Blog Coordinator

« Clark on Waller's Semi-Compatibilism (in Reverse) | Main | Attributability and Daddy Issues »



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

Tamler, you should totally test whether an experimental philosopher would react like that.

I think you make some telling points. But I'm not convinced by the main claim. Why is the intuition that x-philers test the wrong one? It's not the one you want them to test, but why should they care? They're not testing whether penguins have free will either. Personally I think the penguin question is cool, but I'm not expecting them to take any notice. In fact, the question they are testing seems to me the right one, given my views. I care about whether someone like Jerry Coyne is right in saying that ordinary people think that free will is incompatible with determinism, and I think that asking ordinary people whether they believe that is the right way to go. I also care (a tiny bit) whether the consequence argument is valid, but I doubt very much whether polling the folk is going to help me to discover whether it is. So I think the x-philers are probing just the kind of questions on which probing might be useful. Given the not implausible assumption that philosophers formulate arguments like the consequence argument to vindicate their prior intuitions, probing these intuitions might also bear, indirectly, on metaphilosophical issues. Again, they won't bear all that directly on first-order philosophical issues, but that's a limitation of x-phil, not of this or that way of doing it.


seems relevant.

Neil, I'm not asking anyone to test what I think is cool. But when you say that the point of a particular study is to test the intuitions that philosophers are appealing to in their arguments, then you should probe for those intuitions, right? If they said that the goal of the study was to test Jerry Coyne's claim about ordinary people's belief about determinism--assuming he makes that claim--then I'd be fine with that question too.

Tomkow, why does your post seem relevant to you? Because you mention experimental philosophy? Otherwise, I don't see any connection.

I hope this is clear from the dialogue and from all my previous work: I have no problem with X-phi in general. Not at all. I just have a few specific problems with the way most people are going about testing intuitions about free will and moral responsibility. And make no mistake, Josh Knobe is God.

Tamler, now I want evidence that the x-philes take themselves to be probing something other than what they are actually probing. As I read the Nichols/Nahmias debate, for instance, its about the Jerry Coyne question, not about a first-order philosophical issue within the free will debate. I don't mean that as a criticism, btw - philosophers do lots of things that have very little connection - before they get interested in them - with existing debates. So I don't mind x-philes probing the Jerry Coyne question. I don't even mind if they call that philosophy; in fact, I think they have every right to call it philosophy. It bears roughly the same relationship to the fw debate as, say, the debate over theory of mind bears to theories of mental content, which is to say, just a bit. But in neither case is that a reason not to engage with these issues.

Wow, that bit about Joshua Knobe not being God...that was delightfully humorous. Great share!

That was a wonderful dialogue! I just can't stop laughing. Just a remark: i think that the idea that we should also test for the premises of arguments and not just for the conclusion is already in Nahmias et al. 2006.

Neil, the other big problem with probing for the compatibility question is that it leaves you vulnerable to the 'who cares' question? The whole point of arguments is to give people reasons to endorse a conclusion. That's why incompatibilists and compatibilists develop them. If you just ask people what they think about the conclusion before they've heard any arguments, then I think the 'it's polling, not philosophy' objection gains credibility. But if you probe for intuitions about the crucial and controversial premises, then you find out something about the plausibility of the arguments.

Nick and Florian, thanks. Florian, yeah, Eddy and those guys discuss this as a possible objection in that paper but their response isn't (in my view) satisfying. I discuss why in my Phil Compass paper 'Experimental Philosophy and Free Will.'

"Why should I care about mirror neurons? That's neuroscience, not philosophy."

"Why should I care about developmental systems theory? That's biology, not philosophy."

"Why should I care about Arrow's theorem? That's economics, not philosophy."

"Why should I care about what the folk think? That's sociology, not philosophy".

The right answer in each case is "why should I care whether you care? That's psychotherapy, not philosophy."

Neil, very catchy, but you're either intentionally or unintentionally refusing to address the substance of the criticism. People care about a lot of strange things and I don't need to know why. The issue is whether it's worthwhile to test intuitions about the compatibility question itself. Nothing you've said explains why you think it's worthwhile.

