Blog Coordinator

« Nahmias and Murray on Bypassing | Main | Flickerers Getting Funded »



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


This is extremely helpful! This was just the sort of thing I was hoping to get, although I didn't imagine that it would be so detailed and worked out.

btw, Ravizza and I are proponents of the WRE approach--we say this explicitly in Responsibility and Control. I think part of the issue has to do with just how seriously we want to take untutored folk intuitions. I agree that they are part of the equation, but I'm inclined to think that in seeking the WRE, we have a certain amount of leeway here. That is, as theorists, we don't need to be held in a straightjacket of ordinary intuitions. But I agree that we don't want our theorizing to be TOO free of ordinary, perhaps reflective intuitions (I like Rawls' term, "considered judgments").

Ok, I'll do my homework and think about your very thoughtful and comprehensive post,

Just to be slightly more explicit: in using the metholodology (or really family of metholodogies) that aim at a WRE, we might be more or less inclined to hold fixed intuitions. Some will want to hold fixed the intuitions (at least initially) and will be more inclined to adjust their general principles; others will want to hold tight to their principles and will be more inclined to give up (untutored) intuitions. So there is a lot of leeway or variations within the rubric of WRE-type methodologies.

From my perspective, even if it were quite clear that most of the folk think that, if a mental state is in place as a result of causal determination, it can't really be the individual's or is in some way not genuine, I would not place much weight on this view. I would place more weight on theoretical considerations and other intuitions.

Presumably most of the folk, and most of us highly educated (but no doubt nevertheless screwed up in various ways) academic folk think that moral responsibility requires freedom to do otherwise (alternative possibilities). And yet, upon reflection, I want to give this view up. I give it up in light of intuitions about Frankfurt cases and other theoretical considerations.

I think the appropriate methodology gives some weight to considered judgments and some weight to theoretical considerations, and it is irreducibly holistic and not subject to encapsulation in an algorithm

Not sure if this benefit has been covered elsewhere, but it's the benefit that most excites me:

Understanding what our folk intuitions are is important, because then we can try to understand why our folk intuitions evolved to be slightly out of step with reality. An understanding of the evolutionary cause of the specific mismatch could shed light on many other questions.

Although I'm not a Leeway Theorist w.r.t. the first-order conditions for moral responsibilty, I think that from a methodological p.o.v., we should have a lot of leeway with initial intuitions. So, for example, consider famous thought-experiments, such as the Trolley Problem and Thomson's Violinist Case. About the Trolley Problem, after considerable reflection over the years, I have come to the conclusion that, arguably at least, there is no moral difference between the Bystander Cases (I and II, i.e., five on the track ahead and one on the track ahead) and FAT MAN. Of course, this goes against what JJT reported that "everyone she has talked to" thinks--i.e., everyone initially thinks it is NOT ok to push the fat man over in front of the oncoming trolley. And maybe that's right; but I don't see how to distinsuish FAT MAN from the Bystander cases. (I argue this in "Thoughts on the Trolley Problem.)

Also, consider JJT's Violinist Case. Almost everyone initially thinks it is perfectly ok for you to unplug yourself from the famous violinist. But upon reflection over the years, I've come to the conclusion that it is not morally ok for you to unplug yourself (I argue this in "Abortion and Self-Determination" and "Abortion, Autonomy, and Control of One's Body"). I further argue that, because of crucial differences between the violinist case and the context of a pregnancy due to rape, it certainly does not follow from its being impermissible to unplug yourself in the case of the violinist that abortion would not be permissible in the case of rape (even on JJT's working hypothesis that the fetus is a person).

So here are just two cases of thought-experiments in which I am willing to depart from my own initial considered judgments and also presumably most reflective, professional philosophers' considered judgments. So afortiori I would not wish to be bound by "folk intuitions". Let's say I'm driving on the freeway to LA, and I text a driver in a blue Toyota in the lane next to mine, and I ask, "Do you think causal determinism is compatible with moral responsibility", and she replies, "Hey, dude, watch the road, but, well, no." Why should I care exactly?

Again: worm can-opener in use: sorry--I wish I could be as cool as you Experimental Philosophers!

