Why Should We Believe We're Morally Responsible?

In a 1990 Analysis paper, Saul Smilansky claims that Peter Van Inwagen’s argument (in An Essay on Free Will) for the existence of libertarian moral responsibility is weak   I would agree.  The argument focuses largely on his claim that he can’t take anyone seriously who denies moral responsibility, because in the next breath they’ll assert that stealing books is “shoddy.”   People who deny moral responsibility, according to PVI, contradict themselves with monotonous regularity.  I won’t go into the problems with this view here (but see Smilansky’s paper, available on his webpage, for one good response.)   Rather, just assume Smilansky is right.  What are the implications of this?  According to Smilansky, the whole structure of An Essay on Free Will collapses.  I believe that’s correct too.  Because here in a nutshell is PVI’s argument for the existence of libertarian free will.

(1)   We must have libertarian free will in order to have libertarian moral responsibility.

(2)   The claim that we lack libertarian moral responsibility is absurd.

(3)   Therefore we must have libertarian free will.

Without good arguments for claim (2), PVI has no right conclude (3).   At best, he can claim that either hard determinism or libertarianism is true.  But that is certainly not his goal in EOFW.  On the contrary, he claims that if he learned determinism was true, he would abandon incompatibilism.   I had noticed this gap in reasoning as well, and I asked PVI about it when he came to Duke a couple of years ago.  His response was roughly: “I simply think that the  probability that we lack moral responsibility is lower than the probability that incompatibilism is false.”   That seemed fairly reasonable to me at the time.  Lately though I’ve begun to wonder: what is that probability assessment based on? 

This is not a rhetorical question.  I’m really interested to know.   What grounds do we have to assign a high probability to the truth of ‘people can be morally responsible’ (in the strong desert-entailing sense)?   I think that this question can be asked not just of PVI and other libertarians but of anyone who asserts the existence of desert-entailing moral responsibility.  What are you basing this belief on?  Does the belief have strong support?   And to what extent does your theory of moral responsibility depend on a good independent justification for its existence? 

Comments

[James Gibson]

Tamler: why can't the belief that we are morally responsible be basic? If it is, then it doesn't make much sense to ask "what are you basing this belief on?" if you are searching for reasons that justify moral responsibility. For many people, I think, they would consider others to be morally responsible as an intuitively obvious feature of the world, similar to many of the other doctrines of common sense (e.g., that there is a past, other minds exist, etc.). My guess is that for PVI, it is more clear that people are responsible than that any particular explanation of the type of freedom that grounds ascriptions of responsibility is true. This is how I understand his claim that if he learns determinism is true, then he will abandon incompatibilism.

If moral responsibility (the kind to which you refer) is basically held by one person, then this fact of basicality may not do much to convince someone else that does not hold this belief in a basic way or the belief at all. Sure, it is ideal that one could have such arguments to convince you, but I can't see that it is necessary that one do so in order to hold the belief rationally. And, as you know, holding a belief basically does not preclude developing arguments for that belief. There is still dialogical value in such arguments, and it would be a mistake to ignore the search for them simply because one holds the belief in a basic and, if there are no undefeated-defeaters, rational way.

For PVI, if he tweaks his view a bit from the EFW, there may be a sense of absurdity in denying moral responsibility *even if* one does not contradict oneself in denying it, much like one does not contradict oneself in denying the world existed more than five minutes ago (at least the contradiction is not clear to me). In the end, I guess it looks as if what grounds our ascriptions of responsibility (the kind to which you refer) are our intuitions - and their overall strength - that we simply can't give up or pick out like we can choose hats.

[Tamler Sommers]

James,

All good points. I think you're right that the strength of our intuitions leads us to assert the existence of moral responsibility and to feel that not much justification, if any, is required. But this leads to two further questions.

(1) If we can find other explanations for why we have those intuitions, explanations that make no reference to the belief they support being true, then don't we have reason to regard the belief in moral responsibility with suspicion? And

(2) Isn't there something a little intellectually dishonest, or anti-philosophical, about calling a belief as contested as the belief in moral responsibility "basic"? After all, no one really believes that the world was created five minutes ago. And there are no good arguments for that belief. (At most, it cannot be proven to be false.) But as Smilansky says in his essay, people really have doubted the existence of moral responsibility and still do. Calling the belief in MR basic, then, is just a conversation-stopper, no?

[James Gibson]

Thanks Tamler. Regarding (1), I don't believe we should be necessarily *suspicious* about the truth of a belief simply because we cannot find after much effort powerful arguments in support of that belief. We should if the belief doesn't strike us as *immediately* obvious. But some beliefs do. Search as I may, I know of no noncircular proof of the external world that I find convincing, and yet I still think I am rational in believing that there is an external world. I think what we should be is humble about our basic beliefs in allowing them to be defeasible. So long as I can't find any extremely powerful reasons to reject that belief that rely on more intuitive beliefs than the belief in moral responsibility, I will continue to be a commonsense realist about that belief.

Regarding (2), I hope that I'm not being intellectually dishonest in calling the belief basic. If I am, it certainly isn't like the time I went all-in with fifty bucks hoping the other players didn't realize I only had 2/7 off-suit! You're right that no one really doubts the existence of the past, and that some people really doubt whether we are morally responsible. And there are also no good arguments for believing creation occurred five minutes ago, whereas some find the arguments against MR convincing. But how does it follow that one is being intellectually dishonest in claiming that the belief in MR is basic for oneself (and perhaps many others) simply if others reject that belief - the belief in moral responsibility, or even the belief that the belief in moral responsibility is basic? I don't see how this is the case. What would make it the case is if the belief does not strike the cognizer as basic but he claims it is basic anyway. *That* is dishonest.

So as I said, while I may consider myself rational in the absence of defeaters in believing in MR, even though this belief is not based on further argumentation, this is not much to convince you - unless you think I'm a great authority on what to believe :) Nevertheless, when the person who denies MR demands an argument for believing in MR, unless I'm convinced of some evidentialist requirement about justification, I will reject that demand. The conversation *between us on this point* stops there unless I present some postive arguments for MR, or you present some arguments against MR, or you present some case why I must have arguments, or I present some case why it is logically absurd to deny MR (or anyone for that matter).

[Heath White]

Here is a refined version of an argument popular with my students. I am not sure what I think about it. But it goes like this.

(a) We cannot really imagine what it would be like to live in a society in which no one is held morally responsible. (b) We also cannot coherently imagine routinely holding people responsible while thinking that they are not really responsible; if we are holding people responsible, then whatever our criteria for that practice are, that's what our view of moral responsibility *is*. So (c) we cannot really imagine believing that there is no such thing as moral responsibility. [end argument]

The libertarian thinks that we hold people morally responsible, and that our criteria for this include that they have libertarian free will. But should it be demonstrated that we have no LFW, then we'll continue to hold people responsible, changing our criteria, and thus changing our view of moral responsibility. So that is a way in which one might say that, faced with denying moral responsibility or denying incompatibilism, one goes with denying incompatibilism.

[Rob]

"That a theory is refutable is, frankly, not the least of its charms: this is precisely how it attracts the more refined intellects. The theory of 'free will,' which has been refuted a hundred times, appears to owe its endurance to this charm alone --: somebody will always come along and feel strong enough to refute it."

Nietzsche (Beyond Good and Evil, Section 18)

[Tamler Sommers]

James,

You're right--intellectual "dishonesty" is not the right word. (Although, I don't think it's a good description of your all-in bluff either--that's just part of the game.) I guess what I mean is that to call a controversial belief basic seems to violate the philosophical spirit, the the questioning of one's own intuitions in the disinterested pursuit of truth--and so on. Also,if you call the belief 'basic', could any argument against MR convince you that the belief is false? Or would you just think that it had be wrong somehow...

Heath,

That's an interesting argument--I've never come across it. But (a) seems implausible to me, too strong. Could Brave New World serve a counterexample? We may not have found the society in BNW appealing, but it was coherently imagined. And I think it was a society in which no one held is held morally responsible in the strong sense. (If I'm remembering correctly, even the heretics were just sent off to an island where they could pursue their heresies without influencing anyone else.)

[Fritz]

Just one peripheral question (for now at least) --

by "libertarian moral responsibility" do you mean the conjunction of libertarianism about freedom
(= incompatibilism and Det and Freedom + acceptance of freedom) with the thesis that we are morally responsible?

Or is a commitment to "libertarian moral responsibility" a commitment to some particular understanding of what moral responsibility is?

[Tamler Sommers]

Fritz,

That's Smilansky's term, but now that you bring the issue up, I'm not sure I want to keep using it. I think I just mean: "desert-entailing moral responsibility" (DEMR), with no particular commitment concerning type of free will (libertarian or compatibilist) would make it possible.

