Why is the Consequence Argument a Big Deal?

Over the past six months or so, a number of well-meaning, well-read, and intelligent people have asked me the question that is this blog post's title, or made a related remark in conversation. Normally, I'd dismiss this as happenstance but given the frequency with which it has been coming up, I'm starting to wonder if there isn't something else going on. Perhaps we are starting to see one of those gradual and inevitable changes that strike philosophical communities, a kind of literature event horizon where the problems and achievements of roughly 30-40 years or so ago begin to look puzzling and then eventually become unappreciated or invisible to those entering the field now. Below the fold are some musings about this issue and a response (okay, mine) to at least one person I spoke to about the significance of the Consequence Argument.

Something like the literature event-horizon effect was well in effect with much of Quine's work when I was entering graduate school. Our professors tended to assume everyone had read Quine's most famous papers and perhaps some of his lesser ones, that we knew about their importance, and that we appreciated why each was an achievement in its context, and so on. In contrast, many of us then-graduate students had read very little of Quine or the debates of the 1950s. We tended to have absorbed many of the basic lessons and the surviving Quinean legacy as part of the general philosophical background radiation with almost no explicit instruction. When we did read Quine, it seemed so completely familiar as to be almost unremarkable or so obviously wrong in light of the subsequent literature so as to not require much attention. Now I'm exaggerating a bit, and of course there were students who did read a lot of Quine or had particularly interests that gave them a deeper appreciation for Quine, as well as professors who thought it was important to have us read lots of Quine. Quine isn't really the point but just an example of how figures and positions tend to go from being widely shared and known to gradually less familiar over generations. So, if you will, just substitute whatever figure or issue you like that makes you nod your head in knowing agreement about the literature event-horizon effect.

There are, of course, lots of interesting arm-chair sociological explanations to invoke here, having to do with what problems and figures are important in different graduate schools and when, and there is the basic issue of how canons shift around in graduate education, etc. And, of course, there are texts that stave off the event horizon for much longer than their other contemporaries, and even texts that get swallowed for a while only to return later on. How and why all this happens has something to do with arguments, of course, but presumably also with ground level sociological processes.

At any rate, I find it pretty alarming that we may be entering the beginnings of the event horizon for the Consequence Argument, in part because I do think it is a Big Deal and in part because it has structured so much of the literature that informs my own thinking about these issues. (I should also confess that I got interested in the free will problem because of a seminar I took from van Inwagen, so maybe there is some autobiographical element to my estimation of things— but the size of the literature on it is surely evidence that my reaction is not just driven by autobiography.)

At any rate, here's a lightly edited version of the exchange I did have with someone about these issues. I'll start with the question that prompted it, because it does a good job of highlighting the elements I've most consistently been hearing:

The GFP (and a good deal of introductory literature on free will) frequently mentions how important the Consequence Argument was in getting people to become libertarians. I genuinely cannot understand why this might be. While I realize there is more to it than this, the argument really strikes me as formalizing the problem of free will and determinism as it has existed since the time of the ancient atomists. I really don't see what a convinced compatibilist would have found troubling after reading the Consequence Argument that s/he had not already considered. What am I missing?

(name withheld on the assumption that the author doesn't want credit; if I'm wrong just email me I'll be happy to give credit where it is due).

My own reaction:

Context matters— at the time Peter van Inwagen wrote it, almost everyone who wrote on free will was either a compatibilist or viewed as an anti-scientific libertarian nut (I'm way too quick here, of course, but bear with me). So, even if people realized there were deep problems with the conditional analysis, only a handful of (right minded, one might think) outliers to the literature had any sense of why anyone sensible would be an incompatibilist. (Again, not strictly true. There was that great Wiggins paper, important work by Ginet, for example, and various other things I’m neglecting to mention. But uptake depends on a lot of things, not the least of which can involve working out the details.). PVI didn't change the widespread view that libertarians were generally nutty (that was a group effort that seems to have really coalesced in the 1990s with the usual suspects deserving citation here), but he did get folks to see why one might sensibly be moved by quite ordinary convictions to think that incompatibilism must be true. I think it is fair to say that compatibilist up until then had thought incompatibilism was rooted in common sense in some superficial way, but that it was a perhaps a bad piece of theorizing about our natural convictions. Van Inwagen made it much harder to say that it was just bad theorizing from superficial commonsense convictions. Of course, compatibilists still thought it was ultimately bad theorizing, but now they had to point to some or another explicit place where the bad theorizing was happening, and van Inwagen's argument didn't leave a lot of places to launch the criticism that there was bad theorizing going on. So, the formalization you note wasn't just gratuitous— in the philosophical context it did real work to change how people thought about the dialectical situation.

I think part of its ongoing power is that even the most celebrated reply to the argument —I’m thinking of David Lewis' — is itself complex and relies on convictions that are, to my mind, anyway, much less obvious than the ones that PVI's argument seems to trade on. A related but important point is that David Lewis’ argument trades on a story about counterfactuals that was independently motivated— good news for Lewis and his supporters, because the argument is thus not ad hoc, but bad news for anyone not enamored with the Lewis approach to counterfactuals or unwilling to accept its machinery. Here too, though, Lewis' argument can look like a sophisticated way of making what many people think is a simple point: if you are an antecedently convinced compatibilist, the Consequence Argument shouldn't move you. But Lewis' accomplishment wasn't just that, or even that. His accomplishment was in showing exactly how to respond to PVI's argument in its own terms. My own diagnosis of the sociology is that whatever the plausibility of Lewis’ reply, it is hard to shake the sense that the Consequence Argument is latching on to some deep and relatively straightforward convictions, convictions that seem comparatively easy to access, in contrast to Lewis’ reply. Lewis might be right (ask Kadri or PVI) but it is also clear that whatever the right answer is here, the Consequence Argument and replies to it have structured the stakes for a lot of us.

So, in short, I think the answer is partly sociological and partly dialectical: after the Consequence Argument, it got a lot harder to say that incompatibilists were sloppy thinkers being seduced by a nonsensical metaphysics with no deep connection to common sense.

So, what would you have said in reply to the question at the top of this post?

Comments

[Justin Capes]

Manuel,

As a big fan of the C. Argument myself, I think your response is right on. But here's perhaps another reason why people think the argument isn't (any longer) a big deal: the rise of semi-compatibilism.

Prior to Frankfurt's paper and Fischer's subsequent development/defense of semi-compatibilism, all--or rather most--parties in the free will/moral responsibility debate took it for granted that the sort of freedom at issue in the consequence argument was the same sort required for responsibility. However, given the rise of semi-compatibilism and other Frankfurt inspired positions (e.g. certain forms of source incompatibilism)perhaps people no longer see the consequence argument as playing that big of a role in the debate. In some ways, it seems to me that the debate over Frankfurt cases and semi-compatibilism is prior to the issue of the C. Argument. For if what we are really concerned about in the free will debate is moral responsibility, then we first need to figure out whether the control required for responsibility is to be identified with the sort of freedom at issue in the C. Argument. If it is not, then that argument looses much of its importance. Of course, there are other reasons why we might be concerned with free will other than moral responsibility, but certainly responsibility has provided the main impetus for wanting to get clear on the compatibility issue.

