Free Will

Bryan Caplan at EconLog started an interesting discussion about free will.  My contributions:

Now this is odd. I find myself in complete agreement with Bryan here. Both his conclusion-- that there is free will-- and his grounds for it-- introspection-- are perfectly correct. What is strange is that I first learned these arguments in the book Miracles, by C.S. Lewis, in which he makes the famous Argument from Reason which refutes materialism and affirms God...  As someone who accepts Lewis's argument more or less in full as well as his theistic and Christian conclusions, it is pleasant and strange to hear this sort of argument from an anti-religious writer like Caplan...  Since I agree with Caplan I'll contribute to the cause by answering some of his critics.

Jason writes:

"The falseness of uncaused (or 'self-caused') will has nothing to do with behavior genetics. It's not an empirical truth, it's a logical truth. 'Free will' defined as something neither random or mechanistic is simply incoherent."

As Jason makes no further defense of this claim, the answer is simply that since it is not at all self-evident, the burden of proof rests on him, and he has not discharged that burden. As an aside, the idea of randomness seems at least as logically problematic as free will, yet it is generally credited. Jason goes on to make a completely different argument:

"The anti-materialist free-will argument you are implying is identical to the Intelligent Design argument of "Irreducible complexity": Point to a hole in current understanding, suggest, erroneously, that it can't be solved, and thus the scientific answer for the puzzle is... magic!"

But Caplan is making no such argument. He makes it clear that introspection, not the existence of gaps and residuals in the data, is his basis for believing in free will. And he has also pointed out the error in the heads-I-win-tails-you-lose game the determinists like to play:

"The believer in free will can simply say, 'The good twin and the evil twin just made different choices.' The determinist, in contrast, can only ask for a blank check: 'One day, we're figure out the hidden forces that caused them to be so different. Until then, bear with us.'"

Exactly. Determinism is a sort of fideism. It illegitimately transforms a necessary postulate of scientific research (for the 'laws,' so called, of nature must exhibit a high degree of consistency for experimental or, more generally, inductive methods to be of any use) into a truth, dogmatically believed, and treated as a conclusion of science rather than, what it really is, an assumption.

A 'God in the gaps,' or 'free will in the gaps,' argument should indeed be prepared to retreat in the face of decisive evidence that the gaps are explicable. A theist who had pointed to miracles as evidence for the existence of God, and who is then confronted with a scientific explanation of the supposed miracle by natural causes, may legitimately say, "While I concede that, in light of this discovery, the evidence for God is somewhat weaker than I had supposed, I still think it is dominant." But if he is wholly unmoved by the discovery, we are justified in doubting his intellectual honesty. A believer in free will who uses the lack of deterministic explanations for some mental events as evidence should concede that his case has been weakened if deterministic explanations come to light.

But what we are faced with here is something quite different. There are still plenty of unexplained mental events, yet the determinists critique 'free will in the gaps' arguments (whether really made by or merely imputed to their opponents) on the grounds that science will presumably someday come up with deterministic explanations of such events. I am not sure what to make of the rather rare situations, e.g., hypnosis, when deterministic processes seem to operating in or at least influencing human behavior where I would have expected free will to be at work. I am not sure if a comprehensive success by science in explaining human behavior would disprove free will or not. What is certain is that science has not comprehensively explained human behavior, or come anywhere close to it.

Peter Tweig writes:

"So to me this says is that as long as there's some uncontrolled residual, that this residual provides evidence for free will."

But again, Bryan is not, I think, making this argument.

Meng writes:

"Without it, one's decisions would be held completely to chance and not to any quality of the person supposedly making the decisions. Causal determinism is not a cage and I wish that people wouldn't treat it as such."

No, free will is not the same thing as chance. This is based on introspection and commonsense: to say "he did it on purpose" or "he chose to do it" means the opposite of "he did it by accident" or "by chance."

Troy Camplin writes:

"The indeterminism of quantum physics is erased at macrophysical levels."

No, it is not. Think of a boulder exactly poised at the top of a hill. It may roll to the left or to the right, and the tiniest force can upset the balance. There is no limit to the degree with which micro causes can lead to disproportionate macro events.

Daniel recognizes this ("we expect nature to be indeterministic"), but makes a different mistake:

"But indeterminism isn't enough to get you free will--a radium atom may not be determined to decay at any particular time, but that doesn't mean that it freely chooses to decay at the time it does. Randomness is just as inimical to free will as determinism."

No, randomness is not "inimical" to free will; it is different from free will, to be sure, but there is nothing inconceivable or improbable about a universe in which both free will and randomness exist, are fundamental, and interact with each other in complex ways. In fact, that is more or less the world of commonsense.

Robin Hanson writes:

"Unless your theory is that humans somehow have some special additional free will not held by all other stuff or even all other life, it seems silly to focus on human behavior for tests. It is far easier to instead test electrons or amoebas."

But obviously the theory is that humans have free will, not that everything does. That is clear enough based on the evidence to which Bryan appeals, namely, introspection. Bryan is a human being, not an electron or an amoeba. His introspective evidence for free will pertains specifically to human free will.