Tamler, you're right I was being flippant. I was being flippant, in part, because I doubt that the "why should I care" objection needs a good response. Sometimes we can give one - you should care about X because you care about Y, and X is relevant to a proper understanding of Y. My point is that in many of these cases, the link between some things that some philosophers (=people who work in philosophy departments) do and more traditional philosophical questions is tenuous. I meant, say, DST as an example of such an X which is tenuously connected to traditional Ys. When we can give an answer to the "why should I care" question, that's all well and good, but when we can't, that's fine too. I don't care whether you care about some of the things I do (professionally). Indeed, I confidently predict that you don't care about some of the things I do professionally. I don't think that's an objection to what I do; if you do (not that I think you do, really), my answer is just "get over it". I can prove it's interesting, after all (I can point to people who are interested).

I would both agree with Neil and Tamler. It is true that we should not only test whether people directly consider free will and determinism to be compatible, and that we should investigate their intuitions about sophisticated philosophical premises.

Nevertheless, I think that Neil is right in saying that there are reasons to care about whether people are prima facie compatibilists or incompatibilists. There are well-know dismissal of compatibilism that are based only on the fact that incompatibilism is prima facie obvious (e.g. Kant or William James). Now, I understand that in the current debate about compatibilism, no one longer argues this way. But I think this is true only for specialists of the free will debate, and that many non-specialist philosophers are incompatibilist because they just think incompatibilist is obvious. I have (for example) written a book in French about X-Phi in which I analyze french philosophy textbooks (for high-school students). The most striking things was that, 100% of these textbooks took incompatibilism to be an obvious premise and did not mention compatibilism as an option (except Spinoza-style compatibilism, that I consider more to be hard determinism). This parallels my own experience of teachers beginning a lecture about free will with the proposition that is was obvious and undeniable that free will and determinism were incompatible.

For this kind of reason, I think it can matter whether we show that incompatibilism is not obvious. If we can, then I think many non-specialist will consider compatibilism more seriously.

Another remark, about why the industry has focused on direct intuitions about compatibilism: my thought is that he made sense to first investigate whether people are directly compatibilist or not. It seems a sound first step, investigating per se. The problem is that Nichols & Knobe's Nous paper showed that the question was not this simple and has generated a psychological problem that has in itself interest and has drawn much attention.

Nevertheless, I got your point. I promised you'll see soon some experimental philosophy that takes it into account.

Neil, like I said, the issue is not whether I care--I don't care about a lot of things that are worth caring about for other people. The issue is whether it's worth caring about, whether it's an effective way to deepen our understand of the topic. I guess we just disagree on a very meta-level about when a project is worthwhile, which has nothing to do with X-phi in particular. This reminds of that discussion on Leiter about Kitcher's essay "Philosophy Inside OUt". I think that philosophers have some sort of obligation to explain the value or their work, why it is of interest. (I'm pretty sure a lot of X-phiers agree with that).

All that said, I agree with Florian (and a number of people made this point at the APA too) that there's value in finding out starting intuitions on the compatibility questions. Florian, you're right that many people take incompatibilism as self-evident and it's good to show them that it isn't. Not only that, but if it turned out that people were, say, incompatibilists at the outset, but found certain crucial compatibilist premises intuitive, then the starting intuition would still matter for reflective equilibrium or some sort of all-things-considered judgment. I still think though that the proportion of studies devoted to the compatibility question is completely out of whack (in large part because it's become 'industry standard). But maybe that's changing--sounds like it is. Anything in the works on your end?

I tend to agree with the “who cares” position about folk intuitions concerning the compatibility question (actually, about any question). I’ve never heard of a topic that has been moved forward by considering base intuitions completely uninformed by arguments or evidence. Granted the results might be interesting for themselves, but still completely irrelevant to the truth. I’ve done some quick math in my head, and it’s only a rough number, but I think 97% of all folk intuitions are completely wrong and useless. And if we’re right, we’re inevitably right for the wrong reasons. And if we happen to be right for the right reasons, well, that is what we call dumb luck. In any case, intuitions are the eternal enemy of reality.

