Tuesday, November 23, 2010

XPhi

Experimental philosophy (XPhi) has received some press in the last few years. I am skeptical of its purported powers to offer insight on philosophical questions so far. Maybe in the future there will be novel and interesting ways that do offer insight into the philosophical questions but I don't see the way it is being used now doing that.

XPhi uses surveys to ask a "representative sample" ("the Folk") of the population what their philosophical intuitions are in the hypothetical thought experiments used by philosophers to generate intuitions used in philosophical debates. If the Folk's intuitions tend to lean one way, it is interpreted by XPhi proponents as some (but not conclusive) evidence that that way is the right way. In other words, XPhi philosophers are sometimes suspicious of their own intuitions and will give certain weight to the intuitions of the Folk. And there is evidence from XPhi that the intuitions of philosophers tend to run differently than the intuitions of the Folk on many issues.

An example of one instance where XPhi experiments has used to give credence for one perspective over another in contemporary debate is the debate over what are lies. Don Fallis has used experiments showing that the Folk consider bald-faced lies to be real lies. This is counter to many philosopher's claims that bald-faced lies are not lies because they don't involve an element of deception as is required with genuine lying. Experiments show that the Folk considers bald-faced lies to be lies contrary to the classical definition of lies accepted by most philosophers. Fallis has adduced this as evidence that the classical definition must thus be false. My reasons that XPhi as it currently is used is not that interesting and relevant to philosophical questions are the following:

1. Many of the samples used in XPhi experiments are not representative samples (they are the undergrad students of XPhi philosophers who are taking phil 101 type classes).

2. Much like individuals often get similar but different concepts confused, it's not that much of a stretch that sometimes, a majority of the population will similarly conflate two different but (superficially) similar concepts.

3. There may be very good reasons why philosophers have the intuitions they do. They may do so because they realize a subtle difference that the Folk do not that must be upheld in any definition. There are good reasons for including a deception criteria in lying for example as I have pointed out before and also see here.

4. It's been known for a long time by cultural psychologists and anthropologists that surveys, intra and especially inter cultural ones, can vary widely in the responses given simply by changing subtle wordings in the questions asked or the context in which they are given. The answers sometimes do not reflect true differences in cognitive thought processes and profiles but rather in "surface" linguistic differences in how words and scenarios are interpreted.

Perhaps one day XPhi will be used in a way that is insightful but the experiments so far do not seem convincing to me in settling any philosophical issues.