Blog Coordinator

« Free Will Skepticism, Just World Belief, and Punitiveness | Main | Are Compatibilists like Creationists?? »



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

Hi Gregg,

I'm a compatibilist, in what I call a broad sense, since I think MR is compatible with a broad range of naturalistic theses, of which determinism is just one. I have defended positive conditions on MR, but I have no specified account of freedom. I actually don't have any intuitions about free will. I have some judgments about acting freely, but I find these to be largely insulated from my judgments of MR.

I accept some form C-Basic Desert. In this way, I see myself as opposed to many MR skeptics, as you note. In my view, Basic Desert only requires a kind of "right reasons" restriction (cf. Pereboom's exclusion of consequentialist and contractualist considerations). I favor an approach modeled on fittingness, and reject the idea that basic desert necessarily involve anything punitive.

I am winging this here (in between sets at the gym), but I accept 1-5 and somewhat accept your definition of FW, though I don't think FW=your definition (I think that's a feature of FW, but I reject the largely consensual view that FW must be understood in terms only of MR-grounding abilities, e.g., there may be self-regulative features of FW abilities that species may possess but those species might not be appropriate targets of either ascriptions of MR or MA, moral agency). So your definition captures something we normally find central but might not be essential to the concept. I think all the versions in your catalog capture something right about FW and MR, but I have a similar view of each as I have of your definition. I have a very complex way of collecting all those elements, too complex to simplify here, but a gross simplification might suffice. I see autonomy as on a scale of self-regulative causal/functional abilities the variations in grades of which explain the elements noted in your catalogue of positions and the intuitions associated with them. At the weakest link End of the scale only 5 applies, but at the strongest end libertarian intuitions apply. I'm an autonomist, same as a soft compatibilist: autonomy is possible regardless of the truth of D.

Hi Gregg, OK I'll try to play along and use your labels. I'm comfortable defining FW in terms of the capacities for control that ground MR, and even to use MR in the 'basic desert' sense if you allow lots of room for debate about what the hell we mean by 'basic desert'. To begin I'd want to emphasize that (5) is important (esp if we're letting in some contractualist and Strawsonian stuff, not just utilitarian considerations of the JJC Smart sort that motivated Strawson in the first place). If the stuff in (5) comprises the bulk of what our concepts and practices are useful for, then skeptics are in the tricky position of trying to tell people we still have all that--it's just this sliver of other stuff we lack. Again, this is where the rubber hits the road when trying to distinguish some compatibilist for some skeptics.

Beyond that, I'll accept CDF-Compat (1) in both the source and leeway senses, being careful to note that it's the alternatives relevant for MR that are compatible with determinism, and there are unconditional abilities that are, of course, incompatible with determinism (see Chris Franklin's nice essay "Everyone THinks an Ability to Do Otherwise is Necessary for FW and MR"). And there is also a sense of causal sourcehood (I'm developing with Oisin Deery) that is consistent with determinism (but not manipulation) and can ground FW and MR, but of course there's another sense of ultimate sourcehood that can be defined as incompatible with determinism--and as impossible for any finite agent since it requires self-creation (that's the sort of FW that we need to justify deserving eternal damnation, but we can and should shed that sense now and retain a relevant (finite) sense of desert).

And I'm a C-Freedom Compat, but a "Cagey Compat" who thinks there's lots of limitations on the capacities required for FW and MR, limitations we discover through empirical investigation (e.g., psych research on our capacities for self-knowledge and self-control).

And given all of the above, I'm a CDMR-Compat in the leeway and source and desert ways, but Cagey here too, and inclined to think that we have less FW and MR than we think, and that criminals are often more limited in the relevant capacities and opportunities (many have mental disorders, after all), such that we need to reform our punitive/penal practices, though without giving up all talk of desert (including outside our penal practices).

I hope that clarifies rather than muddies...

Eddy, thanks for playing along! That doesn't muddy the waters at all, in fact it helps a lot. Despite your cageyness, you sound like a traditional compatibilists to me.

Matt and Rick, you both sound like traditional compatibilists as well. (Rick, while you don't fully share my definition of FW you do accept (1)-(4) so I take that to be a rather standard view).

