ogingat:

nostalgebraist:

urpriest:

nostalgebraist:

[snip]

I feel like you could capture this entire notion by saying you have an analytic temperament, rather than a positivist one, though.

I guess I find “analytic” too broad – it refers to a whole tradition with many aspects, not all of which I like, and some of which I strongly dislike.  “Positivist,” even if it’s not quite technically right, gives a much sharper idea of what I am trying to get at…

I am starting to become very confident that when you say positivism, you mean naturalism.

I’m aware of that term, but I’ve never been sure whether it fits the bill.  Anyway, I can try to say a few things, and you can decide for yourself.

(Under a cut because it’s very long)

I think my opinions are very similar to jadagul‘s.  I think that verifiable claims have a special place, not because they are the only useful ones, but because they are the only ones which are not modeling choices.  When we think about things, we are playing around with models that correspond in some way to the observed facts.  I don’t think there is a real “fact of the matter” about whether the terms in one model “exist” while those in another don’t, or whether one model is a “correct account” in a way another isn’t, except insofar as one model might actually contradict verified claims.

For instance, mereology looks for a correct account of objects and their relationships to their parts.  I don’t think this a question with an answer; I think that we have concepts like “object” and “part/whole” because these are terms in a modeling system we find useful as a simplification of the kinds of environments we tend to be in.  Being a simplification, this system runs into problems with edge cases and so forth, but this does not mean that we should be able to find the “correct” account of objects and parts that handles all these cases well.  It may just be that the model ceases to be useful in some cases.  (Fluid dynamics breaks down at small scales, but that doesn’t mean we should look for “the true fluid dynamics” which doesn’t.  It just means we should stop using fluid dynamics below a certain scale.)

So, statements about (say) objects are not meaningless, but questions like “what is the correct account of objects?” are meaningless, in a sense.  The disagreements here are not about real things.  There is no “fact of the matter” that could resolve them.

So this is one way in which I feel somewhat distant from some of the kinds of philosophy called “naturalism,” even if I may be a naturalist.  I do not think there is a correct account of non-verifiable or non-scientific things which just so happens to say they supervene on verifiable or scientific concepts.  I think that the idea of “correct accounts” disappears when we step away from direct observation, and we’re left simply with models which may be more or less useful.  (One might say, “there are reasons for making statements which positivism might not allow, but there is no ultimate fact-of-the-matter about whether these statements are true or false, merely contexts in which we may find it useful to make them.”)

Another (related) way I feel uncomfortable with aspects of naturalistic philosophy is that I do not feel especially interested in what is most fundamental.  That is, say, I believe that fluid dynamics “supervenes on” particle physics and “can be reduced to” it, but I don’t think this is especially interesting, among the various modeling statements one can make.

Let me explain a bit of what this entails.  Let’s ask, “do we live in a monistic world, or a dualistic one?”  Are there two sorts of thing, or just one?  One sort of dualism is, of course, mind-body dualism.  And many people would say, “mind-body dualism has now been refuted, because we know everything can be reduced to physics.”

But let’s think about what the alternative could have looked like.  Suppose we had discovered that in addition to physical matter there was a mind-substance, which behaved in different ways, not reducible to physics.  Then we might have studied this substance, and found a kind of “physics” for it, or at least for its statistical properties.  After all, human thought is not somehow transcendently irregular – it has various tendencies which can be described.  Now, how would this picture have looked different?  Well, surely these new laws would have been different from the ones we have now, being laws of mind and not laws of matter, right?

But consider: if you step into a mathematical neuroscience classroom in our world, you will not find the laws of fundamental physics written down on the blackboard.  You will instead find things like the Hodgkin-Huxley equations, which describe neuronal action potentials.  And if you saw these equations presented abstractly, with no context, it would not be obvious that they necessarily described some sort of electrical conductor.  They would just be equations, ones quite different from those of fundamental physics.  They might as well be irreducible “laws of mind.”  (One cannot get anywhere in neuroscience by talking in the terms of fundamental physics, and so our mathematical neuroscience class will be conducted in terms that are pretty much indistinguishable from those used in a world in which Hodgkin-Huxley, or whatever, were the irreducible laws of mind.)

Now you might say, okay, but physics constrained what neuroscience could look like.  In principle, this is true – but the range of things that can be implemented in physics is vast.  Nature has build proteins and cells which do many things that, if described as abstract systems, look like worlds of their own and not obvious consequences of fundamental physics.  Indeed, one can argue that anything we understand in a precise way can be implemented in one way or another on a Turing machine, and thus that any conceivable fundamental laws (such as those from a mind-body dualistic world) can be implemented in our physics.

