Gregory Bateson defined explanation as the mapping of experience onto a tautology.
Other key words here: rationalism, empiricism …
I quote an email from bkMarcus of 2003 11 14 (the setting follows):
You quoted [The Bateson definition of explanation] for years, and for years I didn’t get it.
I think, however deliberately GB might have defined his terms, that he picked the wrong words to convey that idea.
I wish GB hadn’t used the word ’tautology’ — which has the connotation now of circular reasoning, i.e., meaningless formalism.
As I now understand it, tautology (in GB’s usage) just means a logically self-consistent model. And an explanation is a consistent mapping of (a posteriori) experience onto a self-consistent (a priori) model, thus giving us a consistent overall model for our experience, or rather, for a subset of our experience.
This goes a long way to identifying the rift between rationalism and empiricism over the past few centuries. The debate, as I now understand it — having shifted my own emphasis from a pk- and Bateson-influenced (British) empiricism toward a more (continental) rationalism — is about how far we can get in the a priori realm before we have to appeal to experience.
I now believe that experience plays a greater role in checking our logic than it does in helping us to choose between otherwise equally logical models. And the emphasis in research seems to be so much on data, that logic feels like it’s been left behind, seen as necessary but not really important. (Which means that its necessity is itself sometimes ignored.) [emphasis added]
Brian and I were corresponding about my debut consumption of The X Files, a TV series he had followed with interest while I remained virgin to it. I finally got Season I, episodes 1-4 (1993) on DVD, and notified him of my exposure: mocking the show’s man-in-the-street ignorance of science. Poor Gillian Anderson has to play a woman with degrees in physics and medicine who yet doesn’t seem to understand the first thing about science. Naturally, I iterated a few Bateson citations.
pk quotes Bateson’s definition of explanation in several places. Additionally, it will come up in a number of random quote generators. Only today with Brian’s response do I see a pressing need to additional comment: and, for the moment, I allow Brian’s own comment to suffice. His contrast between English and Continental empiricism certainly deserves additional consideration as does the entire subject (or set of subjects). Thus there’s room for plenty more here: but not today.