1. One cannot do modal logic, let alone modal metaphysics, without both modal concepts and 'modal intuitions.' One has to start from a pre-thematic understanding of modal concepts such as possibility and necessity and how they are interrelated and also from certain prior convictions about what counts as possible and necessary. (The same is true in other disciplines such as ethics: if you don't grasp the distinctions and interconnections among the permissible, the impermissible, the obligatory, and the supererogatory, and have some reasonably firm intuitions about what particular actions and ommissions fall under these categories, then there is no point in doing ethics.)
2. For example, I am now seated. But I might have been running now, where 'now' in its last two occurrences picks out (rigidly designates) the same time. In other words, I am seated at t, but I might have been running at t. If you don't agree with this intuitively, then we are not going to get anywhere. We must start somewhere, with 'modal data.' In general, one cannot do philosophy without data, without certain things one takes for granted, not uncritically of course, but provisionally. One cannot start from thin air. Let me put it another way. It is actually the case that I am seated, but it is possibly the case that I am running, hence not seated. If you can't swallow this, then modal logic and modal metaphysics must remain non-starters for you. I would say you are modally blind, or in Pee-Cee jargon 'modally challenged.' And note that the question is not whether, given that I am seated at t, I can be running at a time t* (t*>t); the question is precisely whether I could have been running at t given that I am seated at t.
3. My running now is a possible STOA (state of affairs). Indeed it is a merely possible STOA. By definition, the merely possible is that which is possible, but not actual. Everything actual is possible, but not everything possible is actual: the possible 'outruns' the actual. The portion that does the outrunning is the merely possible.
4. Most of us have modal beliefs. I believe, for example, that I might have been running now rather than blogging now. I believe that that is a real possibility: it is not merely something I am excogitating or dreaming up. I am thinking about it, of course, but it is not my thinking about it that makes it possible. My thinking is grabbing onto something independent of my thinking. The possibility is grounded in rerum natura. Nor is the possibility of my running now something merely epistemic/doxastic. It is not a reflection of my ignorance of all the causal factors involved in my being seated.
It is hard to make this point in a nonparadoxical way. I am tempted to say, paradoxically, that unrealized possibilia are part of the fabric of the real. A merely possible individual or STOA is not actual, obviously. But it is not nothing either. It is something definite, but not something merely excogitated or mind-dependent. 'I might have been running now' is true. What makes it true? No actual STOA makes it true. And nothing mind-dependent makes it true. So what makes it true? We can do modal logic without solving this problem. But it looms large in modal metaphysics.
But suppose are a Spinozistic determinist: you hold that my blogging now is broadly-logically necessary: not just causally necessary, necessary given the (logically contingent) laws of nature together with initial and circumambient conditions, but such that it is broadly-logically or metaphysically impossible that I be running now.
How do I know that my modal belief is true and that yours is false? That is an interesting and difficult question, but it belongs within the epistemology of modal knowledge, rather than within its logic or metaphysics. But perhaps I should say something to rebut modal Spinozism which in effect collapses the actual, the possible, and the necessary. Ayn Rand, by the way, appears to be a latter-day modal Spinozist when it comes to the non-man-made.
5. Consider a morally relevant action such as driving recklessly. That is a piece of behavior for which one who does it is morally responsible. But one can't be morally responsible unless one possesses freedom of the will, a notion that (arguably) must be unpacked in terms of 'could have done otherwise.' But one couldn't have done otherwise than drive recklessly unless it is possible that one not drive recklessly at the very time that one did drive recklessly. So if you believe that we are morally responsible for some of our deeds, then you have a reason to believe that there are unrealized possibilities. (To make the argument rigorous one would have to have a refutation of compatibilism, the view Kant labelled a "shabby evasion.")
6. Note the following point of logic. (P & Poss ~P) is not to be confused with Poss (P & ~P). To say that I am driving recklessly at time t but might not have been driving recklessly at t is not to say that it is possible that I drive recklessly at t and not drive recklessly at t. The second proposition, but not the first, is a contradiction.
7. Of course, I haven't definitively proven that there are unactualized possibilities, but I have shown that this is a reasonable belief. But then virtually nothing can be definitively and compellingly proven. Dogmatists take note. Now let's say that you accept that there are merely possible STOAs. A merely possible world can be defined as a merely possible STOA that is maximal.
The idea is that possibilities are not isolated, but come in 'world-sized' packages. Suppose I am running now instead of blogging. That will change all sorts of other things: my phoneline will be free, my running shoes will be pounding the trail and killing tiny critters, girls will stop to admire my fine form as I stride past . . . everything will be different.
Thus we speak of the merely possible worlds in which I am running now. Possible worlds are just total ways things might have been. The actual world, which is of course one of the possible worlds, is just the total way things are. There is and can be only one actual world. And the world that is actual is contingently actual. If the world that is actual were necessarily actual, then the other worlds would not be possible. After all, to be possible is to be possibly actual.
8. We can go on to ask more technical questions about what possible words are: sets? maximal propositions? maximal mereological sums of concreta? There are two main conceptions, the abstractist and the concretist, but more on this later.
This a topic which I find interesting. The first time a logician told me that 'such-and-such' was an element of a possible world, (even though it seemed to be fairly absurd from my perspective), I told him I thought that, "there is only one possible world- and this is it." My ontological intuitions are such that idea of merely possible worlds 'feels' pretty fanciful to me.
Causality (at least at the macro-level) is central to my view of the world, so that a merely possible STOA is one which for me is carrying an albatross of contradictions around its neck. A LaPlacean demon might tell you that the only way that such a STOA could have been true would be if you had rewritten all of the preceding global (or as you put it 'world-sized') STOAs going back as far as time will reach.
Once while he was on a book tour promoting "Cosmic Jackpot: Why Our Universe is Just Right for Life", I heard physicist Paul Davies discussing how it would have been possible for the universe to have developed with different laws of physics that would have made the development of life (of any kind whatsoever) physically impossible. From that starting point, he arrived at the conclusion that the existence of life and consciousness actually are responsible for the universe having the laws which were necessary for them to come into being.
Given that he was presenting this as science rather than philosophy, I got irritated and actually argued with him a bit. (oops) The reason I mention this is that if you can generate possibilities (like his premise) which contradict so much of the way that the world is, then it makes me wonder just where you have to leave off on calling something a possible world.
I think that I more or less get what people mean when they speak of possible worlds. I may be wrong. It seems to me that a possible world is a global STOA the description of which contains no logical contradictions.
Is there more to it than that?
Posted by: anonymous | Friday, January 30, 2009 at 05:47 PM