I'm pretty sure the point still holds. PVI--correct me if I'm wrong-- believes (1) that we can't have DEMR without libertarian free will, and (2) that the view that we lack DEMR is absurd. But he'd give up the belief in (1) before he'd give up the belief in (2), even though the arguments for (1) are much much stronger.

[Kip Werking]

I would disagree with the argument you mention. And I want to explain why, because I think it motivates much compatibilist and/or libertarian sentiment. The notion that we can't imagine what it would be like to not hold people responsible represents, in my view, just a failure of imagination.

What would such a world look like? Society, in that world, would be, quite simply, "medicalized." This is Dennett's term and he dismisses the prospect of a medicalized society after two short paragraphs near the end of Freedom Evolves. Dennett's writing here is the most explicit description I've found of an alternative to a society based upon freedom and responsibility. In the literature, those who deny the existence of FW and MR typically just focus upon attacking the possibility of free will without (in Paul Davies' words) "suggesting what might replace it." What might replace "free will" is the belief that people do not become bad; they become sick. Badness, evil, criminal and immoral conduct all just become varieties of illness. Just as the brains of people with brain cancer can become sick, so too the brains of criminals can become sick. Of course, this part of the brain will not be their *rational* part, because criminals (assuming they are not mentally ill) can be and usually are rational. Rather, the *appetative* part of their brain goes bad. Just like a brain tumor, whatever-it-is-about-a-criminal-brain-that-makes-it-criminal (let's call it X) is something that is typically (i) not the subject's fault and (ii) something society will not tolerate.

Perhaps most importantly, if wicked just means sickly, so too punishment will just be a species of medicine---crude and painful medicine at that. Unless society can develop painless "moral pills" for criminals, it will need to continue hurting them and putting them in prison; but as soon as it does develop such "moral pills" society may be morally obligated to use them.

I try to sketch, in further detail, what this alternative conception of criminal justice would look like in the last section of my article "Who's Afraid of Creeping Exculpation?" (including an abundance of footnotes citing other relevant articles), which you can find here:

http://people.wm.edu/~ktwerk/exculpation.htm

[James Gibson]

Hey Tamler. Right, the bluff is part of the game. I was just trying to disassociate myself from anyone else that may be committed to some particular belief B, such that the person is willing to entertain any strategy whatsoever in order to keep that belief. For me, the belief is basic - as opposed to me simply claiming it is basic in order to stop dead the challenge (the belief was basic prior to the challenge, and it is one I've had for a very long time).

Anyway, I guess you want to know what would it take for me to be convinced that we aren't deeply morally responsible since I consider the belief to be basic. I want to emphasize that the belief's *basicality* in no way makes the belief any more difficult to defeat in principle. I have lots of nonbasic beliefs that I hold very stongly too, some of them more strongly than other basic beliefs of mine. Those basic beliefs may be easier to defeat. So, it is not as if basicality provides some protective layer against defeaters in general. Basic beliefs too are always in season for sport.

Could any argument convince me? Let me just preface this by saying that as of right now, I am a foundationalist, but I find a position like Sosa's or Keith DeRose's direct warrant realism very attractive. So, I'll provide responses for each. First, a foundationalist response: sure, an argument could convince me if it employed premises that struck me as more intuitive than the belief that we are morally responsible and I could find nothing wrong with the argument there. So whether it is possible to convince me otherwise depends on whether anyone (or myself included) presents some such argument. Now me as a foundherentist: in this case, I would eschew my prior claim about basic beliefs and say any belief considered individually is not sufficiently warranted to be held in a rational way. But, my beliefs receive support from the rest of my beliefs, and together, they become rational to believe. In this case, it would require a large change of my beliefs to convince me that we are not morally responsible. It is not as if my beliefs are open target independent of other beliefs that I hold. I think it would require a legion of arguments that strike me as very powerful. Consider: I take my first philosophy class and am confronted with Descartes' Meditations. I also think Descartes' Cogito is insufficient to answer the skeptical challenges, and being a first year student, don't know of any other good responses to global skeptics. What should I believe? I conclude in that situation that there must be something wrong with the argument given by the skeptic, or some of the presuppositions used to develop that argument. I hold far too many other beliefs that strike me as obviously true to give up my belief because of one argument that I am less sure is sound. I guess in either case, you may have your work cut out for you.

[Heath White]

I am skeptical that a completely medicalized society is feasible. The reason is that I think issues of accountability are part of what it means to view someone as a person. In particular, that's part of how you view yourself. This is complicated and I have to run to class, but let me try to illustrate.

Suppose there is some dispute about how I should act. If I'm listening to reasons from other people on the matter, then I'm holding both of us accountable to some trans-personal standard--we're not fully medicalized yet. Moral issues are just a special case of this. We make the problem too simple by carving out criminal punishment, or "moral" responsibility, as a small region of life to medicalize. Responsibility is pervasive: any kind of reasoning depends on it.

Suppose on the other hand I refuse to listen to reasons from other people, perhaps on the ground that if they differ from me, they're sick and need my help. Well, now I have no conversation partners. There is no way to perpetuate a society with people who think of themselves and others in these terms. How would you raise your children? Or negotiate over how to raise them with your spouse?

In cases like Brave New World or 1984, what you typically have is a ruling elite who medicalizes the treatment of everyone else. But how does the ruling elite treat one another? Presumably they reason together--but that will mean holding each other responsible for good and bad thoughts, and good and bad decisions. Presumably, too, they have ways of dealing with those who get out of line, but they can't think of themselves, facing these disciplinary procedures, as being treated for illness.

I suspect this is not fully persuasive but maybe it shows the direction in which I'd wish to argue. The gist is this: to be a person is to lay oneself open to the force of the better reason, which is a form of accountability or responsibility; but in a completely medicalized society this makes no sense. So there are no persons. We have no idea what that kind of life would be like...I suspect because it's incoherent.

I suspect this is not

[Tamler Sommers]

Heath,

A couple things. First, I want to stress that I don't believe a medicalized or Brave-New-World type society is the only kind of society one can imagine where people don't hold each other responsible (in the strong sense). I think there are plenty of other more appealing possibilities. I only brought up Brave New World because I thought it was a clear counterexample to the claim that we can't even CONCEIVE of a no-MR society.

Second, you write about this about the ruling elite in BNW : "Presumably they reason together--but that will mean holding each other responsible for good and bad thoughts, and good and bad decisions."

I realize this is a Big Topic, but that seems like a huge non-sequitor to me. Why would them reasoning together entail that they hold each other responsible for good and bad thoughts, good and bad decisions? (Of course, I have this same question of Susan Wolf and many others...) I reason with my students all the time without holding them morally responsible for their thoughts and decisions. I may admire good reasoning skills, and lament poor ones, but I don't hold the students themselves accountable for their reasoning abilities, either as a whole, or in a particular case. (e.g. maybe they're hungover so they uncharacterstically affirmed the consequent...) Why couldn't the ruling elite do the same?

(Again, let me stress, I am not endorsing the creation of a ruling elite that runs a Brave New World. I'm just claiming it's a coherent possibility.)

[Heath White]

Tamler,

The idea is this. (And yes, it's a Big Topic.) Consider calling someone on a contradiction. They say p, and then later say ~p. You reply, "Hey, but earlier you said that p--you need to get your act together." You are, in essence, calling them on a violation of rules, holding them accountable at one time for something they said at another. And reasoning in general is like this: if you are critiquing someone, you are saying they're not playing the game right; if you are open to listening to others critique you, you are open to being held accountable.

I think you can treat your students' reasoning in a "medical" manner, viewing it as the result of a hangover or childhood religious training or whatever. But insofar as you are doing that, you are not opening yourself to criticism from them. You are the "ruling elite" in the classroom, which is fine, but not something that can be generalized to every case of conversation and still have a recognizable society. Conversely, if you are open to their criticism of your views, then you will not be viewing their reasoning in a "medical" manner, i.e. as merely the product of causes.

Now the claim is that this kind of accountability is of the same general kind as moral accountability. Put differently, I am very suspicious when anyone insists on the locution "*moral* responsibility" because I don't think "moral" is a natural kind of responsibility. There is only holding one another accountable to norms, whether of consistency or altruism or sticking-to-agreed-decisions. (I suspect that disagreement on this point is the heart of the issue.)

You haven't *completely* medicalized society until you've medicalized all kinds of divergence from norms. But my attempting this kind of mad-dog medicalization means that I refuse to be held accountable by anyone else. That is sustainable only in the very short term.

Hope that helps.