I'm really wondering if those who reject (4) (i.e., Basic Desert Compatibilism) and only embrace (5) (i.e., Forward-Looking Compatibilism) are *really* compatibilists at all. I guess anyone is allowed to label themselves as they wish, but I don't see much different between (5) and what the free will skeptic maintains. I also don't see many compatibilists who reject C-basic desert as being very forward about that! It seems to me that that is disingenuous. If you really believe people *do not* justly deserve to be praised and blamed for their actions, you should say as much and come out just as strongly against retributivism as the free will skeptic does. But I don't see Forward-Looking Compatibilists doing that. I see Forward-Looking Compatibilists as changing the subject and sometimes even obfuscating their rejection of basic desert MR.

As Tom Clark mentioned yesterday: "Dennett has declared himself a consequentialist, non-retributivist compatibilist (at least in personal correspondence, not sure about in public yet), but still wants to talk about just deserts when it comes to punishment." I find this utterly confusing. I also think it muddies the waters so thoroughly that its not helpful. Like a magician, Dennett looks like he is presenting you with one position, but upon closer examination he's really performed a slight of hand. What do others think?

Wow, there is a lot here! I'm going to make a few comments in separate posts. First, I don't think this is a good way of distinguishing source and leeway (or what I would call 'classical') compatibilists: "The former see determinism as compatible with agents being the source of their own actions, while the later see determinism as compatible with the ability to do otherwise." In my book, I argue that all compatibilists, indeed all free will theorists, think that sourcehood is important. What distinguishes them is whether or not they accept the classical thesis: the ability to do otherwise is necessary for sourcehood (or up-to-usness). The source compatibilist says the answer is 'no' but the classical compatibilist says it is 'yes.'

I don't particularly like the term 'leeway' since it suggests that the kind of options we're interested in are libertarian options but obviously this is not the case with the classical compatibilist. A dispositionalist, for instance, might think that I was able to do otherwise because I had a set of dispositional powers (even if determinism is true). 'Leeway' does not really capture this intuition well.

Joe, thanks for those corrections! Forgive my sloppiness in distinguishing source views from those that believe the ability to do otherwise is required for FW and/or MR. I accept your way of stating the distinction.

Let me know what you think of the more substantive issues regarding basic desert MR.

Gregg, I agree with you. A real compatibilist about responsibility has to endorse some form of basic desert, however mitigated and qualified. Otherwise, the differences between the skeptic and compatibilist are mostly (if not entirely) terminological.

And for what it's worth, I see Strawson as a BDCNRAPSSOI (basic desert is constituted by natural reactive attitudes and practices, so stop overthinking it) compatibilist.

Eddy, what happened to your neurotic compatibilism? That was a good label (unlike "willusionism")

Thanks Tamler! Your agreement makes me feel better--I was starting to think I was going crazy.

I accept your BDCNRAPSSOI in replacement for Basic-Desert-isn’t-going-anywhere Compatibilism ;)

I don't like putting connotations of 'moral responsibility' into the definition of free will. Clearly, we could all lack moral responsibility yet have free will. The latter is at most a necessary condition for the former. As these things go, I'm less bothered by your definition: free will is "a kind of power or ability to make decisions of the sort for which one can be morally responsible."

Still, this definition is problematic. Consider Gideon Rosen's skepticism about moral responsibility. He writes: "The skeptic might simply claim that given certain contingent and possibly temporary limits on our knowledge, confident positive judgments of responsibility are, for the moment at least, unwarranted." Supposing that Rosen is correct, no one can make decisions of the sort for which one could be morally responsible. Yet there is no obvious reason to think they lack free will, or lack the power to perform actions that are up to them.

It is better to define 'free will' more cleanly and then claim that it is necessary (but not sufficient) for moral responsibility.

Hi Gregg-

FWIW, (4) and (5) seem to me crucial issues that haven’t been well sorted out in the literature.

Re: (4), I actually don’t think we have a good handle on desert attributions and their role in a theory of moral responsibility. In particular, I’m somewhat skeptical whether ordinary desert attributions are sufficiently thick so as decide between basic and other notions of desert. And, I have yet to hear any explanation of why “mixed” accounts (ala Hart/Rawls, or now, me!) aren’t sufficient to capture existing phenomena of desert ascriptions. Moreover, as Neil Levy has noted, there are “quasi basic desert” options lurking here.

