Moller on Moral Risk

Moral issues are complex.  Likewise the philosophical arguments that bear on these issues.  What are the practical implications of this complexity? In a recent article ( Dan Moller argues that the practical implications are more significant and far-reaching than we commonly recognize.

Moller focuses on the moral permissibility of abortion (MPA).  Moller’s argument runs roughly as follows.  MPA is an issue over which reasonable people disagree.  Reasonable disagreement is possible with respect to MPA in part because the philosophical arguments for and against MPA are very sophisticated and appeal to moral principles that are in principle contestable or far from obvious.  That being said, suppose I have reviewed all the philosophical arguments for and against MPA and by the light of my reason I have concluded that abortion is in fact morally permissible. I am confident that my belief is correct, but of course I also recognize–I would be unreasonable not to–that it is possible that I am mistaken. I am not an infallible reasoner, and, moreover, I could be wrong in my assessment of the relative weight of moral principles (autonomy vs. beneficence, say).  Indeed, I’m also aware that there exist people much smarter than myself who disagree with my conclusions, after examining all the arguments I have examined.  So I’m confident–I’ve examined the arguments as scrupulously as I can–but admit the possibility of error.

Now suppose that I get pregnant.  After alerting the tabloids, I begin to think about whether I want to keep the child.  Given my moral beliefs, abortion is an option, and indeed the option I select after considering the significant costs of raising the child on my own (the child’s mother has run off with someone else).  But then I recall that my moral beliefs may be mistaken, and recognize that if they are mistaken, then in having an abortion I will in fact be committing murder (or its moral equivalent).  That seems like a serious moral risk.  Am I so confident in my moral belief about MPA that I am willing to risk committing murder?  Well, what’s the alternative?  The alternative is that I ‘play it safe’ and have the child, perhaps putting it up for adoption, perhaps not.  Either way there will be costs, but these costs would hardly justify committing murder, which it is at least possible that I will be doing if in fact my moral beliefs about MPA are mistaken, which they may be.

So: wouldn’t it be immoral of me to have an abortion under these circumstances, despite my carefully considered belief that abortion is morally permissible?  It does seem somewhat reckless, doesn’t it?  In electing to have the abortion, I would risk violating another human being’s right to life, whereas no equivalently bad outcome would be risked if I elected to bear the child.

So what Moller offers us is a kind of second-order argument against the morality of abortion.  Abortion may be morally permissible, but we’re not in a sufficiently secure epistemic position to know this beyond reasonable doubt, and so we ought to err on the side of caution.  In fact, we’re morally obligated to err on the side of caution, for otherwise we would be acting in a morally reckless manner, which is (intuitively) wrong.

I am going to post some thoughts on this interesting argument shortly.  In the meantime and after, I hope to hear from you.

This entry was posted in Uncategorized. Bookmark the permalink.

7 Responses to Moller on Moral Risk

  1. Nates says:

    Suppose I want to drive to the grocery store. Given my beliefs about driving, driving is a safe option. I’m a licensed driver, and I have a reliable history of arriving in one piece at my destination. I’m reasonably confident that I’ll get to the store safely, without running over any pedestrians. But then I recall that I am not a perfect driver and that driving always involves some uncertainty. Even good drivers get into accidents, so I have to acknowledge that my trip to the store could cause a serious accident. Indeed, I could well end up killing innocent bystanders. That seems like a serious safety risk. Am I so confident in my driving skills that I am willing to risk killing pedestrians? The alternative is to ‘play it safe’ and walk to the store, lugging my groceries back home. My feet will certainly get sore, and my nose may get frost-bit, but these costs would hardly justify killing innocent pedestrians. And it is certainly possible that I would be doing this if in fact my driving skills failed me, which they might. Therefore, I should never drive to the grocery store.

  2. David says:

    Hmm. In your example, Nates, the risk is that a morally permissible act (driving to the store) will have bad consequences. In Moller’s example, the risk is that you will be performing an action that is seriously morally wrong. That seems to be a difference between Moller’s case and the case you imagine, although I’m not yet sure if it is a morally relevant difference.

    In any case I think I share your skepticism. I suppose I want to say something like this: as moral agents, we are not responsible for avoiding wrongdoing *simpliciter*, but rather for avoiding ‘culpable’ or ‘blameworthy’ wrongdoing. In other words, the best we can do is act as conscientiously as possible and hope for the best. It doesn’t make sense for morality (or other moral agents) to demand more than that.

