Monday, November 07, 2011

The Crazyist Metaphysics of Mind

... a new essay of mine, now in circulating draft. Comments welcome, either on this post or by email.

Crazyism about X is the view that something it would be crazy to believe must be among the core truths about X. In this essay, I argue that crazyism is true of the metaphysics of mind. A position is "crazy" in the intended sense if it is contrary to common sense and we are not epistemically compelled to believe it. Views that are crazy in the relevant sense include that there is no mind-independent material world, that the United States has a stream of conscious experience distinct from the experiences of the individuals composing it, that chimps or the intelligent-seeming aliens of science fiction fantasy entirely lack conscious experience, that mental events are causally inefficacious. This is by no means a complete list. Well developed metaphysical theories will inevitably violate common sense, I argue, because common sense is incoherent in matters of metaphysics. No coherent and detailed view could respect it all. With common sense thus impaired as a ground of choice, we lack the means to justifiably select among several very different metaphysical options concerning mind and body. Something bizarre must be true about the mind, but which bizarre propositions are the true ones, we are in no good position to know.


Unknown said...

Natural selection as a process essentially undirected by the organism is crazy, yet still persists.

Anonymous said...

I think you're on the right track!

Also, I think there's another problem with the standard materialist position, which Stephen Hawking articulates very clearly:

Now, if you believe that the universe is not arbitrary, but governed by definite laws, you ultimately have to combine the partial theories into a complete unified theory that will describe everything in the universe. But there is a fundamental paradox in the search for such a complete unified theory. The ideas about scientific theories outlined above assume we are rational beings who are free to observe the universe as we want and to draw logical deductions from what we see. In such a scheme it is reasonable to suppose that we might progress ever closer toward the laws that govern the universe. Yet if there really is a complete unified theory, it would also presumably determine our actions. And so the theory itself would determine the outcome of our search for it! And why should it determine that we come to the right conclusions from the evidence? Might it not equally well determine that we draw the wrong conclusion? Or no conclusion at all?

He goes on to propose this rather weak solution:

The only answer that I can give is based on Darwin’s principle of natural selection. The idea is that in any population of self-reproducing organisms, there will be variations in the genetic material and upbringing that different individuals have. These differences will mean that some individuals are better able than others to draw the right conclusions about the world around them and to act accordingly. These individuals will be more likely to survive and reproduce and so their pattern of behavior and thought will come to dominate. . . . [P]rovided the universe has evolved in a regular way, we might expect that the reasoning abilities that natural selection has given us would be valid also in our search for a complete unified theory, and so would not lead us to the wrong conclusions.

However, if the theory determined that evolution would happen, and evolution determined that our species would exist, and also that some individuals of our species would do such a thing, then this answer would seem to beg the question.

Unknown said...

Yeah, determinism, another crazy "certainty" that if true, would not have allowed the evolution of needless choice.

Richard Marshall said...


I like Crazyism. I wonder if you think that as Crazyism becomes more widely known the 'common sense' beliefs of folk may alter to mitigate the effects of the theory? There's some evidence I believe that innate language acquisition is pretty much fixed to be unalterable ( so we couldn't acquire a mirror language) but with other innate capacities like morality there seems to be some evidence that these are less unchangeable. So I was wondering whether you think we're stuck with the common sense intuitions we have or whether they could be changed. If they did change for the reason above, would they still be useful to us?

Anonymous said...


I don't think the problem with Hawking's answer is determinism, but rather reductionism.

If evolution ultimately "reduces" to the fundamental laws of physics, with no residue left over - then evolution can't explain *more* than physics does.


Because in that case every statement using evolutionary terms has an equivalent (though perhaps much more complicated) statement that uses only physics terms.

If evolution *doesn't* reduce to fundamental physics - then what is the "extra sauce" that evolution adds above and beyond what mere physics can explain?

I think this is where it Hawking's attempt to resolve his "fundamental paradox" by appealing to evolution fails.

But it doesn't matter whether the fundamental laws of physics are deterministic or probablistic. Only that the theory of evolution reduces to those laws and nothing else (except maybe the initial conditions that the laws act on).

But maybe I'm wrong? Any and all help in setting me straight is appreciated!

Unknown said...

The fundamental laws of physics are, at least as we discern them, regulatory. Therefor, they would appear to serve what would seem to be a logical and anticipatory purpose. Anticipatory systems evolve strategically - they are not reductionist.
Hawkins may not agree but then a number of other physicists will.

Jan said...

This is a great paper!
Some questions and comments:

The abstract is maybe not clear enough on the distinction between metaphysics of mind and metaphysics in general.

