Saturday, May 31, 2008

Guest Post: Naturalism

[Guest post by Barry.]

Let me begin by saying that these remarks are extremely sketchy, just a recording of a recent whisky-fueled conversation. I have hitherto taken myself to be a naturalist, so these remarks are in the spirit of self-examination. Any help will be very warmly welcomed.

Suppose we take naturalism to be the research project whose goal is the reduction of everything to matter. Reduction is supervenience, say; or something tighter. Everything includes all the troublesome phenomena: modality, mind, normativity, maths. Matter is whatever the theoretical physicists tell us the world is made of (never mind that they consider this a deep mystery).

Now imagine some Fukuyama-esque article of the future, triumphantly announcing the end of metaphysics. In this article it will be demonstrated, compellingly, that every single recalcitrant phenomenon has been reduced. (I do not intend the term 'demonstrated' here to be factive.)

What then?

According to our putatively final metaphysics, the world is one big chunk of matter. I think this leaves us with two questions, which the naturalist lacks the resources to answer. (1) Why is the world made of this stuff, and not some other stuff? (2) Why is there this stuff and not nothing at all?

Our explanans must always be explanatorily broader than our explanandum. You cannot explain alpha with alpha; you must explain it with some beta that may include alpha. (The fact that the girl crossed the road does not explain the fact that the girl crossed the road, but it might explain the fact that the girl crossed some part of the road.) As the Aristotelians say, explanations must involve not just the 'that' but also some 'because.'

The worry is that naturalists can explain a lot in terms of matter, but they cannot explain matter in terms of matter.

Perhaps I am being childish; perhaps explanations do have to come to an end somewhere; perhaps the answer to 'why is there something rather than nothing?' is: 'there just is.' Or perhaps this isn't a meaningful question, or it is incoherent some other way. But it doesn't seem to be. It just seems hard.

And here's a possible diagnosis of the naturalist's error. The privileging of matter is just another manifestation of the human mind's craving for fixity - for an Archimedean point. And its perhaps just as misguided as the Cartesian response to this craving, or the religious.

So should we go back to old-school metaphysics? Or embrace some sort of aesthetic scepticism?

- Barry

[See also: Why does the universe exist? -- RC.]

25 comments:

  1. This is perhaps part of the reason that I think naturalism is better understood in terms of the old unity of science picture than in terms of reduction to what's fundamental. The fundamental is surely chimerical. There's no way to step outside the picture and explain and justify everything; there is no outside, and if there were, you'd have to explain and justify that. But it's only from such an outside perspective that anything could be identified as ultimately fundamental.

    The best you can do is to maximize the internal coherence of everything, to tie everything to everything else. That's what reduction should aim at. And the questions of "why is there everything and not something else?" and "why is there everything rather than nothing?" do rather look like pointless pseudo-questions to me.

    ReplyDelete
  2. And the questions of "why is there everything and not something else?" and "why is there everything rather than nothing?" do rather look like pointless pseudo-questions to me.

    Pointless? Pseudo-questions? These are to me the most sublime questions one can ever ask. As J. J. C. Smart once remarked, "That anything should exist at all does seem to me a matter for the deepest awe." I find it incredibly puzzling that other people see things so differently—that far from thinking it a "matter for the deepest awe", they don't consider the questions to be at all awesome.

    ReplyDelete
  3. As a non-materialist naturalist myself, I don’t think we should understand naturalism as being defined by a particular ontology; I would instead propose that this research program is at its core committed to a certain methodology—the methodology of the natural sciences. The link between naturalism and materialism is rather an a posteriori consequence of the fact that the world uncovered by this methodological approach appears to be ultimately made of matter. Yet if it turns out that certain recalcitrant phenomena cannot find a place in this picture of reality, then those same methodological commitments call for an expansion of our ontology beyond materialism. This holds not only for the persistent offenders you list—consciousness, mathematics, normativity and modality—, but also for the ultimate existential questions that you pose.

