Experimental Philosophy Defended (Leiter)

A group of leading contributors to and sympathizers with experimental philosophy responded to my invitation to comment on the recent article in Slate and the critique of experimental philosophy by David Velleman (Philosophy, NYU).  Their response follows:

=========================================================

Responding to a recent article in Slate, David Velleman decries what he calls the “newest philosophical fad,” – the body of research sometimes referred to as “experimental philosophy.”   While not all of us embrace the label “experimental philosophy,” all of us admire the sort of work Velleman appears to have in mind, so it may prove useful to correct some mistaken impressions about the field that Velleman’s critique might engender.   

As we read him, Velleman expresses two concerns about experimental philosophy.  The first is that those who work in this area advocate “substituting [empirical findings] for philosophy altogether.”  We have no idea who Velleman has in mind; all of us reject such a draconian substitution, and some of us have explicitly rejected it in print. (See, for example, Doris & Stich, 2005.) Rather, we argue that empirical work of various sorts may complement other forms of philosophical inquiry – a methodological predilection that Velleman himself shares.

Velleman’s second concern is that “unsuspecting readers” may be unaware that “the ‘discoveries’ touted in the piece in Slate are not exactly news to traditional philosophers.”  We believe that Velleman is mistaken.  One of the findings discussed in the Slate story (and illustrated with a rather tasteless cartoon!) was Jonathan Haidt’s discovery that low socioeconomic status (SES) individuals judge harmless actions which evoke disgust to be morally wrong, while high SES individuals do not judge them to be morally wrong.  Another example mentioned in Slate was the finding by Machery et al. that English speakers in Hong Kong and English speakers (of European background) in the USA had dramatically different intuitions about Kripke’s famous Gödel/Schmidt thought experiment.  A third much discussed finding, not alluded to in Slate, is that English speakers of European cultural background and English speakers of East Asian cultural background report very different intuitions about “Gettier-cases” of the sort that brought about a sea change in epistemologists’ thinking in the middle of the 20^th century.  To the best of our knowledge, none of these discoveries were old news to “traditional philosophers”.   

The work done by experimentally inclined philosophers is relevant to numerous philosophical projects.  Velleman himself concedes that “it is useful to know what most people think about intentional action and moral responsibility” because “in philosophizing on these topics, we can’t stray too far from what people think.” Presumably he would say much the same about knowledge and reference. Anyone with such convictions, it seems obvious, should be concerned to understand what the actual folk beliefs are, rather than resting content with philosophers’ speculations on these matters, buttressed by the dubious practice of assuming that one’s own intuitions are shared by the rest of humankind.  If, as Velleman acknowledges, “it is useful to know what most people think” then there is every reason to pursue the sort of empirical work that experimentally inclined philosophers have been engaged in.  In addition to telling us when professional intuition is at odds with folk intuition, experimental work can shed light on the ways in which the judgments made by both philosophers and lay people may be biased, distorted, or otherwise fallible.  (See, for example, Greene, in press.) And surely this too is something that philosophers should know.

Of course, a number of the hypotheses that experimental philosophers have explored about “what most people think,” or how they go about making ascriptions of moral responsibility, intentionality or knowledge, have been suggested by “prominent philosophers.”  And as Velleman notes,these prominent philosophers “didn't use any data that required approval from the Human Subjects Review Board”.  But Velleman’s comments under this heading suggest what we submit is a deeply problematic  view about what counts as reasonable evidence in this area.  “Traditional” philosophers can use a variety of methods, including consulting their own intuition, to develop hypotheses about the factors that influence people’s judgments on a variety of philosophically important matters.  And some of the hypotheses proposed by these philosophers have been brilliant and prescient.  But to suggest, as Velleman does, that these philosophers have discovered  “that people's description of an action as ‘intentional’ depends on their assessment of the harms or benefits that resulted from it, and whether they were produced reliably or by chance” is at best seriously misleading.  What traditional philosophers discovered were some very interesting and important hypotheses.  To discover whether these hypotheses are true requires systematic empirical evidence of the sort that traditional philosophers, in most cases, simply do not have.

To conclude, we suggest – and we are confident Velleman would agree – that discussion of a scholarly research program will be most productive when it engages the research itself, rather than journalistic portrayals such as the Slate article.  Readers who are interested in learning more about “experimental philosophy” will find a representative collection of articles here.

John M. Doris  (Washington University in St. Louis)

Joshua D. Greene  (Princeton University)

Paul E. Griffiths (University of Queensland)

Gilbert Harman  (Princeton University)

Joshua Knobe  (University of North Carolina, Chapel Hill)

Edouard Machery  (University of Pittsburgh)

Ron Mallon  (University of Utah)

Thomas Nadelhoffer  (Dickinson College)

Eddy Nahmias (Georgia State University)

Shaun Nichols  (University of Utah)

Jesse Prinz  (University of North Carolina, Chapel Hill)

Walter P. Sinnott-Armstrong  (Dartmouth College)

Stephen Stich  (Rutgers University)

Manuel Vargas (University of San Francisco)

Jonathan M. Weinberg  (Indiana University)

=============================

Comments are open; because I am in transit the next couple of days, they may take awhile to appear.  Please post only once.  No anonymous comments.  If the discussion is lively and interesting enough, I may move this thread back to the front once I'm back in the office.

Comments

[John Fischer]

Readers might be interested in the post and discussion of these matters on "The Garden of Forking Paths": http://gfp.typepad.com

[Eric Schliesser]

I encourage work in so-called "experimental philosophy, and I am delighted that it is getting attention within the discipline and outside of it. I agree it can be relevant to many philosophic projects. But some of the claims on its behalf raise methodological concerns. We live in culture(s) in which the sciences and philosophy have had increasing direct and indirect impact on so-called "folk" beliefs and attitudes (and the other way around, of course, too). To investigate what "what most people think" or "what the actual folk" beliefs are and construe this as unproblematic form of "discovery" is naive at best. It is particularly misleading to treat this enquiry as a simple form of hypotheses "testing" or a form of discovery about the "truth" regarding some natural kind: the "folk." There is simply too much room for echo-chamber effects, feed-back effects, self-fulfilling prophecies, etc. And let's not get started on the usual problems besetting social scientific research regarding the normativity of statistical apparatus, framing effects, coding problems, etc.

[David Velleman]

I spoke carelessly when I said that "the ‘discoveries’ touted in the piece in Slate are not exactly news to traditional philosophers". I should have said "*several* of the 'discoveries'." Since I went on to specify which "discoveries" I had in mind, I think my intention was pretty clear.

When I spoke of "substituting" experimentation for philosophy, I was thinking in part about the allocation of resources available to the discipline. Should departments have slots for faculty in the sub-field of experimental philosophy? Should we take time to train our graduate students in experimental design and statistics? As I said in my post, I believe that philosophy needs to inform itself about empirical matters. It's less clear to me that the relevant empirical research should itself be considered philosophy or should take up time and resources available to the discipline.

