Taking Philosophy Seriously

Mr. Waggish has had a fine series of posts provisionally assessing the legacy of the recently departed Richard Rorty. I don’t think now is the time for me to say anything comprehensive about Rorty, if there ever will be a time: but something Waggish says about a way analytic philosophers have of holding fast to their ideas deserves a little consideration. He cites two examples of the seriousness with which analytic philosophers have taken their arguments to heart: “Derek Parfit’s views on (lack of) personal identity, by his own admission, brought him great comfort in facing death. David Lewis, to cite an extreme example, truly believed in an infinite number of alternate worlds for modal purposes.”

As for the first example, I am a little puzzled: Derek Parfit is alive, and I have never heard of any serious challenge to this state — perhaps there was indeed an illness, or maybe Waggish is just invoking the ordinary concerns about mortality most people have. But Lewis’s modal realism is what I want to talk about. Not really the pros and cons of the argument: I’m not a huge fan of Lewis’s oeuvre, and so not much hinges for me on a close reading of On The Plurality of Worlds, or (should the decision go against modal realism) showing how the rest of Lewis’s analyses can be saved from falling with it. What I want to consider is what it would really mean to be a modal realist, in terms of the intellectual environment against which such a view can be defined.

As the reader may know, the formal logic of necessity and possibility was a field that had, until the early 20th century, lay fallow for some centuries. But starting with a few scattered figures (Hugh MacColl, C.I. Lewis), some axiomatic systems for reasoning modally started to appear. In the late ’40s, Rudolf Carnap tied together such research with the already-existing Fregean philosophy of language to develop a semantics that made use of “state-descriptions”, ways of describing how the world might be. A little over a decade later, a very young Harvard student named Saul Kripke rigorized Carnap’s theory by casting it in a purely mathematical form.

Taking a clue from a famous expression of Leibniz, Kripke restyled state-descriptions as “possible worlds”: sets of propositions and valuations for them, which could be collected together in a model and used to evaluate the truth or falsity of modal sentences — a proposition being possible if some world in the model had it as true, and necessary if all worlds in the model did. Modern modal logic, a rich and diverse field, was born. Now, a quick reading suggests that what Lewis did, in saying that possible worlds were as real as anything, did is to take a metaphor and cash it in an inappropriate way — but I think that is imputing a certain unseriousness to him. Let me explain.

Some readers may know that Lewis did not “really” use Kripke-style modal logic, but instead preferred a system he called “counterpart theory”. Counterpart theory was intended, in part, to do away with the problems about identifying objects in different possible worlds as the same thing: instead of worrying about “transworld identity”, one could consider “counterparts” in different worlds that would to a certain degree be like or unlike. However, what Lewis did take away from mainstream modal logic was what, in a manner of speaking, we might call the “concreteness” of possible worlds. Such possible worlds are perfectly determinate, even from no particular point of view; their contents are not relativized to our interests or theories, excepting the incorporation of possible interests and theories in the slightly more complicated “two-dimensional” semantics.

So, it seems to me that Lewis is making a metaphysical argument for a feature of reasoning implicit in the formal apparatus: and this, I think, is a good thing to take “seriously”. When the apparatus one is employing is supposed to have only a heuristic value, perhaps it is harder to really accept its results.