Bryan presumably can't spare the time to respond fully to every commenter on his blog, but it's pretty clear he could emerge from this debate with flying colors. But I would push him farther. If there is free will, must we then reject the impoverished ontology of materialism, with its unwarranted assumption that the entities we encounter in everyday experience can be reducible to those described in physics textbooks? If we reject materialism and accept a richer ontology in which there are such things as rational wills, does it not seem likely that there are rational wills other than those attached to human bodies? In particular, does not the hypothesis that there is a supreme Rational Will which is the source of all others become plausible if not compelling?

And:

Arare Litus writes:

"The apparent fact that we have free will in no way make it seem likely there are other rational wills not attached to humans, it says nothing about this..."

In a strictly logical sense-- in the sense in which the fact that the sun has risen the past 10,000 mornings does not mean it will necessarily rise tomorrow-- I sort of agree with this.

But consider the following analogy. You describe to me a strange type of building, with a circular base and a pentagonal second story, pink covered with blue stars interspersed with pictures of animals, with fountains shooting off the top and landing in a moat roundabout, and lots of other weird details. You ask me the subjective probability that there is such a building somewhere in Los Angeles. I say the probability is pretty low.

Then you tell me that such a building exists in Fairfax, VA. I doubt you, but you drive me a few blocks down Main Street and sure enough, there it is. Now you ask, what is the probability that a similar building exists in Los Angeles? Surely my subjective probability will rise a lot. I don't know why the building is there, but if someone had some reason to build one here, it seems not terribly improbable that the same motive, whatever it was, induced someone to build one in Los Angeles.

It just so happens that intersubjectivity among humans is mediated exclusively through the physical world. If there are nonembodied rational wills, we cannot experience them intersubjectively, that is, any personal experiences that we or others have with them cannot be intersubjectively confirmed, at least not in the usual way. It does not follow that such rational wills do not exist, any more than the non-existence of the strange building in LA follows from my being unable to afford a plane ticket to get there.

Occam's Razor is at most a rule of thumb, and it is often misapplied or over-applied. Materialism is one misapplication: "Explain the facts with the most parsimonious ontology," the materialist exhorts, and then demands that we accept an ontology so sparing that he cannot even pretend it is adequate to explain the facts (at least, "not yet," he justifies himself with his inexplicable fideistic confidence). So here. Most of mankind for most of history has believed in nonembodied rational wills (e.g., pagan gods, angels, devils, etc.) and many, many people in the past and today have reported personal experiences with them. I am by no means saying that all of this should be credited; but any claim that the mere Occam's Razor principle is sufficient to dismiss it it exceedingly odd. It seems to suggest that we can fully explain mankind's preoccupation with the supernatural without conceding its existence. I submit that this is hardly the case.

One thing that occurs to me upon reflection is that the concept of free will may be more fundamental than the concept of causation.  Whether we can ever observe causation is, after all, problematic.  How can we observe "A caused B," rather than merely "A preceded B?"  Do we deduce the concept of causation from our experience of choosing and seeing consequences?  Certainly the claim that an uncaused or self-caused cause is logically incoherent seems groundless.  I'm not sure that even a world entirely without causation, where everything just happens for no reason, is especially inconceivable.

Comments

[Nato]

We're about to step out the door so I don't really have the time to address this properly, and I need to do a bunch of work tomorrow so this is probably going to be holdover commentary until Monday or something. Anyway, I guess the divide between Nathan/JM and Tom/me on the grounds for allocating responsibility is that Tom and I think that someone can be said to me morally responsible for something if the person's actions can be attributed to stable features about that person rather than an external force, while JM and Nathan feel that the person must be irreduceable to other causes or all responsibility must accrue to those causes instead. Further, Tom and I think that the possibilities that we contemplate when making decisions need not be indeterminate in order to be genuine choices, and that a deterministic choice taken is still veridical for all meaningful purposes. Nathan and JM seem to feel that this betrays our sense of what a choice is. I believe Nathan are stretching a natural intuition to include post-theoretical commitments, while Nathan and JM think that I am denying critical parts of those intuitions in order to maintain a compatibilist position.

Does that boil it down pretty well?

[Nathan Smith]

Good summary, except I'd modify this line:

"I believe Nathan are stretching a natural intuition to include post-theoretical commitments..."

It's the maintenance of a compatibilist position that seems to me to involve "post-theoretical commitments," if I understand that phrase correctly. My view is: We know that we have free will, so our theories must make room for that. The compatibilist's view is: We know that the laws of nature explain why things happen, so the experience of free will must be somehow explained away.

[Nato]

Not explained away, just explained.

[Joyless Moralist]

The main point I've been trying to make is that both of us are going to naturally tend towards an interpretation that fits with our more general metaphysical view. However, there is an asymmetry here, because Tom and Nato's view, being more basic, is more readily comprehensible to us than ours is to them. Hence our tendency to say: your view is comprehensible, but doesn't incorporate all the evidence. And their tendency to say: your view is simply unintelligible.

