Intentional Action in Folk Psychology: An Experimental Investigation

Four experiments examined people’s folk-psychological concept of intentional action. The chief question was whether or not evaluative considerations — considerations of good and bad, right and wrong, praise and blame — played any role in that concept. The results indicated that the moral qualities of a behavior strongly influence people’s judgements as to whether or not that behavior should be considered ‘intentional.’ After eliminating a number of alternative explanations, the author concludes that this effect is best explained by the hypothesis that evaluative considerations do play some role in people’s concept of intentional action. Intentional Action in Folk Psychology: An Experimental Investigation People normally draw a distinction between behaviors that are performed intentionally and those that are performed unintentionally. It can hardly be denied that this distinction occupies an important place in folk psychology, but researchers disagree about precisely how the distinction should be understood and what function it serves in people’s lives. Indeed, looking through the existing literature on the concept of intentional action, one can discern two fundamentally divergent viewpoints. On the first of these viewpoints, people’s concept of intentional action is understood as one element in a tacit ‘theory of mind’ (Astington 1999, 2001), which is then understood as something like a scientific theory of human behavior (Gopnik & Meltzoff 1997; Gopnik & Wellman 1992). The basic intuition here is that, by classifying behaviors as intentional or unintentional, people are making a distinction that helps them to predict and explain behavior. This predictive and explanatory role is then held to be the key to understanding the nature of the concept itself. Of course, adherents of this viewpoint do not deny that the concept of intentional action is also used in various other kinds of reasoning (e.g., in reasoning about moral praise and moral blame), but these other uses are regarded as parasitic or secondary, not fundamental to the nature of the concept as such (Mele & Sverdlik 1996; cf. Churchland 1981, 1991). I am grateful for comments from Alison Gopnik, Gilbert Harman, Bertram Malle, Michael Morris, Alfred Mele and Daniel Rothschild. According to the second viewpoint, by contrast, people’s concept of intentional action is bound up in a fundamental way with evaluative questions — with questions about good and bad, right and wrong, praise and blame. Adherents of this second viewpoint claim that people’s concept of intentional action can only be correctly understood when we see that it is used not only to predict and explain behaviors but also to determine the moral significance of those behaviors (Bratman 1987). Taking this second viewpoint to a more radical extreme, a number of researchers have argued that people actually use their moral beliefs when they are trying to determine whether or not a given behavior is intentional (Harman 1976; Lowe 1978; Pitcher 1970). Although these two viewpoints have been discussed primarily by philosophers, it seems clear that many of the crucial issues in the debate can be illuminated by systematic psychological experiments, and that is the approach that I will be adopting here. By studying the precise conditions under which people consider behaviors to be ‘intentional,’ I provide support for an empirical hypothesis about people’s concept of intentional action. This hypothesis is then shown to have implications for the broader questions about the function that the concept of intentional action serves in people’s lives. Intentional Action and Skill The hypothesis that I will be defending here involves a substantial a revision in an account that the psychologist Bertram Malle and I put forth a number of years ago (Malle & Knobe 1997a). I therefore begin by briefly reviewing one aspect of that earlier account. Malle and I (1997a) set out to understand people’s concept of intentional action. How do people determine whether a given behavior is ‘intentional’ or ‘unintentional’? We began by simply asking subjects what it meant for a behavior to be ‘intentional.’ Subjects responded by describing the mental states that, they believed, accompany all intentional action. For example: ‘The person meant to act that way and was motivated to do so.’ Or: ‘Someone gave thought to the action beforehand and chose to do it.’ It seemed to us, however, that these mental states alone were not sufficient to make people consider a behavior intentional. Suppose that Bob is trying to win the lottery and actually does win the lottery. One would not normally say that such a person ‘intentionally’ won the lottery. The problem here is that Bob doesn’t have the requisite sort of control over his performance. There is a kind of match between what he was trying to do and what he actually ended up doing, but this match seems to be too much of a coincidence. He doesn’t have any reliable mechanism by which to transform his attempt into the corresponding behavior. Of course, one could imagine that various other mental states were added in to this story. One could suppose that the agent believed that he would win or even that he intended to win. But no matter what mental state you add, people still won’t call the behavior ‘intentional,’ because the issue here has nothing to do with mental states: it has to do with the amount of control that the agent actually has over the outcome. Or, to put it in the terms of our (1997a) paper, it appears that skill is one component of people’s concept of intentional action. In a series of studies, we asked subjects to read vignettes about an agent who tries to perform a behavior and actually does perform that behavior but lacks the skill to perform the behavior reliably. Most subjects said that the agent was trying to perform the behavior but was not performing that behavior intentionally. These results seemed to lend support to our