From owner-modality@LISTSERV.ARIZONA.EDU Fri Feb 19 06:21:34 1999 Date: Fri, 19 Feb 1999 06:20:07 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Next week To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO My visa still isn't through, and although there are indications that it shouldn't be too long now, I won't be there by next Tuesday. This means I have to rearrange things a little. Week 5 was originally supposed to be on the "tyranny of the subjunctive", on relationships between the 2-D framework and indicative and subjunctive conditionals, and on implications for the analysis of necessity. But that isn't really written up (except very briefly in the Princeton outline); I'd been planning to mostly talk about it in class. So as things stand, we will skip straight to the following topic (the original week 6) instead. This topic involves applying the 2-D framework to epistemic and modal arguments against materialism. The readings are TCM, Chapter 4 -- mostly pp. 131-149. N&N, pp. 144-55. Mind and Modality, Lecture 1 (esp. sections 5-7). The Mind and Modality notes (on the web) have my currently preferred formulation of the 2-D argument against materialism. It's a bit crisper and tighter than the book version, though it comes to much the same thing in the end. The book version has more discursive detail, so you'll probably want to look at the two of them together (the real core of the book argument is pp. 131-36, but pp. 147-49 also has relevant material). It might also be interesting to look at Kripke's anti-materialist argument in N&N, which is similar in spirit, and compare and contrast. (N.B. The Princeton notes have some contrastive analysis that goes a bit beyond what is in the book.) Anyway, consider these the readings for next week's discussion. I'll look forward to seeing your thoughts on them early next week. As promised, there's also an assignment. I'd like you to take two terms that Kripke discusses, and translate his discussion into the 2-D framework. One is "yard", as discussed on p. 76; the other is "cat", as discussed on p. 122 and pp. 125-6. I'd like you to take every significant part of Kripke's discussion here and translate it into the 2-D framework. Characterize roughly what the PI and the SI of these terms look like, at least given Kripke's intuitions. In places where Kripke considers a hypothetical scenario, you should roughly specify the world, say whether it is being considered as actual or as counterfactual, note what the referent of the relevant term or statement is in that world (considered the relevant way, and according to Kripke's stated intuitions), and say what the upshot is the for relevant PI or SI. Try to translate Kripke's more general points in these passages into the framework too, if you can. If you're so inclined, you can say whether you agree or disagree with Kripke's specific and general analyses and why, though that isn't compulsory. This shouldn't be much work (the passages are pretty short). This will be due by next Tuesday at noon, Arizona time. Don't be late. E-mail it directly to me (not to the mailing list). It should be your own work. --Dave. From owner-modality@LISTSERV.ARIZONA.EDU Wed Feb 24 03:11:57 1999 X-Accept-Language: en Date: Wed, 24 Feb 1999 04:04:49 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Brad Thompson Subject: Kripke and the primary intension of "pain" To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO On pp. 150-5 of N & N, Kripke argues that the identity theorist who claims that "pain = C-fiber stimulation" is necessary a posteriori cannot explain away the apparent contingency of the identity statement in the same way as with "water is H20" or "heat is molecular motion". In the latter cases, Kripke claims that the apparent contingency is due to the falsity of a qualitatively analogous statement in a qualitatively identical epistemic situation. So, for example, strictly speaking it is false that heat might not have been molecular motion. But what is true is that someone might have the sensation of heat (that is, be in a qualitatively identical epistemic situation as the one I am in when I sense heat/molecular motion) even though the phenomenon being sensed is not molecular motion. "Pain = C-fiber stimulation" can't be handled in the same way because, Kripke says, being in a qualitatively identical epistemic situation as the one in which I am having pain *is* to have a pain. In the end I might find this satisfying, but it does seem to me worthy of debate/discussion whether or not it is correct to here identify pain with an "epistemic situation". Pains are not conceptual, and there is lots that needs to be said about how they could be properly characterized as constituting a person's "epistemic situation". There is a sense in which it seems that a person (let's call her Jones) who lacked the capacity to feel heat sensations but who had a thermometer built into parts of her body which sent temperature information to her brain could be in the same epistemic situation as a "normal person". That is, she could be in equivalent information states. In the Mind and Modality notes, Chalmers rejects the "epistemic situation" bit in the above for different reasons. But using the notion of a primary intension, I suppose that one would say that Jones does not have a concept of heat assuming that the actual sensations of heat are part of the PI of "heat". Does this seem right? It probably does, but let's suppose that Jones' judgments about the presence of heat is identical to mine in every way (she makes exactly the same discriminations, for example). I can imagine someone (perhaps a Dennett) arguing that Jones does have our concept of heat, and that the PI of heat involves something that Jones and I share (but which I find difficult to characterize). Perhaps something similar could be done with "pain". Might this not defuse the argument against materialism? Of course, it depends on a highly contentious analysis of the PI of phenomenal concepts. At any rate, how we are to characterize the PI's of phenomenal concepts seems to be an interesting issue and one that is extremely important in assessing the anti-materialist argument. From owner-modality@LISTSERV.ARIZONA.EDU Thu Feb 25 00:56:29 1999 Date: Thu, 25 Feb 1999 00:56:17 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Re: Kripke and the primary intension of "pain" To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Brad raises a few interesting questions re "pain" and "heat". First, he wonders whether it's really right to say our epistemic situation re pain really includes pain. Second, he asks whether someone really needs to have heat sensations to have the concept of "heat" and the associated PI. Third, if the answer to this question is no, he wonders whether the point might be extended to "pain", so someone could have the concept and the PI without having sensations of pain. (1) Re the first question, of course I think that epistemic situations are inessential here, and that nothing really turns on them in the argument. But I do share Kripke's intuitions that there's a reasonable sense in which when I'm feeling pain, pain is an essential part of my "epistemic situation". Pain is unlike chairs in that way, for example; when I look at a chair, it's arguable that I could be in the same "epistemic situation" in some sense without there being a chair there. This is just what skeptical arguments trade on -- the idea that my "direct evidence" or some such could be as it is for me right now, without various objects in the external world existing. We might think of "epistemic situation" as being implicitly defined as the situation that is preserved in skeptical scenarios. I can consider a skeptical scenario where I'm in the same epistemic situation but there's no chair, but it's less clear that there's a skeptical scenario where I'm in the same epistemic situation but there's no pain, or no sensation of blue. Arguably, changing those things changes my "direct evidence", and changes the way things are for me now. Of course, this is an intuition and is somewhat loose, but it does seem to get at something central in our thinking about skeptical scenarios and what is held constant. Personally, I think this gets at some deep differences between the epistemology of conscious experience (in the first-person case) and that of the external world. Somehow our epistemic relation to experience seems more direct and intimate than our epistemic relation to external objects. But that's a long story in its own right (there's a little about this in chapter 5 of the book and in the third M&M lecture). Of course such views aren't uncontroversial. Some might argue that we really stand in the same epistemic situation vis-a-vis experience as vis-a-vis chairs. And some might argue that if experiences aren't conceptual, it's not really right to speak of them as being "epistemic" -- maybe our epistemic situation should be restricted to our beliefs, or some such. This might be particularly attractive to someone who thinks that the only thing that can justify a belief is another belief (a coherentist as opposed to a foundationalist, for example). Personally, I think there's a very strong intuition that experiences serve as basic evidence and can serve to justify beliefs even if they aren't conceptual, but there is certainly a lot that needs to be worked out there. Brad wonders whether "Jones" who lacks heat sensations and instead has thermometer readings sent straight to her brain could be said to be in the same epistemic situation. Interesting. I guess I'd say that if the information is all unconscious and unexperienced for Jones, then Jones can consider skeptical scenarios in which the information is different, etc, and can't rule out the existence of those situations in some intuitive sense, e.g. can't rule them out by "direct evidence". So there is a temptation to say that this info isn't part of the epistemic situation, at least not in the direct way that heat sensations are. Something to do with the skeptic-proofness of one but not the other. Still, there's some sense in which one might want to say that the temperature readings are "evidence" in some pretty direct sense. Maybe to a reliabilist, they'll be as direct evidence as anything (I don't know just what a reliabilist will say here, but maybe others can help me out). So anyway, there are some deep epistemological issues lurking around here. (2) Do we need to have heat sensations to have the concept and the PI of "heat". Well, on the Kripkean view, the PI of "heat" is more or less "the cause of heat sensations", so the question of whether one can have this concept without heat sensations reduces to the question of whether one can have the concept (PI) of "heat sensations" without heat sensations. That one in effect was question (3). In suggesting that Jones might have the concept of "heat", I can see three things one might be saying. (a) Jones has the PI "the cause of heat sensations" without heat sensations; (b) The PI of heat is something other than "the cause of heat sensations", and Jones has that PI; or (c) The PI of *our* concept of heat is "the cause of heat sensations", but Jones has a different PI that still qualifies as a concept of "heat". I'm not sure which of these Brad was pointing to, but all are interesting. Option (a) (which reduces to the pain-like case from question (3)) doesn't seem too plausible here, since heat doesn't cause any heat sensations in Jones, so this PI wouldn't pick out anything. But (b) and (c) are left. I think it's plausible that the PI of Jones' "heat" is something like "the cause of heat judgments", or some such (I'm assuming she is led to the judgments in a blindsight-like way). The real question then is whether that counts as a concept of heat, and as the same as our concept. One possibility is that the PI of our own concept is "the cause of heat judgments", not "the cause of heat sensations". That's interesting and not obviously false. I guess it comes down to: what do we say about a (broadly) epistemically possible situation in which we have no heat sensations but still have 'heat' judgments (e.g., the broad epistemic possibility that we are like Jones). If that turns out to be actual, will we say that heat is the stuff that causes the judgments, despite the lack of sensations? Maybe we would. If so, the PI of *our* concept is arguably judgment rather than sensation oriented, and Jones can have that PI. This could be the "Dennett"-like option that Brad gestures towards. It's tricky, though. (I mention this sort of possibility in footnote 41 on p. 368 of the book.) Alternatively, we might say Jones' PI is different from ours, but similar enough that it still counts as a concept of heat. Maybe so, though that would be a largely terminological issue. Personally, I think the previous possibility is more interesting. (3) Can something similar be done with "pain"? Hmm. First, I note that the Jones case doesn't lend direct support, since we saw that that case can't rest on her having the PI "heat sensations" without heat sensations. But maybe something like this could happen anyway. Possibilities: (1) Someone could have our PI of "pain" without having pain sensations. (2) Someone could have a different PI of "pain" without pain sensations, but which would still be a concept of pain. I have to go out now, so I'll get back to this later. All thoughts are welcome. --Dave. From owner-modality@LISTSERV.ARIZONA.EDU Thu Feb 25 22:10:12 1999 Date: Thu, 25 Feb 1999 22:09:46 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Re: Kripke and the primary intension of "pain" To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO On Brad's third point re "pain" etc (continuing from last time). We were talking about the possibility that the PI of "heat" might not essentially involve heat sensations (e.g. maybe just "heat" judgments), and then wondering whether that might be extended to the concept of pain. [(3) Can something similar be done with "pain"? Hmm. First, I note that the Jones case doesn't lend direct support, since we saw that that case can't rest on her having the PI "heat sensations" without heat sensations. But maybe something analogous coult happen anyway? Possibilities: (1) Someone could have our PI of "pain" without having pain sensations. (2) Someone could have a different PI of "pain" without pain sensations, but which would still be a concept of pain.] I take it the interesting possibility is (1), as we are interested in the PI of our concept. Could Jones have e.g. our PI of "pain", despite her lack of sensations, in the way that she arguably has the PI of "heat"? One move might be to suggest that the PI of "pain" should be analyzed as something like "the cause of 'pain' judgments", as in the heat case. If so, then Jones could truly say "I'm in pain", despite her lack of true sensations. Is our PI like this? This depends on what we'd say about the broad epistemic possibility that we are like Jones: what if that turned out to be actual? Maybe there's at least some sense in which we'd say that we still have pain, because we'll use "pain" for what causes the "pain"-judgments. That's not entirely clear to me, but maybe. In any case it seems that one can just re-raise the issue with the concept of "pain qualia". I think it is relatively clear that the (broad) epistemic possibility that we are like Jones is an instance of the (broad) epistemic possibility that we don't have pain qualia. That is, the concept of "pain qualia" will *not* apply in the