I just watched "Another Discourse on Method," a trailer for an apparently excellent movie. It reminded me of something I think about, often unsuccessfully: the foundation of fact. But tonight I remembered something I had concluded about it before: at the core of the issue is emotion. I can imagine someone arguing the contrary to the death, that emotions are explainable in terms of past experience, the collective memory, and possibly physics; unfortunately, I cannot argue against this standpoint, so if it's brought in we've found a circular (once I've set up my alternate reduction, of those other things to emotion) description. So assume it hasn't been brought up. I think you'll find my version palatable and (perhaps surprisingly) at least as solid.

The first thing you should notice (if you've watched the trailer; chances are you haven't) is that the interviewed do a lot of fumbling. They have presumably been asked a question along the lines of "what is the foundation of fact?" and (let me watch it again) they answer saying things like "It's just common..conception and..perception and..consensus that would..would determine what we think, are facts" and (to what makes karma a truth?) "Because it's..inev—it's..it's perv—it happens, it's—I mean, there's no way around it." They expect the answer to be right there for the interviewer: belief in fact probably being a foundation for much of what they do, they expect to be able to explain what it is well. As they fumble along, they continue to discount the possibility that they'd actually have to stop and think for a while before being able to adequately respond.

Catch me on a bad day, and I'd probably do the same. But what I concluded a while ago is that emotions are really the basis for facts. At this stage of the game, we continue to believe that computers can't tell fact from fiction, but why? Might it not have something to do with their lack of emotion? If we programmed a robot with a face to smugly assert, "Yes, Columbus got to America in 1492, I'm sure of it," nodding briefly, would it fool you into believing it could recognize a fact?

If we view mathematics as the study of true statements, which math teacher is more convincing: the one who mumbles articulately, gaze fixed on the board, deriving line from previous lines through standard rules of inference and some "obviously"'s, or the one who stares at you menacingly and asks, "Now do those two lemmas look like they're aching to be used together, or what?"

When Galileo drops a ball down a ramp for the second time with the aid of an orchestra playing presto with distinct thirty-second notes and locks his whole thought and being onto its position, he is preparing himself for an emotional release: Yes, as it reaches the finish line, yes, I do recognize that combination. That combination of that part of the song and the view of the ball right there, it has special meaning to me. I've linked it to success, meaning I set myself up to experience a positive reaction to it, at the same time linking all other combinations of parts of the song with that view to experimental failure. Ah, I recognize this linkage, too: this linkage I have performed, it is the linkage I set up when doing an experiment. I have come to associate the presence of an orchestra, a ramp, and a ball with experiments. I have also set up a pathway whereby recognizing all this impels me to further action. He walks over to his vellum and writes laurus, success. That's indeed how long it takes it to get there from that height; I'll debate it with you to the death on the basis of the particular emotional nature of the vivid memory of the second run: yes, every time I check, it's there. Positive. That was a successful measurement.