I'll leave it to the students of generics to debate just what sort of statistical evidence might be needed to back up the statement "people's description of an action as ‘intentional’ depends on their assessment of the harms or benefits that resulted from it". But how important is it to know what percentage of people fit that description? Testing the statement as an *empirical* hypothesis is not very important, because it serves only as a *philosophical* hypothesis anyway. That was the point of my reference to Aristole on endoxa. The endoxa are useful merely as starting points for philosophical reflection. We don't need to know just how many people believe them, precisely because we're not going to take "people's" word for it anyway: we're going to subject the endoxa to *philosophical* scrutiny.

Like Brian, I am currently traveling and will not be able to participate in the discussion further until next week.

[Brendan]

I can think of three reasonable objectives for experimental philosophy:

1. Proving or disproving claims that philosophers have made about the intuitions of folk.

2. Giving an account of folk intuitions as a project in of itself.

3. Using data collected on what lay person intuitions are to build positive theories.

From my relative lack of knowledge on the subject of experimental philosophy it seems like the following analysis of 1-3 is reasonable:

Concerning 1: In so far as philosopher have long thrown unsupported claims about what the "folk" think, it is important to determine whether or not they are, in fact, true.

Concerning 2: such projects do not seem to really be projects in philosophy, but rather projects in, say, cultural anthropology, maybe? Still seems like interesting work, though.

Concerning 3: I am somewhat skeptical about how much experimental philosophy has to offer in this respect. If what we are interested in is having a true theory of, say, moral responsibility, then how much do the intuitions of lay people really help us? I think this is what Velleman is getting at when he says:

"We don't need to know just how many people believe them, precisely because we're not going to take "people's" word for it anyway: we're going to subject the endoxa to *philosophical* scrutiny."

In so far as the views of lay people might make up some theory of moral responsibility it is not clear why whether or not such and such an "intuitive" view is actually held by lay folk matter if what we are looking for is a true theory.

More promising (for research under 3) is if lay persons intuitions suggest theoretical models that have not been thought of before (like Knobe using his results to provide a novel theory of intentions and intentional action). But, again, it is not clear that we have to know exactly what lay people think (Knobe could still construct a theory out of his results even if the results ended up being flawed in some way: for building a theory, they still would seem useful).

A couple of problems:

1. Making sense of what, exactly, is the category of the "folk".

2. Making sense of what, exactly, intuitions are (this would, no doubt, be related to giving an account of the folk). For instance, do philosophers have intuitions about philosophical topics? If they do, are they different in kind from folk intutions about the same topic?

1 Eric Schliesser addresses above. Concerning 2, It seems intuitions are a kind of belief, but beliefs can be caused by all sorts of things. However, intuitions seem to be caused in a certain kind of way; at least, there is some form of restriction on which beliefs are intuitions. For instance, in my own case, my belief that we do not have free will was caused by philosophical deliberation: this belief does not seem like an intuition. But, if my belief came about in another way, then perhaps it would count as an intuition. This relates to defining the category of the folk, since it is not enough to identify a population that has a certain belief to identify the "folk", you have to identify whether or not their belief in some view is because of theoretical deliberation, or something else.

[not a philosopher]

Dumb question, I am sure, but if you start doing empirical experiments about perceptions and beliefs, aren't you doing psychology?

[Jussi Suikkanen]

I find this passage:

'But to suggest, as Velleman does, that these philosophers have discovered “that people's description of an action as ‘intentional’ depends on their assessment of the harms or benefits that resulted from it, and whether they were produced reliably or by chance” is at best seriously misleading. What traditional philosophers discovered were some very interesting and important hypotheses. To discover whether these hypotheses are true requires systematic empirical evidence of the sort that traditional philosophers, in most cases, simply do not have.'

quite ambiguous and misleading. I wonder what the hypotheses referred to are supposed to be about. Are they hypotheses about what people's descriptions of actions as intentional actually depend on or about whether or not the correct use of the term 'intentional' depends on the harmfulness or beneficiality of the action being assessed?

I find it more plausible that if the former is meant that empirical investigation can provide evidence for the truth of the hypothesis. But as long as there is a possibility that the majority is mistaken in their use of a concept I see no way in which empirical investigation could establish the truth of the latter.

And, I think traditional philosophers (Austin, Ryle, Von Wright, Grice, Anscombe at least) had much more evidence for their views on this point than you credit them for. After all they had decades of experience of using the concept of 'intentional' in their linguistic communities. If their uses had differed from the general use they would have certainly noticed this as Grice points out.

When it comes to trying to establish the truth of the second reading of the hypotheses their arguments really went beyond 'discovering interesting hypothesis'. Grice for instance gives a good explanation for why use of 'intentional' depends on harmfulness and beneficiality of action. 'Intentional' seems to implicate blameworthiness. However, in situations where no blaming is around using 'intentional' for describing an action would flout the cooperative principle 'Do not make your contribution more informative than is required'. That seems to be as good of an argument as any that the correct use of 'intentional' depends on harmfulness.

[Edouard Machery]

I am somewhat disturbed by D. Velleman's post.

I do not really know what the sentence "Testing the statement as an *empirical* hypothesis is not very important, because it serves only as a *philosophical* hypothesis anyway" means. D. Velleman might want to spell out the puzzling distinction between *empirical* and *philosophical* hypotheses.

In any case, I suppose that this sentence does not mean that the truth of empirical premises about folk intuitions in philosophical arguments does not matter. Surely, D. Velleman cares as much as any experimental philosopher about the soundness of his arguments. It is, I think, pretty clear that many important arguments in philosophy rests on empirical premises about folk intuitions. If this is true, the findings in experimental philosophy, or some of them, matter.

If these findings matter for philosophical arguments, I do not understand the point of Velleman's rhetorical questions "Should departments have slots for faculty in the sub-field of experimental philosophy? Should we take time to train our graduate students in experimental design and statistics?" Probably, Velleman thinks that psychologists, not philosophers, should test the truth of the empirical assumptions philosophers often make. It turns out that psychologists don't and probably won't. If findings in experimental philosophy matter for philosophical arguments and if no one will bother doing some experimental philosophy, what is the rationale for *not* having slots for experimental philosophy and for *not* requiring grads to take a class in statistics? (Incidentally, a class in statistics should be required for any philosopher who has an interest in science.)

[jonathan weinberg]

Speaking only for myself, here are how I would respond to some of the questions/comments above.

Several bits of the above commentary can be paraphrased together as: "Is the experimental activity of experimental philosophers really _philosophy_, or is it just psychology?" I think that, when done correctly, it is often both philosophy and psychology -- just as a great deal of good formal work that gets done is simultaneously philosophy and logic, or philosophy and, say, decision theory; just as much good history of philosophy is both philosophy and history at the same time; just as much good philosophy of languge is both philosophy and linguistics at the same time. I don't see why the disciplinary boundary between philosophy and psychology needs to or should be particularly air-tight; indeed, one of the real pleasures of working in the philosophy of psychology these days is the extent to which the philosophical work is continuous with the theoretical science. (I'm not referring here to x-phi, but to the kind of philosophy that gets published in places like _Mind and Language_.) So I just don't see the point of getting worried about disciplinary boundaries here, when they are already so very malleable and porous.