Jones centered world. (In a world of people like Jones, Dennett-ish people saying "pain qualia don't exist" would be right.) So the PI of "pain qualia" applies *essentially* to qualia-like things. This in effect reflects Kripke's observation re "pain". Perhaps one could argue about it re "pain", but then one can just reraise it by explicitly building in "pain feeling" or "pain qualia" or some such into the concept -- i.e. by really specifying that it is a truly phenomenal concept. Once one has done that it seems we have a qualia whose PI applies essentially to pain qualia or something in the vinicity. After all, we need *some* concept to distinguish our own actual situation from the centered world (considered as actual) in which we are something like Jones. (Unless one denies the distinction, in which case one is presumably denying the epistemic intuitions re qualia, zombies, etc.) And "pain qualia" sems to be just the concept. So beyond a certain point, the trick of pushing things out into the world seems to stop. At least that's how it seems to me, but I may be wrong about this, so I'd be interested to hear more thoughts. This ties closely to the sort of issue about the content of phenomenal concepts in "The Content and Epistemology of Phenomenal Belief". There I argue that our most crucial phenomenal concepts have the same PI and SI, each picking out the appropriate sort of quality in all worlds. So the structure of phenomenal concepts is indeed special in a way. At least that's what falls out of accepting the epistemic intuitions, re the conceivability of inversions, zombies, etc. If those intuitions are rejected, it may be a different story. Brad says that this issue re phenomenal concepts impacts the anti-materialist argument in important ways. I think he's right about that, but I'm interested to know how, exactly. E.g., how would it affect the argument I put forward, essentially from the two premises, (1) the epistemic gap and (2) the 2-D analysis of a posteriori necessity? It's not obvious how it affects premise (2). Are you suggesting that it affects premise (1), maybe? If so, I'm interested to hear more. --Dave. From owner-modality@LISTSERV.ARIZONA.EDU Thu Feb 25 22:46:43 1999 Date: Thu, 25 Feb 1999 23:45:49 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Josh Cowley Subject: cognitive deficits To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Keeping in line with my subject title the following was written while suffering from some cognitive deficits. I'm hoping the reaction of others will help me figure out what I'm worried about here. I have a question about the section on cognitive limitations (pp. 138-140) Dave says that there is a weak sense in which we can conceive of a world in which the continuum hypothesis is true. But, since we don't know if it is true, it isn't the strong sense of conceivability that we get when we consider a physically identical world without consciousness. I'm wondering why this sense of conceivability is "weak." I also want to push a worry that Rachael and I were expressing during discusion the other day. Kripke is of the opinion that there is only one actual world, this one. PIs pick out referents at centered possible worlds considered as actual. I take it, that to do this you need to be able to consider any possible world as actual. (First question: Is my last sentence right?) If this is right then consider a world in which I don't exist. (Index I to yourself). I can easily imagine a possible world in which I don't exist (a very unfortunate world). But I'm having trouble considering that world as actual. Now there are some ways of defining actual worlds such that I can consider a world in which I don't exist as actual. It seems like these definitions do one of two things. Either they make me think of different possible worlds as counter-factual or they make me think of lists of propositions true at this world. At the moment I'm unable to articulate what I mean by that second choice. I'll try and fill it out tomorrow. Josh From owner-modality@LISTSERV.ARIZONA.EDU Thu Feb 25 23:30:57 1999 Date: Thu, 25 Feb 1999 23:30:45 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Re: cognitive deficits To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO A quick reply to Josh. >I have a question about the section on cognitive limitations >(pp. 138-140) Dave says that there is a weak sense in which we can >conceive of a world in which the continuum hypothesis is true. But, >since we don't know if it is true, it isn't the strong sense of >conceivability that we get when we consider a physically identical >world without consciousness. I'm wondering why this sense of >conceivability is "weak." This comes down to the distinction between positive and negative conceivability that we will be discussing. Intuitively, we can't rule out the possibility that CH is true, but nevertheless we can't clearly and distinctly conceive of a scenario in which it is true. Negative conceivability goes with not finding a contradiction; positive conceivability goes with clear and distinct conception of a world, in some intuitive sense. Intuitively, we can flesh out a whole world in the "conceptual imagination" in some sense in which there are zombies, whereas we can't do anything like that fleshing out a world for CH (or not CH). It's more the purely formal inability to rule out the hypothesis. If you like, think of positive conceivability as being able to conceive of a world verifying P, and negative conceivability as being unable to rule out P a priori. With zombies, we conceive of two distinct worlds, one with consciousness and one with zombies. But we don't have an intuition of conceiving of two distinct worlds, one with CH and the other with not-CH, in the mathematical case. So only the zombie case goes with a distinct positive conceivability intuition. There's much more to say about this, but this is a start. >I also want to push a worry that Rachael and I were expressing during >discusion the other day. Kripke is of the opinion that there is only >one actual world, this one. PIs pick out referents at centered >possible worlds considered as actual. I take it, that to do this you >need to be able to consider any possible world as actual. (First >question: Is my last sentence right?) That's more or less right. It may be that some worlds don't yield determinate extensions or truth-values, though (especially particularly distant worlds, or worlds where certain background assumptions aren't satisfied -- see "The Components of Content"). >If this is right then consider >a world in which I don't exist. (Index I to yourself). I can easily >imagine a possible world in which I don't exist (a very unfortunate >world). But I'm having trouble considering that world as actual. By "consider a world in which I don't exist", there are two things you can mean. Basically, one satisfying the PI of "I don't exist" or the SI. The first will be a world such that when it's considered as counterfactual, the verdict is that I don't exist. I.e. an instance of the subjunctive possibility that I don't exist. Essentially, that's a world in which this very person, David Chalmers, doesn't exist. The second is a world such that when it is considered as actual, the verdict is that I don't exist. I.e. an instance of the epistemic possibility that I don't exist. That's a very weird world, but in essence it is a centered world without anyone at the center of that world. I'm not sure which of these you intend here, but my guess is that it's the first. If so, that's a slightly odd thing to do, as you're mixing "considering as counterfactual" (in specifying the world) with "considering as actual" (in evaluating the world). So there is the potential for getting mixed up. Still, if you're very careful, I guess you can do that. I can take worlds in which David Chalmers doesn't exist and consider them as actual. E.g., I can take a world centered on Michael Jordan and without David Chalmers. Considering that world as actual, it's more or less an instance of the epistemic possibility that I am Michael Jordan and that David Chalmers doesn't exist (or perhaps of the epistemic possibility that I am Michael Jordan and that the guy with such-and-such DNA structure etc doesn't exist). It's tricky, but one can do it. What makes it so tricky of course is the way that one combines considering as counterfactual and as actual here. In a way, it's best to specify the world one is considering using "considering as actual" terminology all along. For example, one can just talk of the epistemic possibility that P, for a whole lot of P, in specifying a world. E.g. the very odd epistemic possibility that I don't exist. (It's disputable whether this is really a broad epistemic possibility, but I tend to think that it is, as it isn't a priori that I exist, it is a posteriori. That's a tricky matter, though.) Then I can say, if that world (the world with no-one at the center) is actual, I don't exist. So one can evaluate at least some terms and statements there. >Now >there are some ways of defining actual worlds such that I can consider >a world in which I don't exist as actual. It seems like these >definitions do one of two things. Either they make me think of >different possible worlds as counter-factual or they make me think of >lists of propositions true at this world. At the moment I'm unable to >articulate what I mean by that second choice. I'll try and fill it >out tomorrow. Interesting, maybe these correspond in a way to the above. Your first choice seems to be the mixed strategy mentioned above (with both considering as counterfactual and as actual). I'm not sure about the second choice, but maybe it is like the second option above. This is all a tricky business so I'm interested to hear more. --Dave. P.S. I still need to see the minutes from this week's meeting. And lots of people still owe their comments re the week 5 readings and issues. We are falling behind a bit, so I hope people can do this very soon. From owner-modality@LISTSERV.ARIZONA.EDU Fri Feb 26 10:13:13 1999 Date: Fri, 26 Feb 1999 11:11:24 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Anthony T Lane Subject: minutes To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Here are the minutes from this weeks meeting, compiled by Rachael and myself: First, we talked about different positions regarding discussion of possible worlds. Kripke seems to suggest that we cannot make sense of considering other worlds as the *actual* world. Once we have identified, for instance, that water is H2O, it does not make sense to consider worlds in which water is something else. This is distinct from considering the epistemic possibility of discovering that water actually is not H2O in this world. We then debated the plausibility of considering other possible worlds as actual-- in short, whether PIs are, in fact, of primary importance to the meaning of a concept. Then we discussed epiphenomenalism. Some members of the class thought that it is possible to accept the possibility of inverted spectra while resisting the notion that phenomenal experiences are causally inefficacious. Tim suggested that this is not the case, and that inverted spectra leads one down the slippery slope to epiphenomenalism. Then we discussed what is at stake if this is conceded. We then talked about the possibility of identifying particular phenomenal experiences as certain brain states. If this is possible, we debated whether it would be the case that the brain state in question would be, say, the experience of red, even if it were unaccompanied by the corresponding phenomenal experience. Again, we discussed what the crucial *intension* of a concept should be, and debated whether it is just a matter of bashing intuitions if one does not accept the initial idea that the phenomenal experience of having a pain is something that needs to be explained. Finally, Josh made a point about the possibility of himself not existing, which he has since posted to all. Anthony and Rachael From owner-modality@LISTSERV.ARIZONA.EDU Wed Feb 24 12:17:24 1999 Date: Wed, 24 Feb 1999 13:05:41 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Rachael J Parkinson Subject: Re: Kripke and the primary intension of "pain" To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO I have a question regarding Kripke's characterization of heat on pg 133 and how it relates to pain. Kripke says, "Here heat is something which we have identified (and fixed the reference of its name) by its giving a certain sensation."(131) This would seem to pick out the primary intension of heat. But, Kripke points out, we have discovered a posteriori that heat is just molecular motion. Given that heat is just molecular motion, it must be molecular motion in all counterfactual worlds. The primary intension of heat seems to be contingent given what we know about the secondary intension. So if someone where to say that light produces in them the same thing we feel when we feel heat, we would not say that light is heat but that the stream of photons produces a *sensation* of heat. If we were to discover a posteriori that pain is certain C fibers firing in the brain, it seems like that would become the secondary intension of pain, the way heat is molecular motion is the secondary intension of heat. If this was the secondary intension of pain then it would fix pain to be "the firing of certain C fibers" in all counterfactual worlds. Thus, if someone where to say, light produces in me the same sensation that you feel when you feel pain we would not say that light is pain but that the stream of photons produces a *sensation* of pain. If one was to argue that the *only* difference between heat and pain is that we have already discovered that heat is explainable in terms of molecular motion while we have not yet discovered that pain is explainable in terms of the firing of C fibers, it seems like the cases are relevantly similar. What do you think? -Rachael From owner-modality@LISTSERV.ARIZONA.EDU Wed Feb 24 20:12:50 1999 Date: Wed, 24 Feb 1999 21:10:10 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Anthony T Lane Subject: Re: Kripke and the primary intension of "pain" To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Some thoughts on Rachael's note: Kripke urges that, once we have identified heat as molecular motion, we have fixed the reference of 'heat'. As such, 'heat' would still refer to molecular motion if it happened that our nerve endings were different such that we actually had the sensation of cold when exposed to increased rates of molecular motion. I am not sure that I agree with this intuition-- it seems that Kripke is assuming the primacy of secondary intensions in this case. Suppose that one miraculously were to wake up in a world that is mostly like this one. In fact, it is, in many ways, indistinguishable from this one. The sun on ones skin feels warm and snow feels cold. Suppose that some time after relocating to this world one were to discover that heat actually corresponds to reduced molecular motion. The laws of physics in this world, however, are such that things behave exactly as they do in this world. It is not clear to e that one would, in this case, say, "Oh, I guess that what I thought was heat is not actually" upon discovering that heat is not more rapid molecular motion. It seems that this is a case where it is beneficial to think of the PI and the SI and consider which is more important to the concept of 'heat'. I suppose that, in the case of heat, one could ignore the phenomenal component and say that heat really is only increased molecular motion. As I hope my example above shows, however, this does not seem entirely intuitive. I am not sure, in any case, that the experience of pain is a close analog to this one. Suppose that the phenomenal experience of having a particular pain is found to correspond to a particular brain state. Jane, a cutting edge neuroscentist, is satisfied that this is the case. But, suppose that Jane miraculously finds herself in a world slightly different from this one. Jane thinks she's still on earth, but, upon going to the lab and stimulating her brain in such a way that she is in the state that she had identified as 'pain', she finds that her phenomenal experience is actually quite pleasant. It does not seem that this is a world in which pain is actually pleasant. Rather, it seems that Jane would admit to having misidentified pain as being a particular brain state > I have a question regarding Kripke's characterization of heat on pg 133 > and how it relates to pain. Kripke says, "Here heat is something which we > have identified (and fixed the reference of its name) by its giving a > certain sensation."