From chalmers Mon Jan 18 21:31:06 1999 Date: Mon, 18 Jan 1999 21:30:55 -0800 (PST) From: David Chalmers To: aburnette@u.arizona.edu, agillies@u.arizona.edu, akolers@u.arizona.edu, atlane@u.arizona.edu, bradt@u.arizona.edu, chalmers@ling.ucsc.edu, cowley@u.arizona.edu, erikh@u.arizona.edu, erikl@u.arizona.edu, jtismael@u.arizona.edu, kcreath@primenet.com, patrickr@u.arizona.edu, rachaelp@u.arizona.edu, sch@u.arizona.edu, shaughan@ns.arizona.edu, tbayne@u.arizona.edu, tolliver@u.arizona.edu Subject: Philosophy 596B: Mind and Modality Status: RO To participants in Philosophy 596B (Mind and Modality): As you probably know, I won't be in Tucson for the start of this course. Due to a bureaucratic mix-up, I wasn't granted a US visa in time for the start of term, and I'm currently in Australia (at ANU in Canberra) waiting for the visa to come through. I'm told that that should happen before February 15, and could happen as soon as the next few days, but there is no way to know. I hope that we won't miss more than two or three meetings of the seminar; in the worst case, we'll miss perhaps five. Until I'm present in Tucson, the course will be conducted as an online seminar. The first meeting was scheduled for Tuesday Jan 19 (tomorrow Arizona time), so this can be considered a kick-off announcement. My plan is to make the first couple of weeks reading-intensive, with the opportunity for online questions and discussion. I'm currently trying to set up a listserv mailing list for the course and a web site; I'll let you know as soon as those things are operational. If it's possible, I'll try to arrange extra meetings once I make it to Tucson to get everything up to speed. In the next message, I'll send a syllabus for the course, giving topics and readings and an approximate schedule. Note that the first part of the schedule will slightly be revised to accommodate the electronic format, to a degree that depends on the date of my return. In a third message (coming shortly) I'll detail initial topics, readings, and methodology. Here's a list of participants in this seminar (the recipients of this mailing). If you know of someone else who should be on this list, please let me know. And if you shouldn't be on this list, let me know. Enrolled students Angela Burnette aburnette@u.arizona.edu Josh Cowley cowley@u.arizona.edu Anthony Lane atlane@u.arizona.edu Erik Larson erikl@u.arizona.edu Rachael Parkinson rachaelp@u.arizona.edu Brad Thompson bradt@u.arizona.edu Erik Herman erikh@u.arizona.edu Katherine Creath kcreath@primenet.com Auditors/sitting-in/interested: Tim Bayne tbayne@u.arizona.edu Anthony Gillies agillies@u.arizona.edu Scott Hendricks sch@U.Arizona.EDU Avery Kolers akolers@u.arizona.edu Patrick Rysiew patrickr@u.arizona.edu Jenann Ismael jtismael@u.arizona.edu Shaughan Lavine shaughan@ns.arizona.edu Joseph Tolliver tolliver@u.arizona.edu Let me know if you have any questions. Note that the best e-mail address to use for me is my UCSC address, chalmers@ling.ucsc.edu. --Dave Chalmers. From chalmers Tue Jan 19 01:12:36 1999 Date: Tue, 19 Jan 1999 01:12:27 -0800 (PST) From: David Chalmers To: aburnett@u.arizona.edu, agillies@u.arizona.edu, akolers@u.arizona.edu, atlane@u.arizona.edu, bayne@u.arizona.edu, bradt@u.arizona.edu, chalmers@ling.ucsc.edu, cowley@u.arizona.edu, erikh@u.arizona.edu, erikl@u.arizona.edu, jtismael@u.arizona.edu, kcreath@primenet.com, laj@u.arizona.edu, patrickr@u.arizona.edu, rachaelp@u.arizona.edu, sch@u.arizona.edu, shaughan@ns.arizona.edu, tolliver@u.arizona.edu Subject: Initial readings and discussion Status: RO READING MATERIAL: The reading material for the first bit of the course is: _The Conscious Mind_, Chapters 2-4. Jackson, Epiphenomenal Qualia [web]. _Naming and Necessity_. You can also read Chapter 1 of TCM as background. But the course will very much be concentrating on the modal issues, so you might not want to become too distracted by general issues about consciousness. Here's a suggested order of reading, corresponding roughly to the order in the syllabus. Because of the reading orientation, things will proceed a little differently from the way they would in person; for example, I'll leave some of the material intended for the first seminar until I get back. And you shouldn't think of the numbers below as corresponding to weeks, exactly. It might make sense to try to give most of this material a quick reading in the first week or so (assuming you haven't read it already, as some will have), before going back for a more detailed and careful reading to go along with a sequential discussion of these topics. (1) Start with TCM sections 2.1-2.4 (pp. 32-55), for a general introduction to some issues about modality, ontology, and explanation. You might read the rest of 2.4 (pp. 55-71) and 2.5 lightly; we'll return to them later. (2) Next read TCM Chapter 3 and section 4.1, as well as the Jackson piece, for an introduction to epistemic arguments against materialism. (3) Next we come to a posteriori necessity. The relevant material here is _Naming and Necessity_, the rest of TCM section 2.4 (pp. 55-71, on the two-dimensional analysis of a posteriori necessity), and perhaps "The Components of Content" if you want more on the 2-D framework. The order of reading here depends to some extent on familiarity. If you're already familiar with the contents of N&N (as I assume many or most are), it would make sense to first read TCM 2.4, before going back to reread N&N in light of that analysis. If you haven't read N&N, you might want to read it first, perhaps going back and forth with TCM 2.4 to gain a feeling for the 2-D analysis along the way. Either way, the two-dimensional framework will be central to many topics in the course, so it's very important to gain a good understanding of it (I hope that