Since I was still a snotty, coughing bag of infection yesterday, I spent most of it propped up in bed, listening to the Beyond Belief II lectures.

I found the discussion of reductionism by the last three speakers on Wednesday to be pretty interesting. They start at about 2:43 with Stuart Kauffman, followed by Sean Carroll and David Albert.

The basic dispute had to do with whether or not, in theory, it would be possible to describe macro-level phenomenon like love, or the function of a beating heart, in terms of physical interactions at the atomic or subatomic level. Kauffman argued no, that there are properties involved in higher-level phenomenon that aren't captured by the lower-level descriptions, that come about via emergence. Carroll and Albert argued yes, but that it wouldn't be very practical.

I particularly liked Albert's talk about the Copenhagen Interpretation of quantum physics, that our inability to know the direction and velocity of a particle at the same time, says something about the theoretical limits of our knowledge, rather than the current practical limits of our knowledge. Albert argues that such an interpretation is silly (Einstein was one of the sole dissenters), and marvels at how it has been so popular and so widely misused.

The whole batch of talks reminded me of David Marr's levels of analysis, which are different levels at which to talk about and study cognition, from lower-level neural implementation to higher-level algorithmic descriptions.

The whole discussion is really an open question about the ultimate nature of the universe and our ability to understand it, so a definitive answer just isn't going to be there. The reductionist paradigm, assuming that the universe is ultimately causal and naturalistic, has been a huge success. Kauffman basically thinks we're going to get to a point (or have already gotten to a point) where those assumptions break down. Albert and Carroll think we need to keep making those assumptions and just keep working. I think they make a good case that people who have argued along the lines of Kauffman in the past have been put in their place as new theories arise, and that it's premature to talk about theoretical limits to our knowledge.

Reductionists often get depicted in a straw man version as those who think the stock market can be explained in terms of particle physics. Instead, I'd say that higher-level abstract concepts are necessary to talk about macro-level phenomena, because they are an efficient way to chunk up all the lower-level phenomena they encompass. But the big difference is that the non-reductionist seems to think there's some other "stuff" that comes out of the higher-level interactions, while the reductionist simply thinks that there are no new properties emerging, but just more complicated interactions that need to use higher-level concepts to describe them. And I guess that would put me in Albert and Carroll's camp and make me a reductionist.