Modern Day Hobartians?

I've been re-reading Hobart's classic "Free Will as Involving Determinism and Inconceivable Without It."  This got me thinking--what contemporary compatibilists think that free will is compatible with determinism, but incompatible with indeterminism?  I know that Fishcer, for example, intends his view to be neutral with respect to the truth of determinism.  This issue comes up, for example, in discussions of the Luck Objection and the Mind Argument and that many compatibilists think that indeterminism wouldn't help secure free will.  But are there recent papers explicitly defending this kind of Hobartian view where the falsity of indeterminism would, by itself, render all agents unfree?

Comments

[Mark]

I remain very suspicious that indeterminism is freedom undermining, but I doubt I will publish anything to that effect for several years.

[Eddy Nahmias]

It'd be silly to argue that indeterminism is inconsistent with free will or moral responsibility, if indeterminism just means there are some non-deterministic relations in the universe in question. The indeterminism could be (a) in parts of the universe outside the relevant agents' light cone, or (b) such that it affects the agents' world but not in any way that influences their decision-making or actions directly, or (c) "cancelled out" at the macro-level, etc. It's hard to see how any of these possibilities would be relevant to the agents' FW or MR.

What Hobart has in mind, of course, is indeterminism in the places libertarians need it to be, places relevant to decision-making and/or action. But I think Kane has done enough to show that there is no reason to think such indeterminism (e.g., in the brain during "close call" decisions) undermines FW and MR, though there may be every reason to think that it does not increase or preserve FW or MR given the compatibilist conditions libertarians like Kane need to be in place.

I've always felt that there is something attractive about indeterminism in places that affect the course of one's life, that it does preserve the garden of forking paths in an aesthetically pleasing way, but it doesn't seem to help with MR or the sort of free *will* worth caring about (it may help with a sort of *freedom*). This sort of indeterminism would not have to happen in the agents' brain. It could happen anywhere such that it makes it true that things could happen otherwise without the past or laws having to be otherwise.

Question: does anyone know if anyone has ever offered an argument to this effect: even if determinism is true and X occurs, ~X could have occurred, *holding fixed the actual laws of nature* but *without* the entire past being different(and more and more different the farther in the past you go)? Am I right that Lewis prefers the "small miracle" of the laws being slightly different because he thinks that otherwise the whole past has to be different? But it seems like the seemingly small change in laws is only attractive on a Humean view of the sort Lewis accepts.

[Chris]

Eddy,

With regards to Lewis I believe he was primarily driven by concerns of the truth-conditions for counterfactuals. So a world in which a small local miracle occurs before the time of the action in question would be more similar to the actual world than a world in which the entire past was different. I believe it is again considerations of similarity that drives him to posit a counterfactual asymmetry between the past and future: the past is counterfactually independent of the present but the furtue is counterfactually dependent on the present.

[Randy Clarke]

I think Mele's luck argument in Free Will and Luck has the import that, if there's indeterminism where libertarian accounts require it, then agents aren't responsible for their actions. Mele doesn't himself accept the argument. But someone who does, and who also thinks that responsibility is compatible with determinism, would have a view like Hobart's.

[Chris]

If Randy's right about Mele's argument, I wonder if the best understanding of Haji's arguments against libertarianism would also make him a modern day Horbartian.

Another, some what tangential question. I have become a little confused as to what *the* Mind Argument is. According to PVI (1983) there are three strands to this argument which seemingly can be traced back to Horbart, Nowell-Smith, Smart and Ayer; possibly even as far back as Hume according to Pereboom (2001). However, at best it seems only the first two strands of the Mind Argument can be traced back to these authors. The third version which employs Beta seems nowhere to be found.

Is the Mind Argument to be identified with the three different versions PVI presents? Or instead should it merely be loosely associated with arguments the import of which purport to demonstrate the incompatibility of free will and indeterminsim? Or something else?

[Jez]

Haji is a hobartian as I understand him. In his book 'Deontic Morality and Control' he maintains that determinism undermines deontic morality but is needed for moral responsibility (and vice versa). So we can either have deontic morality or moral responsibility but not both. At least, I'm pretty sure that's his position.

