[<< | Prev | Index | Next | >>]

Thursday, May 18, 2006

There is no Moo

Another email excerpt, this time on people's proclivity for underestimating the relevance of (often to the point of completely dismissing) contrapositive evidence.

According to this standard, "I do not see black marks on the wall" is not an observation, while "I see a painted wall" *is* an observation. ("There are no black marks on the wall" is thus an interpretation.)

Right, this is just the sort of thing I'm talking about. That approach is totally misleading, and people use it as an excuse to discount negative evidence when they shouldn't.

Say I'm holding an object behind a blinder, and each round you have to guess whether I turned it over. After a while you realize it's obvious because I move my arm when I do it. We repeat this with different objects, and your scoring is 100% for quite a while.

On each trial at that point how confident are you in your answer?

Now we switch the object out for a moo-can, and every time it turns over it goes "moooo". You find you're doing 100% there too, but you already were, so the mooing isn't really relevant.

Then one round you see me move my arm, but hear no "moo".

There's no painted wall to observe here unless you want to get really cryptic and claim you're observing the air molecules in standard brownian motion.

"I did not hear a moo" is a perfectly valid observation, and quite relevant!

Let me add that the positivist answer of saying you observed silence is not only mere semantics (since silence is defined as the absence of something) but inaccurate: What about the radio playing in the background? The traffic noise outside? Your own breath? The truly relevant observation is not silence, nor any particular pattern of vibrations in the air, but precisely the lack of a particular pattern of vibrations in the air.

[<< | Prev | Index | Next | >>]