Monday, July 07, 2008

Against Metaphysics, Especially the Metaphysics of Consciousness

I worry that metaphysics is a sham -- or at least, that it's a sham if it's thought of as an a priori discipline whereby one discovers a special class of truths while reflecting from an armchair. When you tilt back in your armchair and reflect, there's only one kind of thing you can discover, it seems to me: Facts about your own psychology. What gets called metaphysics, then, is really just a certain kind of self-study -- typically the study of the metaphysician's own concepts.

You can learn about the world by learning about your concepts, if your concepts contain in them important information about the world. Often they do contain such information: My concept of a bird as a type of biological organism, warm-blooded, bipedal, winged, feathered, egg-laying, etc. contains packed in it information about a certain cluster of traits that tend to travel together. Because of this, reflecting on my concept of "bird" can bring forward facts I may not have explicitly considered in the past.

How do my concepts come to contain information about the world? They must do so by contact with the world -- either my own contact (directly by empirical observation or indirectly by hearing the reports of others) or my ancestors' contact if the concepts are innate. I don't see, then, how studying my own concepts could yield a different kind of information than studying the world directly; nor do I see how our concepts could provide an independent source of information about the world orthogonal to empirical observation or immune to empirical refutation. (For a parody of the idea that the truths suggested by conceptual reflection are immune to empirical refutation, see here.)

The weird, science fiction cases that philosophers tend to dwell on in metaphysical discussions are exactly the kinds of cases where we should expect our concepts to be least in touch with the world, aren't they? Consider disputes in the metaphysics of consciousness: Would a silicon robot that behaves just like a human being be genuinely conscious, or would it have no more real consciousness than the computer on which I'm writing this post? Is it "metaphysically possible" (even if in practice unlikely) that my conscious experience is radically different from yours (e.g., red-green inverted or worse) despite all the similarities in our behavior?

If we construe these questions as questions about our concepts or pre-existing ideas, then it's not unreasonable to think we can make progress on them from the philosopher's armchair (though other methods for studying our concepts may be equally or more illuminating, in the spirit of recent "experimental philosophy"). Some people apparently find it impossible to conceive that a robot that behaved like a human being wouldn't be conscious; others apparently find it impossible to conceive that such a robot would have human consciousness. That shows something about their concepts or background assumptions. But how could it be (as Searle and Block and Putnam and Lewis and many others seem to think) that armchair reflection could reveal whether robots really would be conscious? Our concept of "bird" works well for near-home cases but tells us nothing about life on other planets; so also our concept of "consciousness" works well enough for distinguishing waking from dreamless sleep, mundane red experiences from mundane yellow experiences, but how could it cast useful light on robots or inverted spectra?

Metaphysicians often respond to such concerns by pointing to mathematics: In math, it seems, we discover substantive facts about the universe from the armchair, so why not also in philosophy? But is it clear that in studying math we do discover substantive facts about the universe? Not every philosophy of math grants this assumption. Maybe what we do, in studying math, is simply invent and apply rules for symbol manipulation. Maybe we discover facts about the structure of our concepts and invent new concepts. So irrefutable seems to me the view (empirically grounded!) that from the armchair we can discover nothing beyond the circuit of our own minds, that a conservative philosophy of math is mandatory. I find that considering alternative rules of logic (e.g., intuitionist logic or dialethism) and alternative rules of arithmetic (e.g. Boolean algebra) helps me feel the pull of the idea that mathematics is more an invention than a discovery of mind-independent facts.


  1. I find that considering alternative rules of logic (e.g., intuitionist logic or dialethism) and alternative rules of arithmetic (e.g. Boolean algebra) helps me feel the pull of the idea that mathematics is more an invention than a discovery of mind-independent facts.

    While this is a common way of putting the matter, it has always seemed to me to be a pretty clear case of a false dichotomy: there is no reason why it cannot be both, just as there is no reason why advanced theories in physics can't be both inventions and discoveries of mind-independent facts. Indeed, any genuine invention is a discovery of a mind-independent fact, if only the mind-independent fact that the thing invented is possible for the medium or application for which it was invented.

    I think, incidentally, that a more promising course of argument on this point is not the one you give in your last paragraph, but to note that, a priori discipline or not, mathematics is not really an armchair discipline except in the sense that any intellectual discipline is an armchair discipline. There are (among other things) applications (whether intended or unexpected), interactions with physicists and other natural scientists, and community collaboration of a highly integrated sort that puts it well outside armchair reasoning.

  2. there's only one kind of thing you can discover, it seems to me: Facts about your own psychology.

    Why not go further? It seems to me the only kind of thing we can discover in the comfort of our armchairs is a simple play, put on by the myriad players collected from our long evolutionary history -- both biological and cultural.

  3. "the pull of the idea that mathematics is more an invention than a discovery of mind-independent facts."

    How do you explain just how well mathematics works in creating models of the world, models that let us make such accurate predictions about physical phenomena? And the fact that armchair mathematics has, in so many cases, found unanticipated application in creating these models?

    I don't see a bigger question in philosophy than that, if you know the answer, please let me know. ;-)

  4. How do my concepts come to contain information about the world? They must do so by contact with the world -- either my own contact ... or my ancestors' contact if the concepts are innate.

    I see an interesting, albeit tacit, suggestion regarding the bearing of biology on certain epistemic and meta-philosophical concerns.

    If our concepts are understood on the model of prototypes, we have every reason to expect that they--like the white fur of polar bears--encode information about our environment. But what information? Not metaphysical truths, but more earthy information about the behavior of ordinary, macrophysical objects. And perhaps even a theory of mind.