[Tamler Sommers]

Heath,

Not really. I think there's a fundamental difference between what you're calling accountability and the accountability that's at stake in the free will/MR debate. Let me see if I can flesh this out a bit before I go to class and reason with my students in a non-medicalized fashion.

You write: "Consider calling someone on a contradiction. They say p, and then later say ~p. You reply, "Hey, but earlier you said that p--you need to get your act together." You are, in essence, calling them on a violation of rules, holding them accountable at one time for something they said at another."

If that's what you want to call accountability, fine. But my kind of accountability implies not just that the person violated the rules, but that it's the person's FAULT that he violated the rules. And that's what I don't do with my students. If you view accountability this way--call it "strong accoutability", to use Watson's phrase--then it's equally clear that I can be open to criticism by my students without holding myself (strongly) accountable either. If they point out a mistake I'm making--as they often do--I tip my hat to them. I admire their reasoning skills and lament my own (maybe the result of a hangover) without attributing strong accountability to either of us.

[Heath White]

Tamler,

I think we're making progress. You are probably right that we do not (usually) think mistakes in reasoning are someone's FAULT (I like the caps). It's not terribly plausible, after all, that such mistakes are exercises of libertarian free will.

Suppose, then, we found it reasonable to run some kind of analog of this behavior in society. We hold each other accountable, but not STRONGLY accountable; they broke the rules, but it's not their FAULT; in short, we still have responsibility, just not MORAL responsibility.

Wouldn't it then be reasonable to say: well, look, this new practice of holding-accountable or non-moral responsibility or whatever you want to call it is just what responsibility *is*; it's compatible with (perhaps driven by) determinism; so we're compatibilists.

Obviously, whether this is a good reply to your original question depends on whether the sense of responsibility involved is "strong" enough. I'm wondering how to tell; what are the operational criteria? I think it's noteworthy that the kind of (weak) mutual accountability you have with your students is desert-entailing in the sense that someone who utters a contradiction or makes another mistake in reasoning is subject to correction (whether or not they actually get corrected--this can be a matter of mercy as well as justice) based only on what they said, not (say) on whether they're likely to do it again or whether a correction will deter anyone else from making the same mistake. That is, a correction will be valid, whether it is kind or helpful or not. But if, e.g. criminal justice analogs of this weak accountability will also be dishing out punishments based on what a criminal did, not on rehab or deterrence considerations, then this kind of accountability looks as strong as one could want to me.

Kip,

You wrote

"Unless society can develop painless "moral pills" for criminals, it will need to continue hurting them and putting them in prison; but as soon as it does develop such "moral pills" society may be morally obligated to use them."

I take it you don't mean that society would be *morally responsible* for using such pills if they were available...

[Mark Smeltzer]

Heath and Kip,

I think Heath's question is a very good one. Without responsibility and accountability, I have no idea how to make sense of the relation between norms and people -- regardless of whether the norms in question are moral or not.

[Tamler Sommers]

Heath,

I agree, this is very helpful. You wrote:

"Wouldn't it then be reasonable to say: well, look, this new practice of holding-accountable or non-moral responsibility or whatever you want to call it is just what responsibility *is*; it's compatible with (perhaps driven by) determinism; so we're compatibilists."

It would be reasonable only if we never thought that bad behavior was anyone's FAULT. But most people do, at times anyway, think that criminal behavior is the criminal's fault. They do hold them strongly accountable. And so I don't think we're obliged to be comaptibilists.

Your last remark is very interesting but I'm pretty sure I reject the premise. I don't think I'm committed to a belief that the student is "subject to correction" in a non-consequentialist sense. And so I don't think that weak accountability can ground non-consequentialist criminal punishment either. What was it that I said that made you think I was committed to that view (that students are subject to non-consequentialist correction)?

[Robert Allen]

The operative question is not whether or not one can imagine a society whose members do not hold each other morally responsible. (Heath has shown, however, that we cannot.) On the assumption that a view by which one could not live is philosophically suspect, one must ask oneself instead, could I myself, in the here and now, forego thinking of myself and others as praiseworthy and blameworthy? Could I live the rest of my life without holding myself and others responsible for anything? Prof. VI's argument is based on the belief that anyone answering this question affirmatively is a hypocrite. Not even the nihilist Nietzsche managed to pull it off. Witness some of the chapter titles of his autobiography: "Why I Am So Wise," "Why I Am So Clever," and "Why I Write Such Good Books." And Tamler couldn't even sit through a Red Sox game without doing it.

[Tamler Sommers]

Robert,

I agree with almost everything you just said...which has got to be a first. (Even the Red Sox game remark. That's the one area of life where I willingly suspend disbelief about moral responsibility. A-Rod deserves his humiliation, basic argument be damned.) I don't think that Heath has shown (yet) that we can't imagine a society where people didn't hold others morally responsible. But as you say, even if he has, PVI's argument still could be undermined by an individual successful attempt at not holding himself and others MR.

I don't think the Nietzsche example works for you though. The actress who plays GoGo in Kill Bill could write a book called: "Why I am so Hot" without believing she was strongly accountable for being hot . Similary, Nietzsche may have merely been trying to understand or explain why he was so clever without giving himself any credit for being clever.

[SeanD]

If y'all will excuse a hopelessly naive comment:

Don't all of our singular judgments regarding blame/praise in particular cases add 'force' to the case for the existence of MR (though not, of course, necessarily for Libertarian MR)? Wouldn't abandoning all of those judgments constitute a very serious theoretical cost? At the very least, they would need to be 'explained' away, either by an error-theory or by an analysis of moral language which makes blaime/praise attributions non-commital with respect to responsibility.

Is it plausible to suppose that if a hypothesis (i.e., that there is no MR) forces the revision either of a large number of singular judgments or of important theoretical beliefs in another area, then it should be regarded as in some important sense 'less probable' then its more conservative opponent?

[Robert Allen]

"Is it plausible to suppose that if a hypothesis (i.e., that there is no MR) forces the revision either of a large number of singular judgments or of important theoretical beliefs in another area, then it should be regarded as in some important sense 'less probable' then its more conservative opponent?" Sean

According to Quine and Ulian, conservatism is indeed a virtue when it comes to framing hypotheses. But the free will skeptic's hypothesis goes beyond being radical: it is impossible even to fathom, let alone live by, putting its advocates in an untenable position.

Tamler,

The Chapter titles of Ecce Homo are just the tip of the iceberg of megalomania it displays.

[Bill Edmundson]

I think Heath goes a step too far in saying (to Tamler) "the kind of (weak) mutual accountability you have with your students is desert-entailing in the sense that someone who utters a contradiction or makes another mistake in reasoning is subject to correction." Being subject to "correction" does not entail capital-F fault, nor moral desert. For me the really puzzling question is what sense to make of guidance by reasons in the absence of DEMR. I too suspect that the notion of being guided by reasons has to involve some notion of control--but I think something like compatibilist local control can make sense of rational guidance even though it cannot, in the end, make sense of DEMR.

[Anibal]

I think that in many disccusions concerning responsibility (moral or criminal)authors are more easily lean toward a monolithic conception of responsibility, failing to acknowledge a more dynamic conception of responsibility. I have notice via internet of many notable philosophers that are currently developing an ecological approach to responsibility (Hurley). Under an ecological approach to responsibility, responsibility can be seen (i think) as an effective principle or way of organizing brain, cognitive, affective and behavioural systems within persons or between persons to control the answerable, deserving, and reactive attitudinal responses of responsible agents in interaction with other responsible agents (and enviroment) that mutually negotiate or trade to be adscribed or assigned responsibility themselves according to the local nuances of their ecologies (a judge is not equally worthy for their acts than a less demanding worker etc., just for their simple reason of their different "reponsibilities")
Responsibility in this sense, is plotted in a parameter space similar to that defined by Churchland (2005)in which people is or is not "in control" but not in abstolute mannner but rather in a fuzzy manner taking into account several factors. Responsibility and morally conceptions of being responsible, under the perspective of ecological approach depends of the several fluctuatings variables internal to us (e.g. hormones, neurotrasmitters, damaged to prefrontal cortex...) and external (e.g. situational fctors) that change from time to time during a single day, and also in the use we make of our folk psychology as a social tool to manage our conduct (and the conduct of others) in the social world (Malle 2004; Malle Knobe 1997a). That is, we can be responsble in one context and not responsible in another, optimally responsible exercising a particular practice in one context but not responsible exercising that verey particular practice in ohter context, and also even be responsible at the morning but not at night, similar to our biological circadian rhythms but in this case "responsible rhythms" ( hence the notion of ecological responsibility).
Beyond the sphere of an agent-focus asumption of responsibility, responsibility is a necesseary need if we don´t want to fall on big caveats and riddles unssolvable to date, despite evidence in cognitve and empirical disciplines which are now throwing doubts about the illsuion of free wil and thus responsibility. Smilansky, is right in stress that point with philosphically elegant arguments, but i believe that even if total scientific knowledge and the most sound and robust philosophical argument dissacredit the posibility of being repsonsible for something; in the "real world" we are obligued to lay in that illusion because is a "useful" illusion. But this is not to say that libertarianism is true, just only that compatibilism is the correct view. With respect to the political dimension of moral responsibility, i dont think that determinism is more favorable to left ideologies and their attitudes and practices rejecting severe and retributive punishment (or at least exculpating more individuals for their bads acts) and being ethically more respectfull with individuals as a consequence of recognizing the illusion of free will and therefore responsibility. Becuase if we observe that more disadvantage people are in the position they are (behaving antisocially and illegally for cuases beyond their will) why to design social policies to change precisely that, because is totally imposible if determinism is real and in play. Bad people always will be bad people, and good people, good people; nothing can we do.
But if we belive substantively and in basic manner in free will and therefore in moral responsibility, we can expect that people can change their bad outcomes, correct and modify their lifes and of course self-perfect themselves to be better people that make better places, and a better world.

