Blog Coordinator

Knobe's X-Phi Page

X-Phi Grad Programs

« Asymmetry of Intentionality Judgments in Causal-Deviance Scenarios | Main | Where is the Pain? »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Adam Bear

Interesting hypothesis! I guess I have two questions/comments. First, it seems to me that your theory might actually predict that we'd get *greater* differences between the virtuous/reprehensible cases and the neutral cases if the factors you mention play a central role in judgments of what's intentional. For instance, in the train case that I mention in my post, Janet would presumably have a very strong pro-attitude to save the children in the virtuous condition (Meaning 1), but her tripping on the lever in the neutral condition was completely accidental (Meaning 3). And though I realize that you say "only if" in your definitions and that other factors might therefore be relevant or necessary for judgments of intentionality, I get the sense that the pro-attitude and the controllability are playing greater roles in your theory than what our data suggest (and similarly for Janet's lack of reluctance in the reprehensible condition).

The other thing, which may've been hard to glean from my post, is that not only our virtuous cases, but also all of our neutral cases involved an agent desiring to bring about an outcome that might in some sense be expected, at least for instrumental reasons (e.g., Pierre is a chef at a fancy restaurant who wants to please a food critic). Here, I'm not sure if your theory values instrumentally normative reasons as highly as morally normative reasons (that might be what you mean by "statistical normative expectations"), so maybe you would predict that the neutral and virtuous cases both rely on Meaning 1, but to a lesser degree in the former case. Nonetheless, in the train scenario, we did find that subjects rate the *degree* of desire to be the same in the neutral and virtuous conditions. This is, of course, not a measure of *expected* desire, nor does it get at the instrumental/moral distinction that is perhaps relevant. In any case, maybe Fiery and I will someday get to test your hypothesis!

Florian Cova

Good points!

As to your first point, I do not think that all participants use Meaning 1 for the virtuous case and that all participants use Meaning 3 for the neutral case - an hypothesis that seems too unrealistic. My prediction was that the presence of a good outcome (an outcome we expect the agent to desire) will trigger more participants to use Meaning 1, and thus will decrease the number of participants using Meaning 3. Thus, I predict a decrease in intentionality ratings between the two cases but I'm not commited to a given difference size (I wish I could be able to predict the difference size, but I have to say I'm not).

To the second point: I was not saying that you should get different degrees of desires between virtuous and neutral scenarios (in fact, I didn't expect you to). What I was saying is that in virtuous case, the moral goodness of the outcome is so salient that expectations about the agents'desire are preferentially triggered, and Meaning 1 preferentially elicited. In neutral scenarios, the fact that the agent desires the outcome might be a less salient feature of the scenario.

That might seem a little abstract, but here is a more concrete example (discussed in our paper) : Sripada has a version of Knobe's "hitting-the-bull's-eye case" in which he insists on the fact that the agent desires the outcome. In this version, most participants judge the outcome intentional, even if it is a clear case of causal deviation. Contrary to Sripada, I don't think that this shift is due to the fact that people suddenly ascribe more desires to the agent (it was already said in the original scenario that the agent desired to hit the bull's-eye), but to the fact that Sripada's case dwells at length on the fact that the agent desires the outcome, making this feature of the scenario more salient (and thus preferentially triggering Meaning 1, etc.).

Anyway, there might be a prediction you can already check (if you have some time): do you find greater correlation between desires rating and intentionally ratings for the virtuous and reprehensible cases than for the neutral case?

Adam Bear

At this point we have desire ratings only for the train case, and it was from old data that's not referenced in my post, so it's hard to get a good picture of things. Ultimately, I think we'd have to run some more subjects to get a real sense of the interaction between desire and intentionality judgments, which Fiery and I will hopefully get to do at some point in the near future. Right now, we're working on something completely different, but your and others' comments have encouraged me to pursue this work further.

The comments to this entry are closed.

3QD Prize 2012: Wesley Buckwalter