    I would say, however, that one of the possible answers to those questions is that we are simply cognitively unequipped to answer them satisfactorily. There is an answer to why this and why anything exists, but it is an answer that is not accessible to creatures like us. And this hypothesis in fact makes perfect sense precisely on a naturalistic picture of reality. Human beings are nothing special in the grand scheme of things: they evolved in an undistinguished planet on a petty galaxy by a blind process of natural selection, and there is no reason whatever to expect such creatures to have the cognitive capacities required to answer every question to which there is a valid answer. The problem for naturalism, it seems to me, is rather that it appears to imply that we should know much less than we actually do. It is ironically not so much that naturalism has failed to answer a few questions, but instead that it appears to have succeeded in answering so many, that may provide for some critics a reason to question the validity of this methodology.

    ReplyDelete
  4. In Nozick's Philosophical Explanations (maybe it's Explorations, I don't recall), he says something like: "What did philosophers expect? Of course if they keep investigating things, explaining things away, eventually they're going to reach something that can't be explained, an intractable problem: why is there something rather than nothing?"

    I see three possibilities:

    1. The series of explanations is infinite. Every explanandum has its explanans, forever.

    2. There's some starting point, some brute fact of the matter. The world just is, full stop (some sort of recursive explanation might fit here).

    3. It's circular: A is explained by B, which is explained by C, which is explained... which is explained by A.

    I find all of these profoundly unsatisfying, but I know of no other options. And this dilemma shows up in various areas. Think of justification of beliefs: do we have to 1. justify every belieft with some other belief, ad infinitum? Or 2. take some beliefs as brute facts? Or 3. are beliefs justified in a circular fashion?

    Or, think of the creation of the Universe. Was 1. God created by meta-God, who in turn was created by meta-meta-God, ad infinitum? Or is 2. God a primitive? Or 3. did God create us so we can later create him?

    This doesn't much touch on your post, Barry, but it's an interesting, related issue.

    ReplyDelete
  5. Ooh! I recently gave a talk on this topic!

    I'm going to agree with Pablo. You cannot start with ontology first. You'll run into Hempel's dilemma, for starters. Natural means "predictive" (not "physical"), and thats why methodological naturalism works. Physics is natural because it is predictive.

    As to the question of certain ultimate explanations, we must first ask ourselves what constitutes a proper explanation. To cut a long story short... Once we eliminate mere restatements of the explanandum and mere references to explanations we do not yet have, we'll find that every proper explanation is predictive (and, therefore, natural). As Barry says, "our explanans must always be explanatorily broader than our explanandum." In other words, the term 'supernatural' is synonymous with 'unpredictable' and 'inexplicable'.

    Now imagine that I have a set, S, containing all of the laws of the universe. I cannot have another predictive law to explain S without violating my original premise that S is complete.

    Consequently, I cannot explain why there is something rather than nothing because I would need a predictive meta-law (something) to predict the something. I would have to ask why the new something (the meta-law) exists (rather than not existing), and so on, ad infinitum (Scott's point, I think).

    Hence, the existence of the universe (or multiverse, or whatever) is inexplicable. If the universe were an effect (which is a very dubious classification), it would have to be a supernatural (i.e., inexplicable) effect.

    I believe that axioms of rationality also fall into the category of rationally inexplicable facts. We can explain why so-and-so piece of matter acts rationally, but we cannot rationally explain why the rules of rationality have to be what they are (not without circularity).

    To sum up, naturalism is the only option if you want to explain stuff. There may be supernatural effects, but there are no supernatural explanations.

    ReplyDelete
  6. Consequently, I cannot explain why there is something rather than nothing because I would need a predictive meta-law (something) to predict the something. I would have to ask why the new something (the meta-law) exists (rather than not existing), and so on, ad infinitum (Scott's point, I think).

    To be clear, my point was that an infinite series of explanans and explanandum is a metaphysical possibility, one of three unpalatable options. While I find it uncomfortable, there is nothing logically wrong with an infinite series of explanations that I can see. Here you and I apparently differ, as you think an infinite need for meta-laws is unacceptable.

    ReplyDelete
  7. Scott,

    I'm fine with there being deeper and deeper layers of reduction and explanation. However, to explain "why is there something rather than nothing" is to explain why the regression ends (if it does) or why the infinite regression itself exists. I do not think either thing can be explained.

