JPhil). But I do agree with Marquis' conclusion: that contraception does not deprive the combination of sperm and ovum of a valuable future like ours. My reason for thinking this is that during the process of fertilization, the sperm goes out of existence (after it penetrates the ovum, it basically breaks apart). Since the sperm goes out of existence, the "combination" of sperm and ovum does, too (I'm thinking that a combination is something like a set). Since conception destroys the combination of sperm and ovum anyway, preventing conception through contraception does not deprive the combination of sperm and ovum of a future like ours.

  4. Hi Chris - what do you make of fission cases? Take Parfit's car accident case, where I could survive as either brain hemisphere transplanted into an appropriate new body, if the other hemisphere was destroyed in the crash. Since what matters in survival is surely an intrinsic matter, we can conclude that Lefty's successful transplant suffices to secure me what matters in survival regardless of what happens elsewhere -- in particular, regardless of whether Righty is also successfully transplanted. But identity is extrinsic: if both Lefty and Righty survive separately, then neither can be said to be "me" -- I lack a unique "closest continuer" -- even though either one alone could have filled this role in the absence of the other.

    I agree with Parfit that this kind of case establishes that identity is not what matters in survival. Do you disagree? If so, is it because you think that what matters in survival could be extrinsic? Or do you think identity is intrinsic, and have some alternative account of how the identity facts work out in the car accident case?

    P.S. Remember that the case discussed in my original post is one of fusion (two becoming one), not fission (one becoming two). A and B may cease to exist after they merge. They do exist beforehand. I meant to describe a case where you could kill them moments before they would otherwise merge, so they do still exist at this time.

    Now, by merging two near-duplicates, the resulting person C would be psychologically extremely continuous with each source person (A, B): remembering their memories, carrying out their plans and intentions, etc. The only major difference is that C only has half the atoms from each. For anyone remotely sympathetic to psychological accounts of identity/survival, it's hard to see this physical fact as bearing much normative significance. (After all, we think you can survive teletransportation, where all your atoms change.) So it seems like A and B would be severely harmed if they were killed rather than merged.

    As I see things, what matters in survival is that there be future person-stages who are psychologically related to (connected with) my present stage in the right sort of way. Killing someone is bad insofar as it deprives them of such continuants or future person-stages who would continue their "life story" in a desirable way. To "possess" a future of value, in the morally relevant sense, does not mean anything more than this. In particular, it does not require that the continuants be numerically identical to you, as Parfit's fission cases seem to show that numerical identity can be influenced by normatively irrelevant extrinsic facts. In the fusion case, the future stage C ably continues the "life stories" of both A and B. Hence both A and B may be harmed in the above way by actions that prevent C's existence, e.g. the pre-merge killing of each. They are not so harmed (on this account) if they instead cease to exist by merging to form C.

  5. I think Marquis' argument could be made much more simply -- if contraception is wrong, so is abstince. Both prevent the joining of sperm and egg, thus both prevent the potential fetus from having a future like our own.

    An argument against the morality of contraception ends with a Monty Pythonesque conclusion that every sperm is sacred.

  6. That wouldn't help Marquis, dialectically. He'd be left having to explain why his argument against abortion doesn't just as well imply that both contraception and abstinence are immoral.

  7. Incidentally, fusion is not hypothetical; just one form, 'vanishing twins', happens surprisingly often:

  8. @Aaron Boyden: Actually, it appears his main motivation in writing this paper was to provide an account of the moral wrongness of killing. It just so happens that his view is incompatible with reproductive freedom. In conversation, Marquis has admitted that if an alternative account of the wrongness of killing is developed, one that doesn't have the consequence of the immorality of abortion and that can handle the problem cases as well as his account (supposedly) can, then that account is preferable to his. Admittedly this is a little bizarre given the title of the paper.

    @Richard: You're right about the first argument. Marquis needs to claim that gametes are of an entirely different sort than zygotes. In fact, given comments Marquis has made in several lectures, this is exactly what he has in mind. Though Marquis is averse to anything too metaphysical (his typical response is something like "I'm just a simple-minded country boy from Kansas! I don't know what the hell that 'possible worlds' talk is all about!") one might argue along the following lines.

    Unrestricted composition (UC) allows for any (seemingly) arbitrary classes of things to compose an object, regardless of spatio-temporal location. Further, the principle doesn't consider other relevant features: structural organization or restrictions on the kinds of things that can compose an object (you don't just get water molecules from any old kind of atom, they need to be hydrogen and oxygen atoms).

    By contrast a zygote isn't the sort of thing that comes into existence when a sperm and an ovum are mereologically fused. Still, by UC there will be objects that consist of the mereological sum of sperms and eggs, those sums aren't zygotes. The kind "zygote" requires the fusion in the biological sense of a sperm and an egg, not a fusion in the merelogical sense.


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)