(131) This would seem to pick out the primary > intension of heat. But, Kripke points out, we have discovered a > posteriori that heat is just molecular motion. Given that heat is just > molecular motion, it must be molecular motion in all counterfactual > worlds. The primary intension of heat seems to be contingent given what > we know about the secondary intension. So if someone where to say that > light produces in them the same thing we feel when we feel heat, we would > not say that light is heat but that the stream of photons produces a > *sensation* of heat. > > If we were to discover a posteriori that pain is certain C fibers firing > in the brain, it seems like that would become the secondary intension of > pain, the way heat is molecular motion is the secondary intension of heat. > If this was the secondary intension of pain then it would fix pain to be > "the firing of certain C fibers" in all counterfactual worlds. Thus, if > someone where to say, light produces in me the same sensation that you > feel when you feel pain we would not say that light is pain but that the > stream of photons produces a *sensation* of pain. > > If one was to argue that the *only* difference between heat and pain is > that we have already discovered that heat is explainable in terms of > molecular motion while we have not yet discovered that pain is explainable > in terms of the firing of C fibers, it seems like the cases are relevantly > similar. What do you think? > -Rachael > From owner-modality@LISTSERV.ARIZONA.EDU Fri Feb 26 23:28:07 1999 Date: Fri, 26 Feb 1999 23:27:54 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Re: Kripke and the primary intension of "pain" To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO In reply to Rachael and Anthony. Rachael writes: >If we were to discover a posteriori that pain is certain C fibers firing >in the brain, it seems like that would become the secondary intension of >pain, the way heat is molecular motion is the secondary intension of heat. >If this was the secondary intension of pain then it would fix pain to be >"the firing of certain C fibers" in all counterfactual worlds. Thus, if >someone where to say, light produces in me the same sensation that you >feel when you feel pain we would not say that light is pain but that the >stream of photons produces a *sensation* of pain. > > If one was to argue that the *only* difference between heat and pain is >that we have already discovered that heat is explainable in terms of >molecular motion while we have not yet discovered that pain is explainable >in terms of the firing of C fibers, it seems like the cases are relevantly >similar. What do you think? That's an interesting point. I think you're right that *if* we could wholly explain pain in terms of C-fibres, then we might well identify pain (in secondary intension) with C-fibres. The relevant questions are (1) How likely is this explanation, (2) What would follow from the identification? Re (1), I think Kripke is more or less assuming the epistemic gap between physical states and phenomenal states, and is seeing what follows from that. He thinks it's conceivable (or "seems possible") that one could have an arbitrary physical state without a pain, for example. If so, then there already may be a principled explanatory gap between physical and phenomenal. To close that gap, presumably one would have to argue that there isn't really an epistemic gap after all -- either by arguing that there is an a priori entailment from physical to phenomenal after all, or by arguing that such an a priori entailment isn't required for reductive explanation. With heat, on the other hand, it's not clear that there is the same sort of epistemic gap. If we conceive of heat as "the cause of heat sensations", and demonstrate that molecular property M is the cause of heat sensations, then we've arguably explained heat in terms of M. Well, maybe not quite, as heat sensations themselves haven't been explained, but we set that aside as a problem for later. Leaving that aside, we can say that M explains heat modulo consciousness (i.e. modulo problems with explaining heat sensations). Similarly, one has an a priori entailment from the physical story to the heat story modulo heat sensations -- no epistemic gap, except for the sensation gap. When it comes to heat, we don't let the sensation gap bother us, as we're concerned with the external phenomenon. But with pain itself (or pain qualia), it's the very sensation we're concerned with, so we can't set it aside in this way. So there is still the apparent epistemic gap. Of course one could try to argue that there is no epistemic gap, presumably by rebutting the epistemic arguments we considered a few weeks ago. I take it that Kripke thinks that will be an uphill battle, though. In effect, he is here more concerned with the opponent who accepts that there is an epistemic gap but denies that there is an ontological gap. (Arguably the early identity theorists were like this, and many or most contemporary materialists are like this. That's the position I'm calling type-B materialism.) For that opponent, presumably things won't be so easy. Incidentally we can put the points about epistemic gap and explanation themselves in terms of the 2-D framework. When it comes to explanation phenomenon P, it's arguably the primary intension of "P" that's central to characterizing what needs to be explained. To explain heat, for example, we essentially have to explain what causes heat sensations. To explain water, we explain watery stuff. And so on. The SI doesn't emerge until post-explanation. So something similar appies to consciousness. An explanation will need to entail the PI facts about consciousness; or, we'll need an epistemically transparent (a priori) entailment from physical to phenomenal. We seem to get that sort of entailment more or less when a concept's PI is functional or structural (e.g. the heat case is mostlyt functional, and the water case is functional/structural). But if the argument re conceptual analysis is right, it's not clear that there is a functional/structural analysis of the PI of phenomenal concepts. If so, the epistemic gap looms, and explanation has a hard time getting off the ground. (2) Some think that explanation isn't required for identity, and that maybe we could postulate an identification of pain with C-fibres (and of phenomenal with physical states more generally) without a prior explanation. Maybe correlation would give could grounds for the identity, for example. And some philosophers (e.g. Block, Papineau) have argued that identities don't need explanation. If we did this, we could identify pain in SI with C-fibres, and so on. But even here, I think there would be problems to overcome. Essentially, even if we can pin down the SI of "pain" (etc.) in physical terms, there are still problems with the PI. One can argue that the very conceivability of zombies and the like (i.e. their broad epistemic possibility, in the considered-as-actual or PI sense) suggests that there is a world in the vicinity (by the 2-D analysis). If the materialist is right about the SI of phenomenal concepts, then this world shouldn't be described as one without pain (when considered as counterfactual), since the SI of "pain" will be physically specifiable and so will apply to this world. But it will still be a world in which the PI of "pain" isn't present. So there is *something* the world will lack -- essentially the property corresponding to the PI of "pain". We might think of this as the "mode of presentation" of pain. Even if we identify the *referent* of "pain" with a physical state, there is still the "sense" (or the reference-fixing property or the PI) to worry about -- the way that physical state is presented to us, if you like. And if the epistemic intuitions and the 2-D framework are correct, that will correspond to a nonphysical property in its own right. The moral of all this is familiar, I think. To save materialism, one has to either deny the epistemic intuitions, or deny the 2-D analysis of a posteriori necessity. Either strategy remains open to some degree, but a materialist has to do a lot of hard work in biting the bullet on either of them. Anthony writes: >Kripke urges that, once we have identified heat as molecular motion, we >have fixed the reference of 'heat'. As such, 'heat' would still refer to >molecular motion if it happened that our nerve endings were different such >that we actually had the sensation of cold when exposed to increased rates >of molecular motion. I am not sure that I agree with this intuition-- it >seems that Kripke is assuming the primacy of secondary intensions in this >case. Suppose that one miraculously were to wake up in a world that is >mostly like this one. In fact, it is, in many ways, indistinguishable from >this one. The sun on ones skin feels warm and snow feels cold. Suppose >that some time after relocating to this world one were to discover that >heat actually corresponds to reduced molecular motion. The laws of physics >in this world, however, are such that things behave exactly as they do in >this world. It is not clear to e that one would, in this case, say, "Oh, I >guess that what I thought was heat is not actually" upon discovering that >heat is not more rapid molecular motion. It seems that this is a case >where it is beneficial to think of the PI and the SI and consider which is >more important to the concept of 'heat'. Hmm, tricky. This case is complicated by your hypothesis that one actually wakes up in this world. In effect, you are considering the world as actual (wondering what we would say if we found ourselves in it) rather than considering it as counterfactual. I think you're arguably right that if we found ourselves in this world, in which a different property R was causing heat sensations, we'd be tempted to say that R is heat. But that is arguably compatible with Kripke's framework, on which the PI of "heat" is "the cause of heat sensations" or some such. (There are tricky issues here, regarding what one would say on first arrival in such a world vs. later on, etc. After a couple of years there, it seems pretty obvious one could correctly call R "heat". Does that apply even on first arrival? Just say one's nerve endings are tampered with so that temperatures below zero make one feel hot. Will one then correctly say that those temperatures are hot temperatures, or will you instead correctly say that cold temperatures are now making you feel hot. I can see both temptations, but I think there's a good case for the latter. That fits the idea that there's at least a time lag, and that the PI of "heat" is more like "what has typically causes heat sensations in me (and my community)", or some such. But anyway, Kripke's central claim concerns the SI of "heat" rather than the PI. To get at that, we have to consider the world as counterfactual, rather than imagining waking up in it. From our perspective here, is that world (a world where -40 degree temperatures cause hot sensations) a world where -40 degrees constitutes heat. or a world where cold temperatures make people feel hot. Again tricky, but i can see Kripke's case for the latter. At least, the intuition for calling the low temperatures "hot" isn't as strong as when one thinks of waking up in the world (i.e., as when one considers the world as actual). I can still see a case for going both ways, though, which suggests that the SI of "heat" is arguably somewhat ambiguous. >I am not sure, in any case, that the experience of pain is a >close analog to this one. Suppose that the phenomenal experience of having >a particular pain is found to correspond to a particular brain state. >Jane, a cutting edge neuroscentist, is satisfied that this is the case. >But, suppose that Jane miraculously finds herself in a world slightly >different from this one. Jane thinks she's still on earth, but, upon going >to the lab and stimulating her brain in such a way that she is in the >state that she had identified as 'pain', she finds that her phenomenal >experience is actually quite pleasant. It does not seem that this is a >world in which pain is actually pleasant. Rather, it seems that Jane would >admit to having misidentified pain as being a particular brain state Interesting. Your intuition about the case seems pretty plausible. It seems unlikely that Jane would correctly call this a situation in which pain is pleasant. Rather, it would be a situation is which she isn't feeling pain. Of course this again is a point about PIs as described, not SIs. The moral of at least this scenario is that (if the situation is possible and if your intuition about describing it is correct) even if the SI of "pain" can be identified with a given brain state, the PI can't be. (In this world considered as actual, the brain state is present but pain isn't.) So there is still ths issue about identifying the PI of "pain", as above. One could also try to make a similar point about the SI of "pain". Let's take your Jane scenario and considered it as counterfactual instead of as actual. Should we (from our vantage point here) describe this as a situation in which Jane is having a pain (it just feels different), or one in which she is not having a pain at all. There is at least some pretty strong intuition that it should be the latter, I think. If so (and if the situation is indeed possible), then the identification of the SI of "pain" with the brain state will be problematic, too. In effect the relation between these two cases (considering a pain situation as actual and as counterfactual) parallels Kripke's observation re the semantics of pain. In effect, the fact that we have similar intuitions about saying whether a given scenario involves "pain", irrespective of whether it is considered as actual or as counterfactual, suggests that the PI and SI of "pain" (unlike "water", "heat", etc) are more or less the same as each other. I don't think that point is vital to making the anti-materialist case (the argument would still work even if PI and SI are different, as e.g. above and in the M&M notes), but it's an interesting point re the way the concept works. Of course the materialist can try to reply by finding some brain state such that one couldn't imagine Jane having that brain state without pain. Maybe C-fibres are a particularly weak sort of state to