Blog Coordinator

Knobe's X-Phi Page

X-Phi Grad Programs

« Center for Cognition and Neuroethics CFP: Free Will | Main | Moral Dilemmas in Virtual Reality »



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

jonathan weinberg

I don't have access to the piece, but I take it that these sweeping generalizations are offered with little to no specific citations in the x-phi literature? As you note, it's hard to see how (1) survives even casual contact with the actual, like, published papers here.

I would add a number of points to what you say about (3) (but not in any way to disagree with what you do say there!): (i) Much of the best work being done in xphi is _already_ collaborative across philosophy/psychology disciplinary lines, though not with the division of labor that Hansson would hope for, in that the philosophers are usually doing their own fair share of the empirical side of things. (ii) We philosophers have a really great tradition in our discipline on the whole of acquiring whatever skills we may need to address the questions that interest us. If you want to do philosophy of physics, or game theory, or philosophy of music, or ancient philosophy, then you're going to have to learn a lot of physics, or math, or music theory, or Greek -- just for starters. I think one of the best areas of this is in philosophy of language, where so many of our best philosophers of language are fully paid-up masters of linguistics and its methods. (Two of my personal favorites here in this regard are Jason Stanley and Kent Johnson.) I see xphi as just one more instance of our profession's glorious methodological eclecticism.

And regarding (2), that's a move one hears with some frequency, and it's always struck me as simply false to existing philosophical practice, at least in the areas that I am most familiar with. I mean, what you're precisely _not_ supposed to do if you're a reliabilist confronting the clairvoyance cases, is say, "well, my initial judgment is that Norman lacks knowledge, but on reflection, my reliabilism tells me to override that judgment, so, no worries for me there." Right?

There are also plenty of instances in the literature where the author is quite happy to be appealing to something allegedly pre-theoretical, or as part of our ordinary linguistic or conceptual practices.

Shen-yi Liao

Thank you for writing up responses to these, frankly, tired objections. To be honest, I find it incredible that it's 2014 and the "experimental philosophy is just surveys" objection is still getting trotted out. I remember writing a long time ago about this... (hey, it was 2010)!

(The only recent publication that I can think of as falling prey to objection (1) is Richard Kamber's "Experimental Philosophy of Art" (2011). But I really think this is an exception.)

Eddy Nahmias

Without having read the piece, it sounds outdated and less interesting than the critiques raised by people like Kauppinen. I would, however, like to plug survey methodologies (including qualitative data and scales, which are not really experimental). It is useful to find out what people think about philosophical (including moral) issues, which can be studied in ways that go beyond testing how their thinking (or behavior) changes in response to manipulations.

Taylor Murphy

I agree Eddy. The experimental philosophy people in the 50s-60s emphasized the importance of surveys for discovery. The "method of revelation" wasn't a good enough way of collecting the semantic data for one's analysis, they said.

Similarly, Alan Love has an article in Analysis 2013 that points out that along side hypothesis testing in science, there is also experimental exploration, which isn't done for the sake of testing a specific prior hypothesis. It is hard to find a scientific paper nowadays that tests a specific hypothesis instead of exploring an unknown or uncertain area in the field.

Thomas Nadelhoffer

I,too, agree with Eddy. For instance, if one were an applied philosopher interested in legal decision-making and the criminal law, then the survey data on people's beliefs about free will, intentional action, punishment, and responsibility would quite obviously be relevant to one's philosophical project--especially if that project were revisionist with respect to current practices. This is just another way of highlighting why criticizing experimental philosophy in the abstract is such a problem. As Eddy and I pointed out years ago in our piece "The Past and Future of Experimental Philosophy," the field is not monolithic--either in our goals our in our methodologies. Whether survey data is relevant will depend on what one is interested in exploring. But this is a well worn observation at this point (even though some seem to have missed the memo).

Shen-yi Liao

Thomas and Eddy: Could you please point to some of the survey research you have in mind?

I might have missed the memo, or am misunderstanding the terminology. I think there are lots of non-experimental research that are useful -- e.g. ones that involve correlation analysis of SEM -- but they are not really sample surveys in my mind. When I hear "surveys", I think of opinion polls with minimal statistical analysis (i.e. nothing more than a percentage breakdown). And I just can't think of research like that in xphi.

(Scale construction is an interesting case. However, I think insofar as it involve statistical tests for validity, it's quite unlike sample survey too.)

Basically, I think the big distinction should be between works that involve modeling and statistical inferences, and works that don't -- instead of experimental / non-experimental. I can't think of potential philosophical interests for works that do not involve modeling or statistical inferences, but I'd be happy to be proven wrong.

Thomas Nadelhoffer


I take it much of the early work on folk intuitions about free will and moral responsibility was "merely" survey work. For instance, in the work I did with Nahmias and Co., we were giving people deterministic scenarios to see whether people thought that the agents in these scenarios were free and responsible. These studies were, at heart, what you're calling "sample surveys." I think we did a pretty good job motivating the relevance of this type of research in our PPR piece, "Is Incompatibilism Intuitive?" (2006). Perhaps you disagree. If so, I'd be happy to hear why.

Shen-yi Liao

Thanks! I'll read the paper. And my preconceptions will be proven misguided! :)

The comments to this entry are closed.

3QD Prize 2012: Wesley Buckwalter