    We can explain cheese in terms of chemistry, chemistry in terms of atoms, atoms in terms of nucleons and electrons, etc. If this explanatory chain is infinite, then that's fine by me. However, I don't think that pointing to the existence of such an infinite chain would constitute an answer the question. IOW, "why is there an infinite chain of things explaining other things?" seems like it fits the parameters of the original question.

    So, for some questions, I think your option 2 is the only conclusion.

    ReplyDelete
  8. Doctor,

    I agree that '"why is there an infinite chain of things explaining other things?" seems like it fits the parameters of the original question.'

    But I disagree that that necessitates taking the chain itself as brute fact. Our original options simply recur again here.

    ReplyDelete
  9. Here are some comments I posted over at Scott's blog:

    Whether or not we find [Scott's] three options fully satisfying, the fact remains that this is not a problem for naturalism; this is a problem for any kind of metaphysics. The important point is recognizing this so that anti-naturalists can stop bashing naturalists over the head with a problem that effects everyone, including themselves.

    It's perfectly consistent, and indeed, necessary, for a naturalist to admit that there are certain kinds of metaphysical questions that may forever remain unanswered. It would be presumptuous to think otherwise.

    So long as naturalism acknowledges our inability to answer these ultimate questions, it is not vulnerable to this criticism. A naturalist need not--and cannot--commit to the proposition that the world is all there is, because there is no way scientifically valid way to know this. All naturalism commits to is the proposition that the world is all that we know to exist. If we later discover entities or substances outside our existing metaphysics, this will be incorporated into future versions of our physics.

    It is the anti-naturalist position that is problematic: simply positing the possible but unconfirmed, undetectable, indescribable existence of metaphysical stuff is not a serious position grounded in reason. Naturalism allows for the existence of the unknown, but once we discover the unknown, it becomes known. Whereas anti-naturalism embraces the contradiction that the unknown can be known in advance.

    To acknowledge the existence of unanswered questions is not to acknowledge the existence of the unknown, for there may ultimately be no answer to these questions.

    ReplyDelete
  10. Dear Micha,

    I'm afraid I disagree. These hard questions may indeed have answers. Most questions do. We are agreed that naturalism cannot provide these answers. So the naturalist must either ignore these questions or defend a disjunctive methodology - one method for putatively reduction-apt phenomena, another for non-reduction-apt phenomena. Neither response is devastating. However supposing we do not yet abandon hope of answering these questions, one is (naturally) inclined to ask whether some more unified philosophical methodology might be available.

    ReplyDelete
  11. Most questions have answers? How are you counting questions and answers, so as to come up with this fascinating statistical fact? And are there no distinctions among questions, such that some of them might be in categories which are more or less likely to have answers?

    Someone might, for example, suggest that most questions presuppose that we are in a certain broad part of possibility space and ask for help in narrowing that space down. Perhaps all questions of that kind have answers. It is not clear that the questions presently under discussion fit the pattern, though, so any tendency of questions which fit the pattern to have answers would seem to be irrelevant.

    ReplyDelete
  12. So the naturalist must either ignore these questions or defend a disjunctive methodology - one method for putatively reduction-apt phenomena, another for non-reduction-apt phenomena.

    This is, I think, a false dilemma. As I indicated in my comment above, the naturalist can recognize the validity and importance of these questions but claim that we lack the cognitive capacities necessary to answer them. As I also indicated, this claim could be supported by facts known using standard scientific methodology. Given our knowledge of physical cosmology and evolutionary biology, it would be surprising if it turned out that we can know every important question for which there is a true answer.

    ReplyDelete
  13. Barry,

    I don't understand your grounds for disagreement with me. You claim that "These hard questions may indeed have answers." Indeed, I agree. As I wrote in my final sentence in my previous post, "To acknowledge the existence of unanswered questions is not to acknowledge the existence of the unknown, for there may ultimately be no answer to these questions."

    I am not claiming that these questions are ultimately unanswerable. I am claiming that at the present moment we do not have an answer for them, and it is possible that we will never find an answer. So too, it is also possible that we will one day find an answer to them. And insofar as the presence of answer to these questions is incompatible with naturalism, then once we do discover that an answer exists, that will be a blow against naturalism. But we have not yet discovered the answers, only the questions, so how can we call this a strike against naturalism? It seems our status quo position should favor what we know, not what we don't yet (and may never) know.