pick, as we can just imagining wiring the brain so that C-fibres make her go "wow, this feels good", etc. So a better case for the materialist might be one where we have a full functionally-specified state, such that she'll always behave as if she's in pain, etc. Here the anti-materialist will still argue that it's conceivable that she have this without being in pain (cf. zombie cases), but at least it will be trickier for them. Again, the materialist either has to deny the epistemic intuitions (here by going for a functional analysis of the concept of "pain") or reject the 2-D framework (by arguing that the conceptually coherent scenario we are considering as actual doesn't correspond to a possible world at all). --Dave. From owner-modality@LISTSERV.ARIZONA.EDU Fri Feb 26 10:33:53 1999 Date: Fri, 26 Feb 1999 11:31:34 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Timothy J Bayne Subject: epiphenomenalism To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO I want to elaborate on the line I was developing on epiphenomenalism (E), and pretty much retract it. My reply to someone this week is to myself - hope that counts. AT least no one else can accuse me of misrepresenting them. On tuesday Anthony asked why anyone would be tempted towards E. Isn't it obvious that the reason I drive out to watch the sunset is that I like the glorious red qualia that it gives me? Don't I like chocolate, and eat it, because of that chocolatey taste? Qualia seem to be causally efficacious. Here's how I tried to pump the E intuition. Suppose that we accept the possibility of qualia inversion. Although you and I both like watching orange sunsets, the qualia that they produce in me are unlike the qualia they produce in you. Suppose that what it's like for me to see red is the same as what it's like for you to see green; and what it's like for you to tasta ice-cream is what it's like for me to eat chocolate, and vice-versa. Inverted spectra intuitions are, I take it, rather easy to elicit, at least from those who are new to these issues. (Aged and battle-weary philosophers of mind seem not to share them.) I'm eating chocolate because I like its taste (note the reification of tastes). If it doesn't have the same taste to you, but you still like it eating, the you are eating it for a different reason. But how does inverted spectra provide an argument for E? My idea was that the very redness of red makes no difference to what it does, all that matters is that this experience be able to interact with my motor center in the right kind of way, and be the same kind of experience that I have when I see strawberries, fire-engines, and so on. The qualia that I have when I see red things can't be causally efficacious, since you experience different qualia, but act in the same type of way. But this now seems to me to be a bad argument, or at least it's too quick. Two things can have the same causal effects even though they are different things. Thus, in certain conditions a white biliiard ball will cause a window to break, and a red billiard ball will cause the window to break under the same conditions. But a red billiard ball has different properties from a white billiard ball. To rehabilitate the argument we need to talk about whether or not certain *properties* are causally efficacious. Rather than reify qualia, we need to talk about things that have the property of giving one a red appearance. But even here the argument isn't obvious. It may be true that property x and property y have the same causal power in functional system F, but that means that x and y do have causal power within F - not that they have *no* causal power. (I guess the real intuition pump for epiphenomenalism is the zombie argument - but why would you think zombies possible attractive unless you were already an epiphenomenalist? ) Let's back up. The E claim is that the property of having qualia of a certain type are is not causally efficacious. I now think the motivation for it comes from the idea that the causal efficacy of a state/object only depends on its basic physical property, its 'syntax': weight, shape, magnetic force, whatever. (I think that both Scott and Thony said this, or things like it.) From this perspective, *most* properties are epiphenomenal. We say that the age of a vase causes it to be brittle, or valuable, but really these comments are just short-hand way of referring to the causal efficacy of the physical properties of states. The age of a vase causes it to be valuable = people believe that it is old, they believe that old things are valuable, both of these beliefs are physical states in the brain, and in virtue of their physical/syntactic properties, these states cause certain actions such as offering to pay a lot of money for the said vase. Thus, we might say that age, or value, or qualia of certain kinds are causally efficacious properties in a *derivative* sense. Only synatictical properties are causally efficiacious in a primitive or basic sense. It seems that arguments for epiphenomenalism involve fairly large-scale metaphysical commitments re the nature of causation, and the structure of the mind. In short, the only motivation that I can see for E comes from the idea that all causation is bottom up. Talk of people, coprorations, and countries, doing things is really just short-hand for the casual activity of basic material states. The debate over free will starts to creep in the backdoor here, so perhaps I'll shut up at this point. Tim Timothy J. Bayne RM. 213 Social Science Department of Philosophy University of Arizona Tucson, AZ 85721 USA Hm ph. (520) 298 1930 From owner-modality@LISTSERV.ARIZONA.EDU Fri Feb 26 16:51:04 1999 Date: Fri, 26 Feb 1999 17:49:21 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Erik A Herman Subject: TCM To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO There's one thing I need clarification on. One of the objections to naturalistic dualism was: There are properties essential to the physical constitution of the world that are not accessible to physical investigation...the zombie world seems physically identical while being physically different. It seems that your simple response is that these extra properties are phenomenal. But I thought the objection was that there are might be extra PHYSICAL properties (that are presently not available to us). Likewise, regarding your example of electrons that have hidden (protophenomenal) properties-- that we'd still call an electron that lacked these properties an electron. But again, wasn't the objection that there are extra physical properties (that missing, would presumably make it not an electron)? The electron seems an appropriate example for I take this objection to be saying. Namely, that there are more (physical but wierd) properties that have yet to be understood with respect to the electron. Some of the physical properties that are presently being asserted are things like, "if we know this we can't possibly know that (and vice versa)" This isn't a phenomenal property but if it's a physical one it certainly isn't explanatory. This is the kind of stuff I would expect the objector to be referring to when talking about hidden physical properties-- the ones we don't "get", but not phenomenal ones. So the question is, could these weird physical properties be the protophenomenal ones you propose? -Erik H. From owner-modality@LISTSERV.ARIZONA.EDU Sun Feb 28 00:54:14 1999 Date: Sun, 28 Feb 1999 00:54:01 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Re: epiphenomenalism To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Tim and Erik H. raise interesting issues re epiphenomenalism and re the possibility of new intrinsic physical properties. These bear on the question of what the conclusion of the anti-materialist argument should be. I present it in the book as property dualism, and leave open just what sort. I now think that the conclusion of the argument should be a disjunction of three positions: epiphenomenalism, interactionism, or "panprotopsychism". This issue isn't really core to the modal issues we're focusing on, but I'll say a bit about it anyway. Epiphenomenalism is the idea that the phenomenal properties naturally supervene on physical properties and have no effect on them. (One-way psychophysical laws.) Interactionism is the idea that there are two-way psychophysical laws: the physical properties affect the phenomenal properties and vice versa. Panprotopsychism is the "Russellian" position mentioned a number of times in the book, on which the phenomenal is constituted by the unknown intrinsic properties of the physical. I think the anti-materialist argument is actually neutral on these three, so the decision between them needs to be made on further grounds. I don't think there are yet grounds for a decisive preference in any direction. It seems to me, at least, that all have advantages and disadvanatages. Personally, I rank panprotopsychism first (most elegant but also the least clear), epiphenomenalism second (arguably the most conservative, but inelegant), and interactionism third (would require adjusting our view of physics, but not out of the question). But I have days when each appeal. Tim raises questions about epiphenomenalism, and Erik (in effect) about panprotopsychism. I'll say a bit in reply to Tim in the rest of this message and to Erik in the next. Re epiphenomenalism first. Tim says quite rightly that the mere possibility that one can keep Q constant while varying P doesn't show that P is epiphenomenal with respect to Q. It might be that P1 and P2 can both cause Q in different circumstances. So if there is an argument for epiphenomenalism, that's not it. I take it that the argument for epiphenomenalism is more indirect. It proceeds via an argument for property dualism, plus considerations about the causal closure of the physical. If qualia are nonphysical, and if there is a physical sufficient cause of any physical effect, then it seems that qualia are redundant. I take it that few people have the view that qualia are redundant as a premise, but one can see how one could be led to it as a conclusion. E.g., if one has accepted an anti-materialist argument and wants to hold on to the causal closure of the physical. Just say one has causal closure and property dualism. Then presumably for behavior B, there will be a complete physical explanation P. In our world, there are qualia Q around too. But in other worlds, one has behavior B caused by P with different qualia Q' around, or with no qualia around at all. Even so, it seems that P is sufficient for a complete causal explanation of B -- it's not as if behavior is produced by a different mechanism in each case (as was the case in the causal explanations mentioned earlier). If so, then it seems that Q is causally redundant in the actual world. Barring an odd kind of overdetermination, Q seems to come out epiphenomenal. Ways out of this: (1) Reject causal closure (go for interactionism). (2) Reject the modal intuitions (deny the possibility of inversions or zombies). (3) Accept overdetermination (both the physical processes and the qualia cause the behavior, redundantly). (4) Go for the Russellian/panproto view on which qualia are still causally relevant by virtue of their status as the categorical basis of physical dispositions. The fourth would be my own preferred strategy here, as discussed around pp. 153-55 of the book. I think (3) is conceptually problematic, and obviously I don't accept (2). But all are at least respectable alternatives to consider. I do think epiphenomenalism is a respectable alternative too. We have very strong intuitions of causal relevance for qualia, but those can be explained on Humean grounds: what we really know is that there is a constant regularity between qualia and behavior, from which it is natural to infer causation. But all our evidence is compatible with epiphenomenalism. So if there is reason to reject epiphenomenalism, it has to be based on further grounds, I think. A couple of comments re Tim: >causal power. (I guess the real intuition pump for epiphenomenalism is the >zombie argument - but why would you think zombies possible attractive >unless you were already an epiphenomenalist? ) I've heard this said before, but personally I think it gets the order wrong. Hardly anyone is "already an epiphenomenalist" in this context, but plenty of people find zombies conceivable. Maybe there is an argument from that *to* epiphenomenalism, although that is contestable. But in any case plenty of nonepiphenomenalists seem to accept the conceivability of zombies -- e.g. all the type-B materialists (Tye, Loar, Levine, Hill, Yablo, etc), and quite a few interactionists and others. More generally, I think one can assess the epistemic arguments against materialism prior to one's commitment on the metaphysical issue re epiphenomenalism. >Let's back up. The E claim is that the property of having qualia of a >certain type are is not causally efficacious. I now think the motivation >for it comes from the idea that the causal efficacy of a state/object only >depends on its basic physical property, its 'syntax': weight, shape, >magnetic force, whatever. (I think that both Scott and Thony said this, or >things like it.) From this perspective, *most* properties are >epiphenomenal. We say that the age of a vase causes it to be brittle, or >valuable, but really these comments are just short-hand way of referring >to the causal efficacy of the physical properties of states. The age of a >vase causes it to be valuable = people believe that it is old, they >believe that old things are valuable, both of these beliefs are physical >states in the brain, and in virtue of their physical/syntactic properties, >these states cause certain actions such as offering to pay a lot of money >for the said vase. I'm not sure that this is getting at the real crux of the pro-epiphenomenalist argument in the qualia case. The argument you're giving will apply for *any* high-level property, including those which are clearly logically supervenient. I think such arguments will generally be suspect, as there's no reason to think that both micro properties and logically supervenient macro properties can't simultaneously be causally relevant (as e.g. with molecules and billiard balls being simultaneously causally relevant). Of course there is more to say there. But in any case the special force of pro-epi considerations in the qualia case derives from the apparent failure of logical supervenience. That doesn't hold in those other cases, which have no apparent epistemic gap on close examination. But there is at least an apparent gap in the qualia case, which can be extended into an argument against logical supervenience and against materialism, etc, until (under certain assumptions) one gets to epiphenomenalism. Of course there are contestable steps there. But take away the epistemic arguments, and I think the distinctiveness of the considerations re epiphenomenalism here are removed; and conversely, it's the apparent epistemic gap re qualia that gives pro-epi considerations here their distinctive force. >In short, the only motivation that I can see for E comes from >the idea that all causation is bottom up. Talk of people, coprorations, >and countries, doing things is really just short-hand for the casual >activity of basic material states. The debate over free will starts to >creep in the backdoor here, so perhaps I'll shut up at this point. Well, maybe that is the only motivation for a *general* epiphenomenalist thesis, re arbitrary