For example, you’ll recognize the following intuition that gets compatibilists every time:

“Well it sure as heck fire feels like dem dare villins deserve some kinda woopin’. And I donna wanna hear no mention as to that dee-terrr-minitis or whatever it’s called neither. It just don’ feel right leddin’ dem fellas off the hook.”

Nice! As everyone has already pointed out, this dialogue is really astoundingly funny and well written, and it raises some very important philosophical issues that have the potential to move this discussion in helpful new directions.

I'm not quite convinced, though, that it is such a good idea to put so much emphasis on the distinction between 'premises' and 'conclusions' of arguments. It seems to me that this might misrepresent what actually goes on in philosophical reasoning.

For example, suppose I previously believed in free will, but you show me that two other propositions I previously believed jointly entail that there is no free will. I might therefore change my view about the existence of free will, in which case I would probably frame the argument like this:

Proposition 1
Proposition 2
Therefore, there is no free will

But I could equally well take things in the opposite direction, treating this discovery you made as a reason to abandon one of my other two beliefs, in which case I would switch around the premises and conclusions:

There is free will.
Proposition 1
Therefore, Proposition 2 is false.

So maybe the best way of describing the force of the argument would be to ignore the distinction premises and conclusions and simply to put it like this: The argument shows that I have to either change my view about free will or abandon one of those other two beliefs.

If we put things in this way, then it seems that there is just as much reason to consider intuitions about the 'conclusion' as to consider intuitions about the ' premises.' What the argument basically does is force you to choose between accepting the conclusion and rejecting the premises, and to make this choice, you have to consider both types of intuition.


Josh, thanks. I think you're right but only to an extent. Often the conclusion of an argument is far more complex than the premises that lead us there and so we have reason to trust our intuitions about the latter more than the former. Take the Darwinian conclusion: very counterintuitive, right? But the individual premises that lead to it are extremely plausible. Is it just as rational to reject those premises and embrace the intuitive conclusion (Darwinism is false)? If so, then the creationists are right and we should give equal weight to each side.

Or take an example that's more analogous to free will. When I presented the dialogue, Thomas and Al Mele brought up the example of intentionality. I believe it was Thomas who said that when he asks students whether people can do things intentionally but not intend to do them, they all say no. But when he gives them concrete examples of this, they all agree that it's possible. Al then said (I believe) that the intuitions about the counterexamples better reflect what the students really think about intending and intentionality--and I agree. It's a simpler more basic intuition and so we have more reason to trust it. And that's how arguments tend to work. Simpler propositions that lead to more general and complex conclusion.

None of this of course means that we should disregard our intuitions about conclusions entirely. They clearly have some weight as I mentioned my last comment. But it seems like a lot more of the focus should be on the premises.

Hi Tamler,

This last comment of yours is a really helpful one. So it seems like the key distinction is not between premises and conclusions per se: it is rather between simpler and more complex propositions. Then, to the extent that the premises end up being simpler than the conclusions in a particular case, we'll end up paying more attention to intuitions about the premises in that particular case.

While we're discussing this topic, I also wanted to ask you about one other aspect of your thoughts on these issues. (I really do mean this as a question and not as an objection to what you say in the dialogue.)

I agree that experimental philosophers have mostly not been investigating people's intuitions about the premises that earlier philosophers used to argue for compatibilism or incompatibilism. Instead, experimental philosophers have been looking at the reasoning that leads ordinary folks to believe in either compatibilism or incompatibilism.

Still, this work does seem to be examining people's intuitions about certain premises. It's just that the work is looking at intuitions about different premises -- not the premises in the arguments of philosophers but the premises in people's ordinary reasoning. For example, Nahmias and colleagues have been looking at intuitions about bypassing (, while Shaun and I have been looking at intuitions about the nature of the self (

Are you thinking that it is a mistake to start shifting in this direction? For example, are you thinking that it is a mistake to shift from looking at intuitions about modal logic (which figure in philosophers' arguments) over to looking at intuitions about the nature of the self (which seem to figure in ordinary reasoning about free will)?

