Should Ethics Tell Us What to Do?

After posting a comment on Colin McGinn's blog, I was seized by the sudden urge to develop the thought more here. A commenter there (Hugh Millar), criticizing Kant, brought up the example of Gaugin, who discovered that he had to face a conflict between pursuing his art and sticking with his family. Millar's claim is that neither pursuing one's art at the cost of losing one's family, nor holding on to one's family at the cost of abandoning one's art, is a reasonably universalizable maxim. This yields the following criticism:

"Of what conceivable good is an ethic which fails to address such major moral problems?"

Before delving into the question, there is one possible point of confusion to note: It is indeed right that neither maxim is universalizable. Of course Kant did not think that any non-universalizable maxim is immoral, and I think this is a clear case where the moral law simply does not clearly say how one should act. The question, then, is whether the fact that Kant's moral philosophy cannot give a definitive answer to this dilemma should count as a serious strike against Kantian ethics.

That is, should we fault Kant for not telling us how to solve a conflict between the development of one's art and commitment to one's family? This I doubt. It seems dubious, to say the least, to suggest that any moral theory can--or should aspire to--tell individuals exactly how to act in any particular situation. This is so for a number of reasons, chief among which is that such conflicts can arise for very different individuals in very different circumstances. These context-specific details can be important. For example, if the art is terrible and the artist has zero potential, then the conflict arises only because of the artist's ego; a competent person should, seeing this, be able to see that there is a fairly simple solution. Conversely, imagine that the artist's family is a bunch of parasites, feeling no affection for the artist but merely hanging on in hopes of following her to the top. There, too, there is reasonably simple solution (though not as simple as in the previous case), which a competent moral judge should be able to see.

The issue of competence is important: moral judgment is unavoidable. Ethics, as Kant tells us, is concerned not with legislating actions, but maxims of actions. Though there is debate over how closely maxims and actions are supposed to be aligned for Kant, it is reasonably clear on his account that different actions can instantiate the same maxim (Anscombe brings this up as a criticism of Kant, as if he'd somehow missed the point). This means that, no matter how clearly the moral law dictates our maxims, there will still be a disjunct between the maxim and the action it demands. Moral judgment is supposed to fill this gap by finding the appropriate action with which to fulfill the maxim.

A moral theory should, ideally, facilitate moral judgment, e.g., by providing values by which to orient oneself, by pointing to common sources of distortion (personal desires, self-deception and rationalization), etc. But these are only guidelines for moral judgment; they cannot and should not tell us exactly what to do in any particular situation. For example, the conflict in question involves such diverse elements as duties to honor legal contracts (like those involved in marriage agreements), duties to perfect oneself, and duties to seek the happiness of others (like one's family members, though perhaps also various art enthusiasts). Recognizing that these values and not, say, the value of satisfying one's desire to paint are the morally relevant features to take into account places limits on the moral judgment involved.

Let me here make a distinction between two kinds of moral judgment. First, there is value-determining moral judgment, i.e., judgment about which values (or ends, or maxims) are morally relevant. Second, there is ad hoc moral judgment, i.e., judgment about how to act in a particular case. What the criticism I am examining here seems to come to, then, is one of whether or not a good moral theory can stick to value-determining moral judgment but shy away from ad hoc moral judgment, or judgment about precisely how to act morally in any given situation. The criticism is supposed to tell us that the answer is no. An ethics that cannot tell us how to resolve such particular problems as Gaugin's dilemma (and an ethic of universal maxims is such an ethics) should be thrown out. A moral theory that leaves ad hoc moral judgment underdetermined is pointless. But I wonder whether the criticism is coherent. If I am getting it right, the criticism is roughly this:

P1. A good moral theory tells us how to resolve concrete moral conflicts.
P2. A moral theory that appeals to universals cannot resolve concrete moral conflicts.
C: Therefore, a good moral theory will not appeal to universals.

The problem is that if we get rid of universals altogether, then we will lose the guidelines by which to judge concrete moral conflicts. Not only that, but we still will not have a moral theory that tells us how to resolve every moral conflict, since without appeal to some universal "ought"s we will have to judge each conflict entirely on its own merits. (Of course we might still appeal to some regional values--commitment to family or commitment to art, for example--but if we can't adjudicate between those values by any more universal criterion, this doesn't get us very far.)

There is an obvious response here: I've painted a false dilemma, since I am begging the question against utilitarianism. I am assuming that without value-determining judgments, we get stuck with a theory of ad hoc judgments that is entirely decisionist in nature. But there is an alternative: we can have a moral theory that, by fixing a universal value (the greatest happiness) provides a criterion for making ad hoc moral judgments in each case. In practice, this likely stands no chance of giving a definite answer to Gaugin's conflict (calculate the hedonic value of family commitment vs commitment to art; now try this again in act-utilitarian terms!), but it seems to at least theoretically solve the problem. In theory, if not in practice, the ad hoc judgment is the outcome of a deliberation that is determined by a theory, not left hanging after theory has done its work.

But how does that work? What determines the judgment? Why would utilitarianism help? Again, unless we really think it is possible to perform a hedonic calculus here, we still don't get a clear moral judgment. That may not seem like much of a problem. After all, the fact that we can be wrong in the judgment we make does not mean that the criteria of making the judgment should be thrown out. There does seem to be an advantage to having criteria that at least in theory lead to a determinate judgment over criteria that in principle cannot lead to such a judgment. But what is the advantage? It seems to be this: the utilitarian, in making an ad hoc judgment, at least knows where to look--to the greatest happiness principle. He understands exactly what end he is trying to maximize. The Kantian, on the other hand, is stuck considering several different values.

The upshot of this is going to be question begging, I suppose. But let's ask what fits our intuitions about moral judgment. Calculation, or mediation between values? Trying to follow a definite criterion, or weighing among different criteria? To make this slightly less question begging: the difference is between two kinds of complexity. For utilitarianism, the complexity is in the calculation. For Kant, the complexity is in the relative weight accorded to morally significant features. Which kind of complexity better satisfies what we might think of as moral judgment?