"But shouldn't the psychological work be left to the psychologists, who have the relevant training? And do you really think that philosophy programs should take on the job of training people to do experiments?" For experimental philosophy to be done well, it must indeed be performed in accord with the standards & norms of empirical psychology. (Or sociology, or neuroscience, or whatever science might be relevant to the particular piece of research.) So, to do it, one must get at least a bit of that training oneself and/or work in close collaboration with those who do. But how is this so different than, say, if you want to be a serious Plato scholar, you need to learn a whole heckuva lot of Greek -- and that you'll almost certainly get most of that training outside of a philosophy department, and in a classics department? Or if philosophers of logic get a certain amount of their training from math or cs departments? And so on. Philosophers get whatever tools they need from wherever they can get them, and experimental philosophers are surely no different in this regard.

(Having said that, as philosophers do come to have training and experience in these methods, I see no reason why they couldn't begin training their students in it as well, just as some logic does of course get taught in philosophy departments as well. But I suspect that we're not really anywhere near that stage at this point.)

Jussi asks, "I wonder what the hypotheses referred to are supposed to be about." If you look at the sentence of Velleman that we quoted in the relevant chunk of text, you'll see that he is talking there about claims concerning _people's descriptions of actions as intentional or not_. Those are empirical claims, and our overall contention is that empirical claims should be taken seriously as empirical claims, and where that means we (meaning "we philosophers") need to go run a few surveys or whatever, and no psychologist has been interested in running them for us, then we should go run them ourselves. (That's part of the reason why we can't just leave the psychology to the psychologists -- what is psychologically of interest to psychologists is not necessarily what is psychologically of interest to philosophers!)

Now, the question about how one gets from empirical claims to positive philosophical theses is a tricky one, and I think you'll find that in general it's not something that we have tried to pursue. Much of the reason we haven't tried to pursue it, is that a lot of our work -- and all of mine -- has been of a negative or debunking sort: we find places where philosophers have presupposed a uniformity of judgments, and we've found various sorts of systematic diversity. I believe that philosophers in general have not paid sufficient attention to the ways in which various empirical presuppositions and commitments get tangled up in our philosophizing, and so the evidential or explanatory relations between the empirical and the philosophical are still uncharted. But one contribution that I hope experimental philosophy can make to philosophy on the whole, is both forcing the discipline to get clearer on that question (of the relationship between the empirical and the philosophical), while at the same time helping it out with some of the tools and knowledge that it will need to do so.

[Eddy Nahmias]

For those interested, a lot of discussion relevant to the issues raised by Velleman and others has gone on at the Experimenal Philosophy blog as well as the Garden of Forking Paths blog.

The Slate article did not present experimental philosophy particularly well or clearly, so some of Velleman's and others' critiques seem to derive from lack of familiarity with the actual work in this area. I encourage those interested to look at some of the actual articles rather than assuming that because we do polls, we think polls are all there is to do. Quite the contrary. The work in this area is new and its practitioners have taken quite different approaches towards what philosophical implications the results may have (e.g., shifting burden of proof, determining when revisionism is called for, suggesting intuitions are of dubious evidential value, suggesting the psychological sources of intuitional conflicts and philosophical debates, challenging claims made by philosophers about what is intuitive, etc.). Exp phils have been very explicit about *not* claiming that our work *replaces* substantive philosophical debate, though we have tended to leave it open just what our work may suggest vis a vis substantive philosophy and simply raised a lot of really interesting meta-philosophical (methodological) questions about the nature of intuitions, whose should matter, and what role they have played and should play in philosophical debates.

To give you a taste of these points, let me quote a few paragraphs from my paper with Thomas Nadelhoffer, Stephen Morris, and Jason Turner, "Is Incompatibilism Intuitive?" (forthcoming in PPR):

"Of course, if philosophers were concerned exclusively with a technical philosophical concept of free will, then appeals like those above to ordinary people’s intuitions [i.e., incompatibilist claims that incompatibilism is intuitive to non-philosophers and that the counterintuitive compatibilist position has the burden of proof] would be entirely irrelevant—just as they would be irrelevant for logicians debating the concept of validity or mathematicians analyzing the concept of infinity. But there is a reason why philosophers appeal to ordinary intuitions and common sense when they debate about free will: they are interested in developing a theory of freedom that is relevant to our ordinary beliefs about moral responsibility. Given that most philosophers are concerned with the kinds of free will “worth wanting” (Dennett 1984), an acceptable theory of free will should elucidate the abilities presupposed by our practices of attributing praise and blame, our expressions of reactive attitudes such as indignation and gratitude, and our systems of punishment and reward. Often, such a conception of freedom is also tied to our sense of dignity, individuality, creativity, hope, and love. Because the free will debate is intimately connected to ordinary intuitions and beliefs via these values and practices, it is important that a philosophical theory of free will accounts for and accords with ordinary people’s understanding of the concept and their judgments about relevant cases. Minimally, any theory of freedom that conflicts with such intuitions should explain both why our intuitions are mistaken and why we have those misleading intuitions in the first place."

"It is important to keep in mind that we are not suggesting that any philosophical theory would be demonstrably confirmed (or disconfirmed) just because it aligns with (or conflicts with) folk intuitions and practices. After all, such intuitions and practices may be mistaken or contradictory and hence in need of elimination or revision. (Of course, to know the extent to which they need to be eliminated or revised, we must first know what these intuitions and practices actually are.)"

"The upshot is that, to the extent that a philosophical theory (e.g., of intentional action or free will) aims to account for and accord with ordinary intuitions, it is unclear what its proponents should do in the face of conflicting intuitions. One might try to explain the conflict with a contextualist approach, as we suggested above. One might also develop a theory that calls for the revision of some of our concepts and practices. Some theories of free will are revisionist in this way, suggesting that the more metaphysically demanding aspects of the ordinary conception of free will can be eliminated but that most of it can be preserved (see Vargas, forthcoming)."

"Minimally, we believe that empirical data about folk intuitions should (a) encourage philosophers to state more precisely whether or not they are interested in ordinary intuitions about free will and moral responsibility and why, (b) prevent philosophers from appealing to the wide-scale intuitive plausibility of their theories unless these claims can be empirically substantiated, and (c) encourage philosophers to re-examine some of their own assumptions concerning the role of intuitions in philosophy. And in the event that a particular theory fails to settle with ordinary intuitions, the onus will be on its proponents to explain why we should care about a technical notion rather than the ordinary one—especially when understanding the latter is an important philosophical goal in its own right."

