What are Experimental Philosophers Doing?

I am usually not very interested in second-order debates about the nature of philosophy or about the nature and goals of experimental philosophy, but  a few recent posts here and there have attracted my attention (see particularly the exchange between Ron and Thomas).  So, what are experimental philosophers doing? And whatever they are doing, is that philosophy?

1.  One answer to these questions has been recently proposed by Joshua Knobe. In substance, he argues that experimental philosophers are studying empirical questions--how people think about intentionality, how people make causal judgments etc.  To do so, they use, as they should, experimental methods.  Furthermore, this counts as philosophy, because experimental philosophers address the very questions that were of interest to philosophers before the rise of analytic philosophy, i.e., of philosophers such as Hume (e.g., the role of emotions in moral judgments, how people think about causation, etc.).

Now, this is certainly a seductive move and it has considerable rhetorical power. Still, I find this a pretty unconvincing reply. Philosophers such as Descartes, Hume, and before Aristotle and Plato were certainly interested in the questions  highlighted by Josh. But, they were also interested in biological questions (remember Aristotle's work on the growth of the chick), in cosmological questions (remember Descartes's Le Monde), in geographical questions (some of Kant's work), and so on, and so forth. Now, if Josh were right, by getting a PhD in developmental biology, I would be engaged in philosophy, because I would be addressing, with modern tools, the very questions that were of interest to Aristotle. But this is silly.

2. Another answer, which seems to be embraced by Thomas and Eddy, is to say that experimental philosophers bring  new (and, maybe, better) tools to bear on the traditional project of conceptual analysis. Eddy and Thomas call this experimental analysis.

I don't find this project very promising. As I have argued in my Mind & Language paper, this project supposes a capacity to distinguish conceptual competence and conceptual performance. In turn, this supposes a theory of concepts. But, unfortunately, nobody knows which theory of concepts is correct.

So, what are experimental philosophers doing? And is that philosophy?

3. For what it's worth, I have a pretty clear idea of what I am doing.

Some of my work (Semantics, Cross-Cultural Style, Against Arguments from Reference, and the paper on causation by absence with Jonathan Livengood for the Midwest Studies) belongs clearly to the debunking tradition of experimental philosophy. Here is how it goes:
- Identify philosophical arguments that hang on empirical claims about folk intuitions.
- Test these claims and show that they are false.
- Evaluate the consequences of these findings for the philosophical arguments at hand.

In other papers, I am more concerned with psychological questions. The paper on the trade-off hypothesis attempts to causally explain the judgments people make. My current work on conscious mental states with Justin Sytsma is an extension of the traditional research on theory of mind. My recent cross-cultural work on the folk concept of race studies whether people's concept of race belongs to their folk biology.

Now, is this philosophy? Here's is why the answer is clearly affirmative. Current philosophy is extremely diverse. It includes formal modeling (Glymour and colleagues' work on causation, Skyrms on the social contract, etc.), logic, philosophy of language closely inspired by linguistics, philosophy of biology, philosophy of physics, etc., in addition to intuition-driven philosophy. (Much of these fields do not rely on intuitions at all. If you wonder what philosophy could be if it were not allowed to rely on intuitions, I suggest you open Philosophy of Science, BJPS, Linguistics & Philosophy, JSL, etc. )  It is likely that any characterization of philosophy that is wide enough to include all these fields will also include experimental philosophy.

To put the same point differently, consider some salient examples of  philosophical works and debates that are thoroughly empirical, e.g., the debate between simulation theorists and theory theorists in the philosophy of psychology, the debate about the units of selection in the philosophy of biology, the debate about quantum information in the philosophy of physics. Anybody who wants to argue that experimental philosophy is not philosophy faces a dilemma. She might propose a notion of  philosophy according to which these debates count as philosophical. But, then, the (psychological, empirical) debate about what causes people to make the judgments about intentionality they make (etc.) will also count as philosophical. There are no salient differences between the later and the former debates. Alternatively, she might say that these debates are not philosophical, maybe because they are empirical. Much of my work, then, will not count as philosophical according to this definition. But given what gets then excluded (very exciting stuff) and given what turns out to count as philosophy according to this definition (no comment), I am glad my work does not count as philosophy.