Re: (5), I think the “no moral responsibility” crew haven’t really adequately reckoned with what conceding (5) means for the terrain. As Eddy notes, the more one is prepared to grant that consequentialist and contractualist resources can capture ordinary practices, the less significant the eliminativist view becomes. (This is a point I try to drive home in a forthcoming Phil Studies piece; see section 3 of “Moral Responsibility and Desert,” available on SSRN .) I also blogged a bit about it last year, under the guise of questions about what a parameter for a debate versus what’s a live position. And here I can’t resist provocation: I just haven’t seen any serious engagement with the various views that maintain that there is almost nothing left over for eliminativists to claim as terrain in our actual practices once we see what sophisticated contractualist and consequentialist views can do.

That said, I would classify myself as a basic desert compatibilist, so I think you are right about that intuition. Those who accept (5) are compatibilists-in-name-only, which was one of the points PF Strawson tried to make (as Eddy noted).

But it does make for an interesting issue. Why not just see compatibilists and free will skeptics as on a continuum -- we've got classical compatibilists, source compatibilists, and skeptics who accept (5) -- just to name a few. Is the disagreement a quantitative one -- one says 'yes' the other 'no' to the existence of free will? Or is it more of a qualitative difference, a matter of degree? I argue for the latter, which is why I'm a Cagey Compatibilist, like Eddy.

Two more thoughts (and thanks for getting this great conversation going!)

Here's a worry about “No Real Scotsman” objections in this literature, connected with my above remarks re: (5). We all like to define theoretical constructs and then fit theories into them. But in the case of moral responsibility, at least, there is a relatively clear set of practices and attitudes that one can think are well-founded or not. But despite the influence of Strawson, we don’t much try to hitch our theoretical categories to these actual, real world things.

(Or, we advertise that our theories are about these things, but quickly change topics to talk about technical notions without defending the implicit claim that our technical notion about desert, ability, or what have you, is the one that figures in practices.)

So we end up with debates about whether someone is an X or Y, with mostly tacit commitments about the extent to which ordinary practices or debates have adequate justification. That is, we end up with weird situations where “compatibilists” concede that we have to give up core aspects of our (retributive) practices and where “eliminativists” offer positions that are (in principle) compatible with the vast majority of our responsibility practices being in good normative standing.

My sense is that we do better to ignore labels and focus on commitments w/r/t actual practices, attitudes, and judgments. On this (pragmatist, Al White tells me) approach, most of our existing practices, at least in their general form, are basically in defensible shape (so I argue). The normative claim—i.e., “defensible shape”—doesn’t presume that the defense goes quite the way some folk think they do. But I think we can get ordinary desert claims, and most (but not all) retributive impulses and practices.

Does the fact that I think I can get retributive impulses, and basic desert understood one way (but maybe not another way) mean I’m a compatibilist? Or an eliminativist? Partly, this turns on issues about reference for which their is reasonable disagreement. But notice that in putting claims about everyday practices at the forefront, we are less likely to be worried about the best application of quasi-technical categories whose meanings have shifted repeatedly and whose connection to real phenomena is somewhat unclear.

So Joe, how do *you* define free will?

My reason for defining FW in terms of MR is twofold: (1) First, it provides us with neutral definition that all parties can agree to. The debate can then focus on what powers/abilities/etc. are required for FW/MR. If you define free will in term of compatibilist capacities, for example, that's simply begging the question from the outset. (2) Secondly, by defining FW in terms of MR, it highlights the practical importance of the debate. It's hard to see what the FW debate amounts to if we remove all of its practical concerns and considerations. *Why care* about free will if it is not related to MR, punishment and reward, etc.? I believe Vargas has written eloquently (more eloquently than I) on this second point. (Perhaps he's reading this and can jump in.)

I should note, however, that even if you do not accept my definition of FW, I tried to provide a taxonomy that leaves open others ways of defining FW--so, for example, my varieties of compatibilism captures the semi-compatibilisms of Fischer and Waller.

While we continue to debate the definition of FW, I would also like to hear what you think of (4) and (5)

A person has free will provided that at least some of her actions are up to her.

Good discussion! Off to work!

Wow, in the time it took me to respond to one of Joe's comments there were four other comments posted! I'll try my best to respond.

Manuel, I was hoping you would jump in! (Perhaps you would be willing to comment on my rationale for defining FW in terms of MR. If I recall correctly, you provided a similar rationale somewhere.)