[Steve Darwall]

This is a fascinting discussion. I agree with Tamler that the sense of accountability that is normally in play when we have what Pettit and Smith call something like "serious conversations of the intellectual kind," although involving a form of accountability, don't yet have the robust notion of moral fault. But for some arguments that our notion of moral wrong and moral obigation do conceptually involve the idea of moral accountability with this robust notion, I might point you to a paper I gave at the Watson Conference at Riverside last year: "Morality and Accountability": http://www.philosophy.ucr.edu/conference/Darwall-Obligation.pdf

This paper just argues that these notions are conceptually independent. Of course, we might give them all up. But I believe that we commit ourselves to them when we take up what I call a second-person standpoint. I have a book currently in press at Harvard (The Second-Person Standpoint: Morality, Respect, and Accountability) that argues this and that will be out next fall.

[Steve Darwall]

Sorry. That should have been "fascinating," not "fascinting." BTW, I should add that I don't regard my arguments for moral responsibility as necessarily being arguments for *libertarian* moral responsibility.

[Joe Campbell]

Good question, Tamler!

I tend to side with James’s comments, way at the top. Recently, I’ve come to think that the beliefs that we have free will and that we are morally responsible for our actions are natural beliefs: they are not supported by reason but neither can reason take them away. It can’t be ‘charm’ alone that makes our beliefs in free will and moral responsibility endure (see PF Strawson’s work here). Nor can it be that we are all under some kind of massive illusion. To make such claims is to suggest that all humans are irrational but if that is true, then the term ‘irrational’ seems to be stripped of any significant meaning.

Global arguments against free will and moral responsibility — like G. Strawson’s ultimacy argument — are even worse, for they make broader claims. In these arguments it isn’t just all humans that are judged to lack free will and moral responsibility but the conclusion is that no creature can possibly have free will and thus no creature can possibly be morally responsible for his or her actions. I am at a loss to determine exactly what significance these words could have given that they apply to all possible creatures. It is one thing to imagine a world, like the Brave New World, where all creatures lack free will and moral responsibility. In that case we have the creatures in our world to compare to the creatures in BNW. But when we conclude that no creature could possibly be morally responsible for his or her actions, what is the basis of comparison? And without any such basis, what could the claim mean?

As I see it, van Inwagen’s work comprises a global argument against the free will thesis, similar to G. Strawson’s ultimacy argument. (To make things easy, let’s suppose that free will just is the freedom-relevant quality necessary for moral responsibility).

1. If determinism is true, the free will thesis is false (supported by van Inwagen’s many versions of the Consequence Argument as well as his subsequent responses to criticisms of the many versions).
2. If indeterminism is true, the free will thesis is false (supported by van Inwagen’s Mind Argument and his Mystery Argument, given in his essay of the same name).
3. Therefore, the free will thesis is false.

Van Inwagen himself admits that the conclusion is absurd. Since he thinks that the arguments in support of (1) are better than the arguments in support of (2), he supposes that we have libertarian free will. (Thus, I see the argument underlying van Inwagen’s work – all of his work on free will, not just his book – to be a bit different than the way that you characterize it, Tamler).

The mistake that van Inwagen makes (if I dare to make such a claim!) is that he fails to note that OF COURSE the arguments in support of (1) are stronger. All of van Inwagen’s arguments rely on certain transfer principles, which transfer properties like unavoidability or non-responsibility across entailments. With the thesis of determinism we get a nice, tidy entailment between propositions about the past and the laws and propositions about the future, making the transfer of the requisite properties much easier to establish. In the case of the Mind and Mystery arguments things are a lot messier, because the thesis of indeterminism is much messier. But these arguments show that once you buy into the assumptions of the Consequence Argument, the road to free will denial is conveniently paved. Strawson’s ultimacy argument only makes this point clearer. The response, thus, isn’t to suggest as van Inwagen does that there is something-I-know-not-what wrong with the Mind or Mystery arguments but that there is something-I-know-not-what wrong with the assumptions lurking behind the Consequence Argument. The proper response should be to accept compatibilism, since denying the free will thesis is absurd, a point suggested by both Anibal and PF Strawson, I believe.

(Sorry for not responding to your reply to my previous post, Kip. Some of those comments are contained here. I’ll get to the rest of it later this week. Thanks!)

[Blake]

Joe: I just don't see how you get to compatiblism from PVI's view -- or the view that denying moral responsibility is absurd. The Consequence Argument, as you admit, is more compelling and simple -- and frankly intuitive -- than the Mind or Mystery arguments. For one thing, agent causal accounts seem much more compelling and feasible to me than any compatibilist account and avoid the primary problems of the Mind argument. Work on emergence seems to me to go a long way toward anwering the Mystery problem. Further, it seems much more likely to me that the full dynamics of free will may be beyond our capacity to fully explicate than that we don't have moral responsibility. By my lights, no one yet has given a satisfactory account of praising and blaming and deserved consequences that doesn't entail some form of libertarian free will -- that requires both source incompatibilism and leeway incompatibilism.

It seems to me that belief in free will (of the type requrring alternatives) is just as basic as moral responsibility. Don't we all actually engage in the activity of considering alternatives and also feel that we have alternatives open to us when we consider our choices? How can we shake the basic belief that we in fact choose among open alternatives except by an argument showing that determinism in fact obtains? Yet there is no such argument. I know you think that Frankfurt counterexamples undermine PAP (and its numerous mutations), but it seems to me that these Frankfurt stories fail as counterexamples (for among other reasons discussed by Kevin Timpe). That's the road to libertarianism.

[Kip Werking]

“I take it you don't mean that society would be *morally responsible* for using such pills if they were available...”

This is an excellent question. It reminds one of the anecdote Tamler relates about Van Inwagen. As Tamler wrote, “The argument focuses largely on his claim that he can’t take anyone seriously who denies moral responsibility, because in the next breath they’ll assert that stealing books is ‘shoddy.’”

To answer your question: first of all, the distinction between society and individuals does not seem relevant in this context. So suppose that society had the moral pill but chose not to use it. Would I hold it morally responsible? No. Instead, as I described earlier, I would regard this society as being sick and being in need of treatment.

Is there friction between this “medicalized” view of state sanctioned crimes and my earlier statement that “Unless society can develop painless ‘moral pills’ for criminals, it will need to continue hurting them and putting them in prison; but as soon as it does develop such ‘moral pills’ society may be morally obligated to use them.” Perhaps. In general, I prefer a utility-focused consequentialism (although I’m not a utilitiarian: my happiness is more important to me than a stranger’s). Many scholars or philosophers have considered such a teleological ethics to be more conducive to skepticism about free will, and the two may even be ultimately compatible. But the intersection between metaethics (or ethics) and the free will problem is a thorny one, and I would need to do much work/thinking.

One major source of this friction, I suspect, is that sophisticated compatibilists, when pressed, might say that free will and moral responsibility fit within this medicalized view of the world. Suppose an agent commits a murder. Suppose further that society gives this murderer a moral pill (which protects society much better than punishment does), in order to cure him of this “disease.” I can imagine sophisticated compatibilists who would insist that this agent freely murdered and is morally responsible, without deserving any punishment. In other words, criminal/immoral behavior may be just a subset of mental illness, and the behavior retains its own identity as criminal/immoral.