before long you'll be living and breathing it). You'll want to go over that material very carefully to be sure that you understand it. "The Components of Content" has some further material (though applied to the semantics of thought rather than language) that may help you if you're not sure about things. After gaining a familiarity, try to translate all the central examples and arguments of N&N into the 2-D framework; that should prove to be a useful exercise in gaining an understanding. (I will probably make something like this an official assignment at some point.) (4) Next (if we are still online) will be material on the application of the 2-D framework to the argument against materialism: the rest of TCM Chapter 4 (especially 4.2), and Jackson's "Finding the Mind in the Natural World". But I'll leave the details on that until later. If the seminar is in the flesh by this point, we'll have some material on subjunctive and indicative conditionals vis-a-vis the 2-D framework first. ONLINE DISCUSSION: We will start with a discussion of the material in (1) above, especially TCM 2.1. This is the closest thing easily available to an introduction to foundational issues concerning modality, ontology, and explanation. Try to get clear especially on the notions of logical and natural necessity, the various notions of supervenience, and the definition of materialism. The material on reductive explanation is secondary for our purposes, but you should still look at it. Post any questions and comments you might have to the list. Ideally, I'd like to see everyone (i.e. everyone officially enrolled in the course; participation by others is encouraged but optional) post some initial comments or questions in the next week (the sooner the better), so we can get an initial discussion going. (N.B. I wouldn't now lay things out exactly as I do in those parts of TCM, but it should work as a first approximation. Hopefully a lot of things will be clarified and elaborated in discussion.) I hope to have a listserv mailing list and/or web site for discussion up and running soon, but in the meantime, send all public messages to the list of recipients at the top of this message. N.B. My previous two messages had a couple of e-mail addresses wrong, so this is the one to use. BACKGROUND: It would be useful to get a sense of the class's familiarity with relevant issues. I'd appreciate it if everyone officially enrolled in the course could send me a note indicating their background in metaphysics, the philosophy of language, and the philosophy of mind: courses taken, relevant works read, and so on. It would be particularly useful to know whether and to what extent you're familiar with the ideas in and around (1) _Naming and Necessity_ (2) _The Conscious Mind_ (3) Lewis's _On the Plurality of Worlds_ and (4) two-dimensional semantics. You might also mention your areas of main philosophical interest, your year, and any other relevant information, to help me get a sense of things. And you should feel free to take the opportunity to ask any questions that you might have. People who aren't officially enrolled in the course should also feel free to send such a note. (I was going to suggest sending such notes to the whole list, but don't want to clog people's mailboxes too badly, and some relevant information about background is probably best regarded as private. But people should feel free to send messages to the whole list to kick off discussion and to start building our little online community.) --Dave. From josh@ame2.math.arizona.edu Sun Jan 24 17:12:45 1999 From: Josh Cowley Subject: Re: Initial discussion To: chalmers@ling.ucsc.edu (David Chalmers) Date: Sun, 24 Jan 1999 18:11:56 -0700 (MST) Cc: aburnett@u.arizona.edu, agillies@u.arizona.edu, akolers@u.arizona.edu, atlane@u.arizona.edu, bayne@u.arizona.edu, bradt@u.arizona.edu, chalmers@ling.ucsc.edu, cowley@u.arizona.edu, erikh@u.arizona.edu, erikl@u.arizona.edu, jtismael@u.arizona.edu, kcreath@primenet.com, laj@u.arizona.edu, patrickr@u.arizona.edu, rachaelp@u.arizona.edu, sch@u.arizona.edu, shaughan@ns.arizona.edu, tolliver@u.arizona.edu Status: RO Hi folks, I guess I will get the ball rolling. I would like to note that I'm not really committed to anything I'm about to say. I have some concerns about the use of "facts" and properties in the various definitions surrounding supervenience. My worry is pretty general, but I'll draw it out by focussing on the definition of a possitive fact or property. "A possitive fact in W is one that holds in everyw orld that contains W as a proper part; a positive property is one that if instantiated in a world W, is also instantiated by the corresponding individual in all worlds that contain W as a proper part." "(pg 40) In footnote 14 it is suggested that we ought to take the containment relation between worlds as primitive. Here's my problem. Possible worlds are complicted and unnatural entities. Facts and properties are less complicated and more natural things. My inclination is to define possible worlds in terms of properties or facts. Something like, 'Two possible worlds, A and B, are identical just in case all the facts that hold in A hold in B and vice-versa. Two worlds differ in so far as the facts in A are not in B.' Now there is nothing inherently wrong with defining possible worlds with facts and still saying that containment is a primitive notion. But it does seem to add some tension. I would be inclined to define containment by making reference to facts (or maybe properties) rather than using containment to define certain types of facts. What do people think? Josh From chalmers Wed Jan 27 04:24:41 1999 Date: Wed, 27 Jan 1999 04:24:28 -0800 (PST) From: David Chalmers To: aburnett@u.arizona.edu, agillies@u.arizona.edu, akolers@u.arizona.edu, atlane@u.arizona.edu, bayne@u.arizona.edu, bradt@u.arizona.edu, chalmers@ling.ucsc.edu, cowley@u.arizona.edu, erikh@u.arizona.edu, erikl@u.arizona.edu, jtismael@u.arizona.edu, laj@u.arizona.edu, patrickr@u.arizona.edu, rachaelp@u.arizona.edu, sch@u.arizona.edu, shaughan@ns.arizona.edu, tolliver@u.arizona.edu Subject: Re: Initial discussion Status: RO Here's a reply to Josh's note. I'll hold off replying to Brad's note for a day or so to give other people a chance. I may not have said explicitly enough that everyone enrolled in the class should have posted some initial questions and comments by now. They don't have to be deep and profound; even simple clarifications or dumb questions are fine. But it's only through posting this sort of thing that we can get a good discussion going (hopefully on multiple fronts at once). We need to move on to the next material fairly soon. So I hope to see something from everyone within the next day or so. Re Josh's point on the appeal to facts in the definition of containment. The worry is that if worlds are defined in terms of facts, then it may be troublesome to define positive facts in terms of worlds. Personally I'm neutral on whether possible worlds should be defined in terms of facts/properties or vice versa, but I'd like to think that even in the former case, the definition of positive fact is OK. I guess the order of priority would be: (1) take the notion of a property (or a fact) as primitive; (2) build up the notion of a possible world out of these; (3) use the notion of a possible world so defined to characterize the class of positive facts. Of course one still needs to appeal an unreduced notion of containment relations between worlds; but it seems that we have clear intuitions about the application of this notion, and it doesn't seem especially problematic as unreduced notions go. Maybe Josh's idea is that it is less natural to apply an unreduced notion (containmenmt) to an reduced ontology (worlds as facts) than to an unreduced ontology (worlds as basic), but it isn't clear to me that the former is problematic. Again, as long as we have clear judgments about its application. I suppose the alternative would be to have a "fact-based" definition of a positive fact, or just to take the notion as primitive. I wouldn't wholly object to the latter, but personally I find the world-based definition (with containment as primitive( much more intuitive, and I think one ends up at least informally appealing to such a notion in explaining even a primitive notion of positive fact. --Dave. P.S. We are going to be particularly concerned with questions about the nature of and relationships between notions such as logical possibility and necessity, metaphysical possibility and necessity, a priori, a posteriori, natural possibility and necessity, etc. So any questions (including simple clarificatory questions) or comments there would be particularly welcome. From bradt@u.arizona.edu Wed Jan 27 00:40:33 1999 From: "Brad Thompson" To: "David Chalmers" , , , , , , , , , , , , , , , , Cc: Subject: Re: Initial discussion Date: Wed, 27 Jan 1999 01:34:18 -0700 X-MSMail-Priority: Normal X-Priority: 3 Status: RO Hi all! Here's a question about the reading for last week. As I wrote this, I answered this question to my own satisfaction (the error I was making was obvious later)! But, in the interest of contributing to discussion and in the hopes that my own error will be illuminating to others, I will send the message anyways. No jeers or hate mail, please! :) I find it difficult to conceive of 2 worlds that are identical with respect to their chemical facts but different with respect to their physical facts. In general, I'm worried that for all cases in which B-properties supervene on A-properties, it will turn out that A-properties supervene on B-properties. This would be odd, because it would entail that, for example, physical properties supervene on biological properties (since biological properties supervene on physical properties). I think that the way that Dave specifies what it is for 2 worlds to be identical with respect to a set of properties is supposed to rule out this problem (see footnote 4). What we want is to be able to say that there are two possible worlds that are identical with respect to their biological properties but different with respect to their physical properties--this would rule out physical properties supervening on biological properties. So, let's imagine a biological duplicate of our world which, by stipulation, has a different physics. Wouldn't the structure and dynamics of that world's physics have to be the same as ours in order for it to have the same biological properties? But if so, it isn't clear that we've really imagined a world with a different physics. Note: I don't think we can solve this problem simply by saying that there are possible worlds which are biologically identical and physically identical to our world except that there is extra physical stuff in that world. This would be analogous to arguing that biological facts don't supervene on the physical facts because of the possibility of a world with extra biological facts (like ectoplasmic angels). * Here is where I went wrong. There *are* possible worlds with the same biological properties but with fewer or just different physical properties (as long as those absent or different physical properties belong to individuals which do not possess any biological properties). So, for example, there are possible worlds with the same biological properties as this one but in which my shirt is red rather than blue, or in which there simply is no Statue of Liberty. Perhaps, despite this obvious blockade to the objection about supervenience above, my worry that a biologically identical world to this one