Edouard

Comments

[Mark Phelan]

Hey Edouard,

Interesting post. Two Questions:

1: (Regarding the second project, from Thomas and Eddy) Doesn't traditional conceptual analysis require a capacity to distinguish conceptual competence from performance too? This doesn't differentiate the project T & E suggest from traditional conceptual analysis, then, and so doesn't distinguish it from philosophy. (Though both may be equally hopeless by your lights.)

2. Consider the reason you cite for why the kind of experimental work you do is philosophy: "Current philosophy is extremely diverse." In particular, you think it's impossible to draw the bounds of philosophy broad enough to allow in the debates you mention from philosophy of science and philosophy of psychology, but narrow enough to exclude experimental philosophy. But, do you think we can draw the boarders broad enough to allow in those debates AND NOT debates from, for example, developmental biology proper? (And, if so, how?) If we can't, of course, one can run the same reductio you run against Joshua's answer to the experimental philosophy question. (Of course, this is consistent with your response to the other horn of the dilemma faced by the philosophical purist...namely, the 'then count me out of philosophy' response.)

[Joe Paxton]

In attempting to defend Joshua's answer, it's important to point out that the empirical questions that Aristotle, Descartes, and Kant addressed in their lesser known works are of a qualitatively different sort from the kinds of empirical questions Joshua highlights. More specifically, the kinds of empirical questions that Joshua highlights are psychological questions (intentional action ascription, causal judgments, etc.), whereas the kinds of empirical questions that Aristotle, Descartes, and Kant addressed (at least the ones that Edouard mentions) aren't psychological questions.

Furthermore, psychological questions are connected to certain persistent philosophical issues in a way that most biological, cosmological, and geographical questions are not. More specifically, it's arguably the case that certain persistenet philosophical issues (especially those related to morality and aesthetics, possibly also causation, personal identity, consciousness, etc.) are, at root, psychological questions. The same is not true of biology, cosmology, and geography. There are relatively few persistent philosophical questions that are, at root, biological, cosmological, or geographical questions.

I'm not sure what conclusion is supposed to follow from this, though. I want to draw the conclusion that there is no clear distinction, in many cases, between philosophy and psychology. So, if you get a PhD in psychology, it's not as odd to think to think that you're engaged in philosophy as it is to think you're engaged in philosophy by getting a PhD in developmental biology. But I get the feeling that some folks (Edouard?) might be tempted to draw the conclusion that there are very few (if any) philosophical questions. If many persistent philosophical issues can be addressed through means that clearly are not uniquely philosophical, then why bother with calling any question "philosophical," other than to acknowledge certain historical figures who addressed similar questions and called themselves "philosophers?"

[Thomas Nadelhoffer]

Edouard,

Thanks for the interesting post. I am happy to once again have the opportunity to flesh out these metaphilosophical issues with you along with the other readers of the blog. For now, however, I am going to punt with the promise that I will enter the fray soon enough. I nevertheless wanted to point out that I do not currently place myself in any of the camps Eddy and I discuss in our PE piece. I suppose it's accurate to say that whether I am an analyst, cognitivist, or restrictionist depends on the domain--although, as I said in my last comment in Ron's post, most roads seem to lead to the skeptical camp of the restrictionist. In any event, my worry about EA is driven by the points you raise in your M&L piece. My worry about EC is that it becomes difficult to see how to keep most of neuroscience and social psychology from counting as philosophy. Finally, my worry about ER is that once the empirical spade work is done and the salient traditional views have been debunked, the retrictionist project winds down--i.e., to the extent it's proponents are successful, they eventually leave themselves with nothing to do.

So, for now, more hand-wringing...and more coffee...

[Thomas Nadelhoffer]

Oh yeah, one more quick point: The reason to care about whether what we're doing is philosophy--broadly construed--is that if it's not, we really ought to change the name of the movement, blog, etc. to avoid falsity in advertising! I, for one, don't mind being simply an experimentalist. But then my urge to go back to grad school in psychology returns with a vengeance!

