Sunday, September 11, 2011

Reading Changizi on Color Vision

I'm now reading about color vision in Mark Changizi's The Vision Revolution. It's slow going, but NOT because it's difficult or poorly written. On the contrary, it's very well written and the material is pretty straightforward.

But it's NEW to me. Though I'm not an expert in color perception, I have read a good deal about it over the years and have thought about it quite a lot. I've even written several blog posts about it.

It's the joint product of my knowledge and my level of interest that makes the sledding rough. As I've said, what Changizi says is quite different from the literature I've read. All that other stuff’s in Cincinnati, as it were. Changizi’s writing from Timbuktu. That's one thing.

The other thing is that I really want to understand what he's saying, because I regard him as easily one of the best psychologists of his generation. Well, perhaps that's not right, as I'm really not in any position to judge psychologists of his generation. But I've read lots of thinkers in various disciplines, and he's one of the most compelling.

Because I find him compelling, I want to really understand what he's saying. That means I've got to hook it up with everything I've known and thought about color. I can't treat his ideas as self-sufficient capsules. It takes time to make all the hook-ups. I certainly can't make them in real time while reading him. I'm going to have to sleep on it, etc.

It’s like reading Latour. I have to take it slow.

So, one idea is that color vision isn't about picking out colorful fruit against a green background, an old idea that's one of the things I thought I knew. Tossing that overboard is easy. Zhrrt! It's done. And it’s easy to slip Changizi’s idea into place, provisionally.

His idea? That we have color vision so we can sense other people’s moods as indicated by blood flow in the skin. Zhrrt! It’s been installed. Provisionally.

What’s deep, what’s blowing my mind—excuse my French—is an idea he’s developing along the way: The monkey doesn’t see the hump on his own forehead. Well, that’s not what Changizi says; it’s an old Sesuto proverb my father picked up from Blaise Cendrars, L’Anthologie Nègre, translated as The African Saga by Margery Bianco. But it’ll do as a gloss on what he’s saying.

Changizi’s talking about color terms and notes that we don’t have any good way of designating skin color. All these colors, all these ways of designating them, but no good words for flesh tones. Why?

Well, I’m only halfway there yet. Had to stop to decompress while writing this note. Then I got interrupted by a phone call from my sister. That irritated me for 30 seconds, but then my sister captured my attention with her interest in the up-coming object-oriented ontology meetings. She’s the one who reminded me of that proverb, which was one of Daddy’s favorites.

The idea is that once we’ve named the color of the thing, we forget the color itself and, in effect, replace it with the prototype color that goes with the name. That process makes us relatively insensitive to color nuance. Because we don’t have effective color names for skin tones, Changizi argues, we’re more sensitive to nuance.

And it’s the nuance that tells us about blood flow. So, the lack of color terms for skin tone protects us from nuance blindness.

Neat, no?

Do I believe it? Yes. Provisionally. I’ve got to read the whole argument, and think about it. That’ll take time.

But I’m going into it on the provisional assumption that Changizi’s right.

Why make that assumption?

1.) It’s the right thing, the charitable thing, to do. It’s the most effective way of getting into the thick of it. Cutting things off is easy.

2.) I’ve read enough of his stuff (e.g. his latest book, Harnessed) to know that he makes a good argument.

Film at 11:00.

No comments:

Post a Comment