    Why? Metaphysical (and microphysical?) acumen would appear to have little bearing on our fitness. At least on the face of things evolution seems to imply that we should be antirealists about certain categories of concepts.

  5. When you ask, "is it clear that in studying math we do discover substantive facts about the universe?", I think the word "substantive" is misleading. Nobody thinks that mathematical facts are contingent, so there is at least some sense in which we don't gain new "substantive" knowledge when we learn mathematical facts. From the discussion you go on to make, I assume that by "substantive" you mean something like "objective". Lots of people, though not all people, do think that mathematical facts are objective.

    I wonder how far you're willing to let this go, though. I think that everything on this list is both objective and knowable from the armchair:

    Four is greater than one.
    Triangles have 180 internal degrees.
    No square has five sides.
    All brothers are siblings.
    All penguins are penguins.
    If all penguins are happy, then no penguins are unhappy.

    Do you really want to say that none of these are objective facts?

  6. "What gets called metaphysics, then, is really just a certain kind of self-study -- typically the study of the metaphysician's own concepts."

    What is worse is the anti-naturalistic spirit of so much contemporary analytic philosophy where, even if coherent and highly truth-conducive, if a philosophical theory is unintuitive, it gets abandoned. The summum bonum of any metaphysics is one that is both truthful and intuitive. But, like you, I really fear that the latter requirement of it all really stagnates philosophy in general and is one of the main reasons for the perception that "philosophy doesn't advance." Reality is not commonsensical or intuitive; we live in what Dawkins calls "Middle World"--a Newtonian, macro-world where we ramshackle organisms evolved over millions of years in trees and savannas. Valuing the intuitiveness of a philosophical theory too much reduces the whole enterprise of philosophy to a goddamn battle of mere prejudices.

    I find that considering alternative rules of logic (e.g., intuitionist logic or dialethism) and alternative rules of arithmetic (e.g. Boolean algebra) helps me feel the pull of the idea that mathematics is more an invention than a discovery of mind-independent facts.."

    The Platonist in me wants to cry blasphamy, but I think you're on to something--something many nominalists also find intuitive (har har). Don't forget, that bastion of rationalism and "a priori" reasoning was falsified empirically by Fourier and other non-Euclidian mathematicians. Take that, mathematical Platonists!

    As such, the powerful suggestion by those Platonists that mathematics is discovery of "timeless," universal, necessary, abstract objects by way of the "Unreasonable Effectiveness" of mathematics in the sciences, I think, get it backwards: it is not that mathematics consists of a priori discoveries, but how the mathematics is used and applied to the sciences and all natural phenomena that is the remarkable and amazing discovery being "Effectively" exemplified. As such, Brandon is right: mathematics is not a purely rationalistic enterprise as Platonists would hope.

    I'm with you: Metaphysics be dammned. We should then ally with and all become "natural philosophers". Long live metaphysics!

  7. It seems to me that many metaphysical possibilities can be known from the armchair, since their epistemic basis is in observation. Thus, inferences from actuality to possibility are justified. So are inferences from a sort of folk theory about physics, psychology, etc., about how the actual world works to non-actual cases. So are inferences to possibilities that are relevantly similar to actualities.

    Call any thought experiment about a non-actual possibility that can be traced back to such grounds a low-flying thought experiment. Would you accept the thesis that low-flying thought experiments are justified (even though they're done from the armchair)?

    I wonder how many philosophical thought experiments are low-flying thought experiments, or could be revised to be such. The Nogot/Havit Gettier thought experiment seems to be a low-flying thought experiment. So does one similar to the counterexample to Aristotle's "featherless biped" theory of human nature (imagine a chicken that's been de-feathered). I think certain Frankfurt-examples are low-flyers as well.

    Do all thought experiments invoke metaphysical possibilities? What about those refered to in per impossibile reasoning? So, for example suppose I'm thinking about the claims that God is the ground of abstracta, that I also believe that propositions are abstract objects, and that I accept the correspondence theory of truth. I then think to myself:

    "Suppose, perhaps per impossibile, that God doesn't exist. Then it be true that God exists? If so, then there is at least one proposition that doesn't depend on God. So, theistic activism is false."

    Now whether there are other problems in the reasoning here, the point is that in this case, it doesn't seem that I need to appeal to metaphysical possibility to do the armchair reasoning here. Is there something illegitimate about this?

  8. Mathematics is not a metaphysical necessity, things could be otherwise.

    Just because we have the right kind of mind/brain cognitive architecture we apply a given kind of mathematics to the world that in turn works extremely well to the "real world" applications.

    Concerning the a prori and a more empirically way of doing things i'm with Kant's awakening from the dogmatic slumber: why we cannot interwine both, perhaps the psyhcology of every individual is common to the psychology of the rest, although, influenced by the acculturation and other social factors.

  9. When people say stuff like "math is just pushing symbols around" they ignore the real world successes of mathematics, eg: the application of calculus in science and engineering.

  10. Eric,

    Seems to me, you came to the conclusion that one can't figure out things from armchair, from your own armchair :)

  11. Thanks for all the thoughtful comments, folks! Good stuff!

    Brandon: I think the course you offer in your final paragraph doesn't really undermine the idea that one discovers a priori, mind-independent truths through mathematics -- it's just that community interaction can help us discover such truths (contra the myth of the lone genius). So I do want to stick by the stronger claim.