To continue. I do not take what I have said in my previous contributions in any way to denigrate the importance of getting data on intuitions in philosophy. That's important for all the reasons that have been mentioned by Eddy and others (so helpfully). But I do wish to say that it is just a starting point, or one input, into theorizing--we shouldn't end there or take (initial) intuitions (or perhaps a particular subset of reflective intuitions) to be dispositive of philosophical questions. This is perhaps not very contentious, but sometimes I feel it gets lost in the methodological shuffle and excitement about this new study or that one.

I wish I could be cool! I wish I could be cool!

STOP THE PRESSES! John Fischer now believes "we have a certain amount of leeway"! :)


But I explicitly said, "Although I'm not a Leeway Theorist w.r.t. the first-order conditions for moral responsibilty, I think that from a methodological p.o.v., we should have a lot of leeway with initial intuitions." So don't get your hopes up, Kevin!

A boy can dream, John. :)


This is a great post. Thanks for putting so much time into it. I don't have strong feelings about X-phi one way or the other, but for the sake of discussion, let me respond as a skeptic might to some of your reasons.

Before that, though, let me make two points. First, I suspect when people ask, "Why should I care about folk intuitions?", there is a lot of emphasis placed on the *I*: "Maybe people foolish enough to make assertions about what the folk think should care about what the folk actually think," (these skeptics might continue) "but why should *I* care?"

Second, it's not always clear whether the skeptics are claiming that it wouldn't matter what the folk think even if we could figure it out, or whether they are claiming that the methods of X-phi are just never going to figure it out. (Both would be reasons not to care about X-phi, I suppose, but they are distinct.)

With these two points in mind, let's look at your reasons.

1. "Because most philosophers in the debate care about what the folk think, and if they don’t, they sure do a good impression of someone who cares. So, we might as well test their claims for accuracy."

But why the 'so'? It seems to me that most times philosophers make (ill-advised) claims about what the folk think, those claims are not essential to the philosophical point being made. The quote from van Inwagen you give is an example of this, I think -- presumably it doesn't *really* matter to him whether everyone's experience teaching undergrads is the same as his. But if that's right, then why care about whether these empirical-sounding claims are accurate?

Moreover, I'm doubtful that when philosophers appeal to intuition to support premises or thought-experiments, they are actually trying to make a claim about what the folk think. They seem rather to be simply asking their interlocutor: "Well, doesn't this seem true to you?" And if they really are trying to say, "A reason to believe this premise is that most of the folk think it's true", then they shouldn't be.

2. "A good theory of free will needs to know what our beliefs and practices are regarding attributions of MR and the sort of control and choice capacities we think are required for such attributions."

This sounds like the best reason to me -- or at least the one most immune to X-phi skepticism, so I'll skip over it.

3. "If WRE is a good way to go, then ordinary intuitions and practices will serve as one of the inputs to be considered in the process."

But doesn't this presuppose the value of ordinary intuitions rather than support it? That is, someone who advocates WRE will only think that ordinary intuitions should serve as one of the inputs if that person already thinks that ordinary intuitions are worth something (right?). Moreover, a skeptic might think that although there is something in the neighborhood of "ordinary intuitions" that is a worthwhile input to the process of WRE, the sort of untutored opinions we get from X-phi surveys aren't that something.

4. "Because if we think that people’s intuitions conflict or are muddled in such a way that revisionism will be required (see Vargas), then we need to know what the intuitions actually are that need to be revised and how they need to be revised, and we may also need to know the psychological sources of the intuitions to know how revisable they are and how to pull off the revision."

Okay, but again this seems to presuppose that we are already interested in folk intuitions. If I'm a skeptic about the value of X-phi, then I'm not going to be interested in how exactly to revise folk intuition. And should philosophers really be in the business of trying to revise folk intuition anyway?

5. "Because it is required to explore the psychological sources of people’s intuitions, beliefs, experiences, and theories of free will and moral responsibility."

But -- to put the skeptic's position tendentiously -- aren't these questions of psychology? The X-phi skeptic is, I would think, someone who is also skeptical of the relevance of psychology to philosophy, and so this doesn't quite seem like the right kind of reason.