I don't know how to start a thread, so I'm going to plonk a question here in the hope that someone might help me.

I've been wondering whether agent-causation is compatible with determinism. I assume it is because I assume that one can have substance to substance causation, and that such causation could be deterministic. Am I right in these assumptions? Thanks for any help.

[Jonathan Farrell]

Jez, Ned Markosian certainly thinks that agent-causation and determinism are compatible. See his 'A Compatibilist Version of the Theory of Agent Causation' in Pacific Philosophical Quarterly (1999).

[Randy Clarke]

Jez, I think that if substance causation is possible at all, then it could exist without indeterminism. I argue for this in a couple of places in chs. 9 and 10 of Libertarian Accounts of Free Will. (I wonder why you say "substance to substance" causation, though. Isn't the effect a change or the continuation of a state?)

[Mark]

Jez,

I maintain a flavor of agent-causal compatibilism, but I haven't published anything on it yet. As Jonathan pointed out, Markosian is a self-admitted agent-causal compatibilist. That said, the reasons that motivate me to adopt the agent-causal label may be sufficient to motivate a large number of other compatibilists sympathetic to semi-compatibilism.

Semi-compatibilists believe that responsibility-conferring actions issue from an agent's own MRR mechanism, right? Well, if you do some of metaphysics of personhood and discover that the MRR mechanism in question is a necessary component of agency, then the semi-compatibilist definition is now about agents: responsibility-conferring actions issue from (some part of) the agent.

I have a feeling that many semi-compatibilists would favor that move. For instance, I've heard from trustworthy sources that Fischer considers an MRR mechanism a necessary component of agency (he may have published something to that effect as well, but no reference jumps to mind).

[Kevin Timpe]

I raised this question because I'm thinking about a tension between two different compatibilist objections to libertarianism. The first of these is John Fischer's objection that our view of ourselves as free and responsible agents should not hang on scientific discoveries like whether or not the thesis of causal determinism is true. We could call this the "held hostage" objection. (See, for example, pages 44ff of My Way; something similar is, I think, at least partily behind the 'flip-flop' charge often raised to PvI's view.)

The second objection I have in mind is the Luck Objection. It seems to me that there is a certain degree of tension between these two objections. Imagine laying out compatibilist views along a scale. At one end of the scale would be the view suggested in Hobart's title (but which, Eddy is surely right, Hobart didn't actually embrace); on this view, any indeterminism would undermine free will. On the other end of the spectrum would be the view that free will is compatible with both deterministic and indeterministic causation at any point in the causal history of an action. I think that most extant compatibilist views would fall somewhere in the middle of these two positions along a continuum.

Here, then, seems to be the tension. The closer one is to the first pole (presumably motivated by worries about luck), the more one's view is 'held hostage to science' insofar as the discovery of indeterminism in the right spots would undermine free will. On the other hand, the closer one is to the second pole, the more resiliant one's view is to the discoveries of science, but the less luck or indeterminism would be seen to undermine free will. To put it a different way, if one thinks that free will is compatible with both determinism and indeterminism at a wide range of places causally relevant to a particular action, the less dialectical force the Luck Objection will carry—for if indeterminism per se doesn’t undermine free will, then the driving intuition behind the Luck Objection will be undercut.

[Mark]

Kevin,

In Fischer's summary article in the Journal of Ethics a few years back, he clearly states that he believes the rollback argument combined with indeterminism presents a special problem for responsibility. He does not enumerate much more than that, but it is clear that Fischer isn't entirely comfortable with indeterminism.

He is clearly worried about the problem of luck here: if we hold the agent, the laws, and the past fixed, then we rollback time, and we get a different outcome, it may indicate that something fishy is going on that bypasses the agent's MRR mechanism entirely. For example, if it were possible for nerfarious Dr. Black to loop the rollbacks until he gets Smith to commit murder, isn't this just an extreme form of a Frankfurt case?