    ReplyDelete
  14. Barry, Micha,

    I think we can go further, and say that the question is definitively unanswerable.

    Suppose that T is the theory that predicts something rather than nothing. Then we will be in a position in which we have to ask "why T rather than nothing?"

    In other words, in the original question, "something" is so broad a placeholder as to incorporate any theory that would answer the question.

    How about this: why is there an answer to the question 'why is there something rather than nothing?'? That answer is, after all, something.

    ReplyDelete
  15. doctor logic,

    Suppose that T is the theory that predicts something rather than nothing. Then we will be in a position in which we have to ask "why T rather than nothing?"

    Not necessarily. If the universe exists because God created it, this would answer why the universe exists. And the question why God exists may also be answered by that fact if, as some people believe, the existence of God is metaphysically necessary. The fact that something is metaphysically necessary is an answer to the question of why this thing exists which does not itself raise any further existential questions.

    How about this: why is there an answer to the question 'why is there something rather than nothing?'? That answer is, after all, something.

    This answer, though true, does not itself exist. If its truth raises any questions, they are questions that differ from those raised by the existence of something.

    ReplyDelete
  16. Pablo, on most versions of modal logic it is necessary that there is something rather than nothing. It is metaphysically necessary for David Lewis, for instance, and so for the handful of modal realists who follow him, and also for the vast hordes of philosophers who think modality is just like he says except for the realism part. I believe it's L-true that there's something in Carnap's system, though I admit I haven't checked. And there are many other approaches which agree with Carnap and Lewis on this point. Thus, I think it is much more plausible that it is metaphysically necessary that there be something than that God is metaphysically necessary (a bizarre claim which becomes ever less coherent as it is more closely examined). If you're happy with "it's metaphysically necessary" as an explanation, it seems the best course is to accept that as the explanation of why there's something rather than nothing, rather than deriving the existence of something from some other, much more dubious alleged metaphysical necessity.

    ReplyDelete
  17. Pablo,

    And the question why God exists may also be answered by that fact if, as some people believe, the existence of God is metaphysically necessary.

    Necessary for what?

    If its truth raises any questions, they are questions that differ from those raised by the existence of something.

    Hmmm. Are you saying that the "something" in the original question does not apply to things that are not made of matter? For example, does the question not apply to mathematical relationships or physical laws? What do you think the "something" does not refer to?

    ReplyDelete
  18. Thanks Micha,

    I don't really think we disagree then. We agree that if there turns out to be an answer then naturalism is an incomplete philcal methodology. That might not be a reason to avoid it now or ever. My motivation for presenting the issue in this way was simply to undermine some of the headier philosophical aspirations for naturalism, aspirations I myself once held. I used to think that the world at the end of the garden was comprehensible, and that philosophy was just a matter of fitting consciousness, normativity etc. into that picture. The current point is that this may not suffice. (I don't claim any novelty here.)

    On your final point: 'It seems our status quo position should favor what we know, not what we don't yet (and may never) know.'
    I'm not sure whether I quite agree. Prima facie I think we should accept methodology that is most likely to reveal to us what there is or may be, not what we currently know.

    Pablo, thanks for your comments. Sorry not to have responded to them directly - I am on holiday. I agree with most of what you say. However, on your last point: Suppose that naturalism did indeed manage to explain the human shortcomings in virtue of which we are foreclosed from answering The Question. Wouldn't this be a naturalistic answer to the question whether we can answer The Question, not a naturalistic answer to The Question itself? If so, I think this result may be best filed under 'scepticism', or perhaps 'aesthetic scepticism'.

    ReplyDelete
  19. Aaron,

    n most versions of modal logic it is necessary that there is something rather than nothing.

    Are you saying that on most systems of modal logic it is metaphysically necessary that some concrete object exists? If so, then to the extent that we have reasons to believe in one of these systems, we have reasons to believe the the question 'Why does anything exist?' is answered in a way that does not raise further questions.