[quantum anthropo]

Speaking philosophically, the first question for any attempt of experimental philosophy is How an experimental philosophy is possible? Hence, in Kantian manner: How a priori syinthetic is possible? Etc,etc.etc. Initial assumptions,thus,to my mind, are very important before any kind of "philosophical experiment".
Hence,logically,taking Philosophy seriously is not experimental ethnography.

[Antti Kauppinen]

Edouard,

In essence, you're offering a partners-in-crime defense for why empirical work on empirical issues should count as philosophy. For that to work, these lines of study should be both partners and criminal. It seems to me the examples you mention fail on either one or both counts. That is, the questions like theory-theory vs. simulation are not "thoroughly empirical" (and so not in the criminal business) and (indeed, by the same token) not partners either. Or they are partners, but only coincidentally pursued by philosophers (rather than exclusively empirical scientists), and so don't provide entry to the club - surely anyone agrees that the question of what the units of selction are is at best borderline philosophical. (Which is not to say that philosophers couldn't contribute to the biological debate in virtue of their specific training, as they in fact do.)

It is true that the methodological diversity of contemporary philosophy makes any border-drawing enterprise harder, but that as such provides little comfort for the experimentalist. The claim that "there are no salient differences" shows, I guess, that what is salient depends on where you're standing.

The real problem with experimental philosophy is not, of course, mislabeling - you can call it what you want. The problem is giving bad answers to misunderstood questions. Much recent empirical work on 'moral judgment' is a good example of this, as I hope to show in work in preparation, and I've already argued that the debunkers misidentify the assumptions that philosophers make when they appeal to intuitions, and thus fail to test them.

[Edouard Machery]

Thanks for these comments. I reply to them in turn.

Mark:

1. Yes, you are right. Both traditional conceptual analysis and experimental analysis require a distinction between uses/inferences/judgments that are constitutive of a concept and those that are not. Both are in the same boat.

2. This is a clever point, but there might be a way of excluding phds in, say, evo-devo from philosophy. Rather than focusing on the questions or topics addressed by philosophers and non-philosophers, a possible strategy might be to focus on the methods used by philosophers: These methods tend to be less empirical.

Of course, experimental philosophy tends to blur the line. Still, it remains that the use of empirical methods do not have the systematicity that they have in science. Experiments by experimental philosophers are typically enmeshed in theoretical arguments. Naturally, under this view, the difference between experimental philosophy and science is a matter of degree (more or less systematic experimentation, more or less theoretical arguments). But, being a Quinean, this conclusion pleases me.

Joe

1 You are right that the questions highlighted by Josh are psychological, while the questions I highlighted are not. that's the very point of my comment. Simply being a question of interest to a philosopher (particularly, in past centuries) does not make the question philosophical, otherwise issues in developmental biology turn out to be philosophical.

2. You write: "psychological questions are connected to certain persistent philosophical issues in a way that most biological, cosmological, and geographical questions are not". That's just not true for biological and cosmological questions. Consider issues in the philosophy of biology and in the philosophy of physics. Many issues about development, nature, and so on, have clear antecedents in Aristotle.

3. For full disclosure, I think questions that are not empirical (however indirectly) are empty: They have no truth-value.

Thomas:
Points taken.

Edouard

[Edouard Machery]

Antti

I can't let this pass:
"surely anyone agrees that the question of what the units of selection are is at best borderline philosophical"

The question of the units of selection has been at the center of the philosophy of biology for now 30 years. Dozens of papers have been written on it by philosophers (including in some of the recent issues of Philosophy of Science); that is a classical topic in any grad seminar in the philosophy of biology, there is regularly a session on this question at the meeting of the Philosophy of Science Association.

I suspect that your notion of what is philosophical is too narrow to include those debates, controversies, etc. that dozens of other philosophers (if not hundreds--go to PSA) count as philosophical.

And if that's the notion you want to use, I am happy not to do philosophy.

A detail: I don't understand why you say that "the questions like theory-theory vs. simulation are not "thoroughly empirical"." Both positions are empirical claims about how we ascribe mental states to others--how could the debate between them fail to be empirical?

Edouard

[Eric Schwitzgebel]

Alison Gopnik wrote in 1998 (in a paper with me, but it's her line) that philosophy is just very theoretical anything.