There's no way to make this analogy that doesn't sound self-laudatory, but nonetheless it's the best way I can think of to explain what I think is going on. What we're doing here is much analogous to the situation in Plato's Allegory of the Cave. (I'll assume you're all familiar with it, though if not I'd be glad to explain.) For the people tied up watching shadows on the cave wall, talk of colors or 3-dimensional objects, or, even worse, sunlight, is simply unintelligible. It's pure nonsense-talk, and people in such a position will insist on trying to explain all intuitions or experiences through the limited vocabulary of cave-shadows.

Meanwhile, people who have explored to some degree the more metaphysically complex world outside the cave will still have some understanding (albeit perhaps a clumsier one) of the logic of the cave dwellers, but they also clearly perceive that there are some things that the cave-dwellers' vocabulary and logic cannot describe or grasp. So when their metaphysically impoverished comrades demand that they explain these other entities in cave-language, this demand simply cannot be met. The logic of this higher realm will not be intelligible to them until they've done some credulous exploration of it on their own. The skeptic will necessarily be stuck in a trap of his own making.

[Nato]

To expand on my original drive-by comment at Nathan: Incompatibilist Determinists are the ones who think the experience of free will must be explained *away*, Compatibilists think that philosophical/theological treatments of free will define the term inappropriately. It surprises me that Nathan appears to consistently ignore the central feature of my argument by using "free will" that assumes incompatibilism without caveat.

Also, if free will as Nathan uses it is antecedent to all constructed physical knowledge, then no amount of empirical discovery could ever change his position. Even if a neurologist could describe, electrical discharge by electrical discharge*, the exact process that brought Nathan to believe that free will is as he describes, that could never rise to a level that could challenge his intuition because it would be less fundamental. The theory must be wrong even if no empirical flaw could be found because it conflicts with the intuition. Is this correct, or is there some point at which empirical science could pertain?

*Or, substitute with an experimental spiritualist who was able to interface in some intersubjectively testable way with Nathan's notionally-deterministic soul, describing just exactly what soulful dynamic was at work. It doesn't really matter.

[Nato]

"The logic of this higher realm will not be intelligible to them until they've done some credulous exploration of it on their own. The skeptic will necessarily be stuck in a trap of his own making."

I can understand this if theistic philosophers (et al.) have private experiences secularists like Tom and me don't, meaning that we're simply unaware of some phenomena that need explanation. This seems plausible but a little dubious given the quantity of people who were religious and then lost their faith, and the various religious sects who believe in predestination (e.g. Calvanist strains). Another issue is that I lived for plenty of time as an unexamined incompatibilist before unknowingly re-creating (a crude form of) compatibilism in the course of responding to William James. I even believed in (an unsophisticated form of) God when I was little. Perhaps I have merely suppressed access to my true intuitions since then, but it really seems to me more akin to when I corrected my intuition that heavier things *must* fall faster than lighter ones.

[Nathan Smith]

"It surprises me that Nathan appears to consistently ignore the central feature of my argument by using 'free will' that assumes incompatibilism without caveat."

That's true. Compatibilism seems like an effort to sugarcoat determinism, to disarm the natural resistance to it by a semantic sleight-of-hand. One could, I suppose, define two separate concepts, compatibilist-free-will and incompatibilist-free-will, in which case I'd be arguing for the latter. But (a) I don't know what the former means and how it's supposed to be different from determinism, and (b) I'm quite sure that the latter is what ordinary people mean by free will.

"Even if a neurologist could describe, electrical discharge by electrical discharge*, the exact process that brought Nathan to believe that free will is as he describes, that could never rise to a level that could challenge his intuition because it would be less fundamental."

Yes, but note that the hypothetical is faulty. How could the neurologist know enough about my thoughts to explain them? He can't access them except through my reporting, which is far too imperfect and slow and intuitive to be suitable evidence for scientific research of the degree of detail and confidence suggested in the example. The division between the domain of introspection and the domain of intersubjectivity, bridgeable only by the strange and wonderful but fallible instrument of language, is a basic empirical fact about our world, by which we can know that the hypothetical Nato mentions cannot occur. As for the other suggestion that a spiritualist could interface in an intersubjectively testable way with my soul, should such a thing occur, I think we would have to call it, in the strictest sense of the term, a miracle.

[Nato]

"(a) I don't know what the former means and how it's supposed to be different from determinism"

It's orthogonal to determinism. One might call me a compatibilist indeterminist and Tom a compatibilist determinist. What might be confusing that perception in this discussion is Tom and my agreement that too much indeterminate causation would undermine free will by weakening the degree to which an individual could be said to cause their actions. Tom and I disagree about the plausibility of basic randomness in existence as opposed to effective randomness, which has very limited bearing on compatibilist decision as to whether an individual was acting of their own free will in any case. Also potentially confusing is that Tom is more willing to cede the term 'free will' to its classic theoretical description, abstracting all the desirable parts thereof under other terms (e.g. responsibility) that whereas I want to keep a place for the everyday uses of the term.