would really have to have the same physical laws as this one is of independent interest. For example, if it makes sense at all to speak of a supervenience relation among types of properties rather than token instantiations of properties, the above consideration would suggest (maybe) that we would need to modify our definition of supervenience in a way that rules out physics supervening on biology. I'm not entirely sure this would be a consequence, since one would first need to figure out how we are to individuate or specify biological and physical property types. From aburnett@U.Arizona.EDU Wed Jan 27 12:20:25 1999 Date: Wed, 27 Jan 1999 13:19:34 -0700 (MST) From: Angela J Burnette To: Brad Thompson cc: David Chalmers , agillies@U.Arizona.EDU, akolers@U.Arizona.EDU, atlane@U.Arizona.EDU, bayne@U.Arizona.EDU, cowley@U.Arizona.EDU, erikh@U.Arizona.EDU, erikl@U.Arizona.EDU, jtismael@U.Arizona.EDU, kcreath@primenet.com, laj@U.Arizona.EDU, patrickr@U.Arizona.EDU, rachaelp@U.Arizona.EDU, sch@U.Arizona.EDU, shaughan@ns.arizona.edu, tolliver@U.Arizona.EDU Subject: Re: Initial discussion Status: RO All: I have a question about supervenience...Brad suggested that we may have to modify our defintion of supervenience to rule out the possibility of the physical supervening on the biological...On a related note, can someone please explain to me how the direction of a supervenience relation is built into the concept of supervenience? Or is this just a contingency that, once established, allows us to use the term when the appropriate relation is discovered? It just seems to me that there is nothing about the concept of supervenience that makes it move in one direction, and not both...so, my question is, what is it about supervenience that allows us to say that B supervenes on A without necesarrily saying that A must also therefore supervene on B? angela From erikl@U.Arizona.EDU Wed Jan 27 12:31:58 1999 Date: Wed, 27 Jan 1999 13:30:16 -0700 (MST) From: Erik J Larson To: Brad Thompson cc: David Chalmers , aburnett@U.Arizona.EDU, agillies@U.Arizona.EDU, akolers@U.Arizona.EDU, atlane@U.Arizona.EDU, bayne@U.Arizona.EDU, cowley@U.Arizona.EDU, erikh@U.Arizona.EDU, jtismael@U.Arizona.EDU, kcreath@primenet.com, laj@U.Arizona.EDU, patrickr@U.Arizona.EDU, rachaelp@U.Arizona.EDU, sch@U.Arizona.EDU, shaughan@ns.arizona.edu, tolliver@U.Arizona.EDU Subject: Re: Initial discussion Status: RO Hi, I'll try to reply to Brad and Angela on the question of the supervenience relation. The worry is that, if B-properties supervene on A-properties, then equally A-properties should supervene on B-properties. This seems, at best, odd--one doesn't want to say that physical facts supervene on biological facts or be committed to that view given that biological facts supervene on physical facts. I think the problem is that the notion of one set of facts "fixing" another set is left unalalyzed in the concept of a supervenience relation. It is not literally true that biological facts fix physical facts, because the physical facts have already fixed the biolofical facts, so there is no more work left to be done. but the notion of "fixed" seems a contingency left out of the strict definition of supervenience--at least insofar as we have encountered it in the readings thus far. The other position is that, while physical facts MAY supervene on biological facts given the concept, such a statement is uninformative. But again, whether or not something is "informative" or not seems another contingency that is left out of the concept of the supervenience relation. What do you think? Erik L. From aburnett@U.Arizona.EDU Wed Jan 27 13:29:19 1999 Date: Wed, 27 Jan 1999 14:28:31 -0700 (MST) From: Angela J Burnette To: Erik J Larson cc: Brad Thompson , David Chalmers , agillies@U.Arizona.EDU, akolers@U.Arizona.EDU, atlane@U.Arizona.EDU, bayne@U.Arizona.EDU, cowley@U.Arizona.EDU, erikh@U.Arizona.EDU, jtismael@U.Arizona.EDU, kcreath@primenet.com, laj@U.Arizona.EDU, patrickr@U.Arizona.EDU, rachaelp@U.Arizona.EDU, sch@U.Arizona.EDU, shaughan@ns.arizona.edu, tolliver@U.Arizona.EDU Subject: Re: Initial discussion Status: RO all, regarding Erik's take on supervenience, that sounds good but, given that the mentioned aspects are left out of the concept of supervenience then we are just using the word as a convenient label? I think that I was categorizing supervenience alongside words like entailment, for example. Establishing that an entailment holds between A and B allows you to do all sorts of things in virtue of the entailment, but if supervenience is just a label, then it won't be helpful in establishing new facts or conclusions about anything in the way that a concept like entailment does, right? angela From chalmers Wed Jan 27 19:27:33 1999 Date: Wed, 27 Jan 1999 19:27:23 -0800 (PST) From: David Chalmers To: chalmers Subject: Supervenience and symmetry Cc: aburnett@u.arizona.edu, agillies@u.arizona.edu, akolers@u.arizona.edu, atlane@u.arizona.edu, bayne@u.arizona.edu, bradt@u.arizona.edu, chalmers@ling.ucsc.edu, cowley@u.arizona.edu, erikh@u.arizona.edu, erikl@u.arizona.edu, jtismael@u.arizona.edu, laj@u.arizona.edu, patrickr@u.arizona.edu, rachaelp@u.arizona.edu, sch@u.arizona.edu, shaughan@ns.arizona.edu, tolliver@u.arizona.edu Status: RO Some very interesting points re supervenience. Angela asks about whether directionality is built into the supervenience concept. I would say that it is built in, as the definition is asymmetrical. Take the first definition: B supervenes on A if all A-indiscernible worlds are B-indiscernible. One can see that the definition is asymmetrical by noting that if the A-indiscernible worlds are a proper subset of the B-indiscernible worlds, B will supervene on A but not vice versa. Something similar applies to the modified definitions given