To be more precise, I think that there are certain elements of our moral practice which are, upon reflection, untenable. The first would be retributivism. Another would be incompatibility with total life design: if a Grand Designer designed your entire life, you cannot have free will. A third might be that criminal/immoral behavior and mental illness are mutually exclusive; the former cannot represent a subset of the latter, rather, the latter automatically excuses one’s actions from belonging to the former. Free will and moral responsibility, I think, cannot exist on any conception that maintains (at least) all of these elements. But a sophisticated compatibilism might abandon all three. A sophisticated compatibilist might consider an agent who, according to divine fiat, murdered a third party, and was treated with painless brain surgery, and conclude that the agent nevertheless had free will and moral responsibility—even if it is clear to me that, at that point, those terms have lost all utility and significance.

Perhaps it was with this sophisticated compatibilism that I sympathized when I wrote that society is morally obligated to give others a moral pill. Or perhaps, I just meant “it would be better if society gave others moral pill,” which does not imply any freedom or responsibility on society’s part. In any case, between this “medicalized” view and my own previous statement, if one of them must go, it would be the controversial statement. But I am not sure that one of them must go.

[Eddy Nahmias]

Blake writes, "It seems to me that belief in free will (of the type requrring alternatives) is just as basic as moral responsibility. Don't we all actually engage in the activity of considering alternatives and also feel that we have alternatives open to us when we consider our choices?" PvI makes similar claims about deliberation and Galen Strawson makes these sorts of claims in Freedom and Belief (1986) when he writes: it is “in our nature to take determinism to pose a serious problem for our notions of responsibility and freedom” (89) and the fact that “the basic incompatibilist intuition that determinism is incompatible with freedom … has such power for us is as much a natural fact about cogitative beings like ourselves as is the fact of our quite unreflective commitment to the reactive attitudes. What is more, the roots of the incompatibilist intuition lie deep in the very reactive attitudes that are invoked in order to undercut it. The reactive attitudes enshrine the incompatibilist intuition” (88).

I find these claims highly implausible. The reactive attitudes are clearly a part of human nature (which I am not claiming makes them impossible to question or give up). Similarly, deliberation and accountability both clearly involve a sense of the agent's being able to do otherwise. But how to interpret that ability to do otherwise--as an unconditional libertarian sense or a conditionalized compatibilist sense (though not the old-fashioned conditional analyses)--is far from clear. And there is no reason to think that the commonsense or intuitive sense of ability to do otherwise required for deliberation or accountability is the unconditional libertarian one. I would argue that the experimental work done so far suggests the unconditional sense is *not* the intuitive one but that's contentious. It's certainly not clear that the post-philosophical intuitions of philosophers point to the unconditional sense being the requisite one, given the number of compatibilists there are.

I would argue that it is only incompatibilist arguments and libertarian theories of free will that suggest that the ability to do otherwise must be unconditional. The same point could be made, I think, by pointing out that the transfer of non-responsibility principles are not nearly as pre-philosophically intuitive and commonsensical and natural as are our commitments to the reactive attitudes and the sense of moral accountability to which they are related.

Put simply: our pre-philosophical commitment to moral responsibility/accountability is much deeper and evident that our pre-philosophical commitment to either the unconditional sense of ability to do otherwise or the transfer of non-responsibility principles that, I think, are required to argue for incompatibilism. Whether that pre-philosophical commitment has any evidential value (or what sort it has) is a different and difficult question.

[Joe Campbell]

My comments were given in response to Tamler’s original characterization of van Inwagen’s argument for libertarian free will. I have some qualms about this characterization but here it is.

(1) We must have libertarian free will in order to have libertarian moral responsibility.
(2) The claim that we lack libertarian moral responsibility is absurd.
(3) Therefore we must have libertarian free will.

In response, I gave something like the following argument.

(1') If incompatibilism is true, then the free will thesis is false.
(2') It is absurd to think that the free will thesis is false.
(3') Therefore, compatibilism is true.

I also suggested that (1') is supported by the claim that the Mind and Mystery Arguments are as persuasive as the Consequence Argument. Similarly, I claim that the Ultimacy Argument is as persuasive as the Consequence Argument. Thus, I think that if you are led to believe that the Consequence Argument is sound, you should be led to believe that the free will thesis is false.

I think now that it was wrong to suggest that the Mystery Argument is on a par with the Mind Argument. A more accurate suggestion is that in the Mystery Argument, van Inwagen tries to achieve the same goal that he gets by combining the Consequence and the Mind Arguments, for the Mystery Argument begins with assumptions that drive the Consequence Argument and then shows that, even if indeterminism turns out to be true, there is no reason to think that we have free will.

You claim, Blake, that the Consequence Argument is more intuitive than the others but I think that the same intuitions that drive the Consequence Argument may be used to show that the free will thesis is false. It is not the Consequence Argument is more intuitive, it is that the thesis of determinism makes it easier to express these intuitions.

Here is a nice summary of the Mystery Argument from van Inwagen’s “The Mystery of Metaphysical Freedom.” Here he is trying to argue that even angles cannot have free will.

“… either ‘the sheaf of possible futures’ relative to each moment has only one member or it has more than one. If it has only one, the world of angels is deterministic. And then where is their free will? (Their freedom is the freedom to add to the actual past. And they can only add to the actual past in accordance with the laws that govern the way angels change their properties and their relations to one another with time.) If it has more than one, then the fact that one possible future rather than another, equally possible, future becomes actual seems to be simply a matter of chance. And then where is their free will?” (van Inwagen, 191–2)

There are similar quotes in G. Strawson’s version of the Ultimacy Argument from “Free Will.”

Libertarians “have to show how indeterminism (the falsity of determinism) can help with free will and, in particular, with moral responsibility.” (§ 2)

“… it is foolish to suppose that indeterministic or random factors, for which one is ex hypothesi in no way responsible, can in themselves contribute to one’s being truly or ultimately responsible for how one is.” (§ 3)

“… indeterministic occurrences cannot possibly contribute to moral responsibility: one can hardly be supposed to be more truly morally responsible for one’s choices and actions or character if indeterministic occurrences have played a part in their causation than if they have not played such a part. Indeterminism gives rise to unpredictability, not responsibility. It cannot help in any way at all.” (§ 5)

“… indeterministic occurrences can never be a source of true (moral) responsibility.” (§ 6)

All of these quotes, together with van Inwagen’s quotes from the Mystery Argument, suggest an even stronger argument for (1') than the one that I originally gave.

(a) If determinism is true, then we have no free will (supported by the Consequence Argument).
(b) But indeterminism cannot increase the level of freedom or moral responsibility in the world.
(c) Thus, if incompatibilism is true, the free will thesis is false.

I’m pretty sure that I haven’t convinced anyone of anything yet but I hope, at least, that my point is a little clearer.

[Joe Campbell]

I meant to say "Here [van Inwagen] is trying to argue that even ANGELS cannot have free will," not "angles"!

[Anibal]

Dr. Campbell i´ve been honored because you put my name close to the name of Sir Peter Strawson, and that´s in a nutshell what i trying to say.

[Tamler Sommers]

Thanks everyone for all the responses! This was exactly what I was hoping for. I'm sorry I haven't had a chance to reply to the comments directed my way, but I've been playing the role of single Dad this weekend. My daughter is not as interested in justifying moral responsibility as she is in scolding our dog and redistributing the items in our kitchen and bathroom drawers.

Before replying, let me see if I can assemble the responses to the post's question into a couple of fairly broad categories.

(1) We are justified in believing in MR because we cannot conceive of a society or an individual life where such a belief was not held. Therefore, it doesn't matter how strong the arguments against MR are. We cannot coherently imagine our lives without it.

(2)A life without MR might be conceivable, but it is deeply unnatural (the P.F. Strawson way..) and counterintuitive. The belief in MR is basic, and therein lies its primary justification. Renouncing it would cause a great deal of upheaval in our web of beliefs. Furthermore, our intuition that we are MR is much stronger than any intuition that fuels the skeptical arguments against MR.

Does that seem right? (I haven't read Steve Darwall's paper yet, but I gather it would be a sophisticated version of the position in the first category...)
And are these categories more or less exhaustive?

[Bill Edmundson]

Tamler, A third category would be:
(3) A life without MR (in the DEMR sense) is not only conceivable, but naturally accessible, and tutorably intuitive. Renouncing (DE)MR would indeed cause upheaval in our web of beliefs, but would also make possible a new equilibrium. Our intuition that we are (DE)MR is a vestige of our evolutionary past that cannot withstand rational scrutiny.