[Blake]

It seems to me that the consequence argument also remains important because it isn't at all clear that Frankfurt counterexamples are successful. I personally don't find them to be persuasive. Nor do I believe that semicompatibilism is persuasive. Further, the consequence argument seems to work even if one adopts source incompatibilism and there are decent arguments about that suggest that source incompatibilism entails leeway incompatibilism as well.

Manuel's comments seem right on to me. However, if libertarianism was viewed as nutty, that didn't affect in any way the results of the consequence argument since it was directed primarily at compatibilists. As a libertarian, I am still waiting for an argument that doesn't beg the question in favor of a particularly narrow naturalism as opposed to a naturalism that truly seeks to account for human experience rather than discount it (with a nod toward the usual suspects).

[Cihan]

Manuel, I share your anonymous source's befuddlement over the importance of the Consequence Argument. I, too, had thought (perhaps still think) that the CA wasn't anything more than the common (incompatibilist) intuition explicated.

Van Inwagen made it much harder to say that it was just bad theorizing from superficial commonsense convictions. Of course, compatibilists still thought it was ultimately bad theorizing, but now they had to point to some or another explicit place where the bad theorizing was happening, and van Inwagen's argument didn't leave a lot of places to launch the criticism that there was bad theorizing going on.

Even so, this just makes me wonder what pre-CA compatibilist literature was busying itself with. Weren't they arguing for the compatibility of determinism and FW/MR and if so, addressing CA someway? If not, how did they so without addressing CA, one wonders.

I always thought that CA didn't have much philosophical ingenuity but it brought everyone on the same page on how various strands of compatibilism argued for the compatibility of determinism and FW. Perhaps, the real value of CA is just clarifying the issues...

[Michael Drake]

I'm with the no-big-deal contingent. The CA is a nice statement of the argument, but it doesn't seem to add any content to the more classical lines of argument.

[Kip Werking]

I generally agree with Manuel's observations, but not (seemingly) with the importance he places on them. As best as I can tell (and he and others are in a much better place to tell), the CA did exactly what Manuel says: made incompatibilism respectable in a time when it was unpopular. But it still is just a formalization of ancient ideas and arguments. Perhaps it's greatest importance is on making us focus on beta-like principles, and argue for their truth or falsity---something I don't think happens nearly enough (perhaps because it is so hard, and it's not clear what would even count as arguments for or against them).

I think Nichols & Knobe's recent research is a good example of a profound and important result, and so I'm glad to see they won a spot on the philosopher's annual. But perhaps this just reflects my general liking of Nichols' work.

[John Fischer]

Nice post, Manuel; I see you are using your fellowship time wisely!

I agree with what you say. Also, it has puzzled me that some are not moved at all by the Consequence Argument (or what lies behind it). It seems to me that the basic idea is best put by Carl Ginet: my freedom is the power to add to the given past, holding fixed the laws of nature. That is an incredibly simple, natural, and powerful idea. I have been puzzled by those who just don't feel the pull of this idea (or associated picture).

It seems to me that when I act when I act, my act extends the given past; I don't have it in my power to act in such a way, for example, that John F. Kennedy would not have been assassinated, or Bill Clinton never would have existed, or whatever.

Those who dismiss the basic ideas behind the Consequence Argument must reject this simple, basic picture. And, as you know, this picture (and the idea of the fixity of the past) has been around for millenia; the basic argument for the incompatibility of God's foreknowledge and human freedom to do otherwise relies precisely on this picture.

[Cihan]

Those who dismiss the basic ideas behind the Consequence Argument must reject this simple, basic picture. And, as you know, this picture (and the idea of the fixity of the past) has been around for millenia; the basic argument for the incompatibility of God's foreknowledge and human freedom to do otherwise relies precisely on this picture.

True that but I must also ask your opinion on this: Does the CA, as formulated by PvI, contain any philosophical innovation or whether it is just a strikingly clear formulation of (ancient) wisdom?

I am with Kip and Michael Drake in that its value rests in its expressiveness but then I don't know very much.

[John Fischer]

Cihan,

I think that's correct, that is, the CA as formulated by PvI is valuable insofar as it captures in a clear and challenging way basic ideas that have been around for awhile. Also important are formulations by Wiggins and Ginet.

I think this is parallel to what Nelson Pike did in his "Divine Omniscience and Voluntary Action" (Phil. Review) for the ancient argument for the incompatibility of God's foreknowledge and human freedom.

[Andrew W.]

It has always seemed to me that the value of the CA is that it nicely sets out basic libertarian intuitions. But it follows from this that the best version of the CA is the first one, not the third (modal) version. The first one trades on familiar intuitions in a way that the third does not.

Having so much literature devoted to the 3rd argument's Beta principle always seemed like a colossal waste of time, in my view.

[Joe Campbell]

Great post, Manuel! I have nothing to add to your post but three things to say about the subsequent comments.

1/ Respectfully, I think Justin has the cart before the horse: the rise of semi-compatibilism is due in large part to compatibilists like Fischer (see above) thinking that the consequence argument is good, or at least much better and more intuitive than the replies to it. Without the consequence argument, there would be no semi-compatibilism.

Here is one way to see this point. (I think that Kadri has made this point, too.) If you thought that there was a response to the consequence argument, you would never have been convinced by Frankfurt cases in the first place. In other words, if you thought, for instance, that persons can do otherwise in cases of full-blown causal determinism, then of course you'd think persons can do otherwise in Frankfurt cases.

But note that another important part in the development of semi-compatibilism was the (presumed) failure of compatibilist analyses of "can."

2/ As many note above, the consequence argument has been around for a very long time. Here is an Epicurean argument directed toward the Stoics: "... if everything happens by fate, everything occurs by an antecedent cause and if impulse [is caused], then also what follows from impulse [is caused]; therefore, assent too. But if the cause of impulse is not in us then impulse itself is not in our power; and if this is so, not even what is produced by impulse is in our power. From which it follows that neither praise nor blame nor honours nor punishments are fair." (B 76.40)

Van Inwagen never thought of himself as the creator of the argument (he notes Ginet's 1966 version, for instance) but there is no doubt that his version has a kind of elegance and simplicity that other previous versions lacked.

3/ With that in mind, I want to make a case for the third argument over the first, against Andrew. The big difference is that to date there is no generally accepted counterexample to beta. But there is a clear counterexample to the beta-like principle of the first argument, given by Perry in "Compatibilist Options."

Here is the principle: If S can render R false, then if Q entails R, then S can render Q false.