    Thus, I think it is much more plausible that it is metaphysically necessary that there be something than that God is metaphysically necessary (a bizarre claim which becomes ever less coherent as it is more closely examined).

    What's incoherent about the claim that something exists in all metaphysically possible worlds?

    If you're happy with "it's metaphysically necessary" as an explanation, it seems the best course is to accept that as the explanation of why there's something rather than nothing, rather than deriving the existence of something from some other, much more dubious alleged metaphysical necessity.

    Since I'm not myself a theist, I don't think that the question 'Why does anything exist?' is answered by invoking God. But I do think that, if God existed, this fact would answer that question.

    As I said, to the extent that we have reasons to believe in a form of modal logic on which it is metaphysically necessary that some concrete object exists, then to that extent we have reasons to regard the question as being decisively answered.

    Notice, incidentally, that we are dealing here with only one of the two fundamental questions that Barry raised. The other question, 'Why does this world exist rather than some other world?' requires a different answer. (Although some answers to this second question are also answers to the first.)

    ReplyDelete
  20. Micha,

    Necessary for what?

    I'm not talking about any purposive necessity here, but rather as this concept is interpreted in possible world semantics.

    What do you think the "something" does not refer to?

    I think that, in the question that Barry raised, 'something' refers to concrete things. (Above I talked about concrete objects, but since objects are generally understood to have determinate identity conditions, I think that's too restrictive a way of putting it.)

    ReplyDelete
  21. Suppose that naturalism did indeed manage to explain the human shortcomings in virtue of which we are foreclosed from answering The Question. Wouldn't this be a naturalistic answer to the question whether we can answer The Question, not a naturalistic answer to The Question itself?

    Yes, I agree with you.

    If so, I think this result may be best filed under 'scepticism', or perhaps 'aesthetic scepticism'.

    Maybe, but the point is that the naturalist is not required either to ignore the questions or to embrace a disjunctive methodology. We can take these questions seriously, accept that they might have answers, claim that if they do have answers we probably can't know them, and explain why, on a naturalistic worldview, this claim is likely to be true.

    ReplyDelete
  22. I'm not sure what you mean by "concrete" (I suppose it isn't metaphysically necessary on any but hyperdeterminist theories that there are hard aggregates of cement and sand, but I'm guessing that isn't what you mean). Lewis's principle of plenitude says there's a world for every combination and arrangement of things. Nothing is not a combination or arrangement of things. Same for Armstrong, now that I think of it, though he is one of those who drops the realism from the Lewis view. Most forms of first order logic make it a theorem that there's at least one thing (though some people worry that they shouldn't); it is for this reason that I am pretty confident the existence of something is L-true for Carnap, since all theorems of first order logic are. Which, if any, of these would have concrete things as opposed to things of some non-concrete kind would depend on what you mean by concrete, of course.

    ReplyDelete
  23. "Since I'm not myself a theist, I don't think that the question 'Why does anything exist?' is answered by invoking God. But I do think that, if God existed, this fact would answer that question."

    Based on what you've said here you are what I refer to as a shallow atheist.

    A deep atheist understands that even if god existed, that fact would not explain anything.

    Firstly, god existing wouldn't mean the same thing as my car existing. As imagined god is not an object that exists within the material world. So even if god were somehow was metaphysically neccesary his mere existence would not establish the neccesity of existants within universes.

    He could have decided not to create a universe after all. In which case the normal kind of existence whouldn't have occurred.

    So the following sentence by Pablo is false: "If the universe exists because God created it, this would answer why the universe exists."
    No it wouldn't answer why the universe exists. It would answer what created the universe but not why.

    Furthermore, the concept of god as concieved and used as an "answer" for these questions is totally unsatisfying. It's unsatisifying because it doesn't preclude anything.

    No matter what state of affairs that existed in the universe one could as why and then get the answer "Because of God". Hell if the universe didn't exist at all then the answer to "why" is "God did it".

    God fails as an explaination at every level. Theists recognize this and have actually prebuilt into their religions the concept of not being able to understand gods mind. The answer to why is ultimately "God works in mysterious ways".