high-level properties. I think that motivation would be suspect, though it's an interesting issue in its own right. But in any case I think the motivation for epiphenomenalism re qualia rests on distinct considerations. --Dave. From owner-modality@LISTSERV.ARIZONA.EDU Sun Feb 28 01:36:23 1999 Date: Sun, 28 Feb 1999 01:36:09 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Re: TCM (panprotopsychism) To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Now to Erik H.'s messages and the issues re "panprotopsychism". Erik wrote: >There's one thing I need clarification on. One of the objections to >naturalistic dualism was: There are properties essential to the physical >constitution of the world that are not accessible to physical >investigation...the zombie world seems physically identical while being >physically different. > >It seems that your simple response is that these extra properties are >phenomenal. But I thought the objection was that there are might be extra >PHYSICAL properties (that are presently not available to us). There are a few issues here. First we need to get clear on the nature of the new "hidden" properties. Perhaps this talk of "new" physical properties might suggest that they are just properties that will be invoked by a completed physics that we haven't yet discovered but eventually will, by scientific investigation and the like. Or perhaps that will never be discovered, but still have the same sort of general shape as existing properties. That would in effect by the "appeal to new physics" strategy considered in Chapter 3 of the book, which I argue there doesn't work. The trouble is that the new properties will still be introduced in a structural/dispositional way, and there will be no a priori entailment from these structural/dispositional properties to phenomenal properties, at least if the general epistemic considerations re qualia are valid. Those considerations seem to apply to any structural/dispositional properties, even as yet undiscovered one. The move I'm suggesting here is a more radical one. Maybe in imagining a "physically identical" world, we are imagining a world that is the same in all structural/dispositional respects, but not the same in all *intrinsic* respects. That's to say, the "accessible" properties that one holds constant are the structural/dispositional ones, but maybe there are further "hidden" intrinsic properties that are crucial to the physical world. And maybe those intrinsic properties entail the existence and nature of qualia. So we now at least have some sort of property dualism, i.e. a dualism of "accessible" dispositional properties (I'll leave out "structural" from now on) and of "hidden" intrinsic properties. One might argue that this is out of the frying pan straight back into the fire, as the new protophenomenal properties seem to be epiphenomenal to the dispositions we know and care about. After all, the structural/ /dispositional properties seem to make up a seamless causally closed network. If the hidden intrinsic properties are indeed epiphenomenal with respect to the others, then it seems we at most have a terminological variant of the original property dualism (we're calling the new fundamental properties "hidden physical properties", but it doesn't really change the shape of the view). But I think there is one way of looking at this view so it at least has an interestingly different shape. The interestingly different view is the one that locates the "hidden" intrinsic properties as the categorical bases of the dispositional properties. It's often held that dispositions need categorical basis -- where one has a disposition to cause Y, one needs some underlying intrinsic property X that actually does the causing of Y. E.g., a glass's fragility is a disposition, and its categorical basis is the molecular structure of the glass or some such. Arguably, that applies all the way down to microphysics. Russell argued that physics tells us all about the *structure* of causal relations in the physical world, but it doesn't tell us what ultimately does the causing. Mass, for example, is treated by physics as a sort of disposition (the disposition to resist acceleration, be attracted by gravity, etc), but physics doesn't say anything about the basis of those dispositions -- what mass is "in itself". And more generally, one might argue that matter needs to have some intrinsic properties, not just dispositions, but physics is silent about that intrinsic nature. So the interesting hypothesis is that we *need* certain "hidden" properties of the physical world to serve as the intrinsic basis of physical dispositions. These intrinsic properties will be what underlies the mass disposition that physics talks about, the charge dispositions, etc. Physical observation doesn't tell us anything about these properties, as we know about them only by their effects, and as far as the effects are concerned, arguably, any intrinsic basis will do. The hypothesis that physics needs hidden intrinsic properties is controversial, but many philosophers have found it attractive. And note: *this* sort of hidden physical property is not epiphenomenal with respect to physical dispositions. The hidden properties here serve as the basis of physical dispositions, and arguably it is these that ultimately do all the causing. In fact the world might be seen as ultimately consisting in a bunch of causal and other nomic relations between a whole lot of intrinsic properties, which reveal themselves to us only structurally. So the hidden properties are at the very basis of the causal network. So, now we have an interesting possibility re qualia. Maybe it's the case that the structural/dispositional physical properties don't entail qualia, but perhaps the hidden intrinsic properties do? After all, we have no idea what those properties are like. Maybe they are themselves phenomenal properties, or some other more fundamental properties that entail phenomenal properties -- what I call protophenomenal properties. If so, then when imagining a zombie world, we're imagining a world with the same dispositional properties but different (or no) hidden intrinsic properties. So the conceivability of zombies is compatible with entailment by dispositional plus intrinsic physical properties. The cost of course is the presence of fundamental phenomenal or protophenomenal properties at the very fundamental level of the physical world. This is the position that I call "panprotopsychism". (In the book, "Russellian monism" or some such.) In a way you can see this view as motivated by the presence of two metaphysical problems simultaneously: (1) What are the unknown hidden intrinsic properties of the physical world?; (2) How can we locate intrinsic phenomenal properties vis-a-vis the physical world? On having things put this way, it's natural to try to solve both at once: if the hidden intrinsic properties are themselves qualia (or protoqualia), then we both have a hypothesis about their nature and an integrated location for qualia in the physical world, as the categorical basis of physical dispositions. Of course the view is highly speculative and has a number of problems it needs to face up to. But it has clear attractions also, not least the promise of an integrated, relatively monistic world view. Anyway, with all this in hand, I'll get back to Erik's questions. Erik asked first, why can't the new properties just be ordinary physical properties, rather than phenomenal or protophenomenal properties? I think the answer is: first, the properties need to be intrinsic rather than dispositional, to avoid the usual problems with entailment. Second, the properties need to serve as the categorical basis of physical dispositions, to avoid epiphenomenalism. And third and most important, in order to deal with the qualia problem, the new properties have to either themselves be phenomenal properties, or they have to collectively constitute phenomenal properties -- i.e. they have to be phenomenal or protophenomenal. (Where protophenomenal is just a label for novel fundamental properties that are not themselves phenomenal but collectively constitute phenomenal properties.) If the properties don't have this intimate relation to phenomenal properties, we'll still have the usual problems with zombies and the like. I.e. we'll still have the coherence of zombie worlds with the same accessible and hidden properties but no qualia; one could know all about the accessible and hidden properties without knowing about qualia; etc. But if they are phenomenal or protophenomenal, then it won't be coherently conceivable to have them instantiated without qualia, and in principle someone who knew all about the properties in question will be able to know about qualia. So these will need to be special sorts of properties. >Likewise, regarding your example of electrons that have hidden >(protophenomenal) properties-- that we'd still call an electron that >lacked these properties an electron. But again, wasn't the objection that >there are extra physical properties (that missing, would presumably make >it not an electron)? This gets at a question I've been avoiding until now. In what sense (if any) do these hidden properties count as *physical* properties, and so on what sense (if any) is the panprotopsychist view a variety of physicalism? In the book, I mostly treat the view as if it is not a version of physicalism, but even there I say the issue is mostly terminological, and what matters is the shape of the view. I think the issue comes down to: in what sense (if any) is a world with the same dispositional but different "hidden" properties "physically identical" to ours? In particular, does such a world truly count as containing mass, charge, electrons, etc? The neutral answer is: in the vicinity of mass, charge, electrons, etc, there are dispositional properties and intrinsic properties. The world in question has the dispositional properties but not the intrinsic properties. But one might still press, and say -- but does it *really* contain mass, electrons, etc? This seems to be a largely terminological issue. We're considering a certain world -- the world with the same dispositional shape, different intrinsic properties -- as counterfactual, and evaluating the referent of "mass", "electron", etc, there. This comes down to an intuitions about the secondary intension of "mass", "electron", etc. I think it is highly plausible that the *primary* intension of these terms goes with certain dispositional properties (we fix reference to mass as "what plays the mass role", etc). But do the *secondary* intension go with the disposition or with the underlying basis? In the book, I say that my intuition is that the world in question contains electrons, mass, etc. After all, the entities in question look like electrons, quack like electrons, interact like electrons, etc, and there's an intuition that that's enough to make them electrons. If so, the SI of "electron" (etc) goes with the disposition. But I can also see the opposing intuition. This would be the semi-Kripkean intuition that to be an electron (or mass), you have to have the "underlying nature" of electronhood (or mass), and here, that underlying nature is given by the intrinsic property. If so, something with a different intrinsic nature won't count as an electron, and the SI of "electron" goes with the intrinsic property, not the disposition. If one takes the former view, then arguably the intrinsic properties are not physical properties, as they aren't the referent of paradigm physical terms like "mass", etc. And the zombie world in question will be correctly described as having mass (etc), so it will be a physical duplicate of our world. On this view, one might argue that the panprotopsychist view is not a variety of physicalism, as the physical properties are structural/dispositional, not intrinsic. If one takes the latter view, then arguably the intrinsic properties are paradigm physical properties, as they are the referent of paradigm physical terms like "mass", etc. And the zombie world in question won't be correctly described as having mass (etc), so it won't be a physical duplicate of our world. On this view, one might argue that the panprotopsychist view is a variety of physicalism, where our fundamental physical properties turn out to be protophenomenal. Ultimately (as I think I say in the book) I think the issue here is largely terminological, and nothing much turns on the semantic intuitions. We can say that the panprotopsychist view is in a sense physicalist, and in a sense property dualist. It certainly has a dualism of dispositional and intrinsic properties; the only question is whether to call them "physical". I can see the case for going both ways, but in any case the shape of the view is what matters. And I think the shape of the view is pretty clear. If one takes the version of the view on which the intrinsic properties are physical properties, then arguably we have "saved" physicalism in some sense from the anti-materialist arguments. But I think it's clear that the view that results is still quite radical, with new fundamental protophenomenal properties responsible for the emergence of qualia, and in some ways more in the spirit of property dualism than physicalism. At the end of the day it's the view rather than the label that matters. >So the question is, could these weird physical properties be the >protophenomenal ones you propose? Yes, that's exactly it. I intended to gesture toward that idea in the passage you're talking about, but maybe the discussion wasn't too clear. (The view first comes up in considering the physicalist "objection" on pp. 136-38. On p. 138, I mention the ensuing possibility that physical entities have an intrinsic protophenomenal nature as a view that I'll return to. This was in effect a pointer to the discussion on pp. 153-56, which is the main discussion of the panprotopsychist view in the book, though it's also discussed in the context of an informational view on pp. 301-8.) But anyway, the view that fundamental protophenomenal properties serve as the "hidden" intrinsic categorical basis of fundamental physical dispositions is a very important view, no matter how one classifies it. It's a little on the wild and crazy side, but personally I find it the most elegant resolution of the mind-body problem if it can be worked out. It's compatible with all the epistemic intuitions re qualia, and at the same time is compatible a causal role for qualia *and* with the causal closure of the physical (qualia don't so much "change" the physical causal network as lie at its very basis). And the view has more of a monist flavor than the epiphenomenalist and interactionist alternatives, without the "dangling" entities that these views have. So there's a lot to be said for it. On the negative side, one has to solve the problem of just how all these fundamental protophenomenal properties in physical systems come to constitute the (unified, bounded, homogeneous, structured, coherent) phenomenal experience we know and love. (See pp. 305-8 for a little discussion of this.) That's a highly nontrivial problem, and it's not obvious that it's solvable. But it's not obvious that its unsolvable, either. So it's a very interesting topic for further research. --Dave. From owner-modality@LISTSERV.ARIZONA.EDU Sun Feb 28 08:17:49 1999 Date: Sun, 28 Feb 1999 09:15:27 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Timothy J Bayne Subject: Concepts and Properties To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO I asked Josh and Thony the following question last week but our discussion got cut off so I thought I'd open it up. My query is this: in big picture terms, what exactly is the disagreement between Kripke and Chalmers. Here's how I think it goes - but I'm not clear that I've got it right. Kripke elicits a bunch of intuitions about whether we'd call certain things certain names under certain conditions;, i.e. if we found out that x was different in structure, would we still call it x. Kripke, or at least Kripkeans, take the results of these intuition pumps to tell us something interesting about *the world*, or about *properties*. They tell us that certain properties (x, y) are metaphysically linked: although there are possible world in which x and y come apart, they are identical (?) in all the metaphysically accessible worlds. Chalmers, I take it, responds thus: "the intuition pumps are persuasive, but they don't show what you (Kripkeans) take them to show. They tell us something about *language*, about our *concepts* and , and nothing about the world or about the structure of properties (x, y). Kripke's analysis doesn't show that the structure of possibility is other than what we have thought, all it shows is that we were misdescribing it (TCM 134)." Two questions: (1) Is this the right way to characterize the dispute? (2) Isn't the connection between concepts and properties more intimate than Chalmers' line allows? If a concept applies in one scenario and not in another, doesn't that imply that the two scenarios do not share all of the same properties? Don't we generally decide whether or not we are dealing with the same property in different cases by deciding whether or not the same concept applies? If we accept the move from conceivability to logical possibility, then (I think) we are accepting that there is definitely some connection between concepts and properties: a coherent use of concepts implies that the state of affairs (the articulation of properties) picked out by the concepts is logically possible. No doubt there is much to say on how properties and concepts are related, but I'll leave this here. Tim Timothy J. Bayne RM. 213 Social Science Department of Philosophy University of Arizona Tucson, AZ 85721 USA Hm ph. (520) 298 1930 From owner-modality@LISTSERV.ARIZONA.EDU Mon Mar 1 05:06:16 1999 Date: Mon, 1 Mar 1999 05:06:01 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Re: Concepts and Properties To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Tim writes: >I asked Josh and Thony the following question last week but our discussion >got cut off so I thought I'd open it up. My query is this: in big picture >terms, what exactly is the disagreement between Kripke and Chalmers. >Here's how I think it goes - but I'm not clear that I've got it right. Interesting question. It's not entirely clear to me that there is a clear disagreement. Kripke doesn't use the 2-D framework, but it's not clear that he says anything that is strictly incompatible with it. And we have seen that Kripke seems to endorse a certain link between conceivability and possibility, suggesting that when one apparently conceives of a world, there is at least a possible world in the vinicity, though it may be misdescribed. That view can be translated into the 2-D framework too. I suppose one central disagreement is that I am more sympathetic with descriptive views of reference than he is, and this shows up in my explicit use of the notion of a primary intension, something he stays away from. But one might argue that the notion of a primary intension is at least implicit is some of Kripke's discussion, as when he evaluates what the referent of our terms would be if we discovered certain epistemic possibilities to be actual. I suppose that one difference is that Kripke sees epistemic matters such as apriority as *purely* epistemic in some sense and not modal, whereas I think it is often useful to understand them in modal terms. On my view, one can see epistemic possibilities as possible worlds in their own right, where our terms have a distinctive pattern of application to those worlds; and there is a corresponding notion of necessity, 1-necessity or primary necessity, which is in many ways just as important as the notion of subjunctive necessity (2-necessity) that Kripke concentrates on, and which has deeper ties to epistemic notions. Probably in the end these are the most important difference -- a difference of emphasis on PI vs. SI considerations, a willingness to see some sort of quasi-descriptive component in meaning, and a willingness to understand epistemic notions in modal terms. But these don't immediately manifest themselves in first-order issues or doctrines. It is more a different in emphasis, and in the lens through which one looks at the central issues. Personally, I like to think of the 2-D framework as a good way of analyzing and making sense of what Kripke is up to in N&N, rather than as any sort of radical revision of the views there. Of course Kripke might see things differently. >Kripke elicits a bunch of intuitions about whether we'd call certain >things certain names under certain conditions;, i.e. if we found out that >x was different in structure, would we still call it x. Kripke, or at >least Kripkeans, take the results of these intuition pumps to tell us >something interesting about *the world*, or about *properties*. They tell >us that certain properties (x, y) are metaphysically linked: although >there are possible world in which x and y come apart, they are identical >(?) in all the metaphysically accessible worlds. Chalmers, I take it, >responds thus: "the intuition pumps are persuasive, but they don't show >what you (Kripkeans) take them to show. They tell us something about >*language*, about our *concepts* and , and nothing about the world >or about the structure of properties (x, y). Kripke's analysis doesn't >show that the structure of possibility is other than what we have thought, >all it shows is that we were misdescribing it (TCM 134)." Well, I don't disagree with the claims you attribute to Kripke here. It's quite compatible with the 2-D framework to say that Hesperus is identical to Phosphorus in all possible worlds, and that the property of being hot is the property of having fast molecules in all possible worlds. So I can agree that the objects and properties in question are metaphysically linked here; or better, that there is just one object or property in each case, with two different terms picking it out. Of course I think there is more to the story than this. In making the above claims I am dealing with 2-possibilty and secondary intensions. One can also invoke 1-possibility and primary intensions, and explain the a posteriority of "Hesperus is Phosphorus" in terms of the fact that its PI is false in some centered worlds, and so on. But to say this is not to take back any of the above. It's true that I am also inclined to think that some of these issue are vague and depend on the terminology. For example, it's not obvious to me that the SI of "heat" really picks out the motion of molecules rather than the cause of heat sensations in all possible worlds. But again, this isn't to retract the claim about properties. It's just to say that the reference "heat" is somewhat ambiguous between two properties, the molecule property and the heat-sensation property (N.B. this isn't an issue about PI vs. SI; rather, it's an ambiguity merely in the SI). But like any ambiguity, this can easily be cleaned up or stipulated away, and after doing this, we'll have a perfectly good term, "heat" or "heat*" which works just the way kripke says it does, referring to the molecule property, and picking out that very property in all worlds. We can then say that heat (or heat*) is identical to molecular motion in all worlds. So I don't think the point about vagueness and ambiguity is central here. Deep down you are probably right that I am more deflationary about the metaphysics of properties and possibility here than Kripke is, but it's not obvious how to articulate that. Sure, the property of heat and the property of molecular notion are the same property. But the fact that "heat is molecular motion" is an a posteriori necessity here is largely due to our conceptual structure, with two distinct concepts that pick out the same property in different ways. Similarly, it's true that it's necessary that water is H2O, and that fact is entirely independent of us. But the fact that "water is H2O" is an a posteriori necesssity simply arises from the coreference of the terms and the 2-D structure of the concepts. Even here, it's not entirely clear what Kripke would disagree with, so I'm not sure that I've really isolated a substantive disagreement. As for the conceivability/possibility thesis you mention, Kripke shows every sign of agreeing with it, so its not clear that this is a disagreement. >Two questions: (1) Is this the right way to characterize the dispute? (2) >Isn't the connection between concepts and properties more intimate than >Chalmers' line allows? If a concept applies in one scenario and not in >another, doesn't that imply that the two scenarios do not share all >of the same properties? Don't we generally decide whether or not we are >dealing with the same property in different cases by deciding whether or >not the same concept applies? Well, of course I think there are two ways in which concepts can apply to scenarios: the PI way and the SI way (considering the scenarios as actual and counterfactual). We can take these cases one at a time. Taken in the SI way (considering scenarios as counterfactual): if a concept applies in one scenario but not in another, then the SI of the concept applies to one world and not to the other, so the worlds certainly have different properties (as SIs are properties in their own right). For example, the fact that the concept "water" applies to H2O in the H2O-world (considered as counterfactual) but not to XYZ in the XYZ-world (considered as counterfactual) goes along with the fact that they have different properties: H2O has the property of being water and XYZ does not! It will also be the case that we can determine whether we have same property X in two scenarios by seeing whether the concept "X" applies to those scenarios (considered as counterfactual). So there's nothing here with which I disagree. Taken the PI way (considering scenarios as actual): if a concept concept applies in one scenario but not in another, then the PI of the concept applies to one world and not to the other, so the worlds certainly have different properties (as PIs are properties in their own right). For example, the fact that the concept "water" applies to H2O in the H2O-world (considered as actual) but not to H2O in a world where H2O is a coal-like solid (considered as actual) goes along with the fact that they have different properties: H2O in the first world has the property of being watery stuff but in the second world does not! So your first claim will hold good even for PIs. The second claim is a bit trickier. Can we decide whether or not' we are dealing with the same property by seeing whether the same concept applies (to a scenario considered as actual)? Well, in a way yes, in a way no. The concept "water" will apply to first but not the second of the scenarios considered above. But arguably both scenarios contain the property of being water. If so, then the application of the concept "water" (in PI) is not a perfect guide to the presence of the property of being water (the property is present in the second H2O-world, even though the PI of "water" doesn't apply there). On the other hand, the application of the concept "water" to these scenarios (considered as actual) *will* be a good guide to the presence of a certain property, the property determines by the PI of water, i.e. the property of being watery stuff. The concept "water" applies to the first but not the second scenario (considered as actual), and the property of being watery stuff is present in the first but not the second. So even here, the application of a concept is a good guide to the presence of a corresponding property -- it's just that in this case, it's the PI property that's relevant, not the SI property. Let me know if I've misunderstood your claim here. >If we accept the move from conceivability to logical possibility, >then (I think) we are accepting that there is definitely some connection >between concepts and properties: a coherent use of concepts implies that >the state of affairs (the articulation of properties) picked out by the >concepts is logically possible. No doubt there is much to say on how >properties and concepts are related, but I'll leave this here. Right. On my view, the connection between concepts and properties is particularly intimate, more intimate than some opponents will allow. (So I'm not sure why you say above that that connection may be more intimate than *I* allow.) In particular, I think that when one has two a priori distinct concepts, there are two distinct properties in the vicinity, namely the two reference-fixing properties, or PI properties. (If "A = B" is a posteriori, the PI of "A" is a distinct property from the PI of "B".) This issue comes up a bit in the PPR symposium we'll be looking at shortly. Re moving from conceivability to possibility, you're right again. Of course I think there are two such moves in the vicinity: the move from 2-conceivability to 2-possibility and the move from 1-conceivability to 1-possibility. Either way there will be a link between concepts and properties. In fact, I think in both cases, the link between conceivability and possibility and the link between concepts and properties stand and fall together. To see this in a bit more detail: Thesis 1: 1-conceivability implies 1-possibility. Thesis 2: Distinct concepts yield distinct properties. I.e., if "A" and "B" are distinct concepts, in the sense that there are 1-conceivable scenarios to which "A" applies but "B" does not, then the PI of "A" and the PI of "B" are different properties, in the sense that they are not necessarily coextensive. Claim: Thesis 1 <-> Thesis 2. Proof: Left-to-right: Assume 1-conceivability implies 1-possibility. Let "A" and "B" be distinct concepts. Then there are 1-conceivable scenarios to which "A" applies but "B" does not (i.e., "A != B" is 1-conceivable). It follows that "A != B" is 1-possible, and there are 1-possible scenarios to which "A" applies but "B" does not. So the PI of "A" is not necessarily coextensive with the PI of "B", so these are distinct properties. Right-to-left: Assume that different concepts yield different properties. Let "A != B" be 1-conceivable. Then "A" and "B" are distinct concepts, so yield distinct properties, so the PI of "A" is a property distinct (and not necessarily coextensive with) the PI of "B". So there is a world to which the PI of "A" applies but the PI of "B" does not. So "A != B" is 1-possible. Much the same goes for the corresponding claims regarding 2-conceivability, 2-possibility, and SIs. One needs a little more to generalize the right-to-left proof for conceivability-to-possibility links in cases other than identity statements, but it is not to difficult. So, we can see that the sort of 2-D view to which I subscribe (and which i think is independently plausible) has a particularly strong link between concepts and properties (distinct PIs yield distinct properties, etc) and between conceivability and possibility (1-conceivability implies 