Explaining Away Responsibility

Thanks to Neal for posting Explaining away responsibility: Effects of scientific explanation on perceived culpability", by John Monterosso, Edward B. Royzman, and Barry Schwartz.  It’s quite relevant to the “Is Incompatibilism Intuitive?” debate.  Here’s the abstract:

College students and suburban residents completed questionnaires designed to examine the tendency of scientific explanations of undesirable behaviors to mitigate perceived culpability. In vignettes relating behaviors to an explanatory antecedent, we manipulated the uniformity of the behavior given the antecedent; the responsiveness of the behavior to deterrence; and the explanatory antecedent-type offered—physiological (e.g., a chemical imbalance) or experiential (e.g., abusive parents). Physiological explanations had a greater tendency to exonerate actors than did experiential explanations. The effects of uniformity and deterrence were smaller, and the latter only had a significant effect on judgment when physiological rather than experiential antecedents were specified. Physiologically explained behavior was more likely to be characterized as “automatic,” and willpower and character were less likely to be cited as relevant to the behavior. Physiological explanations of undesirable behavior may mitigate blame by inviting non-teleological causal attributions.

The results seem to support compatibilism as the intuitive view, since the subjects exonerated the agents only when the actions were caused in a certain way.  Of course, that certain way was “physiologically” which arguably causes every act, whether the subjects know it or not.   Indeed, the follow-up interviews suggest that the subjects are basing their assignments of responsibility on a dualist view of human agency.  (Eddy, this might be right up your alley.)   Those of us who aren’t dualists, then, could argue that the assignments of blame and culpability are based on an error, namely the error of thinking that some acts are not the result of physiological processes.   

In other words, I think both compatibilists and the incompatibilists can use these results to their advantage.

Comments

[Eddy Nahmias]

Very cool article and experiments. I have a lot to say about this, but can't now. I hope people will read the article and start the discussion.

[Neil]

I disagree with Tamler's reading of the experiments. I think that to the extent we can read them as addressing the compatibilism issue, they tend to support libertarianism. First (though the paper did not address this issue) recall that we have evidence that people think we live in an indeterminist universe (Nichols and Knobe asked their subjects whether we live in a determinist or indeterminist universe; over 90% across all conditions and experiments answer ‘indeterminist’). If this is right, when we ask people to judge responsibility in this universe, they are assuming indeterminism. Indeed, they are probably assuming libertarianism. Monterosso et al. asked their subjects who were inclined to excuse agents in the physiological condition why they were not responsible. Though they did not give figures for this, they say that their subjects tended to give dualist answers: her brain was controlling her, even though she didn’t want it to. Moreover, only in the physiological condition did the extent to which the condition was predictive of the behavior have an independent impact upon attributions of responsibility, which seems to indicate that when the cause is merely environmental, subjects think that agents can exert willpower to overcome the effects of their conditioning. So my reading of what is going on here is that subjects are thinking that when the condition is physiological, it can control behavior independent of the wants of the agents, whereas when it is experiential, agents can control their behavior and the influences upon it. Libertarian free will is effective only against certain kinds of influences.

But I’m not sure how much weight to put on these findings, for this particular debate. The papers which address the compatibilism/incompatibilism question ask subjects explicitly to consider responsibility/free will in universes that are determinist. Of course, as we know, there is a debate about what subjects actually think in these cases. Monterosso et al., however, asked subjects to assess responsibility in one and the same universe, with whatever laws of nature are in place here. One might be a compatibilist (for instance) and want to distinguish between these cases (as Tamler suggests). So by itself I think the experiment is orthogonal to this debate.

[Tamler Sommers]

I agree with Neil, which means either that I disagree with myself or that I wasn't clear enough in my original post.

I do think that these results support, though not definitively, the view that people are natural libertarians. My point was that at first glance you might think that the results support compatibilism as the intuitive view, since determinism by itself didn't rule out culpability. It was only if an action was determined in a certain way that the subjects exonerated the agent. That is the compatibilist position, as I understand it. But if you look a little deeper (especially at the interviews), you find that, as Neil says, the subjects are assuming indeterminism in the cases of experiential explanation. And that allows them to assign blame to the agent. If I had to guess, I'd say they thought that the childhood abuse inclined but not necessitated criminal behavior. (And so there was room for free will and moral responsibility.) Whereas the in the physiological explanation cases, the behavior was thought to be necessitated.

But of course that doesn't mean they're right, especially if they are assuming a dualist view of human agency. So--surprise surprise--I see these results ultimately as supporting an error theory of moral responsibility.

[Neil]

Part of the problem was that I failed to read you properly. My fault. I don't see the results as supporting an error-theory. Like you, I see in the results evidence for one my favorite theses: place very little weight on folk intuitions when doing moral philosophy. I'm coming to the view that commitment to wide reflective equilibrium entails what Weinberg (on the experimental philosophy blog) calls the Platonic approach to metaphilosophy, on which experts have better intuitions than the folk. Why? Well, for a couple of reasons: one, because we're after *wide* RE, and the folk have strange ideas (they believe the universe is indeterministic, that evolution is just a theory, that aliens probe their asses and that Iraq was responsible for 9/11). Two, because I think if we probe the extension of folk moral concepts we will find that they're committed to incompatible intuitions case-by-case. And, three, because I don't want to buy the kind of relativization that the results of, say, Haidt and Weinberg, Stich and Nichols seems to suggest.