Okay, so we agree on the importance of clarifying the difference between (4) and (5). As I said in my post, I think this lies at the heart of the debate. If I understand you correctly, you think we do not know enough about desert attributions to distinguish between the two. I guess I have a question for you before I provided a fuller response: Are we trying to reverse engineer our desert attributions OR are we debating the metaphysical justification for attributing desert? If we are doing the latter, I don't see why we cannot stipulate that we are debating Basic Desert MR?

Your second point was that FW/MR skeptics haven’t really adequately reckoned with what conceding (5) means for the terrain. I don't agree. I think Pereboom, Waller, and others have written extensively on what accepting (5) preserves and what it *does not* preserve! I disagree with you that there aren't enough practical differences between those that accept (4) and those that accept (5). I think there are significant issues at stake! Just consider how compatibilists react when they read one of Waller's books attacking the moral responsibility system--they don't shake their head in agreement, they recoil and defend the retributive attitudes.

I agree totally with you on the point of your second post-- we need to always keep in mind how these debates play out in actual practices, attitudes, and judgments. That said, I think it's inaccurate to say that none of these things would change if we reject retributivism and just deserts. I think we would have to relinquish a number of judgments and attitudes (attitudes that I consider harmful). I also think we would have a completely different grounds for punishment that would require major reforms in our current system. I could go on....

Joe...I will respond in a separate comment.

Joe, I have to run to a meeting soon as well so I will keep my response short. I'm glad you agree with me that "those who accept (5) are compatibilists-in-name-only."

As for your suggestion that we view compatibilists and FW skeptics on a continuum, I would be happy to do so if we all loudly proclaimed our opposition to basic desert, retributivism, and backwards-looking MR. I don't see that happening however. I think people truly deserve praise and blame (in the backwards-looking sense) or they don't. It's hard for me to see a continue here.

Hi Gregg-

Just to clarify: I didn't claim that nothing changes without just deserts or retributivism. I agree that they would (that was the point of noting that complaint about "compatibilists" who give up on retribution are in a funny position- chunks of our practices would have to go).

Re: the provocation. My claim is that I've not seen a lot of engagement by eliminativists with either current wave defenses of our practices on consequentialist/contractualist grounds (i.e., not JJC Smart) or with older classic mixed views (like Hart/Rawls) on which consequentialist resources are used to ground retributive practices. Derk, for example, famously says that he's got no quarrel with views that ground responsibility-characteristic practices in consequentialist or contractualist grounds. But he is (so far as I know) silent on or ambiguous about whether he thinks those non-Derk accounts can get only some, most, or virtually all of our current forms of practices. But (by my lights) the answer to that question is central to evaluating whether eliminativism is mostly the denial of something that doesn't much figure in everyday life or whether, instead, it is really denying the integrity of all that real-world stuff compatibilists and libertarians are on about.

In saying the foregoing, I don't mean to presume that the various consequentialist and contractualist defenses of our responsibility practices do everything they claim they can do, or that eliminativists (were they to seriously engage) would not fail to demonstrate the poverty of these accounts. Perhaps eliminativism will yet carry the day.

My claim is just that there is a body of (mostly "normative," as opposed to "metaphysical") work that tries to ground things like desert claims, retribution, blameworthiness, the reactive attitudes, and so on, in things that don't depend on the kinds of agency normally rejected by eliminativists. If those accounts work—and that is a big if—then the interesting of eliminativism (about at least moral responsibility) is seriously undercut. But for eliminativists to carry the day on those matters, they are going to have to get their hands dirty doing normative theory, responding to work that does—and crucially—showing how ordinary practices can't be preserved or sustained in the ways non-eliminativists have maintained.

Manuel, thanks for clarifying. You are definitely sounding foxy now!

I don't disagree with your comments, to the extent that you point out important work still to be done. I agree that free will skeptics and eliminativists need to more directly address those consequentialist/contractualist/mixed views that attempt to ground retributive practices. But I'm still unclear what these accounts are actually trying to preserve?! I know you claim the concept of "basic desert" isn't sufficiently thick to hang much on, but why is the burden on the free will skeptic to show that their review substantively differs from these other conceptions of desert when these other accounts aren't even clear about what they are setting out to justify! The free will skeptic defines basic desert in a straightforward manner, and then sets out to argue that we do not have it. I find the concept of basic desert rather intuitive. I also find the distinction between backwards-looking and forwards-looking MR rather easy to grasp. Part of my reason for this post was to gain some clarity on what these consequentialist/contractualist versions of MR-compatibilists are actually claiming. I have yet to gain that clarity.