On page 3, you say: "On my way, it is probably..." You should drop the "probably" unless there's a special reason for it!?
P.30: Would you say that in the limiting case of two bizarre, but equally likely theories - would you call them both "crazy"?
p.33: "These disputes can probably be counted on two hands" It would be nice for you to actually do it.
p.36: is section xii. another means of settling the matters of section xi.?
p40: "earth exists. in even more stories, the computer screen ... exists." Could you clarify?

Finally, it would be interesting to hear something about "universal crazyism". Might not ANY well developed theory be crazy - via the crazyness of physics and/or metaphysics?

Anonymous said...

If alternatives X and Y are both contrary to common sense, and we aren't epistemically compelled to accept either individually, but we are epistemically compelled to accept that one of X and Y is right, are they still both crazy? E.g. suppose there's no good reason to think there was a first cause, and no good reason to think there wasn't (anyone who thinks one of these alternatives is obviously right could pick another example), we still have to accept that there either was or wasn't a first cause. Unless maybe the question doesn't make sense. But supposing it does, I don't think it would be crazy to accept one of those alternatives as being the better explanation of the current state of the world, or for whatever reasons metaphysicians have when they accept views that aren't epistemically compelled. (This raises another worry for me - metaphysicians seem to spend most of their time trying to convince each other that acceptance of their particular theory is epistemically compelled, and don't go in for weaker reasons like explanatory strength unless they think they can show that in their particular case those reasons are unusually effective and actually can compel acceptance of their view. So what metaphysician could accept that their theory were crazy in your sense?)

Unknown said...

Crazy or not crazy is not ordinarily an "either or" proposition. Crazy or less crazy is more likely the usual case.

Eric Schwitzgebel said...

Thanks for all the thoughtful comments, folks! And sorry for the slow turnaround on replies: Veteran's Day, etc.

@ Anon Thurs 10:35: Yes, I agree that that's a pretty tepid answer. There's no guarantee -- not even as far as I can see a good reason -- to think that there would be selection pressure toward capacities that would allow us to understand the fundamental structure of the universe, rather than capacities that would allow us to get along in slow, mid-sized Earthly affairs, given that the rules governing the tiny, huge, and fast seem to be very different from those governing the mid-sized and slow.

Eric Schwitzgebel said...

@ Roy: I wouldn't rule out an "anticipatory" system myself. And yes, "crazy" is a matter of degree, because bizarreness is a matter of degree and dubiety is a matter of degree. It simplifies the structure of the essay to speak in yes-or-no, but I very much hope nothing in my thinking on this issue depends on sharp lines.

Eric Schwitzgebel said...

@ Richard: Yes, I believe that common sense can change, as it probably has regarding the motion of Earth. What is crazy can become non-crazy either because we become compelled to believe it (crazy to merely bizarre, e.g., relativity theory) or because common sense changes (crazy to common sense, e.g., motion of Earth).

Eric Schwitzgebel said...

@ Anon Thurs 6:58: I'm inclined to agree that determinism vs randomness isn't the core issue here.

Eric Schwitzgebel said...

@ Jan: Thanks for the careful read!

On page 3: The "probably" is necessary because the account is to some extent an empirically based conjecture. I haven't shown to a high degree of confidence that common sense is incoherent and thus that a non-bizarre theory is possible. I offer that only as probably the case.

On page 30: I am using the <=50% credence cutoff for the purposes of this essay (though dubiety is a matter of degree), so if the two theories are exhaustive of the possibilities and deserve truly equal credence, then neither will rise above 50%, so both will be dubious. But that might not be a very stable situation. As soon as one becomes a tiny bit more credible than the other, then it will no longer be dubious by this criterion.

On page 33: I'm reluctant to do it, since I doubt I would come up with the right list and I don't want to distract the reader with an objectionable list. But some likely candidates are: free will, the nature and existence or non-existence of moral truths, fundamental ontology, the existence of a creator, the existence of realities or universes beyond what we can have ordinary Earthly evidence for, the relationship of mind and matter.

On p. 36: I'm inclined to think that the cosmology section supports worries about method rather than supplies a method. I don't think we can resolve the issues through scientific cosmology -- not in the near to medium-term at least -- so the cosmology section is partly an "even if you thought we could" argument.

On the computer screen: Here's a story with a computer screen but no Earth: You are a Boltzmann baby, randomly congealed from chaos, by miniscule chance, five minutes ago along with one cubic kilometer of your environment. Here's another one: moderate solipsism on which you and your nonconscious mind are all that exists, so the computer screen though it exists is really nothing but a certain kind of state of your own nonconscious mind. (Admittedly, these are strange stories!)

On "universal crazyism": I think crazyism might be justified in more domains than I have discussed in this essay, but I'm not inclined to universalize it at this point. For example, there seem to be some well developed theories in mathematics that (though possibly bizarre) are not dubious. Also, I think that some well developed scientific theories are not dubious enough to go below 50% warranted credence, as long as we are liberal about what the metaphysics underneath might be -- perhaps also with caveats about the spatiotemporal range of those theories.