Here is Perry's counterexample. Let R be the proposition that J does not raise his hand at t. Let t be some day after 1950. Let Q be a conjunction: R & J's mother ate a carrot in 1944. Suppose that J's mother did not eat a carrot in 1944. S (for the sake of argument) can render R false, Q entails R, but S can't render Q false -- since it is already false when S acts!

What beta has over the above inference rule is the added presumption that the relevant propositions are true. The first argument is unsound and the third argument is better.

[Eddy Nahmias]

Interesting post, Manuel. As a compatibilist who has been struggling for years to figure out why the Consequence argument is irrelevant, I have to admit it is a big deal. I think people here have captured why it's a big deal. I'd just emphasize that I think the main contribution of its formulization of incompatibilism is the introduction of the Beta principle, precursor of the Transfer of non-responsibility principles (e.g., If you're not responsible for X and X is logically sufficient for Y, then you are not responsible for Y).

It is these principles which most clearly separate compatibilists and incompatibilists. It's also clear that such a principle seems to be required for most incompatibilist arguments to go through--it certainly is implicit in G. Strawson's Basic argument.

Rather than directly challenge the principle, I like trying to suggest that it proves too much; for instance, it can be reformulated to show that free will is inconsistent with indeterminism, in which case it is not useful in an argument for incompatibilism per se.

This seems to be the approach taken by Lynne Rudder Baker in a recent Analysis article, "The Irrelevance of the Consequence Argument," that I don't think we've discussed here (but is obviously relevant to the title of this post):
http://www.nottingham.ac.uk/journals/analysis/preprints/BAKER.pdf

[John Fischer]

Eddy,

One thing I've argued in various places is that the Consequence Argument can be given in different formulations. I contend that it does not depend in all its (valid) formulations on Beta or a Beta-like modal transfer principle. (Philosophers such as PvI and Fritz Warfield disagree with this.) If I am correct, then simply showing some problem with a beta-like principle would not be enough to exhibit a problem with the CA itself.

So, even if there is some infelicity with some or all transfer principles, what's wrong with this: my freedom is the freedom to add to the given past, holding fixed the laws of nature?

Suppose you have coffee at ten AM and it is now noon. You are deciding between the Mexican restaurant and making a boring old sandwich at home. Isn't it the case that, either way, you'll be acting in a way that is preceded by your having had coffee at ten? Don't we simply presuppose (typically without any fanfare) that our freedom at a given time is the freedom to extend the past, not to jump or "hop" to a different past?

[Tamler Sommers]

Eddy wrote (obviously in a pain-killer induced haze):

"Rather than directly challenge the principle, I like trying to suggest that it proves too much; for instance, it can be reformulated to show that free will is inconsistent with indeterminism, in which case it is not useful in an argument for incompatibilism per se."

Do you mean it is not useful in an argument for libertarianism? Transfer principles are enormously useful in arguments for hard incompatibilism (for FW/MR skepticism), right? Galen Strawson wouldn't say that transfer principles prove too much. They prove exactly enough, no?

[Manuel Vargas]

John, your example doesn't work. It is totally obvious that one should go for the Mexican food and the defects in rationality that would be required for one to have a boring old sandwich instead suggest that the agent wouldn't be sufficiently rational to count as either free or responsible. But I get your point.

[Cihan]

Eddy,

How is "such a principle" implicit in Galen Strawson's Basic Argument? He explicitly says that the argument doesn't depend on determinism or indeterminism and sets out to achieve independently the impossibility of moral responsibility. (It's not exactly an "incompatibilist" argumen.)

[Tamler Sommers]

Cihan,

I think the transfer principle (of non-responsibility) is implicit in the premise "So, if you're going to morally responsible for what you do, you have to morally responsible (in certain respects) for the way you are."

That's only true if you believe that the only way to morally responsible for an action is to be (partially) morally responsible for the factors that cause the action.

[Justin Capes]

Joe,

I see your point, though I think it unlikely that this is true for all semi-compatibilists. Frankfurt's original paper, for example, was published several years before PvI's version of the CA and only a few years after Ginet's. Moreover, one might be skeptical of the CA but grant it for the sake of argument and then go on to argue for a semi-compatibilist view using Frankfurt-style considerations. Indeed, if I remember correctly, this seems to be Frankfurt's approach in the original paper. He seems to leave it up in the air whether determinism really does preclude the ability to do otherwise.

In any event, the rise of semi-compatibilism has, I think, served to undermine the centrality of the CA , especially as it is articulated by PvI, at least in this way: If, as I noted in my original response, the sort of freedom at issue in the CA is not to be identified with the sort of control required for responsibility, then a sound CA would be much less significant than it otherwise would be, since the semi-compatibilist could grant the conclusion but deny that it has any significance with respect to responsibility.

[Eddy Nahmias]

Tamler (if I am in a haze, it has only to do with the amount of kids I have not any pain killers I might have),
Incompatibilism is the thesis that determinism is incompatible with free will. If an argument relies on a principle that is equally effective at showing that *indeterminism* is incompatible with free will (perhaps a big if), then that argument does not seem to me to be an argument for *incompatibilism* per se. Rather, it is an argument for the impossibility of free will (at least if we define determinism and indeterminism as exhaustive options).

Of course, such an argument would also have the *side effect* of being an argument for incompatibilism. But to me it would be misleading to consider it an argument for incompatibilism. But maybe this is a rhetorical rather than a logical point.

Compare:
The President argument is an argument that shows Obama is not qualified to be president (it is an argument for the incompatibility of Obama and the presidency). The argument relies crucially on principle Delta. Now, someone shows that principle Delta equally shows that McCain is not qualified to be the president (and the argument works in roughly the same way, using principle Delta in the same way).
On my reading, the President argument is not really an argument for the incompatibility of Obama per se and the presidency. Rather, it is an argument for the incompatibility of the presidency and any currently viable candidate. (Have I somehow made Bob Barr, the agent causal Libertarian?).

[John Fischer]

Eddy,

I find that pain-killers are a good treatment for children. By that I mean that they are good for the parents, of course!

[Tamler Sommers]

Eddy (in an advanced stage of his hallucinatory drug experience) wrote:

"Compare:
The President argument is an argument that shows Obama is not qualified to be president (it is an argument for the incompatibility of Obama and the presidency). The argument relies crucially on principle Delta. Now, someone shows that principle Delta equally shows that McCain is not qualified to be the president (and the argument works in roughly the same way, using principle Delta in the same way).
On my reading, the President argument is not really an argument for the incompatibility of Obama per se and the presidency."

I disagree. It seems to me that the question of whether Obama is qualified to run for president is mostly irrelevant to the question of whether McCain is so qualified. I think it depends on the context of your argument. If you're trying to decide who to vote for based on that issue, then, yes, you're right, it wouldn't be a good argument to make. But maybe you're trying to figure out whether America is in good shape for the next four years. Maybe you're considering a move to Peru or the Netherlands. The first step would be to show that Obama isn't qualified to be president. The next step would be to show that McCain is also not qualified. Since those choices are exhaustive (practically speaking), we're screwed. On to Amsterdam or Cusco.