[Eric Rovie]

I find the work that the experimental philosophy folks are doing to be quite interesting. I like the idea of taking the intuitions that are often referenced by philosophers and finding out if that's what people REALLY think, intuitively.

I'm puzzled by one thing (which both Brendan and Eric Schleisser mention above), but it may just be my unfamiliarity with the methodology of the project. Here's my question: some of the work I've read in this subfield relies on surveys about intuitions done in undergrad classes on University campuses. Is that really going to give us the intuitions of the 'folk,' or is that going to give us a pretty skewed view of what people "actually" think? Would a sample of students at Princeton, or Wash U, or Chapel Hill give us the same intuitions as those culled from a sample in a grocery store in South Dakota or a bar in Gainesville, Georgia? If not, then what are those surveys really telling us?

I ask not in an explicit attempt to criticize the field, but to get clearer on one of Brendan's questions from above: what is the 'folk' or, more appropriately, WHO are the folk?

[Jeff]

When conducting an investigation certain things are taken for granted. At a minimum these things include assumptions about the integrity of the framework that is used to make the investigation intelligible. The strength of these starting assumptions depends, at least in part, on the type of the investigation that is being conducted. For example, the constraints on the pure mathematician are weaker than those that are imposed in the physical sciences. The philosopher’s situation lies somewhere in between these cases. The following remarks by John von Neumann might resonate with those philosophers who view themselves as being closer to the pure mathematician than they are to the natural scientist:

“As a mathematical discipline travels far from its empirical source, or still more, if it is a second or third generation only indirectly inspired by ideas coming from "reality", it is beset with very grave dangers. It becomes more and more purely aestheticizing, more and more purely l'art pour l'art. This need not be bad, if the field is surrounded by correlated subjects, which still have closer empirical connections, or if the discipline is under the influence of men with an exceptionally well-developed taste. But there is a grave danger that the subject will develop along the line of least resistance, that the stream, so far from its source, will separate into a multitude of insignificant branches, and that the discipline will become a disorganized mass of details and complexities. In other words, at a great distance from its empirical source, or after much "abstract" inbreeding, a mathematical subject is in danger of degeneration.” (JvN)

I do think that empirical work can play an important role in philosophy and some of my reasons for thinking this are similar to what is expressed in von Neumann’s remarks. I have done empirical work. I am employed by a philosophy department. Yet, I do not believe that I am participating in the experimental philosophy movement that is being discussed. The reason I do not believe this is that I am following a rather old tradition and so I must not be participating in this new experimental philosophy. Let me close with an example to illustrate this older tradition. Then, perhaps, someone who is better informed will be able to explain what is new in the new experimental philosophy.

There is a tradition dating at least as far back as Knight and Keynes that draws a distinction ‘measurable uncertainty’ (or risk) and ‘unmeasurable uncertainty’. A related, but more recent distinction, is given by Luce and Raiffa in their distinction between ‘decision making under risk’ (where objective probabilities are available to the decision maker) and ‘decision making under uncertainty’ (where objective probabilities are not available to the decision maker). The subjective expected utility theory [SEU] of Ramsey, de Finetti, and Savage undermines the significance of this distinction since a subjective probability is assumed to be available to the decision maker even when objective probabilities are not.

Daniel Ellsberg’s empirical work suggests that collapsing the distinction between risk and uncertainty presents a problem for SEU. What sort of problem? Well, Ellsberg’s work demonstrates a stable incompatibility between SEU and the reflective choices that are made by a significant number of sophisticated decision makers. I urge you to read Ellsberg’s paper if you are concerned with the use of the term ‘sophisticated’, but I believe that the term must have some presystematic value if norms are to be evaluated in a nontrivial way. Some psychologists have taken Ellsberg’s results as further evidence against the “rational man” hypothesis that has been adopted in certain parts of economics. On the positive side, these psychologists take Ellsberg’s results as a challenge to find better descriptive theories.

In an ongoing research program with Horacio Arlo-Costa, we draw a different lesson from Ellsberg’s results. Specifically, like Ellsberg, we look towards revising the normative account. Several alternatives to SEU have been offered (e.g. Ellsberg, Levi, Gardenfors and Sahlin). In our empirical work we have been investigating the possibility of rationalizing – by appeal to one of the alternatives – choices that have been ruled as deviant by psychologists who maintain SEU as the normative standard. In carrying out this research program Arlo-Costa and I are building on a tradition that we learned from philosophers like Herb Simon, Isaac Levi, and Teddy Seidenfeld. As I said, this tradition is rather old and thus must be different than the new experimental philosophy. Furthermore, I would like to assume that it is precisely this difference that explains why there is no mention of those who might identify with the old experimental philosophy – that is, why there is no mention of the experimental work done at places like CMU, Caltech, Irvine HPS, UPenn, Columbia, Konstanz and LSE.

[Jussi Suikkanen]

Jonathan,

thank you very much for the clarification. There is still a worry though. You may be right that the hypotheses which can be experimentally confirmed to be true are of the actual use of the term 'intentional' by the 'folk'. But, now, these were claimed to be the same hypotheses which philosophers like Ryle, Austin, Grice, Anscombe, Von Wright, and so on entertained about intentional action and which they could be in principle unable to show to be true. I much doubt whether they had the actual use of the majority in mind but rather the use of those people who master the term and apply it correctly given the norms of use. My bet is that their thoughts were on the latter. If there is any evidence that what they were concerned about was the former, then I would be interested in seeing it. And, as long as there is the possibility that the two come apart, then experimental methods seem to be unable to confirm the truth of the hypotheses they had in mind. So, I think the experimentalists and these philosophers are talking past one another here.

[Kristin Andrews]

On allocating resources to an empirical approach to philosophy…

If we can construe experimental philosophy more widely as a combination of theoretical and empirical methods, then it seems to be fairly well established. For example, the philosophy of psychology is grounded in empirical research, as is clear from looking at the bibliographies of Bradford books. What is new is (a) philosophers have started doing their own experiments when the data they want isn’t available, and (b) the method has spread from philosophy of mind to other areas of philosophy. This seems like a natural progression from the collaborations between philosophers and psychologists that have existed over the years.

That being said, since philosophers who do their own research are working cross-disciplinarily, using the methods of both psychology and philosophy, they need to be trained in both these methods. Speaking from experience, it isn’t always easy for a graduate student to get that training. Some cognitive science programs and other interdisciplinary PhD programs do offer students such training, as do some philosophy PhD programs. The training is important, because as has been pointed out there are many pitfalls in experimental research that need to be avoided. But the research isn’t something that can be left up to psychologists, because psychologists aren’t always going to be interested in the issues we’re interested in.