1-possibility, etc), and that these two properties of the framework go together. It seems, then, that it is an opponent who will have to resist this strong link between concepts and properties, at least if they want to resist the move from (ideal) conceivability to possibility. We'll see some instances of that strategy when we discuss the papers by Loar and by Hill & McLaughlin in the PPR symposium. Essentially, these people reject the 2-D framework as a complete account of a posteriori necessity, reject the central claim that an a posteriori necessity must have a contingent primary intension, and deny the strong link between concepts and properties and between conceivability and possibility. It seems that's what an opponent (at least a type-M materialist) needs to do, if they want to hold onto the epistemic intuitions but still hold on to materialism. That's something we'll be looking at in depth soon. --Dave. From owner-modality@LISTSERV.ARIZONA.EDU Mon Mar 1 07:40:05 1999 x-sender: agillies@pop.u.arizona.edu Date: Mon, 1 Mar 1999 08:53:38 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Anthony S Gillies Subject: Re: Concepts and Properties To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Dave recently responded to Tim's question re "the big picture disagreement" between 2D views and those in N&N by (among other things) noting that: >I suppose that one difference is that Kripke sees epistemic matters >such as apriority as *purely* epistemic in some sense and not modal, >whereas I think it is often useful to understand them in modal terms. >On my view, one can see epistemic possibilities as possible worlds in >their own right, where our terms have a distinctive pattern of >application to those worlds; and there is a corresponding notion of >necessity, 1-necessity or primary necessity, which is in many ways >just as important as the notion of subjunctive necessity (2-necessity) >that Kripke concentrates on, and which has deeper ties to epistemic >notions. There may be another way of getting at this point of tension between the two views. This relates to some of Josh's worries about worlds considered as actual. On the 2D view of things, considering a (centered) possible world as actual is straightforward enough, and there is a well-defined and well-behaved notion of necessity and possibility that corresponds to the application conditions of terms in worlds-considered-as-actual. One reason why this is easy to do on the 2D view is that worlds are considered purely qualitatively (see TCM, p.367n30). And so, a la Lewis, the actual world is priveledged not because it is actual. but because we happen to inhabit it: other worlds are truthfully called "actual" by their inhabitants. But Kripke doesn't seem to want to think of worlds in that way. He even scorns the literature on trans-world identity for being confused: these philosophers wouldn't have such difficulty if they didn't think that worlds could only be given qualitatively (p.76). Worlds, on the Kripkean view, are *always* counterfactual worlds---at least that seems to be the way Kripke talks: worlds are total ways the world *might have been* (p.18). So, it's no wonder he focuses on SI's---stricyly speaking, he has nothing else to talk about. Thony "Curious green ideas sleep furiously." From owner-modality@LISTSERV.ARIZONA.EDU Tue Mar 2 00:37:01 1999 Date: Tue, 2 Mar 1999 00:36:41 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Re: Concepts and Properties To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Thony writes: >There may be another way of getting at this point of tension between the >two views. This relates to some of Josh's worries about worlds >considered as actual. On the 2D view of things, considering a (centered) >possible world as actual is straightforward enough, and there is a >well-defined and well-behaved notion of necessity and possibility that >corresponds to the application conditions of terms in >worlds-considered-as-actual. One reason why this is easy to do on the 2D >view is that worlds are considered purely qualitatively (see TCM, >p.367n30). And so, a la Lewis, the actual world is priveledged not >because it is actual. but because we happen to inhabit it: other worlds >are truthfully called "actual" by their inhabitants. Well, I wouldn't go so far as to say that the 2-D view is Lewis-like. Lewis's view holds in effect that all worlds are equally "concrete", and that what distinguishes the actual world is just that we are in it. But it's quite compatible with the 2-D framework to say that only the actual world is concrete while the others are abstract, and this concreteness distinguishes it from the other worlds, not just the fact that we are in it. Or to use other language, we can say that there is a large space of (abstract) possibilities, but only one of the possibilities is *realized*, namely this one. Of course, one can consider any possible world as actual, but that's not to say that the actual world isn't privileged. This world is not merely considerable-as-actual -- it *is* actual! >But Kripke doesn't seem to want to think of worlds in that way. He even >scorns the literature on trans-world identity for being confused: these >philosophers wouldn't have such difficulty if they didn't think that >worlds could only be given qualitatively (p.76). Worlds, on the Kripkean >view, are *always* counterfactual worlds---at least that seems to be the >way Kripke talks: worlds are total ways the world *might have been* >(p.18). So, it's no wonder he focuses on SI's---stricyly speaking, he >has nothing else to talk about. Hmm, lots of interesting issues here. I suspect that the issue vis-a-vis qualitative specification and transworld identity is to some extent separable from the issue about considering as actual and as counterfactual, though there are links. Kripke says that possible worlds *needn't* be specified in mere qualitative terms, though he also allows that they *can* be. So I can if I like specify a counterfactual world in terms of the distribution of qualities, but I can also specify it as one in which *Nixon* does such-and-such, and so on. Kripke thinks this cleanly bypasses any problems about "transworld identity". Of course there might still be a problem concerning identity across worlds given qualitatively, since Kripke at least allows the latter, but the fact that one needn't always specify a world qualitatively means that there isn't a problem about transworld identity *in general*. As for worlds being considered as counterfactual, it's true that insofar as Kripke explicitly talks about what is true in a world, he is always considering those worlds as counterfactual, and thus in effect invoking SI evaluation. He does seem to often *implicitly* consider worlds as actual and evaluate the referent of our terms there (as with considering the epistemic possibility that Godel stole the proof, or thinking about what we'd say if we discovered that the catlike things were demons); but he doesn't explicitly put this in terms of evaluating terms in possible worlds. So at least as far as his explicit framework goes, possible worlds are always considered as counterfactual (though I'd argue that he is implicitly endorsing something more). (You're right that thinking of worlds as ways things might have been goes along with considering them as counterfactual. One way of bringing out the 2-D distinction is to say that worlds considered as actual are ways the world *might be*, while worlds considered as counterfactual are ways the world *might have been*. This goes along with the ties to indicative and subjunctive mood in conditionals and elsewhere. Kripke always focuses on the latter, it seems.) What's the relation between the two points: allowing us to stipulate worlds nonqualitatively, and always considering worlds as counterfactual? The former does not seem to imply the latter: someone might accept that we can stipulate counterfactual worlds qualitatively but also accept that we can consider worlds as actual (whether qualitatively or nonqualitatively). And the latter does not seem to imply the former: someone might always consider worlds as counterfactual but also always consider them qualitatively ("through a powerful telescope"), not allowing nonqualitative stipulation. So the two points seem independent to some degree. Still, maybe there is some indirect link. One way to get at this is to ask whether it is possible, when considering worlds as actual, to stipulate them nonqualitatively as well as considering them qualitatively. To a certain degree this seems to be possible: for example, I can stipulate that I am talking about the epistemic possibility that Nixon was Kennedy's brother, rather than considering the world "given qualitatively" and trying to see whether it is an instance of the epistemic possibility that Nixon was Kennedy's brother. So to that extent Kripke's point holds even here. (Of course, in both cases one needs to be careful that one is not making an incoherent stipulation.) Still, there's a difference here in that arguably the notion of transworld identity doesn't make as much sense when talking about worlds considered as actual. I can consider the epistemic possibility that Hesperus is not Phosphorus (that's pretty straightforward), but presumably that's not a world where *this very thing* (i.e. Hesperus/Phosphorus/Venus) is not *that very thing* (i.e. Hesperus/Phosphorus/Venus). Presumably that would be incoherent. So when we stipulate the epistemic possibility, one is not making a stipulation about the very things from the actual world. Arguably, the notion of transworld identity does not really make sense when applied to worlds considered as actual, as opposed to worlds considered as actual. The reasons for this are subtle but come down at least in part to the failure of substitutivity in epistemic but not in modal contexts (and ultimately in indicative but not in subjunctive contexts). Epistemic possibilities are not "object involving" in quite the same way that subjuinctive possibilities are. So *de re* (object-involving) possibility and necessity may not really be coherent here in this domain. For example, it seems one can't quantify into 1-modal contexts: "it is 1-necessary that x is hot" isn't really coherent. E.g., it is 1-necessary that Hesperus is the evening star, but it isn't 1-necessary that Phosphorus is the evening star, even though both are names for the same thing; and there isn't really any other way to make sense of the claim without such names. So for the epistemic modality, any modal claims involving x seem to be relative to the label under which x is picked out. This doesn't apply to subjunctive modality: "it is 2-necessary that x is hot" makes sense, and any label for x will yield the same truth-value here, as as long as it's a rigid designator. (In a way, this is saying that Quine's critique of quantified modal logic and essentialism was right about epistemic modality, though not about subjunctive modality. So we have a "split decision" in the battle between the likes of Quine and the likes of Kripke. A point like this is made John Burgess in a very nice recent paper in the 1997 Canadian Journal of Philosophy supplement on "Meaning and Reference".) Anyway, nothing here contradicts anything that Kripke says, exactly. >From the 2-D perspective, one can say that he is focusing on the sort of possibility and necessity (namely subjunctive possibility and necessity) where de re possibility and necessity does make sense, and it's quite an achievement of his to make sense of it. At the same time, there are other notions of possibility and necessity out there (e.g. broadly epistemic possibility and necessity) where de re possibility and necessity doesn't make sense. Of course it's no surprise that Kripke, who was such a major contributor to quantified modal logic, was particularly interested in investigating notions of modality for which quantification into modal contexts makes sense (which will be just those for which *de re* modality makes sense). So it's no surprise that he was led to focus on subjunctive modality, and on worlds considered as counterfactual. (More on this in the "tyranny of the subjunctive" material.) But anyway, none of this implies that there's anything problematic or second class about epistemic modality and considering worlds as actual. Things here just work a different way. One can take on board all of Kripke's points about subjunctive modality, and still hold that there are other interesting modal notions and other corresponding ways of looking at possible worlds. Or so it seems to me. But feel to elaborate on this interesting issue. --Dave. From owner-modality@LISTSERV.ARIZONA.EDU Tue Mar 2 00:01:44 1999 Date: Tue, 2 Mar 1999 00:16:22 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Erik J Larson To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO I have a question concerning the possibility of psychophysical laws, given the anti-materialist argument we've been discussing. One of the straightforward problems that one adopts, if the non-(logical supervenience) of consciousness is granted, is the rather unpalatable possiblity of epiphenomenalism with respect to conscious experience. Chalmers' remarks that this may not be so bad, and that, if the argument goes through, we may simply "have to get used to it". But I take it that Chalmers' (self-admitted!) speculation about the possibility of internal, protophenomenal features of an "expanded" physical framework, who's current postulates are all external, relationally specified properties (i.e., of the fundamental pysics variety) is meant to provide a possible future development of a causal role for phenomenal properties. (Is this true?). I have two questions. First, if we have some set of internal (protophenomenal) properties related by psychophysical laws, and still the external, physical properties related by physical laws, then don't we just push the problem of interaction (and thus epiphenomenalism) back a step? Second, if we have the psychophysical laws connecting internal to external properties--protophenomenal to physical--wouldn't this fly in the face of a) the non-(logical supervenience) of the phenomenal on the physical and b) the assumption that the physical world is causally closed? All of this is to suggest that there doesn't seem to be much room for psychophysical laws, at least in so far as they may provide possible answeres to the problem of epiphenomenalism. One response may be that considering psychophysical laws qua answers to epiphenomenalism is not necessary, anyway. But I take it that a "completed" scientific account of phenomenal properties in a theory of the mental would have something like this in mind. (There are really two (or more) possibilities for the role of psychophysical laws relevant here. Qua answer to epiphenomenalism, the laws would be causal. A "weaker" interpretation would assign psychophysical laws the role of correlating internal (protophenomenal) to external (physical) states in a way that yielded information about regularities between the two. I suppose I'll need to think things through more for the weak case, but it seems that the same sort of objections stated above will present themselves.) Let me know what you think. Erik "What our grammarian does is simple enough. He frames his formal reconstruction of K along the grammatically simplest lines he can, compatibly with inclusion of H, plausibility of the predicted inclusion of I, plausibility of the hypothesis of inclusion of J, and plausibility, further, of