later on. So there's at least conceptual room for B to supervene on A but not vice versa. And it's not too hard to find examples. Let A be exact height, and let B be the property of being over six feet tall. Then B supervenes on A but not vice versa. Brad raises the concern that in the case where the A-properties are the physical properties (the usual case for our purposes), it may be hard to find asymmetries. I think one can find at least a few straightforwardly asymmetrical cases by taking e.g. B = the total mass of the universe, or B = the property of having mass, or even B = position. It wouldn't be too hard (I think) to construct a world that is physically different from ours, but has the same total mass, or has the same distribution of massive objects (but e.g. with different precise masses), or has the same distribution of positions (but maybe different particles at some positions). The physics/chemistry and physics/biology cases are at least smewhat tricker, though. Of course things depend on just what we take to be the chemical and biological properties. But there are at least two natural places to look for asymmetry. (i) Entities in our world that have physical properties but no biological (chemical) properties, and which make no contribution to biological (chemical) properties of other entities. Maybe isolated photons, for example. Then there's presumably a possible world a lot like ours but where that entity is absent. That world will be biologically but not physically identical to ours. Brad gives some cases like this. (ii) Entities with biological properties and some "biologically irrelevant" physical properties. E.g. one *could* claim that one could change the position of one particle in my big toenail an iota without changing any of my biological properties. Of course that depends on just what we are counting as a biological property. Anthony also gives the example of switching physical properties wholesale (e.g. matter to anti-matter) while preserving biological properties. [Anthony also asks whether natural supervenience can be similarly asymmetrical. I think the answer is that it is a bit harder to find asymmetries, because we have fewer possible worlds to play with, but we should still be able to find them. E.g. the world without the photon, or with the moved particle in my toenail, are presumably naturally possible. And clearly for examples such as the height case, one has asymmetrical natural supervenience. Finally, someone like me might argue that the psychophysical case is an example of asymmetric natural supervenience, if e.g. a silicon duplicate of me might have the same phenomenal properties -- abstracting away from certain worries posed by panpsychism and the like.] This being said, presumably there are certain liberal ways of construing biological properties that may lead to asymmetry. For example, if one includes things like "being exactly 2.3 light years from something that is alive", and similar properties, then even the photon case might come out symmetrical. I recall there was a paper in the Journal of Philosophy a few years back: Miller, R.B. 1990. Supervenience is a two-way street. Journal of Philosophy 87:695-701. that made something like this point to argue that most interesting cases of supervenience are symmetrical. Personally I think it makes sense to go with a narrower notion of biological property. But this sort of example does at least make the useful point that the mere fact that B supervenes on A doesn't show that B is ontologically dependent on A. In the case above, the physical properties might supervene on fine-grained and relational biological properties, presumably without ontologically depending on them. Some people react by insisting that we need a further clause of "ontological dependence" for the supervenience relation to be useful. I think Grimes made a point like this in Grimes, T. 1988. The myth of supervenience. Pacific Philosophical Quarterly 69:152-60. arguing that supervenience isn't at all useful in understanding ontological dependence. Some similarly argue that at least an asymmetry clause should be built into the definition. Personally I think we should leave the definition of supervenience as it is, allow the conceptual possibility that A and B can supervene on each other in some cases, and consider the question of ontological dependence case by case. I think one can at least say that when (logical/metaphysical) supervenience fails, ontological dependence fails, so supervenience can be considered a necessary condition here. Just what further is requires to give a sufficient condition is an interesting question. Any ideas are welcome! (One might presumably also want to deal with questions about e.g. mathematical facts supervening on anything (as they're necessarily true) without being ontological dependent on those things, etc.) I think Erik L.'s point about fixing and informativeness is related to this point. I think Erik is pointing toward the sense of "fixes" in which A "fixes" B when B ontologically depends (and is determined by) A. So in the cases above, physics might supervene on fine-grained biology without being "fixed" by it in the ontological sense (though I think one could say that it is "fixed" at least in a modal sense). Again, this points to the fact that supervenience doesn't give a complete accounting of ontological dependence. Similarly, one can argue that it doesn't give a complete account of reductive explanation (I think Erik's point about "informativeness" is tied to this). So again people could think about what we need to add to supervenience to get ontological dependence and reductive explanation. I think in most cases of supervenience on the physical, it's plausible that this goes hand in hand with ontological dependence