Sorry for keeping this one short...but I have 100 midterms to grade before tomorrow. I'll try to add more later.

(Derk? Bruce? Thomas? Anyone want to jump in ;)

//How many of you accept C-Basic Desert?//

I do.

//How many of you would be willing to accept something like my definition of FW—i.e., free will as a kind of power or ability to make decisions of the sort for which one can be morally responsible?//

I would.

//Since most free will/moral responsibility skeptics accept (5)...//

Huh? Let's look at 5 again:

//(5) Forward-Looking Compatibilism: This position rejects basic desert MR but maintain that a forward-looking account of MR can be reconciled with determinism (and presumably also luck).//

"[A] forward-looking account of MR can be reconciled with determinism".

Clearly that entails that a forward-looking account of MR can be given. And that entails that some account of MR can be given.

So if you believe (5), you must believe that some account of MR can be given.

And if you're an MR skeptic, don't you have to believe that any account of MR is *mistaken*? That it is, in fact, not actually an account of MR, but only an account of something at best *similar to* MR?

If you actually hold that an account of MR can be given (whether forward-looking or not), then you shouldn't be calling yourself a moral responsibility skeptic. And if you hold that together with the belief that free will is just that power that makes it possible for you to be morally responsible for your actions, then you shouldn't be calling yourself a free will skeptic, either. In such circumstances, you could call yourself a "basic desert skeptic". It's clear and to the point! (Well, clear*er* than free will skeptic, anyway.)

I expect that what you believe is something slightly different than (5), but I'm not sure what. Would the following be a suitable restatement of your belief?

(5') Forward-Looking Sort-of-Compatibilism: This position rejects basic desert MR but maintain that a forward-looking justification for (some/all) MR practices can be reconciled with determinism (and presumably also luck).

If not, what's wrong with it?

//is (5) really a form of compatibilism?!//

It's a form of semi-compatibilism. And combined with your preferred definition of free will, it's a form of compatibilism.

//And if you're an MR skeptic, don't you have to believe that any account of MR is *mistaken*?//

Actually, I think "non-referring" would be more apt than "mistaken".

Mark, thanks for playing along and answer my list of questions.

With regard to (5) you write: "So if you believe (5), you must believe that some account of MR can be given. And if you're an MR skeptic, don't you have to believe that any account of MR is *mistaken*?" Not exactly. Like Pereboom, Waller, and other skeptics, I reject basic desert MR--the kind of MR that is at issue in the free will debate--but am okay with accepting (say) forward-looking accounts.

You write: "If you actually hold that an account of MR can be given (whether forward-looking or not), then you shouldn't be calling yourself a moral responsibility skeptic." Again, I am a skeptic about the kind of MR at issue in the free will debate--the kind that would make agents truly deserving of praise and blame in a non-consequentialist, backwards-looking sense. Perhaps the way the terminology has evolved is unfortunate, but it's a pretty standard usage of the term. See for example, Pereboom's new book in which he defends free will skepticism but also a forward looking account of MR (my précis of it is available here:

Final, why is (5) semi-compatibilism?

Thanks for this post, Gregg (along with the others, of course).

I'm a basic desert compatibilist. Recently, for example, hearing about the testimony of some of the victims (and their faimilies) in the Boston Marathon Bomber trial, I thought to myself,"Boy, that guy deserves a harsh punishment. At least life in prison without parole."

Thanks John--I won't blame you for your retributive reaction!

Great posts, Gregg! And you raise some great questions! I often have many of the same questions for skeptics. You say "Since most free will/moral responsibility skeptics accept (5), is (5) really a form of compatibilism?! "

Couldn't one say in response to the optimistic skeptic "is your skepticism REALLy skepticism given that you think that everything else (anything worth keeping anyway) is compatible with having no free will?"

I guess I'm asking for a nice clear distinction between the optimistic skeptic and the compatibilist. It seems that their positions are nearly identical in many important respects. I wonder if this has to do wth their meta-ethical moral intuitions, but I digress.

Anyway, these questions may have been answered already but I haven't had a chance to read the comments. I would have loved to jump into your first few posts, in fact I'm sure we'll talk about this in the coming months. Solo parenting for 12 straight days gets you behind a bit. Deadlines are swear words for me these days, I'm sure you can relate.

The comments to this entry are closed.

Books about Agency

3QD Prize 2014: Marcus Arvan