"How could the neurologist know enough about my thoughts to explain them? He can't access them except through my reporting, which is far too imperfect and slow and intuitive to be suitable evidence for scientific research of the degree of detail and confidence suggested in the example."

Presumably a complete account of Nathan's introspection would also detail an exact, reliable and sufficient process leading to Nathan's reports. This would, of course, be forever impossible if Nathan's intuitions about the characteristics of choice are correct, but in the case where they are in some way mistaken, it would appear that the neurologist's task is all for naught, because Nathan would perforce still regard the neurologist's account as being wrong from first principles. Of course, if Nathan takes the position that his intuition could not possibly be wrong and represents True information about the world, then he can dismiss the possibility of this scenario ever occurring in the same way that a mathematician never has to wonder if 3+3 will ever equal 7. I don't think that's the way Nathan wants to go, though.

[nato]

I perhaps owe a restatement of my compatibilist definition of free will: Free will consists of decision-making while substantially free of coercion and manipulation. As we gain ever more knowledge and self-control, our wills become freer because we are less and less subject to the vagaries of external forces. In this view, someone saying "I can do no other" because of a moral stance becomes the supreme affirmation of free will, because one has achieved a level of moral self-control that prevents one from choosing the easy wrong over the hard right. Put yet another way, free will is when internal self-control is capable of overcoming external controlling forces. When we wrestle with temptation, that too is an exercise in free will, since we're struggling to establish the supremacy of self-control, and in the long term we're trying to make it impossible for us to yield to those temptations.

This fits very well with the everyday usage of the phrase, and makes most of the distinctions we care about. I think that some people ultimately feel the idea that they themselves are a product of the Universe invalidates distinctions between internal and external, but that's a mistake. I hear, at times, people pointing out that nothing ever really touches; electrons exchange photons and repel one-another, creating tension that prevents one 'solid' from moving through another. Of course, this is a silly way of looking at things, that presupposes that for anything to count as *really* touching two objects must come to the point where there is no space between them in any sense because that's how we experience 'touching' in the middle-scale world we perceive on a day-to-day basis. Of course things look different from up very close, but at the point at which electrons are the most salient objects, we've zoomed far out of the realm in which that intuition about touching applies. That it strikes us differently in one perspective doesn't invalidate the other.

So, then, my compatibilist position is that selves, choices and responsibility are veridical and literal in the same way that chairs are veridical* and literally touch the floor. No particular explanations are entailed, and certainly nothing is explained away.

*And always would have been, even if the underlying physics had turned out totally differently.

[Joyless Moralist]

Thanks, Nato. For myself, I'm pretty familiar with the basic claims of the compatibilist, but compatibilism always seems to me to involve mostly a relabeling and redefining of terms, as opposed to a genuinely significant philosophical shift from the more straightforward forms of determinism. Certainly it does nothing to clear up my worries about moral responsibility.

I think fundamentally, the re-definition of "freedom" just doesn't capture any of what's important in the concept. I mean, sure, as things grow and develop, they become less immediately susceptible to external influences, so we might reasonably say that the "ratio" of external to internal influence shifts towards the internal. You don't just have to look at human beings for that; the same would be true of, say, an oak tree. In its acorn phase, it's beholden to all sorts of minute forces even to get a chance to sprout. Similarly in its sapling phase, it's quite vulnerable to frosts, hungry animals, and all sorts of other things. Once it grows into a mature oak, it becomes more stable, and isn't as much affected by minor environmental changes. It takes something pretty dramatic to keep it from fulfilling its oakish nature. So, in exactly the sense you described, we might say the mature oak is "more free" than the sapling, right? But we're not actually inclined to describe trees or ecosystems or what have you as "free" under such circumstances. And we're certainly not in the least inclined to apply moral categories to them. To blame the tree for falling or praise it for providing shade is simply to misunderstand how moral terms are used.

"I can understand this if theistic philosophers (et al.) have private experiences secularists like Tom and me don't, meaning that we're simply unaware of some phenomena that need explanation."

Well, we might. I certainly think that having faith/a broader metaphysical perspective can affect how one experiences the world. And, by the way, while I don't by any means accept predestination as a doctrine, I don't think such belief is usually motivated by a denial of free will (or a compatibilist understanding of it.) Actually, most such Protestant sects adopt a fairly extreme voluntarist position, I believe. But that is coupled with an understanding of human nature as being deeply depraved, such that the exercise of our autonomous wills can never lead us closer to salvation. Hence our utter reliance on God to gratuitously pluck us out of our miserable state. Hence predestination.

I didn't intend my reference to the cave analogy as an argument, and naturally I don't expect you simply to believe me. (It would almost tell against my description if you did believe me. :)) But there is an asymmetry to the conversation, and that is the only device I can find to describe it. This talk of free will goes way beyond just some isolated impression of Nathan's or mine; there's a whole different way of understanding the world behind this. To explain the elements of this view in the terms of your much more restricted metaphysical view just wouldn't be possible, but that doesn't mean that it's all completely ad hoc. If you wanted to really get perspective on these different outlooks, you'd have to do some extended, credulous exploration of that sort of view -- much more than a transitory childhood belief in a Big Nice Man in the Sky could give you. Otherwise, negative argumentation is mostly all we can offer in terms that you're prepared to accept.

