So I was very excited when I started reading Taylor and Dennett's treatment of possibility, necessity, and causation in chapter 3 of the Dennett's Freedom Evolves - essentially they present a way of thinking about these concepts in terms of possible worlds in order to show that our ordinary understanding of the words are pretty much neutral with regards to determinism or indeterminism - it's an enlightening discussion that, at least for me, clears up some of the conceptual terrain.
I don't actually want to talk about the content, but rather make a quick note about the presentation.
In his book, Dennett mentions that the chapter is that its content is basically a "gentler" version of the argument they put forward in Taylor and Dennett (2001).
What's so interesting is that parts of Freedom Evolves are lifted almost verbatim from the 2001 paper, with the major changes being that Dennett softens the presentation by pretty much removing all of the formalization by rewriting in ordinary English.
So what, in the 2001 paper, was rendered as
There exists (within some set X) a possible world in which the sentence "x (x is Socrates x has red hair)" obtains
is rendered in Freedom Evolves as
There is (at least one) possible world f in which the sentence "There is something that is Socrates and he has red hair" is trueThe really cool thing is that by reading these two versions of what is essentially the same thing, one can really see how valuable formalization can be in philosophy.
Reading through the 2001 paper I was struck by just how much clearer it was than the rendering in Freedom Evolves. The formal renderings (without getting into argument's about just how conditionals are supposed to work) are just much more precise and leave a lot less room for misinterpretation.
I don't believe that philosophy should be a discipline where papers are exclusively written in arcane symbols, but in reading these two versions next to one another my conviction that Formal logic is an essential element in any thinking person's "mental toolbox" has been greatly strengthened. The Formal-logically-type-of items in this toolbox should at minimum include a basic understanding first-order logic, and - perhaps - just enough modal logic to get along.