Again, I'm just trying to express the skeptic's position, not endorse it.


You write: "And if they really are trying to say, "A reason to believe this premise is that most of the folk think it's true", then they shouldn't be."

While I think that your generic statement is often true, I think that it's false for certain species that the statement encompasses.

Specifically, consider the statement:

P1: "Free will is defined as X."

In my view, the fact that most of the folk agree with P1 *is* a reason to believe that P1 is true. Admittedly this is a very curious, and likely rare, situation (in which a proposition becomes more true the more people believe in it), but it is still real.

In other words, I think that common usage is relevant to defining terms, even terms of art, when those terms are shared by people outside of philosophy. I don't think that philosophers have a monopoly on the term "free will" (or moral responsibility, or autonomy, or whatever). And I don't want compatibilists to say that free will exists, if non-philosophers hear that and think that a different kind of free will exists - a non-existent kind.

This would be a minor point except that, once you get to the bottom of the compatibilism debate, it boils down to definitions. If free will is defined as A, then compatibilists win, if B, then incompatibilists (likely) win.

These two points: 1) the importance of definitions to the compatibilism debate and 2) the importance of common usage to definitions, are too often overlooked by the critics of Ex-Phi, in my view.

I don't see why we should conceptualize the philosophical debates about free will as debates about definitions, any more than the debates about any other interesting philosophical term or idea are about definitions. We want a philosophical account of the problematic term or idea; in this instance, we seek an adequate philosophical account of (say) "free will".


Good points well made, and many almost verbatim of what I say to my students!

However, two considerations. One, common usage, while certainly the most practical criterion of word meaning, can wildly change over time. This is particularly obvious in cases where prejudice (and even defiant self-identification) hijacks a common word like "queer"--and it does not rapidly give it back. That seems to show that purely social forces can warp what we mean by a word use, and thus "free will" could conceivably be subject to change by what people take it to mean. (One reason I keep bringing up how FW functions in insanity law is to demonstrate that social and political fashions have tended to reinforce and then erode the place of incompatibilist libertarian ideas of FW in American law, and the watershed is clearly the John Hinckley case.) Of course Manuel might be encouraged here! X-phi can at least show us to what extent a certain kind of concept of FW is "winning" (not the CHarlie Sheen sense), or whether we collectively just wander and and forth between compatibilist and incompatibilist ideas depending on context.

The second point is that I take it that compatibilism and incompatibilism are contraries, unless somehow FW skepticism can be shown false, particularly in a nihilist form. A number of philosophers believe that there simply is no coherent and clear concept of FW.

Alan -

Thanks! I don't remember agreeing with you so much.

I generally agree with the two specific points that you make. Regarding Richard Double style non-realism about free will - I think that surveys can shed light here, too. If there is no common usage, or if there is wide variation in usage, then free will (arguably) doesn't have a clear definition.


You distinguish between A) definitions and B) "philosophical accounts" of terms. It's not clear to me how a definition differs from an "account" of a term, much less how such a difference would be relevant with respect to the value of surveys. It's also not clear to me what the adjective "philosophical" adds to "account", as distinct from "definitions".

I think that, if compatibilists and incompatibilists are, at bottom, offering competing definitions of "free will", then it is absolutely crucial for both parties to admit that, and to conduct research accordingly. This is, in fact, what I think compatibilists and incompatibilists are doing.


As I understand things, a "definition" seeks to capture the meaning of the relevant term. Here a dictionary is helpful, although perhaps there are other resources that can be used with the aim of figuring out what the meaning of a given term is.

A philosophical account need not stick to the actual meaning. The account might seek to capture core features of how we tend to think of the relevant notion, but also to prune out problematic part of how we ordinarily think.

Also, one might distinguish between seeking to capture our ordinary ways of thinking about some notion and what the real essence of the relevant phenomenon is. So, the meaning of "water" and what what is (H20) pull apart in obvious ways. Now I'm not saying that "free will" is a natural kind term, but I think that "free will" picks out a "kind" that is in some ways analogous to a natural kind, and in some ways not, and that as philosophers one of our tasks is to articulate this--to get clear on the role freedom plays in our thinking and theorizing.