the exclusion of all sequences which ever actually do bring bizarreness reactions." -- W.V.O. Quine ---------------------- Erik J Larson erikl@U.Arizona.EDU From owner-modality@LISTSERV.ARIZONA.EDU Tue Mar 2 14:37:06 1999 Date: Tue, 2 Mar 1999 12:57:29 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Angela J Burnette Subject: Re: epiphenomenalism To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO I realize that what follows is actually last weeks topic, but here goes anyway... In an attempt to avoid all out epihenomenalism, it is suggested in TCM and in Metaphysics of Modality that there might be a causal role for the phenomenal if the intrinsic properties of mircophysical stuff is actually phenomenal or protophemomenal, this is a possibility I take it because we understand this stuff only in virtue of its extrinsic properties, which leaves the intrinsic nature up for grabs...if the intrinsic properties were phenomenal or protophenomenal, then perhaps they could form the underlying basis for the physical realm, and therefore somehow be causally related, even though the physical system is causally closed... I have two questions about how this might go, first, if protophenomenal properties are intrinsic, then what access will we ever have to them such that laws relating them might be developed? Also, how is it that "principles associating physical dispositions with (proto)phenomenal bases" is less problematic than laws relating phenomenal properties and physical properties? I'm not sure how one can posit a "transformation operator, for example" given all that has gone before about the utter impossibility of cashing out phenomenal properties in functional or dispositional terms. so, it seems as though epiphenomenalism cannot be avoided without undermining the anti-materialist arguments... angela From owner-modality@LISTSERV.ARIZONA.EDU Thu Mar 4 04:32:01 1999 Date: Thu, 4 Mar 1999 04:31:49 -0800 Sender: "Philosophy 596B: Mind and Modality" From: David Chalmers Subject: Re: TCM (panprotopsychism) To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO Both Erik L. and Angela raise questions about panprotopsychism, vis-a-vis epiphenomenalism. Erik writes: >I take it that Chalmers' (self-admitted!) speculation about the possibility >of internal, protophenomenal features of an "expanded" physical framework, >who's current postulates are all external, relationally specified >properties (i.e., of the fundamental pysics variety) is meant to provide a >possible future development of a causal role for phenomenal properties. >(Is this true?). I have two questions. First, if we have some set of >internal (protophenomenal) properties related by psychophysical laws, and >still the external, physical properties related by physical laws, then >don't we just push the problem of interaction (and thus epiphenomenalism) >back a step? > >Second, if we have the psychophysical laws connecting internal to external >properties--protophenomenal to physical--wouldn't this fly in the face of >a) the non-(logical supervenience) of the phenomenal on the physical and >b) the assumption that the physical world is causally closed? All of this >is to suggest that there doesn't seem to be much room for psychophysical >laws, at least in so far as they may provide possible answeres to the >problem of epiphenomenalism. One response may be that considering >psychophysical laws qua answers to epiphenomenalism is not necessary, >anyway. But I take it that a "completed" scientific account of phenomenal >properties in a theory of the mental would have something like this in >mind. Well, one has to get the shape of the view straight. On the "panprotopsychist" view, one has still has a causally closed physics. In fact physical laws have just the same shape as before. We have just "colored in" physics by specifying the categorical basis of the physical dispositions. Think of it this way. Take physics at face value as true. Even so, one arguably realizes that physics only tells us about the "structure" in relations between things, and tells us about things only be their effects. What is a particle? What is mass? Physics doesn't say. We just know about the effects of these things. More technically, we just know about the *dispositions* in the vicinity of charge and mass and electrons. We know about them solely in terms of their characteristic effects, which are well spelled-out by physics. But arguably, wherever one has a disposition, one needs a categorical basis. That's just to say, whenever one has a causal role, one needs something to do the causing! That thing will be the "intrinsic" nature of mass, or charge, or an electron. Physics doesn't tell us about it -- it is in effect hidden from sight, in a sort of Kantian way -- but arguably we still have reason to believe it exists. On this view, every physical property -- mass, charge, etc -- has a categorical bases, or an intrinsic nature. A fundamental particle will have an intrinsic nature given by the nature of its charge and mass, etc. The world itself is a giant network of causal interactions among these intrinsic properties and the particles that have them. The interactions are structured just physics says -- the laws of physics specify the dynamics perfectly. It's just that physics tells us only about the structural part, and we can't forget the intrinsic part too. Panprotopsychism is the view that these intrinsic properties are phenomenal or protophenomenal. So the physical world consists ultimately in interactions among phenomenal or protpphenomenal properties. It doesn't look that way to us, because we see the properties only from the "outside" (except in our own case!), in terms of the structure of their interactions. But on this view, it's that way all the same. After all, it's no surprise that we don't know the nature of the intrinsic properties of the physical. Now to Erik's questions: (1) Doesn't this just push the problem of interaction and epiphenomenalism back a step? I'd like to think not. The beauty of doing things this way is that we *know* that the intrinsic properties are causally relevant. They are the things that do the ultimate causing! E.g., when mass causes another mass to be attracted, what is really happening is that there is a causal relation between their intrinsic bases, i.e. (on the panprotopsychist view) the relevant protophenomenal properties. The world is ultimately a network of causal interactions among just these properties! So they are causally relevant from the "bottom up". (2) What about causal closure? Another beauty of this view is that one can hold onto causal closure. We haven't introduced new causal "gaps" from phenomenal propeties to fill. Rather, they just play the role of supporting the causation that was there all along. Interactions among particles etc can be just as they were before; it's just that we now know that the ultimate nature of those particles is protophenomenal. (3) What about psychophysical laws? These have a slightly different shape on this view. They don't connect physical properties to phenomenal properties "dangling" outside the system. Rather, they will in effect connect "dispositional" properties (e.g. the characteristic causal role of mass) with the "intrinsic" properties that underlie the disposition (and that are ultimately doing the causing). In effect, they are connecting the "outside" of physical entities to the "inside". Arguably, the truly fundamental laws in this world are laws connecting the basic intrinsic properties to each other. e.g. protophenomenal property 1 (the mass property) and protophenomenal property 2 (the charge property) interact with each other in such-and-such a way. The "dispositional" physical properties just come from our viewing all this from the outside in terms of structure, and the "psychophysical laws" from our trying to fill in the dispositions with intrinsic properties again. But at least from our perspective, it's useful to see these psychophysical laws here, as the dispositional stuff is what we're familiar with (from physics etc), and the intrinsic stuff is all a bit of an unknown. So one can hope that a final science of consciousness might make some progress in connecting the two, and in particular telling us just which intrinsic properties play which dispositions. (4) Logical supervenience on the physical? As I said in the earlier message, there's a sense in which this view sees phenomenal or protophenomenal properties as part of the physical world. After all, they are the categorical bases of physical dispositions, and arguably are the "ultimate nature" of the physical. But still, they don't logically supervene on the physical properties we know and love, because those are dispositional properties, and intrinsic properties don't supervene on dispositions. It would be logically possible to have the same dispositions but a different intrinsic basis, etc. But here, the failure of supervenience isn't due to the intrinsic properties being "outside" the network. It's because it's the intrinsic properties that "carry" the dispositions, serve as their "insides", and ultimately do the causing. Angela writes: >I have two questions about how this might go, first, if protophenomenal >properties are intrinsic, then what access will we ever have to them such >that laws relating them might be developed? Also, how is it that >"principles associating physical dispositions with (proto)phenomenal >bases" is less problematic than laws relating phenomenal properties and >physical properties? I'm not sure how one can posit a "transformation >operator, for example" given all that has gone before about the utter >impossibility of cashing out phenomenal properties in functional or >dispositional terms. > >so, it seems as though epiphenomenalism cannot be avoided without >undermining the anti-materialist arguments... > >angela Good questions. First of all, let me say that I'm not suggesting that the problems you mention are any easier on a panprotopsychist view than on an epiphenomenalist view. But I don't think that they're necessarily harder, either. E.g., for your first problem about "access", obviously we don't have much access to epiphenomenal phenomenal properties, except in our own case. And indeed one can argue that almost any view will have some sort of epistemic barrier of this sort, even a type-B materialist view: one can't monitor qualia in other systems directly, and there are always multiple epistemic possibilities. So the panprotopsychist view isn't obviously worse off here. Similarly for the second problem. Our "transformation operator" isn't going to try to reduce phenomenal properties to dispositional ones. We are just going to try to find a law connecting the two (laws that connect distinct things, of course). That will be hard for an epiphenomenalist, but there's no reason why it should be impossible in principle. Hopefully a panprotopsychist can do the same. How do we find the laws? Well, first of all, by correlating and systematizing first-person data and its relation to third-person data in one's own case. And second, by gathering data about consciousness in others (under certain assumptions, e.g. reliability of verbal report), and relating it to the third-person data. And third, by careful analysis and thought-experiments etc, which can deliver useful conclusions. This way we at least get a stock of regularities and correlations at the "macroscopic" level to start with. then we try to boil these down to a simplest form and ultimately find "fundamental" principles that support these macroscopic regularities. A form of inference to the best explanation, if you like. That might be difficult and speculative, but not obviously impossible. It will be hard enough coming up with one good set of fundamental laws that fits the data, so if we do find a good hypothesis that explains the data we have, we'd have some good reason to believe it. The panprotopsychist can avail herself of all of this. The only extra thing is that the panprotopsychist has strong constraints on the form of the fundamental laws. The basic psychophysical will ultimately connect fundamental physical dispositions to fundamental protophenomenal properties, right down at the microphysical level. Presumably there will be just a few such laws, one for every fundamental physical property. That puts a lot of constraints on the form of the panprotopsychist's fundamental psychophysical laws, which might either be a blessing or a curse. A blessing because the extra constraint narrows down the hypothesis space and tells us something about where to look. A curse because it may be all the harder to come up with laws that do the job. It's not at all obvious that we can derive all the facts about the macrophenomenology we know and love as a consequence of a whole bunch of protophenomenal properties we know and love. That's the biggest problem for the view -- what I call the "combination" or "constitution" problem in the book and elsewhere (see e.g. pp. 305-8). How does all those zillions of protophenomenal properties add up to a unified, bounded, grainless, structured phenomenal consciousness? It's not obvious that they can, but it's not obvious that they can't. *If* this problem can be solved, then I think the panprotopsychist view is the best solution to the mind-body problem by a mile. If it can't, then we have to look at other options. I'm hoping that we will get a better sense of the possible solutions to this problemfor the panprotopsychist in coming years. --Dave. From owner-modality@LISTSERV.ARIZONA.EDU Fri Mar 12 01:20:28 1999 Date: Fri, 12 Mar 1999 02:19:24 -0700 Sender: "Philosophy 596B: Mind and Modality" From: Erik J Larson Subject: Re: This and that To: MODALITY@LISTSERV.ARIZONA.EDU Status: RO I have a brief question regarding the admissability of causal primary intensions. This has come up at least twice; once in Tim's response to Josh's BIV scenario, and then again in our first class meeting. So, the problem is, I don't think primary intensions can ever be appropriately thought of as "the cause of my sensation that" or "that which brought about my experience of" or some such, because in any of these cases the primary intension can vary wildly, depending on what the cause happens to be. So some brain in a vat may have a PI of water that appeals to electrodes and stimulation or whatever, and on earth this of course won't make any sense. So "that which causes my x experience" makes a primary intension the same as whatever the cause happens to be , and not the essential notion or properties of the thing itself, apriori. So I think the primary intension has to be necessarily connected to the essential quality of the thing (at least as it is known apriori or conceptually). What we mean by the concept of water, then, is not what causes the water experiece (which may have nothing to do with the essence of water itself). Anyway, this isn't a major point, but it keeps bothering me. I would be interested to hear any defence of causally-based primary intension, or where I have gone wrong or oversimplified the issues here. Erik L. "What our grammarian does is simple enough. He frames his formal reconstruction of K along the grammatically simplest lines he can, compatibly with inclusion of H, plausibility of the predicted inclusion of I, plausibility of the hypothesis of inclusion of J, and plausibility, further, of the exclusion of all sequences which ever actually do bring bizarreness reactions." -- W.V.O. Quine ---------------------- Erik J Larson erikl@U.Arizona.EDU