Eric Schwitzgebel said...

@ Anon Sat 8:19: With only two non-exhaustive theories to choose from (P or not-P), as long is one is more credible than the other, that theory will not be dubious by the criterion for dubiety that I use in this essay, since the warranted degree of credence would be above 50%. Admittedly, warranted degree of credence of 51% is a very weak sense of "epistemic compulsion"; I should probably address that point explicitly in a revision. I think that low standard for epistemic compulsion is good for the purposes of the essay, since it prevents the response that I am only making crazyism seem plausible by having too high an epistemic standard for belief in a theory.

On your parenthetical point: Metaphysicians in the early modern period often seemed to want absolute certainty. I think the sights tend to be a bit lower now. Some metaphysicians now even seem to think in terms of conjectural best explanations -- though I suspect that's still a minority attitude. I'd be with that minority, though still probably more skeptical than even most of them.

Jackson said...

IMHO Ortega y Gasset's Metaphysics is entirely "sensical," but its "sensicalness" is not yet "common." See his Some Lessons in Metaphysics (Norton, 1971).

Richard Marshall said...


I wonder whether the Crazyism comes about because of traditional philosophy's attempt to present a metaphysical belief as a single unified thought about x. Josh Knobe, Josh Greene, yourself and others prrsent evidence that this might not be so. Prinz, Machery, Carruthers all seem to agree that the mind is not all of one mind! So if standard metaphysical philosophy has ignored this bifurcation and assumed that there will be just one position that can be identified as the belief an agent holds about x. then this seems wrong. If this sort of thing is right then isn't Crazyism just a way of identifying what happens if the traditional position is assigned to people. But without the traditional position, why would we say that a person holds belief x when we think that she will hold an opposite belief if we set the issue up to appeal to a different intuition as is standard in the non-trad approach? Wouldn't we be better saying people don't really hold any metaphysical beliefs, but something much more greyed out, less certain etc

Eric Schwitzgebel said...

@ Jackson: I'm not surprised to hear you say that!

Eric Schwitzgebel said...

@ Richard: Yes, that would be one way to play it out. Everyday commonsense attitudes might be a self-contradictory set in the way you describe. Then conflict with *some* aspects of common sense would be inevitable, for any theory treating the same range of issues, yes?

Richard Marshall said...


Yes, that's the sort of thing I'm thinking. But then I guess the question is whether you think there's anything special about metaphysical beliefs as opposed to other types of belief where the same inconsistency holds. Doesn't this suggest Crazyism is just a general feature of a whole bunch of our beliefs?

Eric Schwitzgebel said...

@ Richard: Yes, I believe there are other domains besides metaphysics in which crazyism could be defended. But it's not just enough that common sense is contradictory or demonstrably mistaken. It also has to be the case that none of the bizarre but live options at the core of that domain is well justified enough to compel belief. There are some adjustable parameters here, esp. how finely the options are individuated, how to think of the "compels belief" requirement, and how bizarre is bizarre enough. Unless the options are at least moderately coarse, the criteria for compulsion weak, and bizarre understood strongly, "crazy" seems too strong a word.

Rachael Briggs said...

Cool paper!

How much stronger is crazyism about X than bizarrism about X (the view that something bizarre is among the core truths about X)? It seems that bizarrism about X, conjoined with the proposition that are core truths about X that we are not compelled to believe, entails crazyism about X. For suppose bizarrism about X is true, and let P be a bizarre core truth about X. Now suppose that there is a core truth Q about X that we are not compelled to believe. What should we make of P & Q?

Since P & Q is a conjunction of two core truths about X, it would seem to be itself a core truth about X. Since it has a bizarre conjunct, it would seem to be bizarre. (You don't make theories less bizarre by adding more claims to them.) And since it has a conjunct that we are not compelled to believe, we are not compelled to believe it either.

Eric Schwitzgebel said...

@ Rachel: Thanks for the thoughtful comment! Yes, on my view, as long as there is one core truth about X that is bizarre and which we are not compelled to believe, and as long as we are warranted in thinking that is the case, we are warranted in accepting crazyism. So bizarre-but-warranted P can be conjoined with bizarre-and-unwarranted Q to generate bizarre-and-unwarranted P&Q, assuming that conjunction works in the usual way such that p(P&Q) <= p(Q) and assuming (as I do for the purposes of this essay) that we can think of warrant probabilistically.

Anonymous said...

Eric @ Sun Nov 13, 11:04:00 AM PST:

Right - Determinism vs. Randomness isn't the core issue. And, ultimately, I don't think the issue is even really reductionism, as such - but rather the implications of lawfulness.