[Blake]

Joe: I know that many accept the a priori view that indeterministic libertarian accounts cannot "increase the level of MR." I agree with this assessment so far as event causal and non-causal accounts are involved, but Randolph Clarke (in Libertarian Accounts of Free Will ch. 9 and "Agent Causation and the Problem of Luck") and Derk Pereboom (in "Is our conception of agent-causation coherent?") and Dale Tuggy's dissertation all seem to me to be successful in showing that agent causation does not have the same problem. The issue is simply whether something beside mere event causation is possible. An agent cause isn't merely indeterminism but a basis of causal control that is not fully determined without the basic agent causal powers that explain the events. So I just don't think that libertarian accounts have been shown to be impossible or to fail to give greater control as you assert. Thus, compatibilism is not the default position if one maintains that we must accept MR.

Indeed, if determinism is true, it seems fairly intuitive that we are merely "biological processors" and thus much (tho admittedly not all) of what we seek to procure by affirming MR must be given up if determinism is true. We cannot hold mere biological processors of causal data responsible -- morally or otherwise.

Finally, Clarke has also shown (in "On an Argument for the Impossibility of Moral Responsibility") that Strawson's argument is critically incomplete and invalid as it stands. So I don't believe that Strawson's argument has much punch unless the holes in it are filled in -- and to date they have not been.

[Robert Allen]

"Another would be incompatibility with total life design: if a Grand Designer designed your entire life, you cannot have free will."

If by 'design your life' you simply mean intend one to do the things that one does, then your consequent does not follow, as I have already pointed out to you several times. Unless God did something to guarantee that someone act according to His will the latter is free in relation to Him.

[Kip Werking]

Robert Allen:

By "design your life" I mean more than mere intention. I mean something like "set up the initial conditions of your biology and environment so that, according to the laws of nature, one's life will proceed upon the desired, pre-calculated trajectory." This is, I think, the same sense that Watson uses in "Soft Libertarianism and Hard Compatibilism". Two psychologists, Greene and Cohen, discuss this same idea (the "Boys from Brazil" scenario) in their article "For the law, neuroscience changes nothing and everything".

[Tamler Sommers]

Bill,

Yes, that's certainly a viable alternative (in fact, a lot of my work is devoted to defending the claims in that category). But it's not a position that justifies a belief in DEMR so I didn't include it...

Am I crazy, or was that by far the worst Super Bowl in the last twenty years?

[Joe Campbell]

Blake,

I admit that agent causal accounts have not been shown to be impossible. But that strikes me as a rather small victory. I think you are wrong in suggesting that they escape the kinds of criticisms that van Inwagen has in mind (as opposed to other kinds of criticisms). In fact, the main target of van Inwagen's criticisms of libertarian free will, especially in "Mystery," is the agent causal theorist. And the quoted passage from van Inwagen is quite general in criticizing libertarian theories. In other words, it is not at all clear HOW it is that agent causal accounts offer a response; it only seems to be 'clear' because we don't really know what the heck is going on when the agent causes her action (and some of us strongly believe that we have free will, and that compatibilism is false).

To drive this point home consider this: If agent causation is so wonderful, why won't compatibilist accounts of agent causation, like Ned Markosian's, work? If it really is instances of agent causation that bring ultimacy into the world, they should work. Answer: Because there is nothing in the account that tells us why we shouldn't still be worried about Consequence Argument considerations. Likewise, there is nothing in the libertarian agent causal accounts that tells us why I shouldn't still wonder how indeterminism can possibly add freedom or moral responsibility to the world. One might say it does but I can't see how it does.

I agree with your views on Clarke's criticisms of the Ultimacy Argument. But there is not one single argument in G. Strawson's "Free Will" paper. There is, first, the Ultimacy Argument and, second, something like what I've been calling 'the Mystery Argument.' (And maybe one or two others.) I think the second is much better and has not been adequately critiqued by Clarke or anyone else. This one moves from establishing incompatibilism to a more general free will denial.

[Mark Smeltzer]

Tamler,

In response to the question in the thread's title, we ought to believe we are morally responsible if and only if we also believe 1) that there are moral properties which are true of us (our moral characters), and 2) we are competent actors within the moral context, such that our actions are generally truth condusive of our moral characters.

I have no reason to doubt (1) or (2), so that provides a powerful basis for refuting other views and constructing a positive account.

Blake,

Joe brought up a good point about Markosian's account of agent causal compatibilism. I am also working on an agent causal account and it is fundamentally compatibilist as well. So, there are certainly some of us who (strongly) disagree that determinism would entail that we are merely biological processors.

[Robert Allen]

Tamler,

In recent memory, Tampa Bay's rout of the Raiders a few years back was much worse, as was the Raven's thrashing of the Giants in 2001. The outcome yesterday actually hung in the balance until the 4th quarter when the officials tipped it in favor of the Steelers with the holding call on Locklear, after which the poor Seahawks simply fell apart. (I was stunned, absolutely stunned, when Holmgren decided to punt with 6 minutes left trailing by 11.) In fact, they were victimized by so many bad calls (the Roethlisberger TD, the nitpicking pass interference call negating their 1st TD, to cite the most egregious blunders) that I refused to pay off on the bet I made my wife. And God help the guy at Ballys if he tries to rub it in.

"By 'design your life I mean more than mere intention. I mean something like 'set up the initial conditions of your biology and environment so that, according to the laws of nature, one's life will proceed upon the desired, pre-calculated trajectory'." Kip

Who would disagree with such a claim? It simply amounts to ‘If A is under God’s control, then A lacks a free will’. But why would anyone think that God is a GD? Why couldn’t He have thrown me into the world instead, to use Heidegger’s apt phrase. That is, my “environment” need not have been arranged by Him to produce my actions, which means that your causa sui argument is a non sequitur: the fact that someone created me does not entail that I lack a free will. Knowing what I will do in S does not entail intending what I do in S.

[Joe Campbell]

The Super Bowl left a bad taste in my mouth. As Robert noted, there were so many calls against Seattle that it is hard to think that the game wasn't fixed. I never saw a Super Bowl with such a one-sided set of calls. Nonetheless, Robert is also correct in noting that Seattle fell apart. Why did they punt? Why did they pick the two-minute warning to start throwing passes over the middle? Why didn't they kick a field goal and try for an on-sides kick? Why did they abandon their game plan in the first place? All of it plus two missed field goals and bad clock management in the last few minutes spells L-O-S-E-R.

But this last point is why you need to pay your wife, Robert. There is no reason to think that Seattle would have won even if the numerous bad calls had gone their way. I think we have a nice example of causal overdetermination here, football's version of a Frankfurt-style example. Even if it was fixed, Seattle is still blameworthy! Pay up, Robert!

[Blake]

Joe and Mark: I agree that the mystery argument has not been fully answered. But I don't see why we should expect mystery in its entirety to go away because a libertarian agent causal account presumes that there is no full explanation outside the agent in external causes so looking for a full explanation is to presume that we must have such an explanation or the account is defective. I reject that. Agent causal powers are basic.

What agent causal accounts explain is why the agent's action is within the control of the agent, not why the agent performed the particular act that was performed. So there are two loci of mystery -- and the first seems to me to be explained but not the second. However, no libertarian account should expect that there is an explanation of the second type of mystery.

I believe that we need to adopt an account of emergence like O'Connor does to come up with an explanation that remains mysterious but doesn't beg the question. Life, mind and causal powers of agents are mysterious only in the sense that the lower levels of organization don't fully explain the downward causal powers or basic powers of organization of the higher levels. But that mystery is given in the very definition and nature of emergence. However, once it is seen that agents possess these basic causal powers, how they exercise control over there acts in a sense compatible with MR is no longer a mystery -- and that is the mystery argument that Joe presumes it seems to me.

I remain very skeptical that determinist agent causal accounts really get us to MR in an acceptable sense and thus they begin attempting to explain what libertarian accounts do but just cannot get there. What they lack is a sense of ultimacy, of the fact that the agent him/herself is the one who is accountable. I've read Joe's accounts and they don't provide the kind of utlimacy needed it seems to me. Though no theorist ought to demand that any agent is absolutely utlimate in self-creation, it seems to me that an agent can be ultimate in the sense that the organization of an underlying chaos of data is explained by the basic agent causal power of agent to organize the prior data into meaningful and novel patterns. Free will is like the power of invention -- not ex nihilo, but a power of organizing prior data in novel ways that are not determined in all respects by those prior data. So the organization is novel but it is not ex nihilo.

Admittedly more work remains to be done with emergence, but if we have MR, it seems to me that type of account of agent causation is required.

Joe: Your point about Clarke not responding to all of Strawson's arguments is well taken. However, I believe that Strawson's second argument begs the question against libertarian accounts by confusing mystery in how an agent has control with explaining the particular act the agent does.