Similarly, one perfectly respectable way to argue for free will skepticism is first to argue that free will is incompatible with determinism (the incompatibility thesis per se). And then to argue that free will is incompatible with indeterminism. Since those options are exhaustive...etc. Doesn't van Inwagen do this (before saying that nevertheless we have free will)?

In any case, the first argument would still be an argument for incompatibilism per se, right?

[Tamler Sommers]

Now that I think aboout it more, Eddy--like Timothy Leary before him--may be on to something.

If the libertarians of that era were perceived of as nutty, it wasn't due to their arguments for incompatibilism. It was due to their positive accounts of libertarian free will. Richard Taylor, for instance, gives a decidedly non-nutty (in fact, quite compelling) argument for incompatibilism before going on to give his positive theory of free will, which, if I'm remembering right, even he thought is a little nutty.

As a libertarian Van Inwagen saved himself from the charge of nuttiness by refusing to give an argument for libertarian free will (unless you count his remark about how a responsibility skeptic he knew once said that stealing books is shoddy). In that sense, unless the consequence argument was an enormous advance over previous arguments for incompatibilism (rather than, as some have suggested, a rigorous reformulation of them), it isn't all that big a deal for the libertarian. Skeptics can be grateful for a rigorous argument that exposed some of the wretched subterfuges out there, but libertarians if anything might find the CA to be a thorn in their side if Eddy is right that it made their task more difficult as well.

[Paul Torek]

John,

Please provide a citation to your arguments that the CA "does not depend in all its (valid) formulations on Beta or a Beta-like modal transfer principle." Online availability is a bonus if I can get it.

[Eddy Nahmias]

Here's the idea again; see if it sounds hazed and confused.

IF it turns out that what makes the Consequence argument (CA) work is some sort of Transfer of non-responsibility principle (along with premises about no power to change the past or laws) AND IF it turns out that that principle and those premises work equally well and in the same sort of way on the assumption of *indeterminism* (call this argument CA*), THEN it looks like CA is really an argument for the impossibility of FW and MR and not for the incompatibility of FW/MR and determinism, and conversely, it looks like responses to CA will be responses to CA* and thus defenses of the existence of FW/MR rather than arguments for compatibilism.

Basically, CA morphs into the Basic argument and as Strawson himself says, determinism is not the relevant issue in the Basic argument.

I take all this to highlight three points:
1. determinism is not really the issue.
2. transfer principles are central to the debate (I want to hear more from John about why he thinks they aren't).
3. a lot depends on our conception of FW/MR.

If anyone cares about the analogy I tried, here's another attempt:
If someone offers a principle X as the crux of an argument against Obama (specifically) for president, there should be something about *Obama* that makes the argument work. If a parallel argument can use the principle equally well against McCain for president (or perhaps anyone for president), then it seems Obama is not really relevant; he can drop out of the discussion of this argument and we can argue about whether the argument succeeds at showing that no one (or none of the viable candidates) should be president.

[Joe Campbell]

John,

You say "my freedom is the freedom to add to the given past, holding fixed the laws of nature." But this, and the fact that the past and laws do not change, can't get you incompatibilism. Don't you need some additional modal claim about the inability to change the past and the laws? And aren't we then just talking about transferring that lack of ability onto everything else (though we might not explicitly be putting it this way)?

Justin,

I didn't mean to suggest that my comments undermined your previous claim that "the debate over Frankfurt cases and semi-compatibilism is prior to the issue of the C. Argument" or your new claim that "the rise of semi-compatibilism has, I think, served to undermine the centrality of the CA."

[John Fischer]

Joe,

Well, "given past" is supposed to rigidly pick out the past in the actual world (or, alternatively, the world in which the individual whose freedom is being assessed is located), and thus I don't really think that one needs more ingredients of the sort you posit. I develop a version of this argument in The Metaphysics of Free Will (Blackwell: 1994); it is in the section on the "Basic Version" of the Argument for Incompatibilism. Carl Ginet also develops such an argument in his On Action.

Cihan (and Joe),

I am traveling and away from my materials, but here are some references. Thanks for asking.

The primary source would be The Metaphysics of Free Will; I don't have the page numbers with me, but toward the end of the book (perhaps in a long footnote) I emphasize that the CA can be given formulations, some of which do not employ or depend implicitly on beta-like modal transfer principles.

Additionally, I develop this point in "A New Compatibilism", reprinted in Laura Ekstroms' Westview anthology on free will; and, with Mark Ravizza, "Free Will and the Modal Principle," Philosophical Studies (1996); and "Critical Notice of J. Howard Sobel's 'Puzzles of the Will'", Canadian Journal of Philosophy (Sept 2001).

[Cihan Baran]

Tamler,

Thanks for the clarification. While we are on the subject of principles like the transfer of non-responsibility, let me ask something about your meta-skepticism. (I hope people don't mind the digression.)

If I understand you correctly, you claim that various sides in the free will debate appeal to intuitions in defending such principles and that intuitions vary across cultures - and are thus unreliable.

On the flip-side, how are you supposed to defend such principles without appealing to intuitions? Every discourse has to start at some point - and what better points to pick then the intuitively plausible ones?

Perhaps, this argument about the unreliability of intuitions really boils down to the ancient infinite regress argument for the impossibility of knowledge in epistemology: In order to know p1, you need to know p2. In order to know p2, you need to know p3... But this forms an infinite regress, so you can't really know anything.

The reason is that, if someone could justify TNR with some other (intuitive) principle, we would still say that the new principle also makes an appeal to intuitions and a similar regress would follow.

But then the notion of appealing to intuitions may be even more fundamental than the infinite regress argument for the impossibility of knowledge because the IRA itself appeals to intuitions (-and Irish nationalism, sometimes).

What I am trying to decide is that whether appealing to intuitions is something specific to the free will debate or whether it is some independent problem about knowledge.

[Joe Campbell]

Thanks, John!

Since I'm not traveling and have The Metaphysics of Free Will handy, I can tell you that the relevant passage seems to be on p. 204, where John writes: "The Transfer Version and the Conditional Version can be seen as first approximations to the Basic Version. I contend that these versions are not equivalent; certain objections to the first two versions are not also objections to the Basic Version. Also, it emerges that certain ingredients (such as the Transfer Principle and the Conditional Version of the Fixity of the Past) are not necessary in order to generate the skeptical conclusion." The argument is continued in fn. 12, pp. 248-9.

I want to make it clear that I am not claiming that all arguments for incompatibilism use a transfer principle, though I have made that claim in the past. Van Inwagen's Second Argument, for instance, does not use a Transfer Principle. It is clear to me now that any English language argument may have a variety of logical forms. All English language arguments have the form P, therefore Q, for instance, since the premises of any argument may be conjoined into a single conclusion.