If a philosophical issue rests on a philosopher’s intuitions about people’s intuitions in general, then the philosopher’s intuitions can be tested to determine whether they are correct or not. Sometimes they are correct. But we can’t know that until the evidence is in. Knobe’s studies are meant to serve as empirical evidence in favor of the intuition Anscombe and others may have had. The fact that others had an intuition that is consistent with the data doesn’t undermine the importance of the empirical research. Empirical plus theoretical evidence for a position is stronger than either one alone. But philosophers are not going to be able to do good empirical work unless they get training in the relevant methodologies. So should some programs offer this training? Yes, just as some programs offer training in mathematical logic. Do all programs need to offer this training? Of course not.

[Antti Kauppinen]

Here’s a proposal for distinguishing between various forms of resistance to experimental philosophy. First, however, let’s get one issue out of the way: ‘experimental philosophy’ is a bit of a misnomer. As all the x-philes here admit, the ‘experimental’ part isn’t itself philosophical. The philosophical part is making use of the empirical data to challenge some bit of conventional philosophical wisdom. “Empirically informed philosophy” is more precise, but suggests misleadingly that traditional philosophizing isn’t empirically informed, which I believe to be false. “Scientifically informed philosophy” would be closer to the truth; the only problem with that concerns the scientific status of the studies actually carried out. So let’s stick to x-phi for short.

Now, here’s four different responses a traditional philosopher might give to the experimentalists:

a) We don’t make empirical assumptions (for example, folk concepts don’t matter).
b) We do make empirical assumptions, but they turn out to be irrelevant to where we end up after philosophical reflection.
c) We make empirical assumptions and they are relevant to where we end up, but you misidentify the assumptions we make.
d) We make empirical assumptions that are relevant to our conclusions and you correctly identify them, but the methods you use to test them are unsound or your interpretations of the data gathered are faulty.

Velleman’s response, as far as I can see, is b. I’ve heard claims of type a as well. What’s in common with them is that they allow for a wholesale dismissal of experimental philosophy. Though this is indeed tempting, I believe that exaggerates the autonomy of philosophical reflection. I agree with x-philes that philosophers do make assumptions that are in some sense empirical and are relevant to their philosophical conclusions. If this is correct, criticism of x-phi must take the form of c or d (or some other option I haven’t thought of here, obviously). In a paper I have under review (http://www.helsinki.fi/~amkauppi/phil/The_Rise_and_Fall_of_Experimental_Philosophy.pdf), I take for the most part route c. I argue there that experimentalists misconstrue the nature of philosophical appeals to intuition, and consequently the survey results they offer fail to confirm or falsify typical philosophical claims about folk concepts. Another way to put this would be to say that they fail to operationalize the ‘hypotheses’ properly – a philosopher claiming that “We would not say that x phi-ed intentionally in S’ is not making a prediction about how most people would respond to a case presented to them, but rather only about the semantically guided responses of competent users of concepts in sufficiently ideal conditions, as I argue. I don’t say so much about option d in the above paper since I’m not a philosopher of social science, but I’ve read enough sociology to be aware of how naïve the reliance on surveys must look to those fed on Giddens, Habermas, Bourdieu, Geertz, or even Weber. Very crudely, the way to find out how ‘most people think’ isn’t to pretend ignorance and present ‘probes’ to test subjects, but to reflectively participate in relevant interaction with them. I also believe that philosophers have in fact always done this, and consequently the claim that they "simply do not have" the relevant sort of evidence isn't only inflammatory but also false.

Whatever you think of these particular arguments, it is clear that taking c or d means adopting a piecemeal approach to criticizing experimental philosophy. This would seem justified in the light of the variety of ‘experimental’ approaches. The methods and data that Harman and Doris use are very different from those of Knobe and Nadelhoffer, and the cross-cultural studies present yet different issues. It could well be that some of these are fruitful and some not. While it’s easy to envision both c and d type responses to these different claims, they must be made case by case. For example, against Harman and Doris I’d argue that they misidentify the assumptions that people doing ethics make about character, and consequently the commonsense view ethicists rely on survives the social psychological evidence (though common sense itself may vary, and anecdotal evidence suggests that belief in rigid character is more common in some parts of the US than elsewhere).

[dustin locke]

In a comment (above), Jonathan Weinberg writes:

“Now, the question about how one gets from empirical claims to positive philosophical theses is a tricky one, and I think you'll find that in general it's not something that we have tried to pursue. Much of the reason we haven't tried to pursue it, is that a lot of our work -- and all of mine -- has been of a negative or debunking sort: we find places where philosophers have presupposed a uniformity of judgments, and we've found various sorts of systematic diversity.”

In defense of experimental philosophy, many experimental philosophers tend to say just what Weinberg has said: well, many philosophers are (or at least seem to be!) appealing to folk intuitions in defense of their theories, so we should go out and survey the folk. But this seems to assume A) that those philosophers ought to have been appealing to folk intuitions in the first place—that is, that the folk intuitions are indeed relevant to philosophical theorizing and B) that reports on surveys are reliable indicators of “intuitions” (as the philosophers who appeal to them use the term).

Now, Weinberg admits that the relationship between the empirical findings and the philosophical theorizing is “still uncharted”, but he claims that:

"one contribution that I hope experimental philosophy can make to philosophy on the whole, is both forcing the discipline to get clearer on that question (of the relationship between the empirical and the philosophical), while at the same time helping it out with some of the tools and knowledge that it will need to do so."

This last claim seems to be at the heart of the issue I'm trying to raise. The "knowledge" in question is presumably the data gathered by the surveys. But does the discipline really "need" to know the results of these surveys in order to "get clearer about... the relationship between the empirical and the philosophical"? The “getting clearer”, I assume, would first involve figuring out whether the surveys are even RELEVANT to the philosophical theorizing. But do we need to know what people will report on the surveys before we can know whether their reports are relevant to philosophical theorizing? I’m not sure we do, and if not, then it seems that we ought to try to tackle the question of whether the surveys are indeed relevant before we start dedicating our time and money to them. Moreover, if we don’t even yet know whether the surveys are relevant to the philosophical theorizing, we certainly don’t know HOW they’re relevant, and thus proceeding with them seems to me like pretty bad methodology. Perhaps I am thinking a bit too idealistically about the sciences, but don’t researchers who claim to be gathering empirical data relevant to some particular theory or theories usually take it upon themselves to know how that data will bear on that particular theory or those particular theories?

[Zeljka]

>One of the findings discussed in the Slate story (and illustrated with a rather tasteless cartoon!) was Jonathan Haidt’s discovery that low socioeconomic status (SES) individuals judge harmless actions which evoke disgust to be morally wrong, while high SES individuals do not judge them to be morally wrong.

While David Velleman is no doubt mistaken that this finding is not news for traditional philosophers this is probably so because the finding is so philosophically uninteresting that traditional philosophers didn’t even care to speculate about it. Any philosopher whose position might have depended in facts of this kind is now (rightfully) long forgotten, if he ever was recognized as a philosopher at all.