    BTW, I've always been of the position that naturalism is a methodological philosophy that recognizes our limitations and tries to overcome them. Materialism is merely a side effect of good methodology. Furthermore it is a tentative belief, like all other beliefs.

    ReplyDelete
  24. So the following sentence by Pablo is false: “If the universe exists because God created it, this would answer why the universe exists.”
    No it wouldn’t answer why the universe exist. It would answer what created the universe but not why.


    As I read what I wrote, I realize I should have introduced some qualifications. I would now state my views as follows. Suppose that someone asked: ‘Why does the universe exist?’ And suppose someone else could rationally persuade this person that some supernatural being exists, that this being exists necessarily, and that this being is the sole creator of the universe. I believe that this reply (a) would provide at least part of the explanation of why the universe exists and (b) would not raise a further question about why this being exists. I grant, however, that it may not provide a full explanation. Moreover, it’s not clear to me whether this being deserves to be called ‘God’ or not.

    It of course goes without saying that it is what Kenny called “the God of the philosophers” that I’m talking about here. Zeus, Apollo, Yahweh, or Allah are mythical beings belief in which is rationally indefensible. My own view is that a person who believed in the God of the philosophers but did not believe in any of the particular deities revered by the established religions would be much closer to the atheist than she would be to the religious believer. The ontological commitments incurred by moving from atheism to philosophical theism are modest in comparison to the commitments incurred by moving from philosophical theism to, say, Christianity or Islam.

    ReplyDelete
  25. Taking ... "And suppose someone else could rationally persuade this person that some supernatural being exists, that this being exists necessarily, and that this being is the sole creator of the universe."

    ... and ...

    "It of course goes without saying that it is what Kenny called “the God of the philosophers” that I’m talking about here."

    Not sure that it goes without saying. Who says a god of the philosophers would be supernatural or even a being?

    I'm about as strong as a "strong atheist" gets without faultering on philosophic fallacies. Yet, I could concieve of a god as a logos.

    It might just be that the theists are presuming something "behind the works" based on effects. Thus, seeing the order in biology they think there is a "designer" and seeing that the universe is orderly to some extent they presume a "creator". These may or may not be one in the same but lets assume they are.

    Certainly we have identified the "designer" behind biological organization. It's a process called natural selection.

    The religious were certainly right in that "there was something more" but they jumped to conclusions that were totally wrong. Sort of like believing the dangers of the volcano are due to some malevolent diety.

    Now I can see how certain rules just have to be the way they are and have certain outcomes. As some have said "God is logics bitch."

    But if that is the case then it is perhaps the cold neccessity of logic and her relationships that is in fact that underlying creator that is being misidentified.

    Perhaps god is in fact logic. And perhaps evolutionary logic is merely a subset of "god" as a whole.


    I would refer to this kind of understanding as a belief in a Logos. It might in fact be that we don't "exist" at all other than as possible logical relationships among concepts, perhaps numerical ones.

    Our whole of existence might in fact be encapsulated by a single gigantic number. One of many possible existences that are specified by other large numbers.

    In such a view all things exist which are possible to logically possible to exist. Logos restricts what can exist more than it creates what does. We exist as potentialities and that is all that existence is. There is no "reality" to the universe at all, per se. The realities are only real in terms of the internal relationships, ala Minsky.

    Other worlds that we will never have the possibility to explore, that in fact may contradict our own where they connected, have as much reality but only from their own perspectives. The are not impermissible by Logos and therefore they are as much real as we are from the outer view of Logos.

    We however do not use the word "exist" in this way so therefore from our internal definition those worlds do not in fact exist. Nor the relationships within them. From our perspective those relationships are mere potentialities, while ours are real, and the vice versa from the perspective of other worlds (realities).

    We might one day be able to figure out enough to discover that we are a merely and purely a mathematical potentiality, at which point we would in fact be able to deduce the existence of these other reality, and also know the nature of the correct formulation of the question "why we are here".

    We would have a "god" of sorts but not the kind one would worship. Praying would not alter Logo's choices as those are fixed.

    One one would however "obey" the rules of Logos if one valued ones existence. Stepping off cliffs tends to cause consequences that Logos enforces in a cold way.

    ReplyDelete

Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)