    Your objection to that stronger claim is that "any invention is a discovery of a mind-independent fact that the thing invented is possible...". This is a nice point, and it forces me to draw out another fairly radical aspect of my view -- that "possibility" in the sense you intend it here is as badly misinterpreted as metaphysics.

    A psychologistic view of metaphysics fits nicely with a psychologistic view of possibility. You can't mean "naturally" or "nomologically" possible in your remark, since it's not plausible to suppose that we discover facts about natural laws (or whatever) by a priori reflection, so you must mean something like "metaphysical" or "logical" or "conceptual" necessity. My response is to collapse all three of these into the last. So yes you discover facts about conceptual necessity by a priori reflection -- but in so doing, you are still only discovering facts about your own concepts or mind.

  12. Anon 6:15: I'm not sure I understand what you mean. What kind of play?

    Kenf: Standard arithmetic works well for dealing with stable, countable objects. Two-valued Boolean algebra works better for dealing with digital computers. Complex or "imaginary" numbers, I'm told, have applications in electrical engineering. Different geometries work better modeling different kinds of physical systems. I take these to be empirical facts about the useful application of concepts and systems we have invented.

    Josh: Yep, I'll agree with that!

    Jonathan: You're right, "substantive" is a bit of a tricky word in that context. I don't know that I meant "objective" by "substantive" though. I guess I meant something like "non-linguistic", "non-psychological", "not merely conceptual". I take "four is greater than one", and I think probably all the statements on your list, to be facts about our language or conceptual structures.

  13. M: I agree with all that (even perhaps the contradiction at the end). Nicely articulated!

    Felipe: I see no problem with what you call "low flying" thought experiments, if they meet one of two conditions: (1.) The possibilities considered are well within the range of our intuitive competence, given what facts about the world we could reasonably expect our intuitions or concepts to we well-attuned to, or (2.) We interpret them only as revealing facts about our intuitions and concepts and not necessarily about the world beyond those intuitions and concepts. The Gettier examples could be plausibly interpreted along either line -- either an exploration of the mundane range of cases where our concept of "knowledge" might be plausibly be latching on to a natural kind or an exploration merely of our concept of knowledge. Same with the featherless biped example.

    The second kind of case, where a little deduction shows a contradiction between two views, I would analogize to mathematical proof. In both cases, I'm inclined to think one is drawing out non-obvious aspects of one's conceptual structures. It can be quite useful to do this, of course!

  14. Howdy Eric,

    "What gets called metaphysics, then, is really just a certain kind of self-study -- typically the study of the metaphysician's own concepts."

    I agree, the study of metaphysics is certainly limited by one's concepts. :-)

    Like any discipline, spiritual or otherwise, it is practice that develops one's concepts. 'Tis certainly true that undeveloped concepts won't get you very far.


  15. Anibal: I agree that Kant might be onto something. I worry a little that he might be trying to have his cake and eat it too, and I'll be darned if I can make sense of the deductions.

    I don't rule out that we might discover, from the armchair, facts about our psychology, facts about how we (nomologically/naturally) *must* see and understand things, given how we're built; but this doesn't reveal how things are in themselves (as Kant stresses), without further assumptions. But cross-cultural psychology or anthropology might be a better means than armchair reflection for discovering what conceptions seem to be merely optional and what humanly required.

    Anon 7:17: I don't deny the real-world success of mathematics. It can be very useful to push symbols around!

    Tanasije: I don't think my position succumbs to this classic objection to logical empiricism. The world could have turned out such that by reflecting in my armchair, I could discover things about Venus that would then be confirmed by astronomers. Or it could have turned out that I could tap into the world spirit and predict coming revolutions in far-away countries. But it didn't. So my claim that all I can discover from my armchair are facts about my own mind is (I think, I hope!) an empirically grounded claim.

  16. Jim/Lifeos: I'll agree with that. And therein is the *use* of metaphysics (and math)!

  17. Eric,

    On the first point, fair enough, but I don't think your response quite catches my point. When talking about something's being "possible for the medium or application for which it was invented" I meant that pretty straightforwardly and very generally. When Edison when through his process of trial and error to find the right filament, he invented the modern lightbulb; he also discovered mind-independent facts about what sort of light bulb is possible given certain design demands, etc. When someone invents a new kind of bridge, he thereby discovers certain mind-independent engineering facts. Likewise, one might say, when someone invents, say, a system of surreal numbers, he is also discovering the mind-independent fact of what is possible given certain parameters presupposed by the process of invention. And so forth. Whatever your view is, it can't be that rigorously-thought-out design (say, of a bridge that no one has actually built yet) or, say, the invention of new search algorithm, is merely discovery of facts about one's own mind (what facts would they be?). But if that's so, then there is a tertium quid between merely invented constructs and merely discovered facts. So the mathematics-is-invented route doesn't seem to be wholly adequate.

    (I think 'nomological possibility' gets used in two different senses; in one sense it indicates possibility relative to the laws of nature. But in another it just indicates possibility relative to some circumstance or condition. In the latter sense one can say that I meant 'nomological possibility'.)

    One thing I intended to press you on originally, but forgot to do so, was the idea that we discover any psychological facts about ourselves from the armchair. Suppose I am armchair-meditating on the concept of X; and suppose I suddenly realize, after much thought, that that particular concept that I had in mind has packed in it information about a "cluster of traits" that mutually exclude each other -- that, in fact, I had been entertaining something contradictory without realizing it. This is not obviously a fact about my mind; if it's contradictory, it's contradictory for anyone who has it. If I can, with however much difficulty, prove that some of my concepts are self-contradictory or self-consistent, or contradictory or consistent when conjoined in certain propositions or judgments with others, it doesn't seem quite right to say that I'm just discovering a fact about my own psychology.