I think that's pretty close to the mark. If you're addressing the most fundamental conceptual, theoretical, normative, existential, etc., questions in any domain -- whether you are informed in your answers by conceptual analysis, empirical fact, religious texts, or whatever, you're doing philosophy.

[Eric Schwitzgebel]

P.S.: I also express this view at The Splintered Mind in my discussion of the philosophy of hair.
http://schwitzsplinters.blogspot.com/2006/07/philosophy-of-hair.html

[John Bengson]

Edouard,

Your post raises a host of interesting issues. I was especially curious about your suggestion that philosophy can proceed without intuition. How do you propose we do this, given that at least some (most or even all, on some views) philosophical questions are modal in nature?

Perhaps you think that the solution lies in the notion of a posteriori necessity: we simply discover necessities empirically. But it's been argued (e.g., Bealer in "The Philosophical Limits of Scientific Essentialism" and more recent work, and Korman in "Law Necessitarianism and the Importance of Being Intuitive") that establishing a posteriori necessities requires the use of intuition. I find these arguments convincing. You must not, if a posteriori necessities are your solution. If you don't mind sharing, I'd be interested to hear what you think is wrong with such arguments.

[John Bengson]

I should add that I ask because I don't find opening the journals you mention especially helpful. All of the (admittedly few -- the last was Grimm's paper in the Sept 2006 BJPS, I think) articles I've read in them rely on intuition, i.e., on how various examples or claims seem.

[Edouard Machery]

John,

It is useful to distinguish two claims:

1 Every problem that currently counts as philosophical can be tackled without appealing to intuitions.

2 There are numerous problems that for at least a large number of philosophers (if, apparently, not for everybody--see Antti's post) count as philosophical that can be, and typically, are tackled without relying on intuitions.

1 is obviously false, partly for the reasons you mention.

I contend that 2 is true. For instance, most philosophers of science rely on case studies (examples of real scientific episodes), on making clear the notions used by scientists (roughly, what Carnap called explication), on developing formal models, or on evaluating the best evidence bearing on specific claims. None of these methods appeal to intuitions--if you take as a paradigm of appealing to intuitions the Gettier cases or the Godel case.

Now, you had a look at BJPS and found people appealing to intuitions. Because I am unclear what appealing to intuitions really mean for you, I cannot really assess this claim. I just note that you managed to choose a paper in BJPS that dabbles in traditional epistemology (referring to the Gettier cases in its introduction). This is an exception more than the rule, I think.

I opened the last issue of Philosophy of Science (with an excellent paper by... myself). The first paper by Bruce Glymour analyzes the nature of the models in population genetics and compare them to the causal models developed in the literature on causal reasoning. Glymour describes or characterizes different kinds of models. The second paper by Gualtiero Piccinini replies to my 2005 paper: The issue is whether the class of concepts in psychology is a natural kind. The 3rd paper is my reply. These two papers mostly clarify specific claims and evaluate the relevant evidence. The 4th paper describes a case study from Pierre Curie in order to distinguish two ways of appealing to symmetry in physics. (Etc.) Moral: nothing even vaguely similar to the Gettier cases.

Upshot: If we are led to conclude that relying on intuitions is not an appropriate methodology, we might have to reject *some* philosophical issues (those under the heading 1 above). But there are other issues that require other methods.

Of course, you might not find these issues interesting. Fair enough. Then, you can hope that the antecedent of the conditional is not true.

Edouard

[Antti Kauppinen]

Edouard,

Are you trying to say that the question whether natural selection acts on genes, organisms, or perhaps also groups is not in the first instance a biological question? That much I would have thought obvious. As it turns out, it is a biological question that cannot be answered without relying on a number of distinctively philosophical assumptions about the nature of causation, the virtues of theories, and so on. So it is no surprise that philosophers of science have deployed their expertise and participated in the debate. But it doesn't by any means follow that when Dawkins says it is genes that get selected for he is making a philosophical claim.