  3. Lime says:

    My initial thought is that it is strange to talk of potential moral bad results in the same way we talk about potential material (if you will) bad results (it seems fine, but I could find out later that I have killed someone). Still, many pro-life oriented abortion counselors take just this line in counseling young pregnant women, though they tend to focus on feelings rather than “reason”: “You may feel like having an abortion is o.k. now, but later you will believe that you have made a big, life-altering mistake.” This approach has always struck me as a bit odd, for reasons David hints at, but it certainly does describe the experience of some women who have had abortions. My usual thought is that such women are in a kind of tragic situation in which all decisions will give rise to some regret.

  4. Josh says:

    I suppose the question is shot whether the language of risk applies to deontological decisionmaking. Is it possible to be 90% certain of a moral conviction, for example? I can be 90% sure I won’t get in a car accident, but my judgment about undertaking a potentially immoral act seems different. It seems much more like an all-or-nothing type judgment. Some others might disagree with me, and they might be more intelligent or better at moral reasoning than I, but presumably I’ve already considered their arguments to the extent possible (assuming I’m deliberating in good faith).

    Otherwise, won’t I be in the position of never being able to do anything either way when there are seious ethical considerations on each side? To use the abortion example, aren’t there compelling moral arguments in favor of abortion too? Or does the fact that the decision to *have* an abortion is potentially murderous, but the decision not to have one isn’t potentially immoral? Aren’t there moral problems with trying to raise the child on one’s own, when it can’t be adequately provided for? Doesn’t that deserve consideration too? Is it just because on one side is “murder” that this seems worse? A more interesting case to consider would be one where there are arguments of equal impact on both sides, instead of murder vs. Something more abstract and intuitively less impactful.

    One of the problem with Considering small risks in a consequentialist calculus is he risk of paralysis. I don’t go outside because I migh be struck by lightning. Isn’t this jet the ontological version of that? I don’t undertake actions that might have severe moral consequences? Does the mere allegation of a severe moral consequence prevent me from undertaking it? If so how could I ever act?

  5. David says:

    @Lime: That’s interesting that this argument has already been put to use (in a way) by anti-abortion activists ‘in the trenches.’

    @Josh: I understand the worry about moral paralysis–I mean, imagine a politician/policy maker taking this argument seriously. What then? But I guess it also seems to me that there are some issues where refraining from an action seems the ‘safe play.’ Forget about abortion and consider another example Moller uses–viz..,the ethics of meat-eating. Suppose that I’ve deliberated in good faith, done my homework, etc., and I am convinced that eating meat is morally permissible (and that raising animals for food is morally permissible). If I’m wrong about this, then I am participating in and sustaining a deeply immoral practice. If on the other hand I’m right about the ethics of meat-eating but I go vegetarian anyway, then all I’ve lost is the pleasure of eating meat. (Perhaps this example works better than abortion because there are fewer moral consideration in *favor* of meat-eating than there are moral considerations in favor of allowing women the right to abort.) Would it make any sense for a person in this situation to refrain from eating meat not on the grounds that he believes it is wrong but rather on the grounds that, as he acknowledges, he MIGHT be wrong?

    I think that’s the crucial question here. If the answer is ‘yes,’ then Moller is on to something, and it remains an open/interesting question whether similar reasons might apply to the abortion case. If the answer is ‘no,’ then Moller’s position is totally wrongheaded.

  6. Josh says:

    David –

    I think I agree. If there is no compellingly moral argument on the one side, so it’s just a debate about whether something is prohibited vs. Permitted (but on no argument required) then moral caution seems like the best move. Abortion just seems like a bad example of such a situation- since it cedes a lot of ground to the pro-life argumen right away. Pro-choice arguments should probably also include a moral justification for abortion, no only a refutation of the “it’s murder” objection. Meat-eating, on the other hand, lacks such a moral rejoinder. The invocation of “taste” seems shallow, right?

    But then I wonder if this just had to do with a feature of arguments involving permissibility vs prohibition, rather than something about expert knowledge.

  7. juan says:

    i just did a quick calculation of expected utility of aborting and not aborting, with what i think are reasonable probability assignments and expectation values for the outcome states.turns out it’s better to abort (at least for my hypothetical agent).the details are left as an exercise for the reader
    of course, i may have been wrong

Leave a Reply to Josh Cancel reply

Your email address will not be published. Required fields are marked *