What is less clear is whether, as van Inwagen says, arguments for incompatibilism (or at least the good ones) "stand or fall together" (1983, p. 57). I guess I would say that they do. (See p. 248 of John's book for his views on the subject.)

For instance, if one could change the past, then both a premise in the Third Argument and a premise in The Basic Argument would be false. It is also clear that if one had a counterexample to the central claim of the Basic Argument -- "an agent can do X only if his doing X can be an extension of the actual past, holding the laws fixed" (p. 88) -- then one would also have a counterexample to Beta. Not that this is a proof that the arguments stand or fall together, of course!

[John Fischer]

Joe,

Many thanks for the scholarly references!

One disagreement between PvI and me is that PvI does indeed believe that all superficial formulations depend at a deep level on a transfer principle (a beta-like principle). That is WHY PvI thinks all such arguments "stand or fall together".

I do agree that all versions employ some sort of fixity of the past principle. Thus, no matter how it is formulated, one could simply reject ANY such (plausible) principle, and thus via this route all the versions would stand or fall together. On that you and I would agree. It is however a very different point than the point about beta-like principles. So, for instance, although I haven't yet read Rudder-Baker's article, it would NOT follow from any problem with any beta-like principle, or any analogue of such a principle, that the CA is "irrelevant". (Again, I haven't read that article, so I'm not commenting specifically on it.)

[John Fischer]

Joe,

Many thanks for the scholarly references!

One disagreement between PvI and me is that PvI does indeed believe that all superficial formulations depend at a deep level on a transfer principle (a beta-like principle). That is WHY PvI thinks all such arguments "stand or fall together".

I do agree that all versions employ some sort of fixity of the past principle. Thus, no matter how it is formulated, one could simply reject ANY such (plausible) principle, and thus via this route all the versions would stand or fall together. On that you and I would agree. It is however a very different point than the point about beta-like principles. So, for instance, although I haven't yet read Rudder-Baker's article, it would NOT follow from any problem with any beta-like principle, or any analogue of such a principle, that the CA is "irrelevant". (Again, I haven't read that article, so I'm not commenting specifically on it.)

[Bob Doyle]

As the last one among many discussants that prompted Manuel to his post, I would like to tell you my personal concerns.

I asked Manuel why he thought van Inwagen was getting so much credit for something that had been one of horns of the dilemma of determinism since ancient times, the two-part standard argument against free will.

But I also noted in my web page on van Inwagen that he did deserve credit for the current focus on compatibilism versus incompatiblism, a focus that to my mind greatly confuses the free will debates.

http://www.informationphilosopher.com/solutions/philosophers/vaninwagen/

As Manuel said:
"PVI didn't change the widespread view that libertarians were generally nutty (that was a group effort that seems to have really coalesced in the 1990s with the usual suspects deserving citation here), but he did get folks to see why one might sensibly be moved by quite ordinary convictions to think that incompatibilism must be true."

The trouble as I see it is that the concept of incompatibilism is complex, not fundamental and simple. As you well know, it contains two opposing concepts, libertarian free will and hard determinism. This is not the kind of clear conceptualization that I expect from hard-working analytic philosophers.

Incompatibilism confused me when I began to read the recent literature, and I expect it confuses many students of philosophy. It has led to a proliferation of variations that further confuse, and it has facilitated changing the subject, following Peter Strawson, from free will itself to moral responsibility.

I will post my analysis of van Inwagen's Consequence Argument and its historical significance.

[Justin Capes]

Bob,

I'm puzzled by the fact that you're puzzled about incompatibilism. I'm not sure what you mean by saying that incompatibilism is not fundamental or simple, but incompatibilism is simply the thesis that determinism is incompatible with free will (and/or moral responsibility): if determinism is true, then no one is free (in the sense required for moral responsibility). As such, incompatibilism says nothing about libertarianism or hard determinism. Libertarianism is the conjunction of incompatibilism and the free will thesis (the thesis that we are sometimes free in the sense required for morally responsibility), while hard determinism is the conjunction of incompatibilism and determinism.

I have to agree with PvI that if anything has brought confusion into the debate, it is these latter terms.

[John Fischer]

As I see it, "incompatibilism" is no more complex than "compatibilism". Guess what: incompatibilism is the doctrine that causal determinism is incompatible with [fill in the blank--maybe freedom]. And compatibilism is the doctrine that causal determinism is compatible with [fill in the blank]. So whatever complexity there is is entirely symmetric.

Now there are libertarians and hard determinists on the incompatibilism side. But of course there is a parallel division among compatibilists: those who believe that indeterminism is incompatible with [fill in the blank] and those who believe indeterminism is (also) compatible with [fill in the blank].

[Alan]

I'm a latecomer to this discussion but I must say I've been impressed both with Manuel's eloquent way of putting this before us as well as the strength of the points of view laid down here.

Here's my two cents, partly in agreement and partly in diagreement with John's last remark. If we can rigorously make an equivlence of some sort between FW and reponsibility, or make responsibility the primary concern of the debate, then of course that's what we want to fill the blanks in with. But I'd argue that because incompatibilists are constained by tying the relevance of determinism to FW/responsibility, their concepts of FW/responsibility have that solid core and any other differences among them form a fog (sorry Manuel! :-) or penumbra of disagreement surrounding that. But compatibilists have many more options, because while they can foggishly agree as a group that compatibilist (or semi-compatibilist!) FW/responsibility does not have anything to do with the incompatibilists' solid core, it is correspondingly open to further elaboration and interpretation just what a given theory of compatibilism means. To me it's a case of being either A or not-A--incompatibilists generally agree on the A, and while compatibilists generally agree on the not-A, they don't have to agree very much at all on what exactly constitutes their version of not-A.

[Roman]

Manuel - I will save you the trouble and out myself as the puzzled writer of the original question. I am glad to hear from you and to see from the comments that I am not alone in my puzzlement. And I thank you for your reply, both in your original e-mail and here.

That said, I wonder whether you should be so concerned that the CA has reached its event horizon. Its arrival at the event horizon (if it has so arrived) might not be such a bad thing. Here's why: first off, as everyone seems to agree, the argument itself is old. Very old. While it is true that overly eager naturalists might simply ignore the incompatibilist intuitions--and I am certainly glad that the CA made that much more difficult, and so gave rise to a higher level of debate about the problem--some version of the argument will most likely always be around. The reason is simple: we have incompatibilist intuitions. Whether they are mere intuitions, as some claim, or necessary conclusions of reason, as Kant thought, they've certainly been around long enough that one might doubt whether giving them the status of "confused folk thinking" isn't just an incredibly lazy way of dismissing an argument one doesn't know how to respond to. Ayer, in thinking that the confusion lies in conflating determinism with compulsion, strikes me as guilty of something of the kind. (And in any case, whatever the source of these intuitions, if they are intimately bound up with our reactive attitudes and our practices of praise and blame, and more significantly of reward and punishment, ignoring them can't be the right way to go.) None of this, of course, is to suggest that we don't also have compatibilist intuitions.