[Justin Capes]

Kevin,

I think this is a good point. A minor question, though. Do you think Fischer takes the "held hostage" objection to be a real objection or more as a motivation for defending compatibilism. I've always understood him to take the latter view. (I'm not sure the answer really affects the force of your point; just wondering).

[Kevin Timpe]

Mark,

I find your roll-back Franfurt case very interesting. In "Recent Work on Moral Responsibility," Fischer does refer to the roll-back/replay argument as "a powerful argument" and "a major challenge to libertarianism" (101f). So there is definitely something to your comment. I think it is best to think of there being a continuum on which compatibilist views fall with respect to these issues, and I think that John's view is not too close to either end. But what I was trying to show is that the more places one thinks the presence of indeterminism would undermine free will/moral responsibility, the more 'held hostage' our view of ourselves as free/responsible agents will be given that science could discover indeterminism in those very places.

Justin,

You're right that Fischer often uses the 'motivation' language when talking about incompatibilist's view of ourselves as free/responsible agetns as being 'held hostage to science'. I don't think he sees this as a 'decisive objection' to incompatibilism. But I think that he thinks this consideration gives a reason to prefer compatibilism over incompatibilism. So I see him as raising it as an objection in some sense; for example, My Way pages 5f. But perhaps I should let John speak for himself on this if he wants to set the record straight.

Finally, I should also say that after my post, I came across an archived post here at the GFP (from 2005, if I remember right) in which Neal Tognazzini raised something very similar to this issue. It is quite possible that my own thinking about this has its source in an earlier reading of Neal's post. If so, credit where credit is due...

[Mark]

Kevin,

My understanding of Fischer's view of being "held hostage" is that compatibilists have the advantage of not being threatened by the question of whether determinism is true.

Gross, flagrant, broad, meaningless indeterminism is threatening to responsibility de facto: even libertarians can only accommodate so much of the right kind. So, we're all held hostage to some degree as regards indeterminism.

We might follow this line to ask to what degree does indeterminism threaten an individual account of responsibility, but I don't think there's any hard distinction to be found between compatibilist and incompatibilist accounts here with one obvious exception: all incompatibilist account accommodate at least some kind of indeterminism.

[Joe Campbell]

Kevin,

Just some comments about the 'Mind' argument. Also, these are comments about van Inwagen's usage of the expression, not about the usage of the expression by philosophers as a whole, which I think is what you were getting at.

In van Inwagen 1983, he writes that the argument "proceeds by identifying indeterminism with chance and by arguing that an act that occurs by chance ... cannot be under the control of its alleged agent and hence cannot have been performed freely. ... Proponents of the 'Mind' Argument conclude, therefore, that free will is not only compatible with determinism but entails determinism." (16)

This is interesting fror a couple of reasons: (1) the use of the expression "'Mind' argument" to designate what we might today call 'the luck argument' and (2) the suggestion that the 'Mind' argument arrives at a kind of Hobartian conclusion.

In van Inwagen 1983, pp. 126-150 he discusses the three strands of the 'Mind' argument that you note above. Note that this is contained in the chapter entitled "Three Arguments for Compatibilism," which explains point (2) above. As Randy said, Mele uses considerations of luck but as part of an argument for compatibilism. The 'Mind' argument is offered here is the third argument for compatibilism that van Inwagen discusses.

In "Free Will Remains a Mystery" (2000; reprinted in Kane 2002), van Inwagen appears to define the 'Mind' argument more narrowly. He writes that "many philosophers have noted, [free will] also seems incompatible with indeterminism. The standard argument for this conclusion (which I have called the Mind Argument ...) goes something like this ..." (Kane 2002, p. 167) He then offers a rather long argument, the conclusion of which is more or less if indeterminism is true, then what one does is a matter of chance and hence not free (p. 168).

