Moral judgement

2. Empirical studies of moral judgement

‘But surely it is an empirical question what kind of psychological state is moral judgement?’ Cognitivism and noncognitivism, as rival answers to that question, had better be rival claims about the way the world really is. If so, the question might seem to be one for psychologists and neuroscientists, not philosophers.

There has been a large amount of recent empirical investigation of moral judgement (see Moral psychology, empirical work in, §3). A central theme of this work has been the strong connections between moral judgement-states and emotions. Such connections seem to run in both directions: hypnotizing people to experience disgust upon hearing a nonevaluative word can increase their propensity to judge that actions described using that word are wrong; brain areas associated with emotion are routinely activated when people make moral judgements. Such findings might seem to favour a noncognitivist view of moral judgement-states; more strongly, it might seem to be the only kind of evidence that could resolve the debate between cognitivism and noncognitivism.

However, to leap from these experimental findings to noncognitivism overlooks two sets of questions. First, it remains unclear just how we should describe the relationship between moral judgement-states and emotional states. Amongst the possibilities that remain open are that certain emotional states typically cause moral judgements, are caused by them, or both; that certain emotional states are often partly constituted by moral judgements or partly constitute them; and that moral judgements are or produce dispositions to certain emotional states. One of the few views that seems to be clearly ruled out is simply identifying the two. (The everyday observation that there can be dispassionate moral judgements is unsurprisingly borne out by fMRI (functional magnetic resonance imaging) studies.) Further empirical work will be helpful in narrowing down the plausible options here. For example, to check whether the link between emotional and moral judgement states is causal, it will help to investigate its disruption, examining whether the causally antecedent state can occur without its normal effect. To check whether the relation is constitutive, it will help to investigate whether removing the constituent state reliably removes the constituted state.

Beyond this lies a second set of questions concerning the nature of emotional states themselves: are these states partly cognitive, or not? Just as an emotional state of fear might be thought to require as a constituent the judgement that I am in danger, so too an emotional state of indignation or guilt might be thought to include as one constituent a moral judgement that I have been wronged or have done wrong. The judgement that I am in danger seems to be a belief; what about the judgement that I have been wronged?

As this suggests, the debate about how to classify states of moral judgement will not be resolved simply by amassing further experimental data. Although the debate cannot sensibly proceed in ignorance of such data, they are not self-interpreting. The items of data about moral judgement most difficult to interpret remain the two mentioned earlier: objectivity-presupposition and motivational connection.

Arguably, where recent empirical work has greater significance for philosophical discussion of moral judgement is in its bearing not on the question of how to classify judgement-states but on the importance to attach to them. How do we know when to take seriously our propensity to make a given moral judgement? Recent work in social psychology suggests that moral evaluation standardly recruits psychological ‘mechanisms’ that are implicated in manifestly unreliable forms of judgement: it is systematically directed by relatedness motives (evolutionarily influential incentives to side with allies), coherence motives (to preserve one’s view of oneself and the world), framing effects, and mechanisms of bias (which focus on citing confirming evidence). Reasoning about our moral judgements often serves the purpose of post hoc rationalization. If so, what kind of further reflective scrutiny do our judgements need to meet in order for it to make sense to take them seriously? When is a moral judgement justified?