>A third much discussed finding, not alluded to in Slate, is that English speakers of European cultural background and English speakers of East Asian cultural background report very different intuitions about “Gettier-cases” of the sort that brought about a sea change in epistemologists’ thinking in the middle of the 20th century.

Again, while it seems natural that some philosophers find this finding entertaining, I see no reason why philosophers should actually be expected to address it. Gettier-cases, like all philosophical problems and puzzles require a mind trained in a particular way to recognize them as such. It is rather obvious that some, many or most ordinary people think that choice is possible or that knowledge is possible or that moral matters (or do not have opinion on any of it at all). Philosophers are distinguished from them because they do not take those intuitions for granted. Philosophical problems are not universal problems of mankind, so no matter how accurate account of universal understanding of them is provided by whatever flawless empirical discipline, that account is irrelevant for solving them. At best it might be a source of inspiration, but only after that inspiration is properly articulated and juxtaposed with traditional problems can philosophy begin.

Which is not to say that experimental philosophy is a bad thing. It is a pretty good thing – for psychology. It is a great improvement over its predecessor in the area of moral reasoning (Kohlberg) which was extremely primitive philosophically. I am pretty sure that many psychology departments would be interested in hiring experimental philosophers and there already exist psychologists who are doing similar things (e.g. Bertrand Malle, Marc Hauser). So I don’t think that there is any danger that this interesting stuff won’t get done unless philosophers embrace experimental philosophy. The question is mainly the one of the cost for philosophers of pursuing somebody else’s (psychologists’) problems instead of their own.

[Travis]

In relation to something Kristin (above) said (notably the last paragraph), I have to say that while experimental philosophy can indeed be fruitful, there arise serious questions about methodology. For example, Doris's citations of studies with relation to the virtues suggested a lack of rigorous methodology, perhaps a lack of how the scientific method generally "works." For instance, citing seminary students as strong candidates for possession of the Aristotelian virtues (assuming they exist) is absurd to many. Because seminary students' behavior was heavily influenced by situational factors, virtues do not exist?

The need for objectivity in the laboratory must carry over into this sort of research if it is to be of any use in the philosophical domain. I don't intend to impugn Doris's character, but a philosophical thesis can gain no credence from studies if the underlying methodology is flawed. It's a line to cross when going from "purely theoretical" philosophy to experimental philosophy, and perhaps not many (surely not the majority) are prepared for making this leap.

[Joshua Knobe]

Just a quick point of clarification about the recent spate of studies on people's concept of intentional action. The chief purpose of these studies is not to reach a better understanding of the true nature of intentional action. Rather, the purpose is to reach a better understanding of *folk psychology*.

What the experiments appear to show is that people's *moral* judgments are playing a role in their folk-psychological concept of intentional action. These results come as a surprise because it had been widely assumed that folk psychology could not be influenced by moral cognition in this way.

Following the experiments discussed in the Slate article, there have been a number of important replications and extensions. But the purpose of these additional studies is not just to determine the precise percentage of people who have one or another intuition. Instead, the purpose is to reach a better understanding of the underlying psychological mechanisms at work here. Hence, the later studies have shown that the original effect still emerges when the stories are given to people who suffer emotional deficits due to frontal lobe lesions (Young et al.), that the effect still emerges when the stories are given to four year-old children (Leslie et al.), that the effect still emerges when each subject is given both stories (Nichols & Ulatowski), that it emerges even when subjects are given stories about outcomes that were due entirely to luck (Nadelhoffer).

Taken together, this body of research offers an important challenge to the view that folk psychology is something like a scientific theory. The results suggest that certain aspects of folk psychology are tied up with moral judgment in a way that would not previously have been expected.

The philosophical significance of these results, then, does not depend on any assumption that the folk judgments are correct. Regardless of whether these judgments are correct or incorrect, it seems that we can use them to address certain fundamental philosophical questions about the nature of folk psychology itself.

[Gualtiero Piccinini]

For those who might be interested, I posted a commentary on the experimental philosophy debate on my blog (link from my name).

[Jussi Suikkanen]

Joshua,

thank you for your further clarifications. I want to ask yet another question about your reply. You say that:

'These results come as a surprise because it had been widely assumed that folk psychology could not be influenced by moral cognition in this way.'

Could you say more about this - where it was assumed and how. I'm quite intrigued. I took that folk psychology includes 'concepts such as belief, desire, pain, pleasure, love, hate, joy, fear, suspicion, memory, recognition, anger, sympathy, intention, and so forth' (list from Churchland). A lot of the concepts here are emotion concepts. I find it hard to believe that someone would have thought that our attributions of emotions to others would not depend on our moral judgments. Many of the emotion terms just seem to have direct moral content. It seems really odd that our moral outlook would not affect who we think feel sympathy or guilt for instance. If someone thought this, I think I would like to try some traditional philosophy against them - especially if this was a fundamental assumption of the philosophical understanding of folk psychology.

[Zeljka]

>Taken together, this body of research offers an important challenge to the view that folk psychology is something like a scientific theory.

The view that folk psychology is something like a scientific theory has been challenged and for the most part discarded many years before experimental philosophy came into existence. Some of the more notable examples of the people-think-like-scientists research programs gone awry were Piaget’s view of logical thinking, decision-science view of probabilistic reasoning, attribution theory view of person perception etc. Thus, I think the best you can do in psychology in terms of novelty is challenge the marginal view that people think like moral philosophers (psychologists’ construal of what the latter think being rather problematic). But I don’t think that the strongest part of your program in psychology is its alleged revolutionary character. The strongest part is your choice of a topic – it is an interesting, important and entertaining topic that has been somewhat neglected by psychologists. Kudos for that and I hope to see more of your stuff published in psychology journals. I hope we all agree that that is where it belongs and where its merits (methodological soundness, proper grasp of literature, theoretical innovativeness) can be properly assessed.

>Regardless of whether these judgments are correct or incorrect, it seems that we can use them to address certain fundamental philosophical questions about the nature of folk psychology itself.

I take it that you can say much more about this because this is a way too weak a claim for philosophical importance. So I would like to know what those “fundamental philosophical questions” are.

[Derek Bowman]

Dustin Locke says above:
"But does the discipline really "need" to know the results of these surveys in order to "get clearer about... the relationship between the empirical and the philosophical"? The “getting clearer”, I assume, would first involve figuring out whether the surveys are even RELEVANT to the philosophical theorizing. But do we need to know what people will report on the surveys before we can know whether their reports are relevant to philosophical theorizing?"

This has generally been my initial reaction to the field and to the only article I've read so far - the "Metaskepticism" paper (available from one of the links above); it seems to me the interesting philosophical questions can be as easily addressed in the hypothetial form - 'suppose we're wrong about folk intuitions; how, if at all, would that change the theories and arguments at hand?'

I've tended to think of thought experiments and the intuitions they evoke to be primarily useful in helping me to figure out what *I* (or my interlocutors) think. And I already know from teaching intro classes to freshmen that not everyone shares my intuitions (I also know that, upon further questioning, many of them do not even seem to share their own espoused views). From that perspective, it's not clear to me how survey results are likely to change that.