If we assume that our thought processes are the result of some lawful framework (e.g., the laws of physics), then that's pretty much game over for metaphysics, isn't it?

Because at that point, we have to admit that our beliefs are those that are "entailed" by the lawful framework, and not by any sort of "free" rationality.

If we are in "the Matrix", then all we can ever know is the Matrix. But we can't even freely investigate the Matrix - instead, the experiments we perform on the Matrix and the conclusions we draw from the results of those experiments are themselves entirely *products of* the Matrix.

Given this, it seems to me that some variant of Buddhist philosophy seems most on target. Yogacara (which you mention in your paper) or Dzogchen Buddhism maybe. Neither is exactly right I think, but in the ballpark.

Alternatively, if our thought processes aren't the product of a lawful framework - what is the alternative? Pure contingency I would think. Which doesn't put us on any firmer ground.

Though - even if we are the products of a lawful framework - why that lawful framework instead of some other? Wouldn't the existence of our particular lawful framework itself be an purely contingent fact?

So, I don't know - does this issue seem to have any bearing on Crazyism, or am I on a different topic altogether?

Eric Schwitzgebel said...

Anon Nov 21: I'm not sure why it would be so bad if our thoughts were governed by a lawful framework. My views on that question are pretty much mainstream compatibilist (Hume, Frankfurt, Watson, Fischer). Regarding the lawful framework itself, if it exists, I think it's worth drawing a distinction between the phenomenal surface (e.g., my images are laid out in two-dimensional space), the scientific study (e.g., what mainstream functionalist psychology does), and the metaphysical underneath (e.g., dualism-materialism-idealism). Generally speaking, I'm more skeptical about the 1st and the 3rd than about the 2nd.

Anonymous said...

I would think that we are stuck on the phenomenal surface...that's all we have access to, isn't it? Even conscious thought is just another aspect of experience - it's included as one of the six basic senses by Buddhism.

I'm more skeptical of materialism than of idealism because it seems to me introducing "matter" doesn't actually solve any problems. Ultimately, materialism is just dodging the question. If the "Physical World Hypothesis" is supposed to explain the orderliness and consistency of our conscious experience, then what explains the orderliness and consistency of the physical world?

It seems that with Materialism we didn't really answer the original question, we just rephrased it. The exact same question about order and consistency still remains, but now we have to ask it of the physical world instead of our experiences.

If the final answer is that the material world just exists, and there is no further metaphysical explanation of it - then why isn't that an equally acceptable answer for the world of experience? Conscious experience just exists, and there is no further metaphysical explanation of it.

As for "scientific study" - this is Hawking's fundamental is science possible when we aren't "free rational beings"?

As Conway and Kochen phrase it in their paper, The Free Will Theorem:

"It is hard to take science seriously in a universe that in fact controls all the choices experimenters think they make. Nature could be in an insidious conspiracy to 'confirm' laws by denying us the freedom to make the tests that would refute them. Physical induction, the primary tool of science, disappears if we are denied access to random samples."

I think these issues blunt a lot of the intuitive appeal of materialism over idealism.

So why are you more skeptical of materialism than of idealism? What is your reasoning there?

Eric Schwitzgebel said...

I am doubtful about one of the premises that seems to underlie your reasoning above: that we learn about the outside world of objects by means of some sort of more direct acquaintance with our stream of experience. In other words, I deny the epistemic priority of self-knowledge of consciousness. (I talk about this in my 2008 Phil Review paper and in my forthcoming paper "Introspection, What?".)

Anonymous said...

It's a good paper, something of a curate's egg that has sprung out from a variety of posts you've done here. I have some vague, somewhat incoherent objections to your main argument, the dualism-idealism-materialism trilemma. Basically, I think materialism is far better epistemically supported than the other two options, while you seem to view them all as equally live. It's something I need to think about more, and I will try to write (lazyness permitting) a proper response sometime in the future. But still, excellent stuff all round.

Btw, there's a typo in your abstract. You say that mental events are "casually inefficacious". Unless I'm mistaken it should be "causally inefficacious".It's been bugging me since you first posted this, but it's only now I've got round to pointing out. :p


Eric Schwitzgebel said...

Thanks for the comment, Juan. I'll be interested to see your response. Please feel free to send it. My own current credences are about 50% materialism, 33% compromise/rejection, and the remainder on idealism or substance dualism. It does seem to me that one might reasonably invest a greater than 50% credence in materialism, which part of is why I am careful to insist in Section IX on a somewhat finer slicing of the options. Another option would be to coarse slice the options to just the four materialism, dualism, idealism, rejection/compromise and then insist on a high bar for "epistemic compulsion". Even if materialism is the most likely option by some margin, the other options might still be live enough for the universal dubiety condition to be met.

Typo corrected. Thanks for catching that!