Note: When I discuss good people and good actions here I am using the term in its ethical sense, to refer to ethically upstanding people and ethically justified actions, not to connote people who are living the good life.
Let us say that we have an ethical theory that allows us to determine the goodness of any action in any circumstances. In other words the typical ethical theory, which focuses on determining which actions we should do and which actions we shouldn’t. Given such a theory how should we judge the goodness of people?
One way would simply be to calculate the total goodness of their actions, subtract the total badness of their actions, and take the result as the goodness of the person. (Assuming of course that our ethical theory at least allows us to compare magnitudes of right and wrong.) Such a theory, however, runs into problems when dealing with one kind of moral luck. Specifically some people simply have more chances to do the right thing than others, or perhaps by chance they find themselves in situations where their actions can do more good. Should we allow that by luck alone some people are better than others, better even than an exact copy of themselves who has lived through slightly different situations? There are very few, I think, who would agree.
But there is an even more fundamental problem with this method of determining the goodness of an individual, which rules it out as a possibility no matter what our intuition is. Specifically it rests on the idea that there is some identity over time, such that the person we are judging now is the same person who previously did those acts. But really all we can say is that there is a continuous chain of people, similar in psychological properties, that leads from those past events to the person we are currently judging. Certainly the person’s past actions will matter to us if we are attempting to judge their goodness, since their actions are basically the only evidence we have about them. However the person we are currently judging is only very similar to those past people, not identical, and so those past actions can only serve as indirect evidence about how they are likely to act.
And these considerations about identity over time rule out another attractive possibility, that somehow we could take the total goodness of a person’s actions and then normalize it with respect to the situations they found themselves in, to give us a basis for judgment that is not affected by the king of moral luck.
But we can consider what the person who exists now is disposed to do in any given situation. Specifically we can judge them as a good or bad person based on the average goodness of the choices they would make were they to find themselves in any given situation. This solution has the advantage of being independent of what that person has done before or what kind of person they will become. Of course this measure doesn’t make any attempt to sum up the goodness of person over their entire life, it is simply a measure of how good they are now. But if we really wanted to, ignoring the fact that to do so may not make much sense, there is no fundamental problem with simply taking the sum of these measures, made throughout the person’s life, as a measure of their total goodness (actually, we would want to integrate them).
But it might seem that even this measure can be influenced by chance. For example, a person’s dispositions towards various kinds of actions may change as a result of some even they are caught up in. But the fact that they are caught up in this event is something outside of their control, they could just have easily escaped it and remained as they were before. This is a kind of second-order moral luck. Is a measure of a person’s goodness that can be influenced by this kind of moral luck equally ridiculous? I don’t think so, because this moral luck does not affect the person’s actions, but rather who the person is, and at a certain point we must accept that even though a person might have become someone else whether they are good or bad depends on who they are, not how they got to that point. If we didn’t we would have no basis for judging people to be either good or bad, because a person’s dispositions are a product, ultimately, of what they have experienced, plus genetic factors, neither of which is under the person’s direct control. (see also this post on responsibility)