  18. I have a hard time making sense of this:

    I take "four is greater than one", and I think probably all the statements on your list, to be facts about our language or conceptual structures.

    It's false that if we'd meant five by "one", then four would have been less than one. You agree with this much, right? So the truth that four is greater than one does not counterfactually depend on the meanings of any words, or any facts about concepts. So it can't be about meanings or words or concepts.

    Where do you reject this reasoning?

  19. Thanks for pressing me on these issues, Brandon and Jonathan. These are the kinds of worries in the back of my own mind that I set aside in writing up this post and in my previous replies to comments. I'm inclined to think the epistemic magic required to discover things beyond my mind from the armchair is sufficiently problematic that a conservative reply to these worries must be possible, but I don't know if my own reply will be adequate.

    On Jonathan's point: As a matter of empirical fact, all human languages that I know of have basically the same counting and adding system, at least for small numbers. The reason for this may be our tendency to interact with objects that are usefully characterized with this system, rather than, say, electric fields. If you had meant five by "one", that would just be a different idiolect within the same general conceptual system, and within that system it's true that four is greater than one, however that fact is expressed. But we could have had an arithmetic system on which there was no well-defined "greater than" relation for simple numbers, or we could have had a two-valued Boolean algebra on which "four" would not pick out a well-defined concept, and so no straightforward translation of "four is greater than one" that would render it true.

    Now is it the case that there is a mind-independent fact that within such-and-such a conceptual system, four is greater than one? This gets, I think, to Brandon's worry too. At least, it's my own biggest worry about the view I've expressed. I've dodged it a bit by insisting that such systems be called "conceptual systems", which makes them sound psychological and so facts about them sound like psychological facts. I'm torn about whether that's really okay, but comforted somewhat by analogy to more clearly invented, psychological concepts. My concept of "bachelor" is such that it's impossible to be married and a bachelor; that's a fact about my concept, and I don't find it compelling to suppose that it's a mind-independent fact about the world. Even more clearly, perhaps, I could define a concept "blimmy" such that it would be impossible to be both white and blimmy, and that's just a fact about my concept. Maybe mathematical facts, though structurally more complex, are no more mind-independent than this.

    This takes me into issues in philosophy of math that I'm not entirely comfortable with. That's why I concluded my post without taking a strong stand in philosophy of math other than to suggest that some conservative philosophy of math seems to me compulsory. Where I'm more comfortable is in bringing the issue back to metaphysics, especially the metaphysics of consciousness. Even if I were to grant that something about the logical structure of our concepts is mind-independent, it's a large and dubious leap from that to the claim that we can discover truths about whether robots would be conscious by reflection on such facts.

  20. I dunno, dipping your toes into phil. of math and saying that "conservative philosophy of math is mandatory" is up there with showing up at day one of phil. of psych and saying that Mind-Brain identity theory is obviously true....

  21. Do we mean the same thing by 'about'? I'm not sure how to define it, but here are what strike me as some paradigm instances:

    The fact that Eric is a philosopher is a fact about Eric.
    The fact that Jonathan is hungry is a fact about Jonathan.
    The fact that if Jonathan eats pizza, he will be less hungry is a fact about Jonathan and pizza.
    The fact that Jonathan is talking to Eric is a fact about Jonathan and Eric.
    The fact that Jonathan is a bachelor is a fact about Jonathan.

    I think those should be pretty hard to deny. (Eric, do you agree?) Once we accept those, though, shouldn't we also accept these?

    The fact that Jonathan is Jonathan is a fact about Jonathan.
    The fact that if Eric were a bachelor, then Eric wouldn't be married is about Eric.
    The fact that all bachelors are bachelors is a fact about bachelors.
    The fact that all bachelors are unmarried is a fact about bachelors.

    Am I reading you correctly, Eric, in thinking that you're denying the claims of this second list? It sounded like you think that these sorts of facts are not facts about you, me, and bachelors, but instead facts about words and concepts?

  22. Thanks for the great initial post and some great discussion!

    "My concept of "bachelor" is such that it's impossible to be married and a bachelor; that's a fact about my concept, and I don't find it compelling to suppose that it's a mind-independent fact about the world."

    Why is the fact that a bachelor can't be married not a mind-independent fact about the world? By "mind-independent fact" do you mean a fact that is made true or false by the world, i.e. that is true or false independently of what any thinker has to say about it? That can't be right since 'x is a bachelor iff x is an unmarried male' is true of some x independently of what I have to say about it. Once the truth-conditions for bachelorhood are set up, what stops the fact that a bachelor is an unmarried male from being a mind-independent fact about the world?


  23. Here's a bad argument:

    I worry that research is a sham -- or at least, that it's a sham if it's thought of as an empirical discipline whereby one discovers objective truths about the world while reading from an encyclopedia. When you lean over your book and read, there's only one kind of thing you can discover, it seems to me: Facts about your book. What gets called research, then, is really just a certain kind of investigation of whatever books you happen to have before you.

    I wonder how your argument against a priori knowledge is different from this one? You responded to apparent counterexamples -- instances of knowledge achieved from the armchair -- by saying that, appearances aside, the new knowledge there is really only knowledge about one's own mind. The parallel sorts of counterexamples here are instances of apparent external knowledge gained from reading a book -- say, the knowledge that the American Civil War was the bloodiest war in American history. The parallel move to yours would be to insist that this fact is not really about the Civil War -- it's really a fact about your book.