[Robert Allen]

Joe,

Consider it done and thank you for bringing me to my senses. (I also enjoyed your other posts in this thread.) But consider this also: I don't know if you caught it, but it was related during the 1st half that the head linesman, a gentleman named Leavy, knew Holmgren from way back when, having officiated high school basketball games in which he coached. Now I'm usually not one for conspiracy theories, but there were so many bad calls against Seattle that I'm wondering whether there was element of payback there for something Holmgren did to rub Leavy the wrong way.

[Blake]

Joe: We agree on one thing. There were too many bad calls against Seattle for it to be mere coincidence. I'm no Seahawks fan, but if the holding call is not made when Seattle would have been on the two yard line ready to go ahead by 4 instead of having 3rd and forever, the game may well have turned out differently. As an ex-middle linebacker, I know how deflating it is to make a big play and have the ref blow the call and set the team way back. And what was that with the touch-down and the failure to overturn the callt when the ball rather clearly hit about 1/2 yard in back of the line?

[Mark Smeltzer]

Blake,

My experience has been that if we start with the common sense presuppositions noted previously*, we can produce a maximally robust account of freedom and responsibility that adequately accounts for agent-causal control, solves the mysteries typically associated with agent-causal control**, and is fundamentally compatible with determinism. Unless one starts with the presupposition that this kind of control is incompatible with determinism, one will be hard pressed to arrive at that belief from those starting points.

The second presupposition is the freedom relevant condition, and there is no necessary connection between agent-competence and the falsity of determinism. The type of control required to facilitate agent-competence is, fundamentally, equivalent to what Fischer calls "guidance control", which is the cornerstone of semi-compatibilism. Thus, even though the presuppositions under discussion are agent-causal in nature, they do not entail incompatibilism.

* The presuppositions were: 1) there are moral properties which are true of us (viz., our moral characters), and 2) we are competent actors within the moral context, such that our actions are generally truth conducive of our moral characters.
** There will still be closely related mysteries left to solve concerning moral realism, epistemology, the metaphysics of objects persisting over time, etc.

[Joe Campbell]

Blake,

Thanks for the comments!

You write: “What agent causal accounts explain is why the agent's action is within the control of the agent, not why the agent performed the particular act that was performed.”

The first part of this claim seems wrong to me. Exactly why is it that the agent’s action is within the control of the agent? Because it is the agent’s action? That doesn’t seem very helpful. To the extent that it is helpful, it seems that the compatibilist agent theorist can give the same answer. Nor is it that I am looking for “a full explanation outside the agent.” I’ll settle for a partial explanation inside the agent.

You have to admit that even if determinism is true, there must be some way to distinguish things that I do from things that merely happen to me – some way of distinguishing between jumping off of a cliff and being pushed off of it. It would seem pointless for the incompatibilist to deny this. What the incompatibilist denies is that this difference can ground a robust notion of control or freedom. You tell me that the agent causes her actions and I believe you. Now tell me how it is that the agent causes her actions in a way that makes her have the kind of control or freedom necessary for moral responsibility.

Here is a challenge to make the point clearer. Tell me what it is that distinguishes instances of indeterministic agent causation – the positive component that grounds the agent’s control, freedom, and moral responsibility – from compatibilist versions of the theory? Of course, you can tell me what it is that the latter has but which the former lacks, e.g., an explanation of the action as an event that is causally determined. And you can suggest that this additional component makes it such that the agent’s act is not under her control and thus that she is not morally responsible for it. But that is something different.

If you’d like I can present the challenge in a different way, one that is more in line with van Inwagen’s Mystery Argument. Suppose that there are two agents, X and Y, both living in worlds that are not causally determined. Both seem to perform some action type z in circumstances that appear to be identical. Thus, it seems to external observers as if X does z' and Y does z" (where z' and z" are instances of the same action type), and it seems to X as if she freely does z' and to Y as if she freely does z". Nonetheless, X is actually the agent cause of z' whereas z" is not agent caused by Y. Y attempts to agent cause z" but for some reason fails and z" just spontaneously happens, for no reason at all. What exactly, according to the libertarian agent theorist, is the difference between the two cases?

You suggest that compatibilist accounts cannot provide an explanation of the kind of ultimacy that is necessary for moral responsibility, yet unless the above challenges can be met I would claim that the libertarian agency theorist cannot do so either.

[Blake]

Mark and Joe: You both suggest that Markosian's compatibilist agent causation is a good example of a determinist agent account. In my view it is multiply flawed. His initial notion of counterfactual causation is not causation in a sense any incompatibilist would recognize as a deterministic thoery of causation. His account results in biting the bullet on several coercive scenarios where he is compelled (pun intended) to say that the agent is morally responsible and free though coerced. His example of agent causation in his scenario of the strongest theory of causation (that the past together with laws physically entails but actual future) is not a valid instance of morally responsible action. I have several criticisms that I am in the process of drafting -- but for now the concerns expressed by Robet Allen regarding Markosian's account seem persuasive (tho not decisive).

Joe: What the agent X does in the incompatibilist scenario to bring about z' that Y doesn't do to bring about z" is to creatively organize the data of the prior moments of experience into a novel choice or act of willing that is not fully explained without X's basic power of organizing data into novel patterns and then to downward cause his/her physical act by this act of choice. What the agent X does is to creatively produce an act of willing that downward causes an effect or act (and I'm with Kant that the ultimate moral locus of responsbility is the will). In the deterministic scenario, Y does not create the act of willing but has the act of willing caused in her (it happens to her) by prior causes. So Y has no control over whether s/he produces either the act of willing or the act that issues of the so-called "act of willing."

[Mark Smeltzer]

Blake,

I haven't read any of Ned Markosian's work, so I cannot comment on the merits of his account -- it's definitely on my to-do list. I was alerted of his work when I started asking around to find out whether there were any other agent-causal compatibilists. As such, my claims should be dealth with independelty of his account.

[Mark Smeltzer]

Blake,

Sorry for the double post, I should have finished reading your post and provided comments all together. Anyhow, I wanted to comment on this statement, "In the deterministic scenario, Y does not create the act of willing but has the act of willing caused in her (it happens to her) by prior causes."

I am curious what you intend to denote by "her" in that sentence. You seem to be painting a picture of an agent who is trapped by deteminism, but that is an obviously flawed interpretation of the consequences of determinism. If determinism is true, then it is part of her, her willing takes place thanks to determinism, and if she is a moral agent, there is a clear sense in which she contributes to the event flow (the effects of her character are imprinted on event streams which flow through her). If determinism is true, it must go all the way down.

Perhaps you could clarify why it is problematic that in the deterministic scenario, she cannot help but express her character (if determinism is true, she obviously lacks that kind of regulative control -- viz., an ability to be other than she is). I see the work of her character is to affect itself in a way that is true condusive of itself. So, if her actions are not a conduit for expressing and ultimately knowing truths about her character, what good do her actions serve?

In other words, why would the kind of freedom involving an ability to be other than she is (the kind of freedom involving regulative control) be valued over the truth-condusive kind offered by guidance control? Pardon me for not seeing the value of arbitrary change...

[Joe Campbell]

Blake,

I claim that (a) this account is not very explanatory -- exactly what does it mean to "creatively organize the data of the prior moments of experience into a novel choice or act of willing" -- and (b) there is nothing about this account that is essential incompatibilist?

Forget Markosian, what about Kant, for instance? Suppose you have dual explanations of the same action: a deterministic, theoretical explanation in terms of prior events and a practical explanation in terms of the agent's creative organization. If the latter should be enough to get me to forget about my worries about spontaneous, uncaused 'actions,' why shouldn't it be enough to get you to forget your worries about determinism?

My point is not that deterministic agent causal accounts are necessarily good but that they are as good as their indeterminitic cousins.

[Blake]

Mark: By "her" I mean the basic event causal structure of her choice-making faculty (whatever it may in fact be). The structure of the faculty was caused by events before her birth and outside of any "agent's" control and, thus, the "choices" that flow therefrom are also outside of her control. In every moment, both what she is and the deterministic results of what she is are the result of prior causes beyond her control and which did not choose. This event-causal structure may contribute to the causes but only as already caused to be what it is -- so there is never a time when what she is and what she does is up to her or within her control. In all respects, the causes merely act through her. That is the problem and it is why the so-called "determined agent" is not really an agent at all if determinism is true. She is a mere biological processor of prior causes.