on the physical, though there may be a tricky case or two we'll encounter later on. (E.g., what if a strong sort of panpsychism is true, and every physical particle is constituted by an inner phenomenal essence?) In any case the central application of supervenience will be in its guise as a necessary condition for ontological dependence; e.g. in the mind-body case where a failure of supervenience would imply the falsity of materialism. The idea that supervenience is a necessary condition for ontological dependence is relatively uncontroversial, I think, at least among those who buy into modal talk about these things. All this ties into Angela's second point about supervenience as a convenient label. It's certainly the case that supervenience won't solve all our problems for us; it may not give an exhaustive analysis of ontological dependence or reductive explanation, for example. On the other hand, it may be able to do a good amount of work in clarifying those notions (e.g. at least as a necessary conditions, and perhaps as part of a sufficient condition); and it certainly has roles in other philosophical areas. I guess the moral is that a technical notion such as supervenience should be seen as our servant, not as our master. (I think more or less the same goes for notions such as "entailment", actually.) It doesn't solve major philosophical problems for us on its own, but we can hope that it at least plays a useful role in clarifying them, and we can judge its application and usefulness on a case-by-case basis as we go. That's enough for one note. All thoughts are welcome! --Dave. P.S. I've used Thony's idea of putting the mailing list in the Bcc label. The only trouble is that it means on can't reply directly to the list. That isn't a problem as long as one has kept a copy of the list around the place. If anyone needs a copy, let me know. Hopefully a listserv will be set up soon. From chalmers Wed Jan 27 20:07:51 1999 Date: Wed, 27 Jan 1999 20:07:47 -0800 (PST) From: David Chalmers To: chalmers Subject: Laws, angels, etc. Status: RO A few other issues that came up. (1) Is there a strict boundary between logical and natural possibility, or a gray zone? (Erik H.) Hmm. I suppose I'd say there's a gray zone for logical possibility at least of statement, because of vagueness in the meaning of words. E.g., it might be vague whether it is logically possible to have a heap with four grains of sand. But that's not a vagueness in the boundary between logical and natural possibility per se. A vagueness in that boundary would require something that is clearly logically possible, such that it's vague whether it is naturally possible. I suppose one could again find examples due to vagueness in language (say it were vague whether photons counted as "particles"; then it might be vague as to whether massless particles are naturally possible). But those aren't very interesting. Leaving such cases aside, a gray zone for natural possibility would presumably involve a gray zone in the laws of nature themselves. I suppose one could try to find such areas of grayness -- e.g. sometimes it's not clear how much variation in initial conditions should be counted as naturally possible, and how much should be seen as built into the laws of nature. But I think for most of our purposes it won't hurt to see the bounds of natural possibility as fairly strict. (2) What if there are nonphysical angels (etc.) in our world? (Erik H.) If that's the case, then materialism is certainly false in our world. Here the supervenience definition in 2.1 seems to give the right result. There's a world physically identical to ours but without the angels (presumably), so angels won't logically supervene on the physical by the account on p. 39 (and p. 40). And the angel facts will be positive facts about our world that don't supervene, so materialism will be false by the account on p. 41. Of course, we don't need to make assumptions about whether or not there are angels in our world for the purposes of defining materialism, as long as our definition gives the right results either way. The particular difficulty is raised by the case in which (1) there are no angels (etc.) in our actual world, but (2) there's a physically identical possible world (not actual) with angels. In this case, the intuition is that materialism should be true in our world, despite the existance of a physically identical world where things differ. Hence the modifications to the definition. In the case where the actual world has angels, the original definition was already OK. Re angels traveling across worlds: I guess I'd say (with Lewis) that if a being can move from one world to another, then those worlds are causally connected, and so aren't really different worlds in the first place. Just one big world with two parts. Of course the same being might arguably be present in two worlds (e.g. a possible world where I'm a mathematician, not a philsopher), but that's not the same as traveling between worlds. (3) Are laws of nature captured in the physical properties of the world, or are they additional? (Anthony.) This comes down to the question of Humean vs non-Humean views of laws. Personally I lean toward a non-Humean view, holding that laws involve more than any kind of regularity. It seems to me that two physically identical worlds (at least identical in their distribution of physical properties) could have different laws (e.g., Tooley's example of worlds that differ in the laws governing the behavior of a pair of particles that never actually meet), and maybe that there could be a world physically identical to ours but with no real laws at all (the Hume world where it's a giant cosmic coincidence). If so, laws (and causation, etc.) involve something ontologically significant in the