On Imperfect Perception, or Perceived Imperfections?

There’s something I need to make right.

The attentive person’s honest, real-time report is an imperfect approximation of her subjective experience, but it is the only game in town.

When a fruit salad, a lover, or a jazz trio is just too imperfect for our tastes, we stop eating, kissing, and listening. But the law of large numbers suggests that when a measurement is too imperfect for our tastes, we should not stop measuring. Quite the opposite — we should measure again and again until niggling imperfections yield to the onslaught of data.

Those subatomic particles that like to be everywhere at once seem to cancel out one another’s behavior so that the large conglomeration of particles that we call cows, cars, and French Canadians stay exactly where we put them. By the same logic, the careful collection of a large number of experiential reports allows the imperfections of one to cancel out the imperfections of another. No individual’s report may be taken as an unimpeachable and perfectly calibrated index of his experience — not yours, not mine — but we can be confident that if we ask enough people the same question, the average answer will be a roughly accurate index of the average experience.

One of the balls bouncing around my head for the last few months grew flubber a few days ago, and the passage I read today spoke directly to it.

Daniel Gilbert’s Stumbling on Happiness was the only non-fiction I could find in the house where I’ve spent the last week, and despite Mr. Gilbert’s overwhelming cleverness (How’s that working out for you?) there are some interesting ideas and data.

I wouldn’t recommend the book, but maybe he’s more to the point in his TED talk?