I do agree with Eddy, when he emphasizes that we can't SOLVE philosophical problems by doing surveys. The data from such surveys is part, but not the whole, of giving a philosophical account of the relevant phenomena.

John -

Who are you, and what have you done with Manuel Vargas?

At bottom, I think that your comments boil down to something like:

"Yes, common usage can be relevant to determining definitions, but philosophers like myself shouldn't be limited so much by common usage."

In response, I can't persuasively argue that common usage is 100% outcome-determinative of the compatibilism debate. My point here is simply that, in my view, skeptics of empirical philosophy have given too little importance to common usage, and to the role that surveys can play in determining it. I think that helps answer your question, which motivated this thread.

Because these skeptics have, in my view, given too little emphasis to common usage and surveys, they adopt an attitude toward Experimental Philosophy that smacks of hiding from empirical testing, at precisely the time when they should be embracing it. To my ear, the tone of your posts here (and elsewhere) is "even if all of the folk define free will as X, I don't want to be bound by what they say." In my view, you should be bound by what they say! A compatibilist who says that "free will exists", but means something radically different than what others mean by "free will", has achieved a hollow victory.

Of course, the devil is in the details, and much depends on subtleties of the philosophy of language...

A few quick points, but I hope some others will jump into this discussion, since it's really about what we think is the best methodology for theorizing about free will, and I'd like to hear more about what methodologies people think are best and why such methodologies do or do *not* require some systematic information about folk intuitions, beliefs, experiences, or practices and their psychological sources.

Neil, perhaps the best way for me to address your skeptic is to make it explicit that my reasons for caring about and testing folk intuitions were meant to be conditional:
If you make claims about folk intuitions, then you should test whether your claims are right.
If you use WRE or suggest revisionism, then you need information about the relevant inputs and how robust or revisable they are.
If you care about the psychological sources and reliability of the intuitions you refer to in your theorizing, then you should study this empirically.
And so on.

And if you don't care about any of the above, then don't you at least care about elucidating a theory of free will that helps ground attributions of moral responsibility (my point 2 that you agree with).

And if you don't care about any of this, then it looks like you're right and this skeptic need not care about folk intuitions or x-phi. But then I want to hear more about how this skeptic thinks we should proceed. So I was also trying to motivate the antecedents of the conditionals.

I worry that Kip and John are talking past one another a bit. In any case, I don't think I want to go quite so far as Kip regarding the role of folk intuitions or concepts in the free will debate. But I do feel like John's description of WRE sounds too insular. I can probe and alter my own intuitions about free will (or trolley or violinist cases) and try to find an equilibrium I find maximally consistent, explanatory, etc. Perhaps I also probe my fellow philosophers' intuitions and take account of them as I seek reflective equilibrium.

But especially if we want our theory of free will to provide explanations and guidance for real-world MR judgments and practices (reason #2), it seems like we should make WRE wider than our own philosophically informed (perhaps theoretically corrupted) intuitions. And to do that, it seems helpful to understand both the content of folk intuitions and their psychological sources, reliability, and robustness.

"But especially if we want our theory of free will to provide...guidance for real-world MR judgments and practices"

I might want MY theory to provide such guidance, but I certainly don't want anyone else's to be put to use that way!


Right. I did say that I agree with Nichols and Knobe, in the intro. to their very helpful book, Experimental Philosophy, that an important benefit of X-phi is to understand underlying mechanisms of thought. I think this is the most important benefit of studying patterns of folk intuitions, and it is considerable.

Also, WRE really means "Wide" RE--I take that seriously in all of the ways Norman Daniels talked about, i.e., embedding one's equilibrium in a wider context of natural and social sciences, and, I presume, an understanding of psychological mechanisms.

Kip: I would distinguish revisionism about concepts from revisionism about conditions of application. So I'm not sure I'm with Manuel entirely. But I do think we should not feel required to take all intuitions, folk or otherwise, to be fixed in our theorizing.


You ask about what our methodology should be. Here are some ruminations, no doubt rambling and inadequate (and not sufficiently developed, lamentably).