So, aside from the "the argument isn't going away even if the CA version of it fades" issue, why shouldn't you be concerned? Well, because so much of the importance of the argument is context-dependent. You mention the historical context, and then go on to talk about all the other great features of the CA. But I'm not entirely sure that these other issues are separate from the historical context, or that I'm wrong in my initial thought that the main contribution of CA was to formalize an ancient argument. These are connected points: Formalization is not essential to a convincing argument. That philosophers came to see formal arguments as superior to non-formal ones is itself a historical feature. And my sense is that it is slowly changing. A good deal of writing on free will, action, moral responsibility, and other cognate topics today avoids formality. And if I am right about that, it may well explain why the CA has reached its event horizon--if it has. Kant's version of the argument in the Third Antinomy was immensely powerful at that time; back then, the CA would probably not have made a dent. And so if we are moving outside a period where only formal arguments are taken seriously, it is natural that the CA should begin to seem more puzzling and less of a big deal. But why should this be a source of concern to someone not specifically attached to formalism?

[Alan]

If I may insert one more remark in response to Roman, who generously revealed himself, I think the source of much of what is called incompatibilist intuition is straightforward. If I define actualist possibility as that possibility which the world allows (I know--this is a non-standard use of "actualist", but if I used "metaphysical possibility" that would be taken too broadly too), then if someone takes what is conceivable as logically possible as also actually possible, then that would naturally lead to an incompatiblist sense of the open future. Of course there is an additional factor of contextualized blending of these possibilities as well. Example: I can conceive prior to my doing so of my leaving through a door into a hall in any number of logically possible ways. I can exit and go left or exit and go right, or I can go straight through an opposing wall, or I can turn invisible as I walk through as well, etc. All logically possible. But my recalled association with the real world drops my interests to what I've done before--I've often gone left and right upon exiting a door into a hall. So my real-world experience drops my interest into the logical possibilities of going left or right. But it is still the case that my conceiving of the logical possibilities of going left or right does not entail that in the next few moments I have an actual ability and opportunity to do both. In short, the human mind naturally conflates what is, shall we say, "locally" logically possible with what the world might actually allow.

[John Fischer]

Roman,

Thanks for the thoughtful post. I agree with the thrust of it--good writing in this area of philosophy, and in most areas, avoids technicality as much as possible.

Perhaps this is just a small point, but I tend to use the term "Consequence Argument" to refer not just to PvI's versions--or versions, since he gives three in the book--but to any of a family of arguments that employ crucially a fixity of the past premise. So, for me, "CA" is meant to pick out any of these arguments, including the "ancient" ones.

I suppose we agree that what is important is not any particular formalization, particularly highly technical ones, but the argument itself and its core ideas.

[John Fischer]

Some further (quick) ruminations.

Consider "Cartesian skepticism" about the external world ("CS"). Now CS has been around a long time, since before Decaretes even; it is, as with the CA, an ancient position, based on certain core intuitions. Now one can seek to formalize the intuitions and argument for skepticism; one can indeed employ a principle, the Principle of Closure Under Known Implication, which is structurally the same as the modal Transfer of Powerlessness Principle (PvI's Beta). This gets at some powerful intuitive ideas, but may then issue in problems. One could reject the Closure Principle but still hold onto the basic intuitions behind skepticism, I think. In any case, one might use the term "Cartesian Skepticism" to refer to any of a FAMILY of arguments or even core intuitions. And even if there is an infelicity with one principle (or others), it would NOT follow that the position, CS, is itself untenable. Similarly with the CA and free will skepticism.

I develop some of the analogies in The Metaphysics of Free Will (Blackwell 1994); I have benefited from the writings of Fred Dretske and Tony Brueckner on these and related issues.

[Roman]

Right. I take it that the argument itself--as opposed to PvI's particular versions of it--is not going away. Well, unless everyone in the world is converted to semi-compatibilism, maybe. And I am not averse to using "CA" to refer to the core ideas rather than PvI's versions in particular. But with regard to the topic of Manuel's post, my puzzlement (and I think this goes for most of those who share this puzzlement) is about the role of PvI's CA in particular. That is: it would be odd for someone working on free will to be puzzled by why the core ideas of CA might be a big deal; but this is different from being puzzled over why PvI's version was a big deal. It was certainly an elegant argument, and stimulated a lot of superb discussions of free will. But what I find more surprising is the extent to which philosophers writing on free will before this version of CA was drafted seemed to really miss the point of the historical debate.

By the way, I didn't mean to come off as anti-formalist. Formal arguments can be great at focusing discussion on particularly tricky premises, and PvI's formulations did just that. But it strikes me that the core ideas are the big deal in general, whereas PvI's particular versions are a big deal mainly given the contingencies of the historical situation in philosophy.

[John Fischer]

Roman,

An additional quick point. I don't think the CA is going away, even if everyone in the philosophical world sits in a circle, holds hands, sings "Kumbaya" (ok, I can't spell "Descartes", much less the name of that song...), and proclaims the truth of Semicompatibilism. Remember, Semicompatibilism seeks to get the best of both incompatibilism and compatibilism; thus, the Semicompatibilist takes very seriously the CA, and can (though need not) accept it. Of course, the Semicompatibilist contends that the CA "goes away" when we seek to attribute moral responsibility.

[Roman]

Yes, that last sentence is what I meant. Although there are lots of other reasons for wanting free will--e.g., the problem of evil, life hopes, etc--I tend to think that the issue of MR is the major historical and intuitive reason for caring about the kind of choice that CA claims is incompatible with determinism.

[Bob Doyle]

Hi Justin,

As I see the Consequence Argument, it is just the first of the two horns in the classical two part dilemma of determinism, the standard argument against free will.

1) If determinism is true, there is no free will.

2) If indeterminism is true, our actions are random and we can't be responsible, so no free will.

J. J. C. Smart was perhaps first to claim these two exhausted the possibilities. (Mind 1961)

So a longer version of 1) is that determinism implies a causal chain going back before Aristotle imagined it going to the first cause. And that is the nub of the Consequence Argument, am I right?

Are these what Roman sees as the core ideas in PvI's arguments?

What I believe van Inwagen was doing was denying this kind of strict causal determinism. And indeed, with quantum physics we now know that there is no such strict causal determinism.

But what he did was attack a position held by "classical compatibilists" since Hobbes and Hume (the position actually goes back to the Stoic Chryssipus).

When he calls his attack "incompatibilism," he conflates his position - in favor of free will - with hard determinism.

That is what I find personally confusing. If he had just tried to deny determinism, though, perhaps his book would not have been so popular?

Had van Inwagen not noticed that most philosophers, except perhaps Ted Honderich, had changed the subject from free will to moral responsibility, since Peter Strawson in 1962 anyway, and become agnostic on the truth or falsity of determinism?