Hi Josh, you ask:

"For example, are you thinking that it is a mistake to shift from looking at intuitions about modal logic (which figure in philosophers' arguments) over to looking at intuitions about the nature of the self (which seem to figure in ordinary reasoning about free will)?"

Right, exactly. I think X-phi should only focus on intuitions about modal logic. Either that or intuitions about mereology, supervenience, and the substance dualism/ property dualism distinction.

But seriously, like you said in the previous comment, the simpler the better. That's what I love about your CEO/environment experiment. When I'm talking about testing for intuitions about premises, I'm thinking more of premises about cases in manipulation arguments--e.g. four case, Zygote etc--or maybe Strawson's basic argument (although Eddy and I tried that with little success). And I have no problem with testing for intuitions about bypassing--I like Eddy's angle there--but I think you can do that without bringing in the compatibility question. (The Sripada study that I mention in the dialogue seems to do a version of that.) And of course I'm all in favor of probing for intuitions about the self too. I just don't think descriptions of determinism need to be involved.

Hi Tamler,

Just a quick clarification about my previous comment. The point is not just that experimental philosophers are switching from intuitions about one question to intuitions about another (say, from intuitions about modal logic to intuitions about the self). Rather, the point is that experimental philosophers seem to be concerned about these intuitions in an importantly different way.

If you look at the studies that Nahmias and colleagues have done on bypassing, it's not just that they are looking at intuitions about certain propositions that would -- considered in the abstract -- entail something interesting about free will. Instead, the claim is that it is these specific premises that explain why people actually are troubled by the free will question. So Nahmias and colleagues say that they are really getting at the root of the uneasiness that so many people feel about this issue.

Similarly for the studies that Shaun and I have done on intuitions about the self. It's not just that we are studying intuitions about the compatibility question and then, separately, intuitions about the self. The hypothesis is that people's intuitions about the self actually *explain* their intuitions about the compatibility question.

I'm not sure quite how to think about these issues, but Nahmias has certainly made a pretty good case for the claim that a lot of the things that philosophers traditionally worried about (deterministic laws of physics, etc.) are just a distraction and that he and his collaborators are getting at the real heart of the issue here.


I have a lot of admiration for Eddy Nahmias's work, including the stuff I have seen on "bypassing". I may simply not have seen the same stuff you have, and I'm certainly and absolutely not an expert on the X-Phi literature. But I frankly haven't yet seen anything by Eddy or anyone, and I would be skeptical that anything could convince me, that "a lot of the things that philosophers have traditionally worried about (deterministic laws of physics, etc.) are just a distraction and that he and his collaborators are getting at the real heart of the issue here". Wow--maybe this is correct, and I look forward to being educated into the "real heart of the issue"--whatever that will turn out to be.

But I'm thinking that this might be just a slight exaggeration, no? But I'm certainly open to learning that most or even all of the traditional questions are distractions; I just haven't yet been convinced.

John, here's a story for why Eddy's work should interest you. Suppose that Eddy explains why many people come to be incompatibilists. They start from intuitions that are caused by a confusion of determinism with bypassing. Those intuitions then bias their response to incompatibilist arguments; though they no longer confuse determinism with bypassing, that confusion has played a role in their finding the consequence argument plausible. Now a story like that could play a role in moving beyond dialectical stalemates. You agree that the issues are tricky and finely balanced with regard to the consequence argument: a debunking explanation which accounts for even some small degree of the plausibility of the CA should therefore shift the debate significantly. Even if Eddy can show only that some significant proportion of individuals confuse determinism with bypassing, that might play a role in one's own assessment of the state of play: each individual should discount their CA intuitions (slightly) in light of the fact that there is some probability that those intuitions owe their power to a past confusion. Of course, this story is contestable; for one thing, it depends on the claim that my intuitions now are affected by past mental states of mine even though I no longer possess those mental states. But that claim is plausible.


I agree with you – the deterministic laws of physics aren’t simply a distraction. Instead, those laws are the *main* issue in the FW debate. Until mankind proves that new forces are an emergent property of human thoughts (i.e., new forces are created that aren’t simply a direct sum of existing forces), all of the other arguments supporting FW won’t hold any water. The neuroscience viewpoint will continue to “net out” those arguments.


