Moral judgement

3. The justification of moral judgement

Turning to this question, the first point to notice about it is that it is ambiguous: the failure to distinguish two readings of it is a source of much confusion in moral philosophy. It can be used to ask either what makes a judgement-content true, or what warrants someone in making a judgement with that content. These two questions are different, since moral judgements, like judgements of other kinds, can be subject to misleading evidence. Suppose I promised to meet you at ten o’clock but wrote eleven o’clock in my diary by mistake. Then when I look in my diary and judge that I have an obligation to meet you at eleven o’clock there is nothing which makes that judgement true: it is false. But, upon now reading my diary, I am warranted in making that mistaken judgement. We can call the first kind of justification ‘determinative justification’, since it concerns what makes a judgement true. The second we can call ‘warrant for judgement’.

These two kinds of justification, although distinct, seem to be related. When you judge that a person is morally good or an action morally wrong, you are judging that there is something good about the person or something wrong about the action. You might not be able to identify what that is, nor even be interested in doing so; but your judgement is the judgement that there is something that makes the person good or the action wrong. If so, the two kinds of justification are connected, in the following way. You are warranted in making a moral judgement when you are warranted in thinking that there is a determinative justification for it.

This can seem to invite the following line of reasoning. When an action is judged to be right or wrong, it is part of the content of that judgement that there is a reason for the action judged to be right or against the action judged to be wrong. But reasons are general: if a feature of one action is a reason against performing it, it must be a reason against performing any other action with this same feature. So reasons can be stated as general principles. Therefore, our moral judgements about particular cases are warranted to the extent that we are warranted in accepting a set of general principles that support them. The job of a moral theory is to produce a coherent set of principles that will support our moral judgements. Where does the warrant for accepting the set of principles come from? It comes from the coherence of those principles with each other and with our moral judgements. We thus arrive at a ‘reflective equilibrium’ account of the warrant for accepting a mutually supporting set of judgements and underlying theoretical principles.

This line of thought, however, should seem very hard to accept. It carries the preposterous implication that only moral theorists make warranted moral judgements. Slightly less preposterously, it might be suggested that moral theorists are directly warranted in their moral judgements, and everyone else can be derivatively warranted in making theirs – just as we can be derivatively warranted in believing the conclusions drawn by scientists whose theories we do not understand. But that seems wrong also. Whether morally wise nonphilosophers are warranted in judging as they do surely does not depend on their having (dubious) beliefs about the success of philosophers’ moral theories.

If the line of reasoning just described leads to an implausible conclusion, where does it go wrong? It makes two contestable assumptions. It assumes that in order to be warranted in judging that an action is supported by reasons, I must be warranted in accepting an account of what those reasons are. But why accept that? Someone skilled in face-recognition need not be skilled in describing the similarities between faces. Likewise, why not accept that a morally wise person’s skill in being responsive to moral reasons does not entail a skill in articulating what those reasons are? The second is that what makes a judgement correct must be expressible as a general principle. We can turn to the issues this raises next.