Apparently, many Americans think that there is no global warming. One might say, "So what?" These Americans are idiots and don't know what they're talking about; surely our theorizing about the climate should not take seriously these "intuitions" of the folk. Now, you will say, but such matters are about natural kinds, and free will is not exactly a natural kind.

Right, I agree. But attributions of freedom do play a distinctive role in our commonsense conceptualization of human behavior as well as our more reflective theorizing. John Perry calls "free will" a "human kind" (as opposed to a natural kind, such as water). Perhaps our task is to give an account of--an explicit articulation of--this distinctive role, including the connections we make between freedom and moral and legal responsibilty, and so forth.

If our aim is to give a perspicuous account of free will or freedom as a "human kind", it is not clear how much emphasis we will want to place on ordinary folk intuitions about free will. I'm not saying that our theorizing here should float free of all such intuitions; it is just not clear that their role should be central. And their role would not be to indicate the "meaning" of the term, "free will".

Anyway, this sketch is just a sketch; John Perry is working on a fascinating book project on free will in which he is developing roughly this kind of picture. By the way, I believe the book is entitled, "The Wretched Subterfuge".

I'll accept Eddy's invitation to jump in, though I share John's caveat that these thoughts are not fully developed.

I've always been somewhat perplexed by controversy over X-phi about free will, for a number of reasons. First, I don't have any intuitions about free will. I don't know what it is - if asked for a characterization, I'd say that capacity necessary to be morally responsible.

Second, and John might have been alluding to this, I find many x-phi results not to be about the folks' intuitions regarding the concept of free will, but about conditions of application - that is, they render judgments about whether or not agents act freely under particular conditions. (This may be unfair of me; I don't claim to have read extensively in x-phi.)

Third, I find in these discussions we are too quick to distinguish folk intuition from trained philosophers intuition. I'm still the folk - insofar as I spent a great deal of my individual development engaged in the practices of blaming and praising and holding others responsible long before I was a philosopher. More than that, I still live with and interact with many (mostly?) non-philosophers. So if I refer to common-sense morality, or intuitive claims about responsibility, I don't take myself to be too infected by my theorizing to have lost a grip on pre-theoretical thoughts.

Finally, a brief word on methodology. I'm interested in moral responsibility. Interestingly, I think there's less controversy about what that is than free will - but I think that's because it's better connected to a host of other concepts, like desert, blameworthiness, praiseworthiness, and the like, and that this gives us checks and balances on being too revisionary or mistaken with any single concept. We should want the best theory of moral responsibility. What's that? The one that holds the most theoretical virtues (simplicity; symmetry; explanatory power). We should want it to best explain the target concept. But that concept isn't in isolation - and so my inclination is to want a theory that will explain how it is that we deserve blame and praise for the things we do only when we're morally responsible for them. This gives me a number of concepts to have to do work on. But the richer and more interconnected the tapestry of concepts is, the more resources we have to investigate them.

None of this is to say that x-phi has no place in developing such an account. I think much of the resistance to it has come out of an impression that it's treated as a magic bullet to solve a host of problems in the debate. It's certainly not a magic bullet; and from my perspective, I'm genuinely unsure what sort of result an x-phi study could produce that would significantly affect my approach to the issue.

Matt, I'm not saying that x-phi has or is likely to produce such results wrt free will, but there are possible findings - my suspicion is actual findings,elsewhere - which ought to alter our approach to philosophical questions. The discovery that intuitions about x are produced by a mechanism that responds to features of x that it is hard to see are the relevant features is one such discovery. Tamara Horowitz and Frances Kamm had a debate a few years ago about whether the distinction between harming versus not aiding was produced by the heuristics concerning loss aversion identified by prospect theory. If it is - and it probably is - we should see the distinction as irrelevant and irrational. I recently argued that the doctrine of double effect is a byproduct of the Knobe effect. Now neither of these claims is indisputable, but if the empirical claims are true the philosophy will have to accommodate them. By the way, though, I doubt that standard x-phi - probing folk intuitions - will produces much in the way of genuinely significant findings. I think Josh's success set a bad example to the rest of the field. It is not necessary to spawn further Josh Knobes. He's doing a perfectly good of being Josh Knobe on his own.