[nato]

"...we're certainly not in the least inclined to apply moral categories to them. To blame the tree for falling or praise it for providing shade is simply to misunderstand how moral terms are used."

This is certainly true, since nothing the tree does can be said to be caused by any sort of conceptualization, much less moral reasoning, whereas human decision-making is suffused with moralistic valuations, whether good or bad.

"[Compatibilism] does nothing to clear up my worries about moral responsibility."

"If you wanted to really get perspective on these different outlooks, you'd have to do some extended, credulous exploration of that sort of view -- much more than a transitory childhood belief in a Big Nice Man in the Sky could give you. Otherwise, negative argumentation is mostly all we can offer in terms that you're prepared to accept."

I suppose I can understand this as a reason JM and Nathan would fail to accept my construal of free will, but I'm not sure what negative argumentation is on offer. Is it just that cognitive scientists cannot give a complete account of the mind? Compatibilist positions mostly bracket mechanics, so the presence of a complete functionalist account wouldn't help nor does the lack hurt. Is the negative argument that we don't specifically have a convincing account of certain intuitions about 'free will' - if so, I can certainly elaborate on extant sketches. Or is there some logic error? If so, I can't seem to find an instance of someone pointing it out.

Of course, there's a certain point at which JM and Nathan might say "Well, your position is complete and fairly well-supported from your perspective, but the deficiency of that perspective means that your conclusion is inadequate." After exhausting negative arguments and presumably justifying Compatibilist responsibility from all perspectives, any inadequacies would have to relate to additional items arising from metaphysical factors. I would already have a naturalist responsibility, but lack responsibility+.

I guess that could be regarded as dangerous, since if there's a working (but presumably inferior) way of justifying morality from a naturalist perspective, then the motivation to engage in credulous explorations of the various religions might decline to the point where access to responsibiliy+ is eventually lost irrevocably. I would hope that there's some reliable, cave-accessible way of demonstrating to cave-dwellers that there's a more colorful, multi-dimensional world outside, especially given the number of available outsides, some of which are trivially easy to dismiss as delusion. "Hey, out here we have a square circle!" some call, and I know that they don't, both because it's logically impossible and because I've seen the sort of malarky such claimants tend to show once I take a look. Others say something more vague, "If you come over here, I can show you something that will really expand your horizons." That's a far more plausible claim, though so far I've been disappointed with that as well. Really the only times I've really had my world rocked is when I've gone into the cognitive philosopher's chamber of the cave, where I saw things I certainly never had before, to the point where it took years to see them.

To make this more concrete, a person who asks an interlocutor to first assume the the truth of their own position in order to get access to the facts undergirding that position has the difficulty of competing with many many competing claimants to such private knowledge. Nathan makes a valiant and plausible case for the results of Christianity, but then has the difficulty of explaining how something that is *not* unitary can make claim to morality as a whole. Quakers were critical to ending slavery, but Nathan isn't a Quaker, he's Orthodox, which was busy taxing peasants as an element of tsarist government at the time. Quakers were also amongst the first to allow women to preach and are frequently on the forefront of accepting committed same-sex couples. Atheists are similarly difficult to group, of course, since they range from Stalin through Russel to Jainists and some sects of Buddhism, but as a negative position it has no particular truths to advance, much less private ones.

[Nathan Smith]

"I'm not sure what negative argumentation is on offer. Is it just that cognitive scientists cannot give a complete account of the mind?"

That is a negative argument against determinism, so to the extent that compatibilism is a branch of determinism, then yes, it would apply to compatibilism too. The basic point is that it's invalid to reject free will just because you want to have a "scientific" explanation of everything when we don't have explanations that reduce everything to a materialist basis and there's no reason to think we ever will. But when Nato writes:

"Compatibilist positions mostly bracket mechanics, so the presence of a complete functionalist account wouldn't help nor does the lack hurt."

that reminds me that as far as I can tell, compatibilism is basically a semantic twist that tries to neutralize the commonsense advantage of free will by appropriating the ordinary vocabulary of will and morality and trying to supply narratives and phraseologies that seem to make that vocabulary consistent with determinism. JM's oak example is a brilliant retort to Nato's effort in this direction.

Nato has seemed at several times to recognize the concept of free will as an independent causal factor, which he calls 'radical' free will as a rhetorical device to marginalize it. I am quite certain that it is exactly this that ordinary people who use the phrase 'free will' have in mind, but ultimately whether the compatibilists succeed in marginalizing it or not is a secondary issue. If Nato and other compatibilists succeed in developing a concept of determinism-compatible free will, I would just say that I'm not particularly interested in it; that I'm confident we have 'radical' free will; and the debate would go on as before between determinism-cum-compatibilism and ('radical') free will.