world over and above distribution of particular physical properties. I discuss this a little on p. 86 of TCM. I think a commitment on this view isn't essential to most of the core issues we'll be discussing, though. I avoid the issue for the most part in the book by stipulating that physically identical worlds have the same physical laws (i.e., physical laws are built into the physical supervenience base). The issue is interesting, though, as another example of the way in which ontology and modality interact, and as another domain in which there are modal arguments with ontological consequences. We'll see later on that there are interesting parallels between the law case and the phenomenal case. One complication is that a smallish minority of philosophers hold that laws of nature are metaphysically necessary and hold in all possible worlds. If that's so, then metaphysical and natural necessity arguably collapse (and so will metaphysical and natural supervenience, etc). On my preferred usage, such laws still won't be logically necessary, as it is still conceivable that they are false, there are conceivable scenarios with different laws, etc; so we'll still have a distinct notion of logical necessity and logical supervenience. (The necessity of laws will be an example of what I call in the book "strong metaphysical necessities".) But this is a tricky matter that we'll be discussing later on. (Again, any clarifying questions on logical vs. metaphysical necessity, etc, are welcome.) (4) Thony's (interesting) question re Kripke and the 2-D framework. Maybe I'll save that one until we're focusing on those issues in a little while. Looking forward to people jumping in and continuing the discussion... --Dave. P.S. I hope I really Bcc'd it this time. From erikh@U.Arizona.EDU Wed Jan 27 08:13:47 1999 Date: Wed, 27 Jan 1999 09:12:59 -0700 (MST) From: Erik A Herman To: David Chalmers cc: aburnett@U.Arizona.EDU, agillies@U.Arizona.EDU, akolers@U.Arizona.EDU, atlane@U.Arizona.EDU, bayne@U.Arizona.EDU, bradt@U.Arizona.EDU, cowley@U.Arizona.EDU, erikl@U.Arizona.EDU, jtismael@U.Arizona.EDU, laj@U.Arizona.EDU, patrickr@U.Arizona.EDU, rachaelp@U.Arizona.EDU, sch@U.Arizona.EDU, shaughan@ns.arizona.edu, tolliver@U.Arizona.EDU Subject: Re: Initial discussion Status: RO OK, here goes, First is a question: by definition, is there a strict boundary between logical and natural possibility, or is there a gray zone? Comment: The idea that there is a world physically identical to ours but has extra nonphysical stuff: Can we postulate anything about that world when we don't know about our own worlds non-physical stuff--if it is bound by logical entailments, etc? (if that makes any sense) Assuming our world has no angels is a strong assumption perhaps. And maybe these angels can travel across worlds? Is that allowed? -Erik From atlane@U.Arizona.EDU Wed Jan 27 14:08:55 1999 Date: Wed, 27 Jan 1999 15:08:10 -0700 (MST) From: Anthony T Lane To: Erik J Larson cc: Brad Thompson , David Chalmers , aburnett@U.Arizona.EDU, agillies@U.Arizona.EDU, akolers@U.Arizona.EDU, bayne@U.Arizona.EDU, cowley@U.Arizona.EDU, erikh@U.Arizona.EDU, jtismael@U.Arizona.EDU, kcreath@primenet.com, laj@U.Arizona.EDU, patrickr@U.Arizona.EDU, rachaelp@U.Arizona.EDU, sch@U.Arizona.EDU, shaughan@ns.arizona.edu, tolliver@U.Arizona.EDU Subject: Re: Initial discussion Status: RO Erik points out that if B-properties supervene on A-properties, it seems that A-properties must also supervene on B-properties. It seems that this possibility depends on the type of supervenience being discussed. If we are talking about logical supervenience, this does not seem to be true. At the global level, given that the world has certain physical properties and natural laws that constitute its A-properties, there is no possible way in which it could have different biological properties. However, the reverse of this does not seem to be true-- one can imagine a world that is biologically identical to this one, but in which there are slightly different physical properties. For instance, one can imagine a world composed of anti-matter, but which has suitably different laws than we have in our own world such that it is biologically indistinguishable from our own world. I am not sure whether this is also true in the case of natural supervenience. Is it possible that, given the natural laws that hold in our world, a world could be biologically identical to ours and yet be physically different? I would hazard a guess that it might be. We can imagine a world identical to our own biologically, except that organisms are based on Silicon instead of Carbon. Of course, it might be impossible that such a world could be biologically identical to this one-- it would have to be a world in which there is a great abundance of Silicon and it must somehow behave chemically just like Carbon does in this world. Perhaps tis is ruled out by the laws of nature. A question about laws of nature: are they somehow captured in an exhaustive list of the physical properties of the world, or are they additional facts that are added on? I expect you (Dave) do not need to say much about them beyond the fact they may be different in logically possible worlds, but not in naturally possible worlds. Anthony From rachaelp@U.Arizona.EDU Thu Jan 28 09:29:31 1999 Date: Thu, 28 Jan 1999 10:28:46 -0700 (MST) From: Rachael J Parkinson To: David Chalmers cc: aburnett@U.Arizona.EDU, agillies@U.Arizona.EDU, akolers@U.Arizona.EDU, atlane@U.Arizona.EDU, bayne@U.Arizona.EDU, bradt@U.Arizona.EDU, cowley@U.Arizona.EDU, erikh@U.Arizona.EDU, erikl@U.Arizona.EDU, jtismael@U.Arizona.EDU, laj@U.Arizona.EDU, patrickr@U.Arizona.EDU, sch@U.Arizona.EDU, shaughan@ns.arizona.edu, tolliver@U.Arizona.EDU Subject: Re: Supervenience and symmetry Status: RO David