Whether this matters or not depends on how much importance you assign to “fundamental-ness.”  The more I learn about science, the less I feel like this is an important concept.  Very little good thinking gets done about macroscopic things by actually using the terms of fundamental physics.  The in-principle reducibility is always there in the background, but nonetheless, a fluid dynamicist can do a life’s work of research that would apply just as well if fluid motion were a fundamental, irreducible property, and neuroscientists can do the same.  We live as if we are in a world that contains many different sorts of things, not just one.

Here is another way to play with the idea of dualism.  Consider that we have two pictures of fundamental physics – general relativity and quantum mechanics – which both appear to hold true in their own domains, but are described in different terms which seem to be irreconcilable.  Isn’t this about as dualistic as you can be – just as dualistic as a world with separate mind and body substances?  But we don’t tend to think in these terms, either because we don’t tend to think of gravitational fields as a kind of “stuff” (is there a reason we can’t?), or because both of these form a part of “fundamental physics.”  But then, in any properly investigated dualistic world, both sorts of substances would equally be part of “fundamental physics.”  Whatever we find, we describe it, and then call it “physics”; if there were a mind-substance, we would have done the same, and then philosophers might still ask whether there are things besides “physical things” (meaning: the matter substance and the mind substance).

For this reason, talk of physical vs. non-physical things feels very off to me.  It’s not that I think something called “physicalism” (say) has turned out to be true, but that I’m not sure what the alternative is.  We say things like “we don’t understand consciousness, so perhaps it is non-physical.”  But if we did understand consciousness, it would form part of our overall world-picture, which we would then call “the physical world.”  The case of gravity vs. the other forces shows that we will see a unified “physical world” even if what we have is two irreconcilable pictures; so long as they are precise enough, they seem “physical.”  Once we understand anything else sufficiently well – even if integrating it with the rest of physics is problematic – we will see it too as “physical.”

This makes the distinction between, say, Chalmers’ views (in The Conscious Mind) and a conventional “materialist” view seem kind of bizarre to me.  Chalmers says: the laws of physics don’t logically imply consciousness, so consciousness must be something else above and beyond them.  Which is fine, but in what way is this not just more “physics”?  We have the observations (self-reports about qualia), we learn how to relate them to physical states, and we have learned something new, which was not logically necessary but is nonetheless true.  This is not unphysical, any more than gravity is unphysical because the rest of physics does not logically imply it.  There are observations; we make models that usefully organize and predict those observations; that is all there is.  A conventional view might be something like “we used to think that the mind was non-physical, but now people think they’ve reduced it to physics; but we don’t understand consciousness, so maybe it is non-physical after all.“  I don’t think “non-physical” is a sensible concept; I think that qualia are a unique sort of empirical fact, and so of course they will need a separate model of their own, but that this is not a very interesting difference.  (Fundamental-ness is not very interesting.  The world where mind-stuff or fluid-stuff is fundamental looks very similar to ours.)

What have I tried to say here?  Well, in part, I just wanted to throw some of my opinions on the table, since we had been talking vaguely about my opinions without me describing what they were.  Another is to suggest why I feel ill at ease with the modern (analytic) philosophers I have encountered, even when they are naturalists.  You mention Humean Supervenience; I have a hard time seeing quite what it would mean for this idea to be correct, or why I would care.  At most I feel like it may be a property of some useful models, but that even in that case it would not be an especially significant feature of these models, relative to their various other features.  Elsewhere, David Lewis is (in)famous for arguing that all possible worlds exist.  Again, I am not sure what it could even mean to be “right” about this.  At most, one could say that this picture is useful for thinking about certain aspects of reality.  But as with all models, once it goes beyond what is implied by observable facts, there ceases to be a fact-of-the-matter that could decide between it and other models compatible with the same facts.  We could say “hmm, this sure is a useful way to think,” but not “modal realism is true!” or “modal realism is false!”

Maybe there are philosophers who think like this, and I have not found them yet.  jadagul suggests Rorty.  I find some things to sympatheize with in (the later) Wittgenstein, and (yes) in some parts of positivism.  But I don’t want a label that suggests that David Lewis and I are in any way interested in the same project.

(via ogingat-deactivated20150801)