"if there's a working (but presumably inferior) way of justifying morality from a naturalist perspective"

There isn't. Ideological naturalists just borrow Judeo-Christian morality, with some modifications, without being able to justify it. They offer untenable arguments and then hope people won't look too close.

"[the] Orthodox... [were] busy taxing peasants as an element of tsarist government at the time..."

That's a bit unfair: the Orthodox Church didn't have enough power to challenge the tsarist government, which had abolished the patriarchate by force under Peter the Great; and the concept of absolute monarchy was pre-Christian in origins, going back to ancient imperial Rome. Also, from the 18th century on, the tsarist government broke with the Muscovite past and emulated the high culture (without being able to understand or develop the institutional foundations) of the West. The church hierarchy often could not resist being instrumentalized by the state to a certain degree, but the doctrine and practice of love and mercy persisted at the popular level and it is thanks to this that Russians look back with at least partial nostalgia on the peasant life of "Holy Rus" in a way that few would look back on being a slave in imperial Rome.

Anyway, tsarist Russia was certainly no worse than the Roman empire, Chinese empire, Aztec empire, Carthaginian empire, etc., in terms of oppressing its peasants; it compares unfavorably only to the West, which had been Christian longer. This is the general pattern: Christian societies often compare unfavorably to other Christian societies, but virtually never to non-Christian ones.

[Joyless Moralist]

"This is certainly true, since nothing the tree does can be said to be caused by any sort of conceptualization, much less moral reasoning, whereas human decision-making is suffused with moralistic valuations, whether good or bad."

But that's circular. The whole question concerns what sort of conditions would be necessary in order to make moral reasoning and moral valuations possible/veridical. Here you just seem to be saying that humans are moral animals because they are.

Let me put this another way. It seems to me that, when we make a moral judgment, we are not simply expressing a "boo" or "hooray" (as the non-cognitivist would contend.) We are suggesting that, quite beyond our own personal feelings of like or dislike, a particular action ought or ought not to be done, according to some sort of objective moral standard. It isn't always necessary for the person being praised/blamed to understand or agree with this standard in order for the judgment to be veridical (though in some cases ignorance may excuse, depending on the reasons for that ignorance.)

That we do make such judgments is, I think, pretty clear (and even within analytic philosophy, non-cognitivism has gone into serious decline after the powerful arguments of such philosophers as Peter Geach.) The question that follows is: are such judgments justified? And if so, why? It's not enough simply to observe that an oak tree doesn't make these sorts of judgments, because *we* could still make them about *it.* There needs to be some categorical difference between us that makes such judgments applicable in one case and not in another.

And the main justification I find in your posts seems to be expressed here:

"Put yet another way, free will is when internal self-control is capable of overcoming external controlling forces."

But that's exactly what the tree *does* seem to have -- the ability to overcome external controlling forces through the maturation of internal powers. So if moral judgments are to have meaning in the human case, we need another account of its origin.

[Nato]

I'm not sure if JM is arguing against someone who doesn't believe in cognition, or someone who denies objective morality, or someone with some other opinion I don't hold. Trees don't cogitate, people do. People reason, trees do not. People feel happy and sad and love and hate and so on. No one seems to think that we don't veridically think. Is JM saying that if determinism is true then we're not really thinking? I thought we were arguing about whether we'd be mistaken to assign moral valuations if determinism is true, not whether we'd be mistaken to even believe we engage in moral reasoning. That would be a very aggressive and somewhat unusual position. Even eliminativists like the Churchlands don't deny that we think, just that our psychological language meaningfully represents underlying neurological events*.

Maybe I'm going off in the wrong direction here; I don't feel like I understand what JM's trying to say. It's just that the comparison of a tree to a human seems so very strange. Humans make moral valuations and those valuations drive their behavior in pursuit of contemplated goals. Trees just grow. It's very hard to decide whether will is free or not if there's no will in the first place. Perhaps JM objects to my asseveration that trees have no will, since we could ascribe one to them, but this applies to everything. If evidence can tell us another human has a will, it can tell us trees don't.

Anyway, that's a bunch of stabs in the dark, none of which I can really imagine being hits. Let me know.

*A position that has become increasingly hard to hold, at least is an aggressive form. Research connects the metaphors and categories of our everyday mental language with actual neurological organization all the time. Even Dennett has been impacted by this as VS Ramachandran's experiments have shown that there are important ways in which peoples' impression of how the visual system works coincides with the way it really does.

[Joyless Moralist]

Basically, I want you to explain what free will is in such a way that it's clear to me why a tree doesn't have one. Because it seems to me that a tree *does* have "internal self-control capable of overcoming external forces," which, according to you, is the primary definition of a free will.

One of the reasons that this is so hard is because I do indeed think the evidence for the non-necessitation of our wills is very fundamental to our experience of the world, such that freedom (of the non-determined sort that I believe in) *does* seem to be a necessary component of thinking, and choosing, and engaging in moral reasoning.