    Maybe it would help me to see your point if you could explain how you think your argument differs from the bad book argument.

    (Incidentally, I've just written a new post on my own blog saying some of the things I've been saying in this thread.)

  24. Well, anon 10:44, therein lies the difference between blog posts and publications, right?

    Jonathan: In response to your first comment, I worry that we're getting tangled up in side issues made by our words here. The key thing is to get clear what the relevant sense of "mind-independent" is and whether facts of the latter class are mind-independent in that sense (regardless of what they're "about").

    Charlie offers a suggestion -- two, actually. One that "mind-independent facts" are made true by the world (outside the mind) and the second is that they're true regardless of what any thinker has to say about it. If you’ll grant me that what we’re talking about are invented concepts (maybe you folks won’t grant this, but that’s a different argument), it may be helpful to clear away the underbrush by inventing a new one: “smeggy”. I hereby declare that something is smeggy by virtue of having eight faces and not being purple. When I conclude that smeggy things must be larger than geometric points, on some particular geometric system, have I discovered a fact that’s made true by the world? It seems to me rather that it’s a fact that’s made true by the concepts and geometry I’m employing. Is it something that’s true regardless of what anyone thinks about it? I’ll grant that it’s true regardless of who thinks it, but I’m not sure I need to grant that there would still be facts about smeggy things in Euclidean geometries if there were no people to have such thoughts. This last point, though, is where I start to feel unsure of my footing. I’m not even sure I fully understand what that counterfactual sentence means….

    I thought I addressed your second point, Jonathan, in the second paragraph of the original post. You can learn about the world by reading the textbook because the textbook contains important information about the world by virtue of having the right kind of empirical connection to the world. Am I missing something?

  25. Thanks for the clarification, Eric. The encyclopedia analogy was intended against your earlier suggestion (or apparent suggestion) that anything we could learn a priori would have to be a fact about concepts; I take it from your last comment, however, that you're backing off of that suggestion, in favor of the alternate claim that we could never learn mind-independent facts a priori.

    Mind-independence, on (one of) your preferred characterizations, is in terms of truthmakers: the fact that p is mind-dependent just in case it is true in virtue of facts about minds. And a fact is mind-independent just in case it's not mind-dependent.

    (I hope you don't mind my glossing it that way; I know you put mind-independence as "made true by the world (outside the mind)". I'm just trying to be as neutral as possible -- leaving open, for instance, the possibility that there could be truths that are true in virtue of nothing at all, or in virtue of something that is not the world (or a mind).)

    It is interesting to see truthmaker talk called into service in the post "Against Metaphysics". The way most people investigate "true in virtue of" claims, I would have thought, is by doing something in the ballpark of traditional armchair metaphysics. Do you have thoughts on how you know which facts are made true by which other facts?

    So, onto smegginess. What did you do when you wrote this passage: it may be helpful to clear away the underbrush by inventing a new [concept]: “smeggy”. I hereby declare that something is smeggy by virtue of having eight faces and not being purple.

    Certainly you've coined a new term, the word "smeggy". Maybe you've created a new concept, SMEGGY. (You haven't if, for instance, nativism is correct, or if a crude concept empiricism, according to which SMEGGY is identical to the conjunction of EIGHT-FACE and NOT-PURPLE is correct.) Let's grant that you've created a new concept. What you have not done, however, is create a new property. My orange eight-faced dragon has been smeggy her entire life -- we just never had a word for it until now. In virtue of what is she smeggy? She's smeggy in virtue of having eight faces and a non-purple color. These are properties intrinsic to my dragon; her smegginess has nothing to do with me or you.

    Smegginess is an intrinsic property -- it obtains in virtue of the intrinsic features of the smeggy body.

    (Compare: I a bachelor in virtue of being male and unmarried. And the substance in my cup is ice in virtue of being H2O and frozen.)

    Onto the counterfactual reading of mind-independence. I think it's obviously true that my dragon would be smeggy even if there were no people, or if you had never made up the word "smeggy". We know what sorts of things would have resulted in her nonsmegginess: if she'd been born purple, for instance, she wouldn't have been smeggy. If her mother hadn't drunk so much, she wouldn't have been smeggy (because she wouldn't have developed the birth defect that resulted in her having eight faces). If there were no people, she would still be a dragon, she would still get hungry, and she would still be smeggy.

    You say you have a hard time even understanding these counterfactuals, but I really don't see why. Just plug in your preferred account of counterfactuals. We can use Lewis's: go to the nearest worlds where this orange dragon has eight faces and there are no people -- are these worlds in which the dragon is smeggy? Sure they are. "Smeggy" means "eight-faced and nonpurple", and this dragon has both those properties in all those worlds.

    Can you say a bit more about what you find troublesome with this counterfactual? (Do you think that ALL counterfactuals which include there being no people in the antecedent are similarly problematic?)

  26. Eric, what worries me about your whole argument is that it just sounds like good old Logical Positivism again, e.g. when you say things like:

    My concept of "bachelor" is such that it's impossible to be married and a bachelor; that's a fact about my concept, and I don't find it compelling to suppose that it's a mind-independent fact about the world.

    Now, Jonathan has already pointed out why this can't be right: "bachelor" does not refer to a concept, but to a certain kind of people (roughly, unmarried adult males). What I find so irritating is that you make that claim as if Quine had never existed and never written "Carnap and Logical Truth", where he already argues that even seemingly trivial analyticities like "Eric is Eric" are as much true in virtue of what the world is like, namely such that Eric is identical with himself, as they are true in virtue of what "Eric" and "is" means. I can see no new argument in what you say that would make Quine's point obsolete, except your "empiricist dogma" that it just cannot be that we learn anything about the mind-independent world from our armchair. Why? Because it just cannot be!