Dustin Locke then says, in response to his own question:
"I’m not sure we do, and if not, then it seems that we ought to try to tackle the question of whether the surveys are indeed relevant before we start dedicating our time and money to them."

Now it's this part of the opposition that I don't understand. There are plenty of people in philosophy doing things I have little interest in doing, and there are plenty who are concerned about the areas that interest me. So I can understand why you don't want to do 'experimental philosophy,' (neither do I); but why should that be a barrier to others doing so?

I suppose if I eventually find myself with some say in the disposition of resources other than my own time I might have to consider the matter more carefully. If some people think it's worth their time to explore questions in this way, why oppose their effort? (Challenging their arguments is, of course, another matter) Even if you're right that it's not worth the effort, we can learn from failed experiments too.

[Tom Hurka]

Joshua:

I haven't read your papers, which may answer this question, but here goes anyway.

Granting your data, why should we think that the folk-psychological concept of intentional action is affected by moral judgements rather than that "intentional" as used in folk discussions is not a psychological term at all but a moral one? For example, it could be used as a moral intensifier, so that "A did x intentionally" means that A's x-ing was more seriously wrong (if x-ing is wrong) or more highly admirable (if x-ing is admirable). I think many people would say, of the businessman case, that knowingly harming the environment is more seriously wrong than doing so unknowingly, even if you don't positively want to harm the environment; but knowingly helping the environment isn't more admirable than unknowingly doing so, if you don't positively care about the environment. If this pair of judgements is in fact common, could experimental subjects not be signalling the difference between them when they apply "intentional" to the first action and not the second?

I don't say this is the right explanation, but how do we decide whether the folk use of "intentional" does in fact concern intentional action (in which case judgements about that subject are affected by moral judgements) rather than having a purely moral and not psychological use? One advantage of the second option is that it is more charitable toward common sense..

[jonathan weinberg]

Zeldjka, the question of the connection between affect and moral judgment is an old & persisting one, and philosophers have not substantially addressed what it would mean if different people in different cultures, SESs, etc. have different such connections; so I don't see how the Haidt et al. result merit the kind of dismissive scorn you aimed at them. I also find your metaphilosophy a bit puzzling; if we're not interested in 'universal problems of all mankind', or at least in problems for more than just us odd birds who go into philosophy, why on earth should they keep paying all our salaries? At a minimum, I'm sure most epistemologists would not agree with you --most of them take themselves to be studying knowledge, not special-philosophers-only-knowledge*. (And when they are doing the latter, they are usually admirably clear about it, like Stroud.)

Travis, I'm not sure I totally understand your comment, but it seems that you are challenging the Darley & Batson study that Doris (and many others) like to cite. If that's right, then I think you have the dialectical obligation to say a bit more, because that's a classic & much-discussed (& much-extended) study. If that's not what you had in mind, then perhaps you could make the target of your remark a bit clearer?

[jonathan weinberg]

Dustin, while I was & am leaving open the question of what the _ideal_ relation between the empirical and the philosophical may be, I said plainly in the bit of text that you quoted what the _current_ relevance of much x-phi work is to many philosophical projects: "we find places where philosophers have presupposed a uniformity of judgments, and we've found various sorts of systematic diversity". This is, as I also said clearly, a debunking project, and which shows why the antecedent of this conditional is false: "if we dont even yet know whether the surveys are relevant to the philosophical theorizing". We do know they're relevant -- they're relevant to attacking that presupposition.

But you say, "this seems to assume ... that those philosophers ought to have been appealing to folk intuitions in the first place". You seem to be arguing that, in order for me to debunk a presupposition, I must first somehow be endorsing it! If you think that philosophers shouldn't even make such presuppositions in the first place, that's fine with me -- we're on the same side, even -- but please take it up with the large portions of the profession who have this commitment. They're the ones you should be fighting with; not us. (You might start with Frank Jackson.)

(As for assumption (B), on the link between our surveys & intuitions, it is addressed at some length in the "Normativity & Epistemic Intuitions" paper, linked above.)

[Zeljka]

>Zeldjka, the question of the connection between affect and moral judgment is an old & persisting one, and philosophers have not substantially addressed what it would mean if different people in different cultures, SESs, etc. have different such connections;

The fact that, at some level of not necessarily justified detail, different people and different cultures are different certainly got a fair share of philosophical discussion. Whether the source of that difference is variability of affect, judgment or interaction between the two might be an interesting psychological question. The answer to that question might have some philosophical implications, too, but unless those implications are spelled out and related to traditional literature this mere possibility doesn’t make the question philosophically relevant.

Btw, once you moved from the specifics of the study to the general psychological question of a relationship between affect and judgment that is more promisingly relevant for philosophy (and you had to do that, otherwise the point of the study would be utterly philosophically irrelevant, as I argued), it is no more clear that philosophers are not actually better off reading psychological literature on this topic (of which there is plenty) than waiting for work to be done by experimental philosophers.

>I also find your metaphilosophy a bit puzzling; if we're not interested in 'universal problems of all mankind', or at least in problems for more than just us odd birds who go into philosophy, why on earth should they keep paying all our salaries?

It seems that you are assuming that only those working on universal problems of mankind deserve any salary at all. The fact that there are a few philosophers, a small proportion of whose have decent salaries, that there is in general no expensive equipment at philosophy departments etc. I think speaks to my claim that, indeed, philosophers are not engaged in solving problems of great importance. However, there is some small proportion of odd-birds and odd-birds in transition (e.g. undergraduate philosophy majors) that happen to think that philosophical problems are interesting and are ready to pay to learn more about them and/or have them solved.

[Travis]

Jonathan, that seems to be the series of studies I meant. My point is that, lacking training in how to perform an experiment ourselves, ensuring that as many variables as possible are eliminated, to simply launch into experimentation is likely to do more harm than good. In Doris's case, he was merely citing the studies (I seem to recall). There is, however, a similar obligation to anyone who would even cite, in philosophical research, a study of this sort.

Unfortunately, I do not have the piece (from Doris) before me, but I seem to recall that one of the studies chose seminary students, put them into groups, and then recorded their behavior. The students who were told that they were running late to get to an appointment did not stop to help someone who had dropped papers and such in their path. The students who were not running late stopped to aid the person.

I do not know if there were other factors that went into the study, but the basic notion from Doris was that not even seminary students have this virtue of beneficence. One seeming problem with the experiment is that there is no great reason to believe that seminary students, as opposed to philosophy students (or better yet, "real" philosophers) would exhibit the virtue. Another problem is that virtue may not be very widespread, so the experiment is of limited worth in that respect.