So perhaps the most *accurate* thing to say would indeed be that in a deterministic universe, we don't actually engage in moral reasoning (except we clearly do, which is a good reason not to be a compatibilist, but...) But I thought it might be easier to focus on the question of what makes moral judgments veridical. It's a commonplace of moral argumentation that a moral "ought" implies a "can." (I think there are a few interesting subtleties of this claim that are worth exploring, but I don't think any are of relevance to this conversation, so I'll just leave the traditional claim standing.) Trees aren't subject to moral judgments because they lack the "can" -- their behavior is necessitated by a combination of external and internal forces or influences (i.e. they aren't free) so it's meaningless to talk about what they morally "ought" to do.

In order to explain, then, why human beings *are* subject to moral judgments, we need to explain why they are free in a meaningful sense that trees are not. To me, the existence of complex neurological processes in the intermediary (call them "moral reasoning" if you like, but it makes no difference) doesn't change the moral status at all if the outcome is still entirely determined. To show me otherwise, you would need to give me an account of freedom that applies to humans but not to trees. But, to escape circularity, you'll have to avoid appeals to mental activities that, on my understanding, presuppose a free (non-necessitated) will. "Humans are morally different from trees because they engage in moral reasoning" doesn't get us anywhere, because, in my view, our ability to engage in moral reasoning is one of the primary evidences of the falsity of compatibilism.

[nato]

"it seems to me that a tree *does* have "internal self-control capable of overcoming external forces," which, according to you, is the primary definition of a free will"

If this is the problem, then it's a matter of context. I assumed that we were already discussing only conscious beings. For conscious beings' wills to be free, they have to be able to overcome external forces. If there aren't any wills in play, we don't have to wonder if they are free or not. The interesting contrast in the discussion was between free and non-free wills, so I didn't feel the need to specify additional premises related to whether a will was present at all. The rather belabored exegesis aside, I'll move on.

"It's a commonplace of moral argumentation that a moral "ought" implies a "can." ... In order to explain, then, why human beings *are* subject to moral judgments, we need to explain why they are free in a meaningful sense that trees are not."

The meaning of 'can' is the big deal here. let me quote my very first comment in this thread:

"The mystery occurs because of the use of a sense of possibility that is either inappropriate or degenerate. That I "can" choose otherwise is interesting in a different respect than the manner in which a quantum random number generator "can" emit '0.534'. We can insist on both of these being the same form of "can," but I see no motivation therefor. If we keep apart capability, stochastic indeterminacy and epistemic incompleteness, I think the problems of free will become far more tractable and require no appeals to mystery."

I didn't flesh this out any further because the discussion went in a different direction, but it is a critical matter. If an engineer says that a specific engine design "can" produce a certain amount of torque, this is somewhat different from saying that engines as a class can produce that torque. In subjunctive forms of the word, an accident investigator might observe that an engine "could" have been the source of speed in a collision, right before introducing evidence that was not, in fact, the case. There's even a very special form of "could" that might apply to quantum events in which the result is not only unknown but independently causal. There's lots of forms of the word "can," and I think all of them may supply the morally-relevant sense of possiblity, depending on circumstances. "Can the genocide be averted?" has nothing to do with whether it's physically impossible, it's a question as to whether there's an acceptable course of action to be taken that could reasonably be expected to result in preventing the genocide. I think we are usually aware of the 'specific engine' form of personal possibility, though we frequently also deal with the 'engines as a class' form when contemplating longer-term possibilities, and so on. Pretty much the only times I can think of ever dealing with the independent causality form are when we get into thinking about philosophy, theology and metaphysics.

[nato]

Also, a note to Nathan: Yes, I agree that the comparison between the Quakers and Orthodox Church is unfair; they were not in comparable situations and we cannot expect to compare outcomes as some kind of verdict on their moral judgments. However, Christianity and other religions have also not been on equal grounds, so we'd want to be careful about comparing them. Without offering any supporting evidence or even thinking very deeply, I would rank the aggregate moral ratings of selected worldviews from best to worst:

1)Modern Buddhism
2)Modern Christianity
3)Enlightenment Humanists
4)19th century (Western)Christianity
5)19th century Buddhism
6)Modern Islam
7)15th century Islam
8)15th century Buddhism
9)15th century (Western)Christianity
10)10th century Islam
11)1st century BCE Roman Republic
11)2nd century (Western)Christianity
12)2nd century Roman Empire
13)10th century (Western)Christianity
14)1st century BCE Judea
15)Joshua 10
16)Nazis circa 1943
17)1950s Yanomami tribes.

[Joyless Moralist]

"Non-free will" is also an oxymoron in my book, so I'm afraid that provision doesn't help much either. I might define a will as being something like, "an individual locus point of free choice/decision." But of course, we haven't agreed on a definition of free choice or decision. So if that's supposed to explain the difference between human beings and trees, then once again we have a circularity problem.

See, this is the difficulty with compatibilism -- it's very hard to discuss it because it gets most of its force by using the same terms the voluntarist uses, only to mean very different things. It makes it devilishly hard to get a discussion off the ground.