  27. I'm sympathetic to the position in this post but I still think an armchair philosopher might be able to find some truth even if it is usually drowned out by the static caused by quirks of his own mind because eternal reality is the inputs to the mind and so some hint of reality should remain in the mind's output.

    So I'd be inclined to think of it a methodology of last resort for those problems that no other methodology can investigate.

  28. The human cognitive brain emerged and flourished in the course of evolution not because it provides absolute truths about the world (how would we recognize an absolute truth if we were confronted with one?) but because it gives us the biological machinery to arrive at pragmatic conclusions about the world. These enable us, in contrast to other creatures, to adapt successfully to an ever wider and richer ecological niche.

  29. Thanks, Jonathan and others, for continuing to push me on this. There's nothing like spirited opposition to help me see the gaps and cracks in my arguments!

    Jonathan: The truthmaker remark was just adapting the language of an earlier comment. It's not my own preferred lingo, though I don't see why it couldn't be given an empirical interpretation.

    I have an unusual view of properties, which is connected to the themes of this post and discussion. Properties, on my way of thinking about things (and I think metaphysicses [pardon the word!] are ways of thinking about things) hold when a concept applies, so that if you have two different concepts that can come apart we can say that two different properties are instantiated even if there is no external reality that undergirds this difference. Thus, for example, I am a "property dualist" about consciousness, but I think of property dualism as a thesis about our concepts rather than about the world behind our concepts.

    So, from the armchair, we can discover relaionships of properties, and we can realize that if an eight-faced dragon were born purple he would not be "smeggy" -- the concept SMEGGY would not apply to him. But we have still, I think, not discovered anything beyond the circuit of the mind.

    In general, I am highly nervous about counterfactuals -- especially what Felipe would call "high-flying" counterfactuals that involve weird or remote possibilities. I have concerns both about natural necessity (since I don't think there is a unique set of true natural laws, only competing models with different virtues and different counterfactual implications) and about conceptual necessity which (as should be clear) I give a psychologistic, deflationist reading of. I am much more comfortable hanging an argument on empirical observations (including observations about the limitations of the armchair) than on counterfactual claims that begin "if there were no people". I don't think such claims are always unjustified, but my "philosophical fog" alarms start to go off.

  30. Joachim: I find it interesting how analytic philosophers will use arguments from authority using Quine as an authority as they will with respect to virtually no other philosopher. I think Quine was right that the analytic/synthetic distinction with regard to natural language is blurry, but I think much less follows from this against Carnap that people ordinarily suppose. For example, I think we can use concepts as technical terms with official definitions such that analytic truths follow from them. We may not, in the end, decide such concepts are useful, if the empirical facts turn out a certain way, but that doesn't undermine the analyticity of certain statements.

    My view is indeed similar to Carnap's in many ways, though I also differ from him in other ways. I don't think metaphysics is "nonsense", only that its truths are truths about the users' concepts. I don't accept verificationism. And I don't hold to a cartoon version of the analytic/synthetic distinction.

    I'd be happy to hear more, though, about what particular arguments from Quine you believe undermine my views.

  31. I agree with you, Genius, with special emphasis on the point that what we learn from armchair introspection is a consequence of facts about the world (resulting from previous encounters or possibly evolutionary selection) affecting our minds.

    And yes, Arnold, how could one disagree with that -- at least without overthrowing evolutionary theory!

  32. Eric, my reference to Quine was not at all intended as an appeal to authority. In fact, I am highly critical of all of Quine's arguments against the analytic/synthetic-distinction, except the one from "Carnap and Logical Truth" I was referring to. And I think I didn't just gesture vaguely in Quine's direction but I also said what the argument is. So let me repeat it in my own words this time:
    A sentence like "John is a bachelor" is not about concepts or our own psychology, but about John and a certain social status that he has, namely being a bachelor. But why, then, should we assume that a sentence like "bachelors are unmarried" is about concepts and not also about some worldly fact, namely that being a bachelor and being unmarried always occur together? It just doesn't make any sense to me to claim that "bachelor" has a different reference in the second sentence compared to the first. After all, there is no indication that "bachelor" could be ambiguos like "bank", or that it might have some hidden indexical component. So, the point that I tried to make, maybe hiding too much behind Quine's back, was simply this: If you claim that "bachelors are unmarried" is about our concepts, and not about bachelors, than you owe as a credible account why all the semantic appearances count against your view, or how you intend to explain those appearances away. The burden of proof, I think, is entirely on your side here, as always when one of two contradictory claims is so much less plausible than the other.

  33. Sorry, Joachim, if I interpreted you uncharitably there! I've so often heard people appeal to Quine's arguments against Carnap and the analytic/synthetic distinction appealed to in argument-from-authority style that I've got a bit of a knee-jerk reaction.

    I worry about the word "about" in your and Quine's argument here, and I regret casting my own point in that way earlier -- though I also find it hard to avoid the term entirely! I might be willing to grant that "bachelors are unmarried" is "about" bachelors, in some sense of "about" -- but still it's true in virtue of the meanings of the terms, isn't it? There's some sense in which one can learn this truth without knowing anything "about" the world beyond our concepts and language, yes?

