Williams on Moral Luck (Part 4)

In this post, wish to examine Williams' account of retrospective justification as its own special type of justification, independent of the type of justification implicated in justified action (as defined in my previous post).

Williams does not describe, at length and in a systematic way, what makes retrospective justification different form other types of justification. However, he does identify the Gauguin case as being special in that his decision involves a life-defining project. Williams unpacks this idea by noting that the evaluative standpoint Gauguin occupies once he has become an artist is quite different from that which he occupied when he first made the decision to travel to Tahiti. This suggests the following possibility for making sense of what is special about retrospective justification. Let us suppose that Gauguin was a stockbroker before he became an artist. I will refer to the perspective from which Gauguin would assess his own life at the time he decided to travel to Tahiti as that of Gauguin the stockbroker. The perspective of Gauguin the stockbroker contrasts with that of Gauguin after he has become a famous artist. I will refer to this second perspective as that of Gauguin the artist.

Following Williams, we can suppose that Gauguin the artist feels some regret about the decision of Gauguin the stockbroker to leave his family. Nevertheless, Gauguin the artist also recognises that if Gauguin the stockbroker had not acted as he did, then Gauguin the artist would not exist. Thus, although Gauguin the artist regrets the immoral actions of Gauguin the stockbroker, he is nevertheless grateful that Gauguin the stockbroker acted as he did.

By Williams’ lights, the gratitude of Gauguin the artist points to a special type of justification, which is both success-dependent and fundamentally retrospective. The fact that Gauguin the artist feels grateful is taken to show that he thinks the action was the right course or (in some meaningful sense) justified. However, since the perspective from which Gauguin the stockbroker’s decision is justified only comes to exist many years after the decision was made, there was no way for the justification in question to exist contemporaneously with the decision. In brief, the justification in question must be retrospective since it is located in a perspective that did not exist at the time the relevant action was performed. This presents us with an intelligible difference between the type of justification at play in Williams’ example, and the type of justification we earlier identified with justified action.

But how does this relate to Williams’ claim that if we hold to (M1) we are forced to give up (M2)? Williams observes that even those of us who subscribe to the moral requirement to take care of one’s family may feel grateful for Gauguin’s achievements. However, since his achievements came at the cost of fulfilling his moral obligations, our gratitude is taken to be evidence that we sometimes rank non-moral considerations above moral ones. This conclusion flies in the face of (M2); the claim that moral considerations are supreme. Moreover, we are forced to accept it so long as we hold to (M1), the claim that morality is luck-free, since this precludes the possibility of retrospective moral justification.

In sum, Williams maintains that there will always be cases in which retrospective justification takes priority over all others, a fact that our gratitude for Gauguin’s achievements illustrates. Since moral judgements cannot be retrospective (a la M1), then on such occasions they will always take second place (contra M2).

Williams on Moral Luck (Part 3)

The lesson to be learned from the failure of the uncharitable reading, limned in the previous post, may be put as follows: If the distinction between right action and justified action is to avoid begging the question against Williams, then our account of right action must make room for the concept of retrospective justification. The goal of the charitable reading, then, is to characterise right action in a way that is in keeping with what Williams explicitly says about retrospective justification, and then to use this common ground as a non-question-begging basis for criticising Williams’ argument. As we already noted, a non-question-begging notion of right action cannot be equated with mere successful outcome. This leaves us with three other possibilities: (1) right action just is justified action, (2) right action is a combination of right action and success, and (3) right action is its own special type of success-dependent justification. I will consider each of these possibilities in turn.

According to (1), Williams’ notion of retrospective justification simply equates right action with justified action. However, if we define right action as justified action, then there is nothing (in principle) to prevent Gauguin from being justified at the time in which he makes the decision to leave his family. Since justified action does not require success—only that the agent be justified at the time of making the decision—there is no need to wait until the outcome of his decision is known. This account of right action flies in the face of Williams’ insistence that the justification Gauguin has is only available if he is successful. Consequently, (1) fails to provide a common ground from which we may criticise Williams’ argument.

According to (2), we may define right action as the conjunction of justified action (i.e., action with adequate reasons) and successful action. On this picture, there may be cases of justified action that fail to constitute right action and there may be cases of successful action that fall short of right action. Since this definition of right action implicates success, it would (when applied to the Gauguin case) take on the retrospective nature of Williams’ justification. However, as we already noted, Williams stipulates that Gauguin’s justification is purely retrospective, which means he did not have adequate reasons for his decision at the time he made it. As such, Gauguin’s decision fails to constitute a justified action and therefore, according to (2), fails to constitute an instance of right action. Consequently, (2) also fails to provide a definition of right action that can accommodate Williams’ notion of retrospective justification.

The final proposal is that right action represents its own special type of justification, independent of justified action. If we define right action along these lines, then it is clearly able to accommodate Williams’ notion of retrospective justification. But what could this special type of justification possibly amount to? This will be the question of the next post on this topic.

The 81st Philosopher's Carnival

is here.