Sunday, April 04, 2010

Anti-Consequentialism and Axiological Refinements

Many common objections to utilitarianism -- e.g. cases involving unexperienced harms or sadistic majorities -- are really objections to a particular theory of the good, rather than to the core idea that we ought to promote the impartial good. So when faced with such objections, it's worth asking not just whether the action seems wrong, but whether the outcome is really desirable in the first place. If not, the consequentialist has a simple response: the act is indeed wrong, precisely because it doesn't maximize what's (genuinely) good.

Consider the case of the perverted doctor who secretly molests his patients while they're unconscious. This seems wrong, even if doesn't cause anyone to experience suffering. But, more than that, it also seems like a bad outcome (in particular, it seems bad for the molested patient). Not only did the doctor make an immoral decision, but he brought about a worse state of the world. The upshot of the example is that there are certain things we consider undesirable, things we don't want to happen, whether or not we learn of them. A plausible theory of the good will need to accommodate this, which rules out value hedonism, but leaves untouched the core of utilitarianism (i.e. impartial welfarist consequentialism).

A slightly trickier case is the "KKK world", where an overwhelming majority of the population consists of racists who derive immense sadistic pleasure from oppressing the minority. Even if oppressive actions maximize both net pleasure and preference satisfaction, they still seem clearly wrong. I agree with this judgment, but again, it also seems like a bad outcome. We can establish this most clearly by noting that it would still seem bad even if the outcome were the result of natural events rather than human agency. That is, suppose a black guy is struck by lightning, bringing malicious happiness to millions of racists. There are no "wrong actions" here, for the simple reason that there are no actions at all. Nonetheless, it doesn't seem like a good way for things to turn out. The upshot, I take it, is that there's no value (perhaps even disvalue) in sadistic or malicious "pleasures" of this sort.

Is there a sure-fire way to diagnose whether one's intuitive response to such a scenario represents an intuitive rejection of consequentialism proper, or just a particular axiology? I'm not sure. A good heuristic -- sufficient but not necessary to establish intuitive disvalue -- is to try to 'naturalize' the event (as in the previous paragraph) and see whether the intuitive repugnance remains. But it may be difficult to do this with some cases. For example, in the doctor case we can imagine a version where he 'touches' the patient as a result of a muscular spasm rather than a choice, but this no longer seems nearly as bad. Sometimes it is precisely the vicious or inconsiderate exercise of agency that contributes to the disvalue of the outcome.

Once consequentialists build the intrinsic disvalue of vicious action into their axiology, we need a more sophisticated test to distinguish them from deontologists (and hence to distinguish genuinely deontological intuitions from mere axiology-refining intuitions). At this point we may turn to agent-neutrality. For while a consequentialist may be concerned to prevent vicious actions, he is ultimately no more concerned with his own actions than with other people's. He will thus consider it worthwhile to perform a single intrinsically bad action himself if this prevents multiple similarly bad actions from others. I take the essence of deontology, by contrast, to be the idea that there are side constraints on action that we simply shouldn't violate, not even to prevent more such violations in future. This is what the deontologist really needs to establish. But of course that isn't nearly as intuitive as the axiological claim that (e.g.) value hedonism is false.

[Update: it turns out that agent-neutrality is not quite the feature I'm really after here...]


  1. I wonder if the lightening case is truly naturalized. It seems that we really take sadistic racist pleasure to be something like an evil act in itself, indeed part of what is wrong with, say, lynching (surely not the biggest part, which I'd say is the outcome).

    So even though the "action" in the lightening case is naturalized, part of what makes parallel actions wrong seems to remain. I wonder if my point here is a non-sequitur or not...

  2. Joshua - right, we take vicious emotional responses (e.g. sadistic racist pleasure) to be 'intrinsic bads'. But my thought is that this is merely an axiological refinement that consequentialists can happily take on board.

  3. My main problem with taking vicious emotional responses as intrinsic bads is that that implies that it might be acceptable to inflict large harms on innocent people to prevent a large amount of vicious emotional responses, even if those responses do not motivate vicious actions.

    For instance, suppose a delivery driver is going to deliver a videotape of a black man being killed by lightning to a TV studio, where it will be broadcast to the TVs of billions of racists, who will then take malicious pleasure in it. If that maliciousness is an intrinsic bad, does that mean that, if the number of racist viewers is sufficiently large, that it is okay to kill the delivery driver to stop the tape from being delivered? (Assume the delivery driver is innocent, and has no idea what he is delivering). That seems strange to me.

    I think a better solution is to consider the fulfillment of malicious desires to be neutral. They don't contribute to value, so one black person's desire to live outweighs the desires of a billion racists to lynch him. But they don't contribute to disvalue either, so it's not desirable to inflict large harms on innocent people for no reason other than to thwart malicious desires.

    1. Yes, good point, that does seem the best way to go in that case. (Some positive badness might instead reside in their pre-existing state of being sadistic racists, rather than any particular emotional expresssion of this. Could it be worth harming one to "cure" a population of their viciousness? *shrug*, maybe...)

      The molesting doctor case is trickier. Given the intuition that intentional molestation harms the victim in a way that a mere muscular spasm does not, should we then be willing to kill one innocent in order to (somehow) cause a million would-be molesting doctors to instead merely touch their patients' private parts via an unintentional spasm? That seems weird, but on the other hand, I'm reluctant to abandon the intuition that molestation is worse (more positively bad) for a patient than a physically similar spasm on the part of their doctor. (To make the case neater, perhaps we also need to remove any desire on the doctors' parts to molest their patients. If they're still titillated by the results of their spasm, that maybe isn't a huge improvement!)

      What do you think?

    2. I think you're onto something with the idea that much of the badness resides in the state of being sadistic racists. It does seem to me that it might be acceptable to harm people in order to prevent a sadistic racist from being born (even if their birth is, in Parfit's words, a "mere addition"). The creation of such sadistic preferences seems a net bad.

      I'm not sure about curing them. If being a vicious racist is a deeply ingrained part of their personality that might be equivalent to killing them and replacing them with a different person. I'm against creating such people in the first place, but if we do screw up and create them it might still be bad for them to die. I consider the fulfillment of their sadistic preferences to be of zero value, but fulfilling other preferences they have, like not dying, preserving their personal identity, and not being in pain, are presumably still valuable, even if in a better world that person would have never existed.

      To me replacing the molesting doctor with the spasm doctor does seem a little better. I know if I found out later that a doctor touched me by accident I wouldn't be angry at all, but if they did it on purpose I'd be furious. If the doctor remains titillated by the results of the spasm I think that would still be better, if the doctor apologized to me for having the spasm, and then guiltily admitted that they enjoyed it, I think I'd forgive them, as long as I knew that they would choose to not have the spasm if they could. As with the racist, I think some of the badness resides in having a doctor who intends to molest people existing in the first place, not just in what they do once they exist (although obviously that contains badness as well).


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.)

Note: only a member of this blog may post a comment.