  34. Eric, maybe you can grant me the use of "about" for now (which seems so hard to avoid, I guess, because it serves as something like a general placeholder for any kind of referential relation between words/concepts and the world - and I really would like to hold on to this kind of minimal semantic realism!). Then I agree with the following: There is a sense in which we can learn that bachelors are unmarried merely in virtue of understanding the English sentence "bachelors are unmarried". This sense was quite aptly dubbed "epistemic analyticity" by Boghossian (in his 1997 "Analyticity"). The other sense of "truth in virtue of meaning", however, is the one of being made true by meanings/concepts or of being about meanings/concepts, which Boghossian calls "metaphysical analyticity". And it is this latter sense that is attacked by the argument from Quine that I was referring to. So, in short, I have no in principle objection against conceptual truth in the sense of epistemic analyticity (in fact, I try to defend conceptual analysis in my dissertation), but I think that metaphysical analyticity is in big trouble, though (but maybe not hopelessly so; if you are interested why, then I can send you the forthcoming paper "In Defense of Metaphysical Analyticity" by Frank Hofmann and me) - and that's where Quine definitely had a point!

  35. I'll tag in for Joachim; I think we're on the same page.

    What Quine demonstrated, I think, which is also what I was arguing earlier, is that there is no interesting sense in which the fact that all bachelors are unmarried obtains in virtue of the meaning of any terms.

    There are some uninteresting senses, of course. Like this one: the sentence, "all bachelors are unmarried", is made true in part by the meanings of its words -- if "bachelors" had referred to husbands instead of to bachelors, the sentence would be false. But of course this kind of "truth in virtue of meaning is cheap" -- the sentence, "Jonathan is a bachelor" is true partly because of the meanings of its terms, too. It would also be false if "bachelor" meant husband.

    Knowledge of concepts and language are, I think, neither necessary nor sufficient for knowledge that all bachelors are unmarried. The necessity claim is easily refuted: many people know that bachelors are unmarried who do not have any idea what the word "bachelor" means. I have in mind, for instance, people who don't speak English.

    The sufficiency claim is trickier and more controversial, but I have become convinced by Timothy Williamson's recent arguments (ch. 4, if memory serves of PoP) that it is possible for someone to have the relevant linguistic and conceptual knowledge and nevertheless fail to know that all bachelors are unmarried, if, for instance, he becomes convinced by faulty reasoning that this proposition is not true.

  36. I posted behind Joachim without realizing it. I'll just add that he's right to separate the epistemic questions from the metaphysical ones, and I should have done a better job of that in my last comment.

    The question whether analytic truths obtain in virtue of meaning is independent from the question whether conceptual or linguistic knowledge is sufficient for knowledge of those truths. I say "no" to both anyway, but Joachim points out that you can say no to the first and yes to the second.

  37. Thanks, Joachim, I'm reluctant to describe myself as a friend of "metaphysical analyticity," and as I briefly mentioned in an earlier reply to Jonathan, truth-makers is not really my prefered idiom -- but there still may be a sense in which I accept what you and Boghossian call "metaphysical analyticity". So sure, why don't you send along that paper!

  38. Jonathan: The necessity claim is at face value false, but there may be a version of it that can be worked out in terms of translation or the like. What is it to believe that bachelors are unmarried? Maybe you need some concept that translates roughly into the concept "bachelor" (or some conceptual architecture that invites such a concept)....

    On the sufficiency claim: What would it be to possess the concept "bachelor", "married", etc., and NOT believe that bachelors are unmarried? I don't think a lapse in judgment due to a bad argument would suffice, but only a persistent (and counterfactual supporting) tendency to misuse the terms in a variety inferences and attributions (this fits in with my broadly dispositional view of belief) -- and I would be inclined to deny that someone with such persistent tendencies of misuse really does have the requisite concepts matching our own. (A different story will have to be told, of course, for false beliefs about complex mathematical facts.)

    So I hold to the view that there are analytic truths that are truths "about" or that are "made true by" (or something like that, I'm not sure how best to put it) facts about our conceptual systems and structures. This is not to say that there aren't problematic cases in natural language (is "stars emit light" uttered pre-scientifically true in virtue of its meaning?), but I have a story to tell about such cases. (Briefly: "Sun" in natural, prescientific language does not pick out a single rigid concept but rather can be rigorized and developed in different ways, and depending on the empirical facts we may want to rigorize or develop it one way or another.)

  39. Eric, your last comment seems to run together the two questions that Joachim helpfully distinguished. When I said that knowledge of concepts or language was neither necessary nor sufficient for knowledge of analytic truths, I made an epistemic claim. You gave an argument against this epistemic claim, then suggested that there is a real metaphysical conception of analyticity. But, as Joachim made clear, both by argument and by serving as an example, you can disagree with my epistemic claim and still reject metaphysical analyticity.

    So the rest of this comment is in defense only of my epistemic claim.

    As for the question what it would be like to reject analytic truths even with understanding of the relevant language and concepts, I'm really going to just point to Williamson here. (For arguments -- I'm not accusing him of being an example!) I recommend the piece in his new book. Very briefly, he gives two kinds of examples: someone with funny views about vagueness, and someone with funny views about quantifiers.

    Someone with funny views about vagueness might end up thinking it's non-true (perhaps false, perhaps indeterminate) that all bachelors are unmarried, because there are borderline cases of bachelors who are borderline cases of unmarried for whom it is not true that if they are bachelors, then they are unmarried (I -> I: I), and that the quantifier can only be true if all instances are determinately true.