The fallacious step in your reasoning is the move from "Philosophers of science spend a lot of time talking about X" to "X is a philosophical problem". Of course philosophers of special sciences, whether biology, physics, or psychology, engage with live debates in those sciences, especially those that take place at such a level of abstraction that they cannot be simply settled by further observation and experiment. If they didn't, they would hardly be philosophers of biology, for example. But that doesn't make those debates philosophical ones. Indeed, the philosophers of biology I know would be delighted if their work contributed to progress within biology itself. I wouldn't want to define such joy out of existence a priori by declaring that their very involvement turns the issue into a philosophical one.

(Clarificatory note: when I say that philosophers of the special sciences bring general philosophical accounts of causation, explanation, and so on to bear on particular scientific debates, I do not mean to suggest that it is a one-way street - the general accounts will also be informed by the actual practice of science.)

[Edouard Machery]

Antti,

I do not see anything fallacious in the inference from: x is a topic/question/problem dealt by group of philosophers Y using philosophical means to x is a philosophical problem. I'd be curious to see what such an argument would look like.

Furthermore, it is fallacious to conclude that if problem/issue/topic x is biological (psychological, physical, logical, linguistic, etc.), it is not philosophical. Obviously, it can be both, can't it? Otherwise, you would define out of philosophy many, many topics and the debates around them.

Edouard

[quantum anthropo]

Let us suppose that real philosophical problem has no history, correspondingly, some unsolved problems of science are philosophical,i.e.they could be intuitively associated with "the limits of reason". Hence, experimental philosophy is transcendental.

[jonathan weinberg]

I don't think there's much room to challenge the philosophical status of the work that Edouard approves of & practices. It's not like philosophy, as a profession or as an intellectual discipline, has any even remotely fixed boundaries. As a general rule, something that philosophers work on, in a manner historically continuous with what has gone on under the banner of philosophy in previous decades(/centuries/millenia) , is going to count as philosophy.

A more important worry, I think, is whether -- regardless of whether what he himself is doing counts as philosophy -- Edouard is throwing out too much philosophy with the bathwater. It's hard to see what room there is for hoary old questions about free will, moral goodness, epistemic normativity, and the like. And I don't see why the empirical challenges to particular philosophical _methods_ should have the implication that all these philosophical _issues_ are themselves hokum. Unless, of course, one thinks that recent analytic methodology is simply _the_ way to attack such questions, but such a claim strikes me as highly suspect.

[Edouard Machery]

Jonathan,

It would certainly be a mistake to claim that *all* the issues that are currently tackled by means of intuitions can't be handled by different methods. At the very least, I do not see how the argument would go.

But, it is unclear how to address at least *some* issues if one does not rely on intuitions. The best case might be the nature of reference.

By the way, don't you think that Steve Stich's and others' attack against the method of intuition is an indirect way of arguing that at least some philosophical issues are hokum, as you say (new word for me). This is certainly how I have always thought of our work on reference.

Edouard

[jonathan weinberg]

At a fairly fine-grained level, there will be some debates that will have been motivated entirely by the kinds of intuitions under attack here, and these debates will indeed come out as hokum, on my line. (I suppose that's my view about the issue of what is required for knowledge beyond JTB.) But this isn't going to be true of any big questions, I think.

As for reference, I take the lessons of y'all's papers to be (i) it's bad to base your theory of reference on intuitions; (ii) it's bad to try to use a theory of reference to answer substantive questions in other areas (like scientific change or the conservation of race talk); and from that it follows that (iii) it's _really_ bad to try to use an intuition-based theory of reference to answer substantive questions in other areas. But that does not show that those other substantive questions aren't good ones. And it does not show that there aren't interesting issues about reference -- I would take it that at a minimum one could treat "reference" on much the same model as, say, "gene", but where its home sciences are linguistics and cognitive science, instead of population genetics and developmental biology.

[quantum anthropo]

Indeed, some philosophical issues are not Wittgenstein's fictions predicted by Minkowski's Weltpostulat. May be they exist into some pre- logical space of transcendental intuition in the form of unsolved mathematical, physical, biological or even psychological problems...
Thus, experimental philosophy as some kind of new experimental transcendentalism can be possible. Hence,probably, main problem of such sort of experimental philosophy is to find Simple and Clear Experiment in new taste to identify philosophical problem.