has suggested that global-logical supervenience is necessary but not sufficient for reductive explanation. It seems clear that supervenience is a necessary condition for reductive explanation- but it is not so so clear that a good explanation of a phenomenon can be cached out merely in terms of the lower-level facts. Searle implies this when he suggests that it is inadequate to explain a cocktail party or a football game merely in terms of the physical facts, though logical supervenience implies that this is possible. If a reductive explanation is to be illuminating, perhaps what is necessary beyond supervenience is what David suggested on page 51. He suggests that the project of reductive explanation requires both explication and explanation. Thus we can explicate a football game in terms of the rules, the players, the plays, etc. and explain it in terms of the lower-level physical facts, eliminating any mystery surrounding the high-level phenomena. My question: can some one expand on why local logical supervenience is too stringent a requirement for reductive explanation?(51)I'm not disputing the claim- I would just like to understand it better. -Rachael From erikl@U.Arizona.EDU Thu Jan 28 10:24:39 1999 Date: Thu, 28 Jan 1999 11:22:59 -0700 (MST) From: Erik J Larson To: Rachael J Parkinson cc: David Chalmers , aburnett@U.Arizona.EDU, agillies@U.Arizona.EDU, akolers@U.Arizona.EDU, atlane@U.Arizona.EDU, bayne@U.Arizona.EDU, bradt@U.Arizona.EDU, cowley@U.Arizona.EDU, erikh@U.Arizona.EDU, jtismael@U.Arizona.EDU, laj@U.Arizona.EDU, patrickr@U.Arizona.EDU, sch@U.Arizona.EDU, shaughan@ns.arizona.edu, tolliver@U.Arizona.EDU Subject: Re: Supervenience and symmetry Status: RO Rachel, the key is "local" logical supervenience. Dave gave the example of two physically identical organisms (ie meeting the condition of local supervenience) with non-supervenient high-level properties (their biological "fitness" say) because of differences in the match of the physical characteristics of the organisms to their environment. So if physical facts are not considered globally, certain facts about individuals may fail to supervene, even though they are entirely physical. I want to say that certain properties about individuals are context-dependent in a way that requires global fixing of facts to cache out the supervenience relation. Searle's quip about the football game is consistent with this. Erik L. On Thu, 28 Jan 1999, Rachael J Parkinson wrote: > David has suggested that global-logical supervenience is necessary but not > sufficient for reductive explanation. It seems clear that supervenience is > a necessary condition for reductive explanation- but it is not so so clear > that a good explanation of a phenomenon can be cached out merely in terms > of the lower-level facts. Searle implies this when he suggests > that it is inadequate to explain a cocktail party or a football game > merely in terms of the physical facts, though logical supervenience > implies that this is possible. If a reductive explanation is to be > illuminating, perhaps what is necessary beyond supervenience is what David > suggested on page 51. He suggests that the project of reductive > explanation requires both explication and explanation. Thus we can > explicate a football game in terms of the rules, the players, the plays, > etc. and explain it in terms of the lower-level physical facts, > eliminating any mystery surrounding the high-level phenomena. > My question: can some one expand on why local logical supervenience is too > stringent a requirement for reductive explanation?(51)I'm not disputing > the claim- I would just like to understand it better. > -Rachael > > From chalmers Thu Jan 28 17:54:08 1999 Date: Thu, 28 Jan 1999 17:54:05 -0800 (PST) From: David Chalmers To: chalmers Subject: Re: Supervenience and symmetry Status: RO Rachel is right that logical supervenience alone doesn't capture everything that is relevant about explanation. I do think that it's captures what's going on with reductive explanation, especially when one thinks of this as a two-stage process of explication plus empirical satisfaction. But it's also true that reductive explanation is far from the be-all and end-all of explanation. Obviously a reductive explanation of a football game or a cocktail party may not provide the most illuminating explanation of what is going on there. It will be a "mystery-removing" explanation in that it will explain how football games and cocktail parties are possible, but even after the process of explication plus empirical satisfaction is complete, there will presumably be a major role for high-level explanations in helping us to understand the high-level phenomena in a comprehensible way. There is still the point which came up last time, concerning cases (e.g. symmetrical cases) where B supervenes on A while not being intuitively ontologically dependent on A, or reductively explanable in terms of A. The supervenience of the physical upon sufficiently fine-grained biological properties might be an example. Maybe to handle this sort of case one needs to build in some further clause about the autonomy of the domain of the supervenience base. It's not entirely obvious how to cash this out, though, and all ideas are welcome. Re local supervenience, I think we often want to say that phenomena are reductively explanable even when they aren't locally supervenient. Fitness is an example, as Erik suggests. Perhaps my being a brother is another one. These things don't supervene locally, as they depend constitutively on environmental factors. But it still seems reasonable to say that we can give reductive explanations of them. In such a case, a reductive explanation will advert to lower-level properties of both me (locally) and of my environment. --Dave.