School days.

A while ago I blogged that the best science education is pre-theoretic; instead of teaching that one or another theory is The Way Things Are, a good science education teaches students to assess whether any theory they encounter is a good theory, and how well it compares to competing theories. Is it falsifiable? How good is its empirical coverage? And so on.

One of the reasons I went into linguistics when I started college was because I didn't want to choose between language and math. Once I got over that, one of the reasons I went into linguistics for real when I started graduate school is because my real passion was in the philosophy of science. Linguistics is an old discipline but sort of a new science; it has not always been approached empirically, and as a relatively new science it's still sort of working out its methodology. For someone interested in math, language, and the philosophy of science -- and that describes a whole lot of the people who go into theoretical linguistics -- it's pretty much the optimal field.

Until the lack of established methodology gets kind of frustrating. And the physics-envy in the field of linguistics gets old. But I have physics-envy too. I am guilty.

But so the philosophy of science. One thing that the variety of linguistics that I was taught really excelled at, by which I mean the variety of linguistics that is associated with Penn and few other places, is that at heart it is pre-theoretic in just the way that science is supposed to be. Even where some theory or other is favored to explain some set of data, the heart of the field is more about how to evaluate competing theories than it is about any particular flavor of explanation.

And so to address a question from a recent lunch conversation I had, that is what I got out of school.