I was just joking about Manuel - I know that there are non-trivial differences between your views.

As for your concerns about experimental philosophy, we could argue about it all day. I'd be glad to share some more thoughts with you over email or phone, or in California when I visit. But I'll try not to hijack this thread.

Hi Neil,

I don't think I was specifically targeting anything you said (I don't see a post from you, in fact), but I think what you say here is fair. Just to be clear, though, I would never deny that empirical findings should be relevant to our philosophical theories. Our theories must be consistent with our best naturalistic picture of the world. But it's one thing to say that developments in psychology or physics or neuroscience could have an impact on theories about moral responsibility (as they no doubt can), to thinking that there's a unique domain, x-phi, that is best placed to handle those issues. I don't find the latter very plausible, especially with respect to probing folk intuitions. And the former ought to be fairly uncontroversial.

I haven't posted to this blog for about a year now since Thomas last scolded me for posting too much. I have been doing some studying, trying to figure out why you people still cling to moral responsibility in light of the evidence coming from neuroscience. From what I gather, most of you find Frankfurt-style cases convincing! Is this your basis for believing in moral responsibility? OK. This is how I see it. It doesn't matter if the universe is completely causally determined or if there is some indeterminacy. If the evil neuroscientist wants me to kill my brother, his brain states are bound by either determinate or indeterminate forces involving his chemical makeup, genome, outside and internal stimuli, etc. He has no control over whether he coerces me or not. He is a biological machine just like me. My actions are bound by either determinate or indeterminate forces (depending on how the universe actually is) involving my chemical makeup, genome, and outside or inside stimuli (including the potential coercions of the evil neuroscientist). Suppose I already have the desire to kill my brother and I do so, sparing the evil neuroscientist the decision to coerce me. How am I morally responsible? My actions are still a function of my brain states which I have no control over owing to determinate or indeterminate forces which govern my chemical makeup. How am I responsible? How am I responsible?

I think you all are just clinging to this primordial intuition that we are morally responsible agents because it scares you to think otherwise. Whatever helps you sleep at night, I guess.

Hi Noah, your comment doesn't seem to be about my post, but I'll say a couple things in response:

1. we people (here at Flickers) do not "cling to moral responsibility." Some of us believe humans have some degree of moral responsibility (free will, desert for actions, etc.), and some of us don't. Thomas, for instance, is a skeptic about free will and desert.

2. Most of us who do believe that humans have some degree of free will and desert do not do so only or mainly because we find Frankfurt cases convincing. Some of us think such cases help illustrate why determinism need not threaten free will or responsibility. Some of us (like me) think such cases actually distract us from the main reasons why we should be compatibilists.

3. I can't give you a brief explanation of why we can be free and responsible (in senses of those words that matter and correspond to what should matter), even though we obviously cannot create ourselves from scratch or be the ultimate source of our own being, character, or reasons. If you are convinced by arguments like Galen Strawson's Basic argument, you will find such explanations unsatisfactory. But for now, surely we can agree that there is a big difference between there being distal (long past) causes of what happens, over which we have no control, and there being proximate (recent) causes of what we do, over which we have no control. Nothing yet discovered by neuroscience entails that all of my "actions are still a function of my brain states which I have no control over." Some of my brain states (e.g., my intention) may be ones *I* have control over in that they are controlled by my other brain states (e.g., my rational deliberations), the ones that matter when it comes to control. If you want that *I* to be distinct from the brain, then this reply won't satisfy you. But even if that *I* were distinct from the brain, it's not clear how it could be the ultimate source (or be ultimately responsible) for its being, its nature, or its reasons. That's why G. Strawson's arguments make neuroscience irrelevant--his premises apply to non-physical souls too.

4. I don't think I'm "clinging" to primordial intuitions or that "it scares" me to think otherwise. But who knows? I don't have perfect access to my motivations. Indeed, that's the threat to free will that scares me more than neuroscience or determinism--that we don't know why we do what we do. But sometimes I think we do. I think I'm replying to your post just to get some points clear. I hope my motivations are not more subversive ;-)

The comments to this entry are closed.

Books about Agency

3QD Prize 2014: Marcus Arvan