    Someone with funny views about quantifiers might think that it can only be true that all bachelors are unmarried if there is at least one bachelor. If he also has funny empirical views about marriage and thinks that there are no bachelors, then he'll reject the quantifier too.

    These are stupid views, but I think Williamson is right that they're consistent with concept possession and knowledge of language.

  40. Jonathan, I also like Williamson's argument a lot, but I'm also struggling to find a way to resist his conclusion. For, somehow I tend to agree with Eric here when he says: "I would be inclined to deny that someone with such persistent tendencies of misuse really does have the requisite concepts matching our own."
    Now, Williamson does argue, of course, that someone could rationally disagree about a supposed conceptual truth like "bachelors are unmarried" and still possess all the relevant concepts. However, Eric rightly emphasized that persistent disagreement is the crucial case here. So, let's take the person with funny views about vagueness (call him V). Supposedly, we can find the true theory about vagueness in a rational way. But if this is so, then we will ultimately not disagree with V but converge on some theory of vagueness, and so we will also not disagree about the bachelor-sentence then, either. However, if we ultimately do not come to agree with V, then there seem to be two possibilities. (1) there is no rational way to find out the truth about vagueness, or (2) there is some kind of conceptual difference between us and V. If (1) is true, then there never was a rational disagreement between us and V in the first place - contra the above assumption. If (2), however, then there really is something like conceptual truth in the epistemic sense. Do you think this could be a promising line of response to Williamson?

  41. Joachim, if I'm understanding you right, the response you're describing is not too car removed from my own view. I think that although possession of concepts and linguistic competence are not sufficient for knowledge of any particular analytic truth, such competences, combined with sufficient rational reflection, are.

    In other words, concept possession doesn't guarantee knowledge of the Gettier conclusion -- but if you're sufficiently rational, it guarantees you enough to get to it.

    By the way, I didn't take Williamson to be arguing that people could rationally decline assent to these analytic truths -- just that they could in fact, consistent with understanding, decline assent. I think Williamson will admit that his subjects are making philosophical errors, and are therefore, in some sense, failing to live up to the standards of rationality.

  42. Jonathan, given that you roughly seem to share my view on knowledge based on conceptual competence, then probably it's already a bit more likely to be true... :-)

    Although I don't have Williamson's argument exactly in mind, I do think that it would weaken his argument considerably if he did not suppose that rational disagreement about supposed conceptual truths is possible. I thought that this was his improvement on Quine's holism argument. Think about his example of the logician Graham Priest who denies the validity of disjunctive syllogism (because he denies the law of non-contradiction). Doesn't Williamson strongly suggest in the book that Priest is someone who rationally disagrees with him or us? After all, at least from my point of view of an amateur logician, it seems as crazy to claim that Priest fails to grasp our common logical concepts as it seems to accuse him of irrationality. Furthermore, if someone did disagree irrationally with us about some seeming conceptual truth, why should this show anything at all with regard to epistemic analyticity? For, if you are sufficiently crazy or obnoxious, you can deny anything whatsoever with any or no reason whatsoever. So, these people might still be disposed to rationally assent to "bachelors are unmarried", but there disposition could be masked by their overriding craziness. But all a sensible defender of epistemic analyticity or conceptual analysis should claim is that, in virtue of the possession of certain suitable concepts (like "knowledge" or "bachelor") we are rationally committed to the ultima facie acceptance of certain conceptual truths (in the epistemic sense).
    Do you think this is plausible?

  43. Eric, it seems to me that anyone who believes in the theory of evolution should agree with your skepticism about metaphysics as a privileged path to "a special class of truths while reflecting from an armchair."

  44. Joachim, I'm not sure about Williamson. I'm sure that he thinks that these subjects are making philosophical mistakes -- it seems to me that if you're making a philosophical mistake, then you're not proceeding rationally.

    But I think we've gotten both off-topic and into sensitive matters of Williamson interpretation, so I'm going to suggest that we take this to email if we want to continue the discussion further.

  45. Jonathan, I agree! So, if you have any further thoughts about my last post or some other issue from our discussion, then I would be very interested to come to know them by email-testimony...

  46. What a great and helpful discussion, Jonathan and Joachim! Thanks so much!

    I don't think I'm so far away from the point on which you two are converging, perhaps -- although we might be being a bit too rough here with "rational"? I probably wouldn't require someone to have the right philosophical views at the end of the day for that person to count as rational.

    Also, I'm not so sure that the views described as "funny" really are so bad. Williamson's view of vagueness strikes me as pretty odd, too! One could have a different set of rules for the quantifier "all", for example. I'm not an "atomist" or "molecularist" about concept possession -- closer to holism (but not crazy holism), where divergences in patterns of use are sufficient to render concepts slightly different -- so we don't necessarily share *exactly* the same concept of "all" with someone who has a different logic for it, and my concept of "bachelor" is somewhat different if I admit a range of vague cases and you do not. So then, what's analytic and rational for you with respect to the concepts we both describe with those labels may differ....

  47. Oh, and Arnold, I'm with you on that -- though there is clever philosophy enough out there to wiggle out of almost anything, so I don't think the argument will be a short one!

  48. If you are interested in a view of how the brain constrains our intersubjective understanding of verbally communicated concepts you might take a look at "The Pragmatics of Cognition" in *The Cognitive Brain*,pp. 300-301.

  49. The word "meta" means "self-aware," which means that established fields of knowledge are distilled phenomenologically and reiterated as subjective theory. Metaphysics isn't a sham but is in the unique position of offering nothing that be objectively proven but everything than can be subjectively useful. Explore my metaphysics tags for good examples.