Van Inwagen also mentioned the other horn of the dilemma, renaming it the Mind Arguement. It did not catch on.

"The Mind argument proceeds by identifying indeterminism with chance and by arguing that an act that occurs by chance, if an event that occurs by chance can be called an act, cannot be under the control of its alleged agent and hence cannot have been performed freely. Proponents of [this argument] conclude, therefore, that free will is not only compatible with determinism but entails determinism." (p.16 of Essay on Free Will)

I personally like this approach, because van Inwagen might have convinced his fellow Libertarians that the willing act itself needs to be adequately determined for moral responsibility.

This point was made by R. E. Hobart in his Mind 1934 article - FREE WILL AS INVOLVING DETERMINATION AND INCONCEIVABLE WITHOUT IT.

Al Mele, starting from a suggestion of Daniel Dennett, has offered his "modest libertarianism," suggesting that indeterminism is involved only in the early stages of deliberation (what Aquinas called the Intellect) and not in the final choice or decision, which must be consistent with character and values and involve no chance.

My question, for those interested in a close analysis of the standard two-horn argument against free will as taught in all our philosophy classes, is this:

Could you feel responsible if your adequately determined will chose from among thoughts of alternative possibilities that included some thoughts that had been randomly generated by free associations in your mind?

What if those random thoughts had been considered and de-liberated by your adequately determined will?

Could a person with such a will feel responsible for subsequent actions also adequately determined?

Or would randomness anywhere (a causa sui) contaminate any actions further along the new causal chain?

"Adequate determinism" is the determinism we have in the world. It includes microscopic indeterminism - negligible for all macroscopic objects, but not for our thoughts, which contain noise.

[Bob Doyle]

John,

Thanks for your argument that the complexity is entirely symmetric between compatibilism and incompatibilism.

You said:
"Now there are libertarians and hard determinists on the incompatibilism side.

"But of course there is a parallel division among compatibilists: those who believe that indeterminism is incompatible with [fill in the blank] and those who believe indeterminism is (also) compatible with [fill in the blank]."

Although I see how you can logically imagine a "compatibilism" between indeterminism and free will, there is historically no such compatibilist position, is there?

Libertarians make indeterminism a requirement for free will. They don't call themselves "compatibilists" on indeterminism, do they?

They get into trouble when they want indeterminism in the decision and action, when all they need is indeterminism in the alternative possibilities.

Since for me "compatibilism" has historically been between determinism and free will, I see "incompatibilism" as including those two opposed views - libertarians and hard determinists, and thus more complex.

To call libertarians "compatibilists" on indeterminism, although a logical possibility as you have shown me, seems only to further complexify and confuse, not clarify a muddled situation.

It seems to me to exacerbate Kant's concerns about "word jugglery" and the "wretched subterfuge" he saw compatibilism to be, if we were to use the term compatibilist to describe libertarians.

Please excuse my naive approach to these things, but I am trying to clarify the history of the free will problem, and seek help from all you Gardeners to make it more useful.

http://www.informationphilosopher.com/freedom/history/

[Bob Doyle]

Manuel,

I like your bringing out the decades-long trends in philosophical thinking on free will.

You said:

"Context matters — at the time Peter van Inwagen wrote it, almost everyone who wrote on free will was either a compatibilist or viewed as an anti-scientific libertarian nut (I'm way too quick here, of course, but bear with me)."

At the time PvI wrote, very few professional philosophers would risk their reputation by writing seriously about free will. The exceptions were mostly American, followers of Roderick Chisholm, like Richard Taylor and Keith Lehrer, both advisers on van Inwagen's thesis.

Free will had been officially declared a "pseudo-problem" by the logical empiricists and analytic philosophers, along with other pseudo-problems like the mind-body problem and knowledge of the external world.

Free will was first called a pseudo-problem by Moritz Schlick in his 1936 essay The Pseudo-Problem of Freedom of the Will.

Ludwig Wittgenstein had convinced Schlick and his Vienna Circle of logical positivists/empiricists that philosophical problems could not be solved, only dis-solved, by careful attention to the use of language.

In C. A. Campbell's inaugural address at Glasgow University in 1938, In Defence of Free Will, he attempted to restore sensible discussion to a problem he regarded as unparalleled in the history of metaphysics.

Later he attacked Schlick directly in his 1951 Mind article - Is Free Will A Pseudo-Problem?

During those decades there were also several "scientific" libertarian "nuts". Beginning with Arthur Stanley Eddington's 1928 claim that "free" electrons might somehow make room for free will, several respected scientists tried to build amplifiers in the brain to get Heisenberg's quantum uncertainty involved in our decisions.

At the time PvI wrote, Daniel Dennett had dispensed with these scientific attempts to explain free will (in his 1978 Brainstorms) and Robert Kane (in his 1985 Free Will and Values) had started his so far unsuccessful attempt to find an "intelligible" event-causal account of libertarian free will using quantum uncertainty.

The crux of the problem was locating indeterminism in the brain. How could the quantum event happen at the right time and place to affect the decision?

As you know, the broadest context of this problem goes all the way back to Democritus and Epicurus, whose introduction of a chance swerve was essentially the same thing as Eddington's "free" electron.

The ancients asked the same question about the swerve, how could it help with freedom if it was random?

My take-away from your "context matters" is to encourage students to examine critically the complete historical context.

Very little new has been added to the free will debates, considering it has been nearly two-and-a-half millenia of philosophizing.

But there appears to me to be something wrong with the standard two-part argument against free will that should have been seen over 2000 years ago, except for what William James called our "antipathy to chance."

http://www.informationphilosopher.com/freedom/standard_argument.html

[Bob Doyle]

Hi John,

One more thought about your feeling that incompatibilism is no more complex than compatibilism - and that whatever complexity there is is entirely symmetric.

Here is a truth table analysis to show that incompatibilism is always more complex than compatibilism, since there are always two ways to be incompatible (FT and TF) and only one way to be compatible (TT).

...............Determinism.....Free Will

Compatibilism........T............T
Incompatibilism......F............T
Incompatibilism......T............F

...............Indeterminism.....Free Will

Compatibilism........T............T
Incompatibilism......F............T
Incompatibilism......T............F

Does that make sense?

Carl Ginet recently asked me to describe him as an incompatibilist, not a libertarian. We know he does not want to be thought a hard determinist. Isn't that a sign of some confusion, some loss of precision in our terminology?

[Saul Smilansky]

Manuel,

This sort of inquiry is really important for the health of the FW debate, particularly since reflecting on why we do what we do (work on what we work on, assume what we assume, etc) often shows that the reasons are not as good as we typically believe.