Then there is a parenthetical qualification: "this is one way to formulate the 'Mind' Argument. Other statements of the argument are available, including some that do not appeal to the concept of chance." (p. 168) Lastly, he adds on this same page: "In an Essay on Free Will, I tried to show that the 'Mind' Argument depended on the 'unrevised' version of rule Beta." (p. 168)

Thus, it seems that van Inwagen's usage of the expression "'Mind' argument" is ambiguous in several respects. Initially (1983) it was intended to be an argument for compatibilism and some Hobartian conclusion. Later (2000), it is used more narrowly as an argument for the claim that indeterminism is incompatible with free will, though one that makes appeal to considerations of luck or chance. And he seems to think that (revised) Beta is essential to the argument, or at least that Beta helps us to provide a version of the argument that is more revealing in important respects, which would explain the eventual identification of the 'Mind' argument with the third strand noted in 1983.

[Chris]

Joe,
Thanks for the helpful clarifications. I have been trying to figure out what the essential features of the Mind Argument are but to no avail. Specifically, I am trying to understand how, if it all, the Mind argument differs from the luck objection (raised most notably by Haji and Mele).

I failed to notice the quote on page 16 from 'An Essay on Free Will'. That is very interesting since, as you suggested, that presentation of the structure of the Mind argument makes it seem awfully close, if not identical to, the luck objection. Moreover, it always seemed to me that the first strand of the Mind argument was simply an underdeveloped version of the luck objection.

I have also wondered about the relation between the "rollback" argument and the Mind argument. I am beginning to wonder if the Mind argument should be identified simply with any argument that attempts to show that free will and indeterminism are incompatible. Under this understanding the Mind argument is an argument for compatibilism only because it gives those of us who believe that free will is a non-negotiable truth, reason to be compatibilists. This understanding of the “sense” in which the Mind argument is an argument for compatibilism seems justified since van Inwagen himself concedes that it is strictly speaking only an argument for the incompatibility of free will and indeterminism (see ‘An Essay on Free Will’, p. 148)

The Mind argument will then be identified with any of the following types of arguments: One might argue for their incompatibility (i) by arguing that indeterminism entails chance and chance is incompatible with freedom (first strand of the Mind argument; this would also seem to be the core claim behind the luck objection); (ii) by arguing that action is impossible since indeterministic causation is impossible (second strand of the Mind argument); (iii) by arguing that free action is incompatible with indeterministic causation since no one would have a choice about any of their actions (third strand of mind argument). The rollback argument has always seemed to me to be an instantiation of the first strand that argues that indeterminism introduces a type of chance that is incompatible with freedom.

I’d be interested to think what you think about this way of understanding the Mind argument.

[Chris]

I'd not only be interested to think what you think, but also to *hear* what you think! Opps.

One more argument that might be worth considering in this connection is Pereboom's (2001, Chapter 3) argument against event-causal libertarianism. He says that it is a verstion of the Humean of Mind argument adapted to his causal history incompatibilism.

[Dan]

EJ Coffman has a very nice paper on the Luck objection to libertarianism. In particular, he thinks the Mind argument and the luck objection are two different strands of reasoning (though they frequently get run together). Absolutely worth checking out. His new website at Tennessee doesn't yet have links to his papers. I'm sure he'd send a copy along to anyone who asks (he's that kind of a guy). Also, he's got another paper (co-authored with Donald Smith) specifically on the failure of the Mind argument.

[Chris]

Dan,

I am familar with Coffman's paper, at the least the version he presented at the Pacific APA. But, in a nutshell, it seems like the differences he lists are merely historical differences. That is, differences in ways the arguments have been presented rather than differences in the ways the arguments must be presented.

Another way to think about this is as follows. I agree with Fischer that there are many different ways to formulate the consequence argument, some that employ a transfer principle and some that do not. All these versions are attempt to crystallize the same core intuition: determinism seems to render us powerless.

I am wondering if there is something similar going on with the Mind/luck arguments. Perhaps there is a core intuition that both the Mind and luck argument attempt to crystallize. All the arguments attempt to show that indeterminism somehow inhibits or at the very least does not supplment agential control.

[Joe Campbell]

Chris,

I think now that you might be right. But let me think some more.