OrangePhilosophy

Another example is the famous race riot one in the ethics literature. I don't remember offhand who first produced it, but I've found it in most introductory ethics texts. To appease the mob, who will begin lyching people and rioting, the sheriff decides to save a lot of effort and just put an innocent man to death. Since all they know is that the rapist in question was black, they just find a black man and kill him. Utilitarians should say that this is the right thing to do, because it only costs one many his life and prevents much worse harm. Yet it just seems unjust and a violation of this man's rights. People standardly give such arguments against utilitarianism, and I've seen two sorts of responses. One is simply to bite the bullet and say that we really just have poor intuitions on these cases, and it really is the right thing to do to use that one person as a means to the greater end. Some utilitarians have done this. The other response is to revise the whole theory. It's not about what acts in particular lead to the best consequences in terms of happiness and unhappiness (or substitute your own intrinsic goods if you want to expand this to a more general consequentialism). The original theory put it that way, but what we really ought to do is look at which types of act will tend to produce good results. Killing innocent people tends to produce bad results, so we shouldn't do it. Torturing innocent people for the sake of others' happiness tends to produce bad results so we shouldn't do it. Then you come up with a whole list of rules and pick the ones that maximize happiness. It gets you out of the problems I've been talking about, because you now no longer get the counterintuitive result. I have two problems with this. First, it seems to do as much damage as it saves. Second, it isn't at all clear what this is supposed to look like, because rules turn out to operate on a continuum from very specific rules to more general rules. It causes as much damage at is saves for the very reasons that utilitarianism is supposed to do better than theories like Kant's when Kant's absolutism seems wrongheaded. Lying, for instance, seems generally wrong when it comes to hiding something you did wrong, deceiving someone for personal gain, or trying to make someone else look worse than they really are. On the other hand, it seems pretty clear to most people that lying is ok when the person being deceived is trying to harm someone, and a lie is the only way to protect the potential victim. Lying to Nazis hunting Jews during the Holocaust is a common example along these lines. Similar examples for breaking promises show up in all the introductory texts. If you promised someone to meet them at a certain time, but it turns out someone will die if you don't break that promise, you don't have any moral obligation to keep it. Utilitarians get the right result on these cases, mostly because they focus in on the act rather than the rule. Kant has an absolute rule against lying and breaking promises, but if you modify the view to make it not about the absolute rule but about evaluating whether a specific action will be universalizable, then you might well end up with a non-absolutist neo-Kantian who says lying or breaking promises is ok in such circumstances as the ones I gave. This comes from having an act-based Kantianism rather the rule-based one Kant himself developed. So it turns out that act-based ethical theories do better with some cases, even as rule-based ones do better with others. How will utilitarianism avoid the problems facing theories like Kant's if they make their rules so general as to include rules against lying and promise-keeping on the grounds that those actions tend to produce unhappiness? Similarly, how does the neo-Kantian avoid the problems with consequentialism if she drops the absolutism and then says to universalize particular actions to see if you would want everyone doing that action in only that kind of circumstance? The "means to end" restriction seems to prevent doing this in the cases of the racist mob and the torturing case, because both involve using someone as a means to an end. The problem seems to me to be that once you go to act-deontology you have to drop the means-to-end prohibition, since that's now got to be applied on a case-by-case basis. The only way out I can think of is recognize that rules can be very specific or more general. If you can resist making your rules too general, they avoid the absolutism of Kant. On the other hand, if you can resist making your rules too specific, they avoid the extremes of act-utilitarianism. Yet this seems to raise so many problems that I'm not even sure where to begin figuring out how to resolve them. Any time you get a difference of degree that makes a huge difference in a theory from extreme to extreme, just saying that we need a via media doesn't solve anything. Maybe the realm of normative ethics has done this and I'm just unfamiliar with the literature on this point, but we need to see how the theory will figure out where to settle on that continuum between very specific and very general rules. I'm skeptical whether there's one place that will solve all the problems of the two extremes. It may well be that settling somewhere will solve some of the problems but not others. I'd have to see this worked out in detail to have those doubts assuaged. Another problem is that I can't even think of how you would explicate such a view. How do you rigorously define how specific your rules need to be? Is there such a way? Putting in hard and fast rules about how general a moral statement needs to be just seems to me to move in the wrong direction in terms of capturing our moral intuitions, which is what many seem to see the process of ethical theorizing. Maybe some will insist that you can't get such a theory, but this seems to me to be giving up the very idea of having an ethical theory. If you're doing that as a way to move toward a virtue conception of ethics, that's fine, as long as you are admitting that your move will bring you into ground that will not involve a complete theory about why certain actions are right and wrong. Some virtue ethicists embrace this lack of theory, but others really worry about it (or at least should given that they think they have a theory). As far as I know, Linda Zagzebski is one of the few contemporary virtue ethicists to try to develop of genuine virtue theory that covers such things, but I haven't gotten far enough into her book to know if she succeeds. These are just thoughts I've had as a metaphysician having to teach this stuff without having had graduate courses that deal with it systematically, so I'd welcome comments from anyone more well-versed with the literature on how people who really do this stuff try to avoid the problems I've raised. I'd be very surprised if these problems have never come up in the literature. Either way, I welcome suggestions on how consequentialists or Kantians can still make use of the act-rule distinction to steer clear of counterexamples. I'm not arguing that the distinction isn't important, just that it doesn't seem to solve the problems it's supposed to solve.