I'm obviously not saying that no philosopher can correctly perform or cite an experiment, but I'm saying that, lacking proper training in the experimental field, philosophers really aren't qualified to engage in this sort of thing. I'm aware that there are philosophers who are thus qualified, but I just wanted to urge caution. Crossing a discipline is crossing a discipline. It's murky water, but I think that there is much to be gained from properly-conducted and carefully-interpreted experiments.

[Zeljka]

>so I don't see how the Haidt et al. result merit the kind of dismissive scorn you aimed at them.

I don't think i aimed dismissive scorn at them at all - I just said their study was not philosophically interesting. I believe that few psychologists would be offended to hear that their work is not philosophically interesting - much of it might not be and was never meant to be. It is only philosophers doing psychology that might reasonably interpret that kind of a comment of their work as, I guess, scorn.

[Keith Douglas]

I regard the investigation of factual ("empirical" ) matters in philosophy to be simply an indication that there is no dividing line between science and a science oriented philosophy. This attitude is actually shared by others regarded as scientists. For example, the textbook Perception, by Sekuler and Blake, includes a chapter called "knowledge and perception". I challenge anyone skeptical about "experimental philosophy" to tell me why this chapter does not affect our conceptions of justification, evidence, etc. As for who actually does the investigations, doesn't matter, so long as they are well informed in all appropriate respects. (See also my paper "Descriptive, Normative, Idealized" on my website for more on this.)

And certainly philosophers in the appropriate areas should learn everything they can that is, too from the adjacent fields. Unless one believes knowledge comes in pieces seperate from one another, philosophy should unify in the limit with other disciplines. This is not just true of "sciency things" but also our aesthetics should link to art criticism, and so on.

As for the specific role factual investigations can play, another use beyond the ones mentioned by others is the "shake up" factor. Sometimes "stepping out of the ivory tower" can remind us that there are alternative ways of thinking, even if they are mistaken. This can be very useful. Similarly, if philosophy is to understand our lives and make a difference we should know what it is we are starting with better.

(I am reminded over the debate over whether there is a scientific method. I'm skeptical of those who claim that scientific method is a refinement of common sense in part because those who become scientists are not common ...)

[Robert Allen]

"We find places where philosophers have presupposed a uniformity of judgments, and we've found various sorts of systematic diversity". Weinberg

Fine, as long as it’s understood that this does not philosophically settle anything. (And that the intuitions in question are not to be understood a la Kauppinen, which isn’t clear.) No philosopher should be expected to abandon his view just because the folk oppose it; although, as Nahmias contends, he does owe us an explanation of why they are wrong. (Should Berkeley have deferred to their judgment; was it reason enough to abandon his view?) No, faced with this fact, he must argue further that they are mistaken, digging more deeply inside HIMSELF for reasons, the same thing he would have to do were a lone colleague to present him with a counterexample. As my teacher Mike McKinsey once said, apropos Descartes, “Philosophy is done in the first person.” Appeals to common sense by a philosopher are entirely superfluous, grandstanding really. His own judgments/intuitions, buttressed by arguments, suffice for his work.

I want the X-ers to tell me: why am I wrong to emulate Descartes in the way I go about doing philosophy? I may be trying to defend what I take to be a folk intuition, but I am really after the truth of the matter, not an understanding of what the folk think (as Zljeka points out at GFP). Prof. Velleman (as usual) has hit the nail on the head: considering endoxa is only the first step in a philosophical investigation. (Think of Socrates in the Euthyphro.) And, at the end of the day, a view may be idiosyncratic yet (turn out to be) philosophically correct. As I’ve said, no philosopher worth his salt is going to abdicate a definition just because it happens to be at odds with the vernacular.

[dtlocke]

Thanks for the response, Jonathan.

I take your point about debunking: If philosopher X is using an empirical claim about the folk intuitions as a premise in his argument, then it seems appropriate to show that there is empirical evidence which apparently refutes that claim. However, my point is that this might not be the BEST way to respond to such a philosopher. If it is in fact true that the folk intuitions are IRRELEVANT to the philosophical thesis at hand, then the best way to respond to him would be to simply point this fact out. Thus, it seems that it's better to decide first whether the folk intuitions are relevant. If they aren't, then we can respond in a much more time and cost effective manner.

Actually, if we don't have (sufficient) reason to believe that the folk intuitions are irrelevant, we can still respond in a time and cost effective manner if we don't have (sufficient) reason to believe that they ARE relevant (which I think we don't). All we need to do is point THAT fact out.

I agree that both of our approaches are debunking--the difference is that you want to debunk by showing that the empirical claim in question is false, whereas I want to debunk by showing that we (at least) don't yet have sufficient reason to believe that it's even relevant. I think my strategy has the advantage of A) being more cost and time effective and B) forcing us--or the philosopher to whom we are responiding--to think about HOW the folk intuitions are relevant (if they in fact turn out to be relevant) BEFORE we go designing and carrying out experiments.

As far as (B), thanks for the reference! I believe I've read that paper before, but I'll be sure to go back and take a closer look.

[jonathan weinberg]

Zeljka, I honestly don't understand what your picture is of what philosophy is for, and what philosophers' jobs are supposed to be. But I think it's fair to say that it's just besides the point here if indeed _some_ philosophers are doing something so specially intra-philosophical that they can cheerfully disregard whether their projects & premises connect in any way to the outside world. That's fine if _some_ do that, and indeed as I noted earlier there are some (like Stroud) who do so & are very up front about it. All _I_ need is that there are also _some_ philosophers, indeed a fair number of them, whose work has the sorts of connections to the empirical world that I have been adverting to. And to demonstrate that there are such philosophers & plenty of them, it's just a matter of reading the literature. I would start by pointing you to the Doris & Stich chapter linked to at the top of the page, and especially their discussion in sections III and V, and you'll see they do an excellent job of drawing numerous connections between the empirical literature (including Haidt's research) to extant philosophical research. In my own areas of philosophical specialization, it is clear that many ongoing projects would not make sense as having philosopher-only relevance: to just take the examples that come easiest to my mind, one can consider the recent contextualism literature in epistemology; or the consciousness literature in philosophy of mind; or the puzzles of imaginative resistance in aesthetics. So there's plenty of philosophy out there that cannot be cloistered in the way that you would have us do it.

I must confess to being rather confused by the following: "...it is no more clear that philosophers are not actually better off reading psychological literature on this topic (of which there is plenty) than waiting for work to be done by experimental philosophers." Since the philosophers in question, as is obvious from an inspection of their their published work, do study the relevant psychological literature very closely, I am not sure just what sort of advice you take yourself to be offering. And I have yet to see why it makes any sense at all for philosophers, where particular empirical results might be relevant, to wait idly on the psychologists wherever the relevant work hasn't been done and isn't being done.

Also, with regard to your statement, "I don't think i aimed dismissive scorn at [Haidt et al] at all - I just said their study was not philosophically interesting" -- look, if you think that Haidt wouldn't be _very_ surprised to be told that his work "was not philosophically interesting", I'm pretty sure you're badly mistaken.