As others have argued above, the Consequence Argument is not, philosophically, a big deal (it's still a neat formulation, with some interesting philosophical advantages and disadvantages; and Peter's book is of course splendid). So it IS interesting that (as you described) this argument played such a big role in waking people up to the possibility that compatibilism isn't just obvious. This was probably a distinct American phenomena. The C argument played no such role in the UK, simply because the discussion there was very different. When I was studying in Oxford in the late 80's it was hard to find a compatibilist: earlier ones (Ayers, Glover) ceased working on the FW problem, while the people working on FW (Peter Strawson, Galen Strawson, David Wiggins, Martha Klein; Isaiah Berlin in the background though not active) were anything but simple compatibilists.

[John Fischer]

Bob,

I don't know how productive it is to repeat myself; I don't have anything to add to my previous post. As far as I see it, incompatibilism is no more complex than compatibilism. Within each camp there are various sub-camps. But I'm also not sure why this is important. Of course, we should distinguish between hard determinists and libertarians; and of course we should distinguish those compatibilists who also require determinism for freedom from those who don't.

[Eddy Nahmias]

Just to add to John's point, we should also remember that compatibilism, like incompatibilism, says nothing about whether any agents have free will, much less whether humans have free will. It just says that free will is possible in a deterministic universe. Now, every compatibilist I know of has assumed that arguing for compatibilism thereby shows that we actually have free will. But this is too quick (and concedes too much to the idea that determinism really should be the central issue in the debate). There may be other threats to free will, entirely distinct from determinism, such that securing the truth of compatibilism would not secure our free will (skeptical or "neurotic" compatibilism is a viable position).

I belabor this point I've made often here just because it makes Bob's truth table less clear (I don't really understand how to read it anyway) and highlights that the names in the debate are less illuminating than we might like for yet another reason.

[Bob Doyle]

Hi Eddy,

I like your focus on the point that compatibilism just says that free will is possible in a deterministic universe.

As an astrophysicist by training, I hope to show that the universe is not deterministic in the classic sense of strict causality, but that our will is adequately determined nevertheless, which is a requirement for moral responsibility.

If quantum mechanical uncertainty were not enough, we now know that the information in the universe today (like that in our blog posts) simply did not exist at earlier times. This is a key element in information philosophy.

A Laplace Demon could not know enough to predict our complex information-rich future.

So van Inwagen's Consequence Argument is relevant. His sense that Determinism is not the case is correct.

But the existence of real Indeterminism does not entail that everything is random.

The microscopic quantum indeterminism we have is negligible in the case of large objects. Our central nervous system is such a macroscopic object, capable of averaging out microscopic fluctuations, as biologists have argued for some time.

But this may not be the case for our thoughts and our memories of experiences.

In any information system, storage and recall are susceptible to noise, both thermal noise and quantum noise.

Such noise could lead to the random association of ideas that give rise to new alternative possibilities for potential actions.

Determinists have always been right that our will must be determined so that, as Hume said, "its actions disclose the character and abiding dispositions of the agent."

But this does not mean that alternative possibilities for action need to be determined.

Far from it, possibilities may be generated by free associations in the mind, based on the information there about our past experiences and our feelings about them.

I am frankly embarrassed by the bizarre suggestions made by many of my fellow scientists about the role on indeterminism in free will. They generally try to amplify quantum events to add a random element to our decisions. Robert Kane's work over the years has always tried to do this.

Daniel Dennett and Al Mele have properly called for indeterminism to be limited to the early stages in the decision process.

I am saying indeterminism is limited to generating possibilities for our will to de-liberate and choose adequately determined actions. Indeterminism does not affect the will and its determined decisions.

You may see that this is quite similar to the role of chance (caused by quantum mechanical events that alter the genetic code in our DNA) in the creation of new species by natural selection.

Compatibilists and Determinists have been right about the Will,
but wrong about Freedom.

Libertarians were right about Freedom,
but wrong about the Will, which is adequately determined.
_______________

I have just been reading Stuart Hampshire's Thought and Action. He found in Spinoza's Ethics a "freedom of mind" - despite Spinoza's reputation as a strict determinist.

See http://blog.i-phi.org/?p=14

Reflecting on Hampshire, I imagine this relation between our thoughts and actions...

Our THOUGHTS Are FREE
- They Are Within Us (Aristotle's ἐν ἡμῖν)
- But They Seem To come TO US.

We DETERMINE Our ACTIONS
- They come FROM US.
- And They Make Our Will Feel "Up To Us." (Aristotle's ἐφ ἡμῖν)

More details here:
http://www.informationphilosopher.com/freedom/cogito/

http://www.informationphilosopher.com/freedom/history/

[Kip Werking]

Saul,

It would be nice, if true, if people projected equally regarding both good and bad actions. But there is at least some evidence suggesting that "perceivers may be inclined err on the side of assuming immorality regardless of mitigating circumstances; inferences of morality, on the other hand, are more carefully qualified."

The following complete quote is from The Paranoid Optimist, which is a brilliant and underappreciated piece of evolutionary psychology (for those of you who like that sort of thing):

"Several sources of evidence support these ideas. A study by Reeder and Spores (1983) demonstrated that people make attributions about morality in an asymmetric fashion. In the study, perceivers inferred that immoral behavior (stealing from a charitable fund) was caused by immoral dispositions regardless of situational inducements (whether the target’s date encouraged the target to steal money or donate money). In contrast, in the moral behavior condition (in which the target donated money to the fund), perceivers’ inferences depended on situational cues; when the target was encouraged to donate money, perceivers inferred lower morality in the target than when the target was encouraged to steal (Reeder & Spores, 1983). These results suggest that perceivers may be inclined err on the side of assuming immorality regardless of mitigating circumstances; inferences of morality, on the other hand, are more carefully qualified."

[Joe Campbell]

Bob,

Maybe you got this point given Eddy's comment, but your truth table test is way off and if your argument that the complexity of the truth table were to show that incompatibilism was "more complex" than compatibilism, the right assignment of truth-values would show the opposite conclusion!

As for the first point, a compatibilist may accept or reject either to determinism or the free will thesis as long as he doesn't reject one because of the other. I myself reject determinism and remain unsure about the free will thesis. In the future, perhaps, I might fully reject both theses. I suppose that if I were born in the 19th Century, I might accept both theses. What I likely won't do -- what makes me a compatibilist -- is reject the free will thesis BECAUSE I accept determinism or reject determinism BECAUSE I accept the free will thesis. (Since "BECAUSE" is not truth-functional, these claims won't register on the truth table.)

Similarly, it is possible for an incompatibilist to reject both theses (Galen Strawson does, for example). The only thing that an incompatibilist can't do is accept both theses. Thus, you get four possible rows for the compatibilist and only three for the incompatibilist. Incompatibilism is more complex -- provided that your truth table test is the right test for this claim.

On to another matter. I like what you say against indeterminists like Bob Kane who try to use quantum indeterminacy to ground free will by adding "a random element to our decisions." Certainly this is more problematic than helpful. But why think that if indeterminism is "limited to the early stages in the decision process" that this will help, let alone that it should be a requirement for freedom? How exactly can indeterminism generate "possibilities for our will to deliberate and choose adequately determined actions"? I don't get it.