"If an engineer says that a specific engine design "can" produce a certain amount of torque, this is somewhat different from saying that engines as a class can produce that torque."

I don't see how this is helpful in the least. Yes, there are sometimes ambiguities about what class of things we're referring to when we use modal operators. But when making specific judgments, one of the first things we want to do is clear those up. If I want to make judgments about the value of a particular sort of engine, I'll be interested in the capabilities of that particular engine. There's no sense in which the engine becomes able to produce X amount of torque because other kinds of engines can do so.

"In subjunctive forms of the word, an accident investigator might observe that an engine 'could' have been the source of speed in a collision, right before introducing evidence that was not, in fact, the case."

Yes, but here the uncertainty being described is epistemic, not metaphysical. Once the definitive evidence to the contrary has been gathered and presented, a person who said that the engine "could" have been the source of speed in the collision would be corrected: no, it couldn't. They've already proven otherwise.

If that's the sort of contingency that's operative in the human case, then we're left with the curious result that moral responsibility will gradually diminish with advances in physics and psychology. Once a thing can be definitively explained, the uncertainty will vanish, and with it the modal operator.

I'm afraid I still don't see any relevant sense in which a person can be said to have been able to do X if it was 100% metaphysically necessary that they would not.

[nato]

"If [epistemic contingency is] the sort of contingency that's operative in the human case..."

While one could correctly conclude from what I said that I believe epistemic contingency to be operative in the human case depending on circumstances, the most common scenario seems to be that of the specific engine. "What have I the capability to do?" we ask ourselves, and enumerate possibilities despite understanding that if we go to try to execute any of those we may discover that unexpected or unconsidered circumstances mean that the course of action cannot actually be completed. Thus, in everyday cases we tend to bracket epistemic possibility.

Of course, JM points out that as we have more and more knowledge about the world, the set of things about which we could be mistaken shrinks, presumably until the omniscient decision-maker only ever has only one option. The mathematics of information theory shows that no closed system can model itself entirely*, but there is no danger of anyone ever knowing their own mind, and nothing in the Universe could determine the precise future faster than the future arrives. It would seem that we are logically precluded from ever knowing our options to be collapsed to one except in degenerate cases like imminent death.

JM still might grant this kind of possibility and still say that it's not free will by my definition, since we can't overcome the Universe. I think this is the basic reason why ordinary people, when asked, want special causality for free will: it seems like the only way to overcome the brute facts of existence. I understand the feeling, but it doesn't really help. Even with special causality the Universe keeps on buffeting us, our wills unable to dictate events. All that we get out of it is being out of the Universe's control as well, which we value because we (for good reasons) reflexively hate the idea of being manipulated.

So, then, we can progressively better about weeding out some impossibilities, this just pushes the locus of uncertainty farther out. We live with this in our everyday lives, as we stop deciding to be princes or frogs when we grow up, and lose any impetus we might have had to blame ourselves for our inability to achieve being noble amphibians. I suppose that's a kind of reduction in moral responsibility, but it shouldn't be worrying.

On the other hand, one might claim that an omniscient viewer would be unable to assign any moral responsibility to anyone in a determinist Universe because there would only ever have been one way anyone could have been. This doesn't appear to trouble engineers, however, when they evaluate the fitness of a set of machines for a certain purpose. If the omniscient entity evaluates us for our goodness, I'm not sure how the presence or absence of metaphysical alternatives could make a difference. We're either good or we aren't. "But we have no control over whether we're good or bad!" one might reasonably object, to which I answer: yes, at some point the regress of responsibility would indeed pass through and beyond us. We finite beings only get to discover later if we have what it takes to be good.

That said, we're part of the Universe, so if we're subject to it, we also own it. We don't dissolve; we take part in the determination of our own destiny.

" I might define a will as being something like, "an individual locus point of free choice/decision." But of course, we haven't agreed on a definition of free choice or decision."

Since you offer, here's mine: I would define a will as being, in the very broadest sense, a coherent system that searches for paths to fulfillment of represented and evaluated goals. Systems with attenuated representation and evaluation are wills in only attenuated senses. I could say that a dog has a will, but its representations and evaluations are going to be far less rich than that of a human, so I shouldn't regard it as being as veridical.

"I still don't see any relevant sense in which a person can be said to have been able to do X if it was 100% metaphysically necessary that they would not."

If a person could (engine-wise), and chose not to, then presumably this says something about who and what they are, which seems like relevant data, whether or not there's anything metaphysically necessary about it.

*At least, not in real time or faster

[Nathan Smith]

re: "3)Enlightenment Humanists"

The French Revolution.

[nato]

The French Revolution was certainly awful, but to the degree we can blame Enlightenment thinkers morally seems to depend on if they should have known better. Arguably that was when we really learned the lesson that no one is smart enough to dream up a total reinvention of society that will really be better than the system that has evolved over the centuries. Now, after than came the Marxists and Anarchists and Socialists etc etc, and they had less excuse. I don't know. Like I said, I didn't actually think all that deeply on it.