Josh can obviously speak for himself but I don't think his point was that "the traditional questions are distractions." He was saying that Eddy's work might show that determinism is not the threat that some people think it is. I would think you would agree with that--esp. when the threat concerns moral responsibility.

Josh, I have no problem with X-phi examining the mechanisms underlying free will related beliefs and worries--on the contrary, I'm a huge fan of that work. I just don't see why the focus always has to be on the reasoning behind intuitions about the compatibility question. After all, most people aren't troubled by the compatibility question. It doesn't really come up. They're troubled by far more specific threats to free will and moral responsibility. Why is it supremely important to explore the sources of intuitions about the compatibility question?

Hi all,
Tamler's lucky his dialogue is so funny, or some people might have found it obnoxious (not me, though!--I slept in the same room with him after he delivered it and didn't even harm him at all, as far as he knows. We were just saving hotel money at the APA; nothing untoward... as far as I know).

Thanks to Neil for explaining some of my ideas better than I. And they're not just my ideas. I want to make sure Dylan Murray gets his creds--he is, in fact, first author on our paper that was just accepted to PPR (we'll post link to online pub once it's up).

In that paper, we address some of the concerns Tamler and others raise, including explaining why folk intuitions matter (drawing on the WRE approach favored by people like John) and explaining why our results regarding choice and the ability to do otherwise are relevant to incompatibilist premises and the Consequence Argument. I agree with Josh and others, contra Tamler, that we need not care *only* (or mainly) about intuitions about premises or principles (like Transfer of non-responsibility), but like others, I have been working more on intuitions and concepts regarding more than just the "broad-level" compatibility question--for instance, questions about how people understand choice and ability, manipulation cases, and Frankfurt cases. There's a sense in which Tamler's dialogue is a relic of the 2000s--people have already moved beyond where he begs us to move beyond.

Finally, as he knows (from our own attempts), trying to probe people on their intuitions about premises in complex arguments is not an easy task. To demonstrate, I'll post one pretty lame attempt I made to test what people think about the sort of skeptical argument advanced by Sam Harris in his new book...

Here's a survey I did on 193 undergrads at Georgia State.

Some scientists argue that (1) free will is an illusion.
They reach this conclusion by arguing (2) that free will requires that you can control which specific thoughts and desires come to mind so that you ultimately control what you think and want, and (3) that it is impossible for anyone to control which specific thoughts and desires come to mind--eventually, they have to come from somewhere beyond your control.
In arguing in this way, they do not claim that your thoughts and desires have no influence on what you do. So, they say that, even though free will is an illusion because what you think and desire is ultimately beyond your control, (4) what you think and desire still influences what you do.
There are three ways that one might react to this argument offered by some scientists, and we are interested in knowing which reaction you have. Please indicate which of the following you agree with most:

(A) They are right that (1) free will is an illusion, basically for the reasons suggested here (i.e., 2-3).
(B) They are wrong about (1). Free will is not an illusion, because (3) is mistaken; you can control which specific thoughts and desires come to mind.
(C) They are wrong about (1). Free will is not an illusion, even though (3) is right, because (2) is mistaken; free will does not require that you have an impossible ability to control which specific thoughts and desires come to mind. Instead, free will only requires (4) that what you think and desire properly influences what you do.
(D) I am not sure if they are right or wrong because I am not sure about how the argument presented here is supposed to work or I am not sure what it requires to have free will.

Here are the results:
A: 27 (19%)
B: 40 (21%)
C: 107 (55%)
D: 19 (10%)

Now, I'm not sure what to make of these results, but I was surprised by how many of them gave the "compatibilistesque" response C and how few gave the "libertarianesque" response B. (I predict if I surveyed people who read this blog, the results would be very similar, except more people would give response D to indicate they think the argument or alternatives are poorly phrased.)

Did I manage to test how the folk think about the premises of an argument? Even if it were better constructed, would this sort of experiment provide better information than ones that test responses to questions about agents in deterministic universes?

The comments to this entry are closed.

Books about Agency

3QD Prize 2014: Marcus Arvan