Joe: I suggest that the phrase "creatively organize the data of the prior moments of experience into a novel choice or act of willing" actually explains a lot. First, this phrase must be imbedded within a usage -- and the usage I have in mind is the process metaphysic and particular theory of Charles Hartshorne. Second, what it explains is a basic human causal power that we know we in fact have, i.e., the power to creatively organize the disparate and chaotic data that goes into our experience into a whole and unified experience that we sometimes bring to consciousness. In the act of perception, we actually organize data according to schema and add a unity to our experience that is not present within the data themselves. So there is an irreducible element of creativity and novelty in this act of organizing prior data into our experience. What we add to the data of the causal past is a creative organization that shematizes our experiences (perhaps along the lines of Kant's categories).

Further, we also have the basic power to organize the chaotic activity of populations of neurons in the cereberal cortex into an ordered process of neural activity. In fact in tests the chaotic neural activity shifts into a self-organizing neural patterns in the act of perception (somewhat like the way chotic water molecules in a tub drain self-organizes itself into an orderly process of a spout or funnel that is much more efficent at emptying the tub).

In the experience of choosing, what we do is take the data of prior experiences and add to these data something not in the data -- and in this activity of ordering the data we create experience, consciousness, choices, thoughts and so forth. The new level of mind is emergent and not fully or deterministically explainable by the lower level causal data. Thus, the break with determinism occurs at the very place we expect it -- the act of creating our consciousness and thoughts and (dare I say it) our acts of willing and choosing among alternatives. The prior data do not explain fully or determine the causal powers of the mind that emerges from the lower level of neural activity. (Such a causal break with levels of explanation is simply what occurs in ontological emergence).

The control comes in the fact that we control the orderly emergence of our mental activities. We organize the data as we learn that we can given the past data available to us -- so it isn't ex nihilo and it is related to but not determined by prior character and prior physical states. So the act of willing is: (1) within our control; (2) a basic power that we have; (3) not determined by the prior causal data; and (4) not fully explainable by reference to lower levels of organization due to its emergent properties. Further, in the act of choosing we both act partly out of our character and also partly reform our charcter - which is what is required if we have any responsibility for what our character is.

With respect to Kant's dichotomy of the world of phenomenal appearances that are deterministic and noumenal realities that are indeterministic, I believe that Kant is methodologically correct. If we attempt to fully explain phenomena in the sense science assumes and strives for, then we assume (without proof) that the world is deterministic. But as agents experiencing our freedom and acting necessarily as choosing among alternatives, we know that it is not fully deterministic. Further, science hasn't come within light years of predicting or giving laws of predicting any given individual's choices. I contend that it won't and cannot reach that level of explanation precisely because libertarian free will emerges with the level of physical complexity that we see in a proper functioning human cereberal cortex (and to less degrees in lower forms of life depending on the complexity of their neural networks).

So I claim that the libertarian view is viable and superior to compatibilism. On the agent causal libertarian view, MR is a feature of proper functioning individuals precisely because such agent causal powers emerge therewith.

[Joe Campbell]

Blake,

Let me think about this more. I like the process philosophy stuff above a lot! I don't want you to think that what you've said is unimportant or uninteresting or untrue just because it still doesn't help me to understand exactly how it is that I can be the agent cause of my actions. But the explanation may well be in the passages above and I just haven't gotten to it yet. So I'll read it over more carefully and get back to you.

But the passages on Kant I think I understand clearly and I still don't see why my actions can't both be free, when viewed from some agent-centered perspective, yet causally determined, when we "attempt to fully explain phenomena in the sense science assumes and strives for." Don't these dual explanatory viewpoints help us to understand why we don't see free agents in a causally determined world even though freedom is compatible with determinism?

[Sergi]

I'd like to announce you all that a new and very interesting book on moral responsibility has just been published. Its title is 'Moral Responsibility: The Ways of Skepticism'(Routledge, 2006), and there Carlos Moya explores the main arguments about moral responsibility and subjects them to veru lucid criticisms. In the last part, he tries to break skepticism by replacing the usual defence of free will on the basis of free choice with a new account focused on agent's beliefs. He defends his proposal as a kind of libertarianism, but I think it has a valuable collection of insights for compatibilists.

[Joe]

Joe: I'll wait for your further response -- but I wanted to comment on Kant's view. First, the notion that the world is deterministic is a mere phenomena, a mere appearance that is merely assumed because of the methods we adopt. William James made the same point in the Dilemma of Determinism -- science must assume determinism but that assumption doesn't make it so. It is somewhat like the fact that we must assume that we are free in a libertarian sense because it is impossible to act otherwise -- and we consistently make choices among altneratives in simply living (Wheaties or Cheerios or skip breakfast, steal it or pay for it). However, Kant regarded the knowledge of choice as an a priori truth because it must be assumed to make sense of our experience at all. So the reality is the world of libertasrian free will, the mere appearance is determinism. At least, that's how I see it and how I read Kant.

[Mark Smeltzer]

Blake,

Again, same question, but new sentence, "By 'her' I mean the basic event causal structure of her choice-making faculty (whatever it may in fact be). The structure of the faculty was caused by events before her birth and outside of any 'agent's' control and, thus, the 'choices' that flow therefrom are also outside of her control."

Since you used the term in question as part of the explanation, I'm not any closer to understanding what you mean by "her" in the determinist scenario. If I had to take a guess, I'd say that it still sounds like you're talking about trapped agents.

I'd also like to hear you answer my question about the value of incompatibilist agent-causal freedom. Why is it (an ability to be another than one is), in principle, to be desired over the compatibilist version (an ability to be who one is)?

[Blake]

Mark: It seems to me that in a deterministic universe, the only structure of personality or character that one can have is the event causal structure -- a given that always precedes any action or existence by the agent. So yeah, I see determined "agents" as trapped within a given event causal structure that merely acts as a processing format or algorithm. We may be very sophisticated processors of prior data, but given that the outcome is determined by the prior data long before we even exist it must be the case that we are merely processing data and not creatively interacting to transform it into something that is not merely an alogrithimic outcome. You'd have to explain to me why you believe that view isn't entailed by materialistic determinism.

Incompatibilist agent causal freedom is to be desired over compatiblist (non)freedom because it provides a basis for an unified self or conscious experience and morally responsible agent that emerges from the basic power to organize the manifold of experiential data. As I suggested above, only the libertarian agent causal account (imbedded within the process metaphysic that I have only gestured toward) can explain how we act partially out of a history of learning and identity created from the past data and also partially reform ourslves in the very act of choosing. It seems to me that a view of agency must be imbedded within a larger metaphysic that provides explanatory background of the underlying assumptions about action, causation, continuity and discontinuity in identity and so forth -- and that is where the process metaphysic that I mentioned comes into play.

One of the things I like about this approach in particular is that I am not stuck in my past; I can choose (within some limits of the potentiality of the past data such as DNA and social constructs) to transcend my past by making new choices. I make new choices by creatively organizing the data of my experience in a novel way and direct it to a new goal or end. As Harthsorne stated, each of us is an artist and the fabric of our art is our experience itself and the beauty of the art is the sum of the choices we make. (Was that too poetic in this analytic arena?)

Do I really need to explain the value of a personal self that chooses in a morally responsible way in comparison to a mere biological processor of algorithms?

[Mark Smeltzer]

Blake,

I never voiced approval of the thesis of (strictly) materialistic determinism. Simply that freedom is compatible with it. As an agent-causalist, I am a realist about agents and persons. So, naturally, I am going to be opposed to any views that reduce persons to mere events. I presume we are in agreement here.

My personal views aside, freedom (that which justifies praise and blame) simply requires that there are moral properties which are true of persons. From there responsibility (that which justifies praising and blaming) also requires that these properties be communicable such that we can (to some degree) be assured that we can reliably come to know an agent's character by examining its actions and thus be in a position to hold that agent accountable according to (not for) the character we believe the agent possesses.

Candidates for satisfying the requirements of this account are emergentist and substance dualist views of persons. I don't see any other alternatives.

Lastly, I see incompatibilist views as fundamentally opposed to freedom and responsibility in the sense delineated here since they, from the start, are grounded in the idea that agents must be-able-to-be-other-than-they-are, which militates against the very idea of responsibility since it is dependent upon a concept of exchanging information about agents. In the incompatibilist scenario, objective information about agents is seen as the enemy, and as such incompatibilism militates against the very practices it is supposed to underwrite (loving, hating, praising, blaming).

[Clark Goble]

BTW Markosian's paper A compatibilist version of the theory of agent causation doesn't appear to be online anymore. Does anyone know if it was just moved or was taken down permanently?

Mark, by "emergentist" do you mean both kinds of emergence? Regular and radical (or ontological)? When you say, "reduce persons to mere events" it sounds like regular emergence would be insufficient. Only the radical emergence that one can find for instance among the process thinkers Blake follows.