Complicated systems, ritual, and the common cold

Revision as of 14:09, 3 February 2011

WR Ashby writes this wonderful stuff arguing for complex systems:

Science stands today on something of a divide. For two centuries it has been exploring systems that are either intrinsically simple or that are capable of being analysed into simple components. The fact that such a dogma as "vary the factors one at a time" could be accepted for a century, shows that scientists were largely concerned in investigating such systems as allowed this method; for this method is often fundamentally impossible in the complex systems. W. Ross Ashby. (1956) An introduction to cybernetics. p.7

It has implications for everyday thinking. People interacted with complex systems before science.

I've got a cold right now. If I want to cure it scientifically I could vary the factors one at a time. I could try lemon for a week, what to get a new cold in exactly the same conditions and then try ginger for a week, then cayenne, echinacea, garlic, tea, vitamin C, prayer, etc etc. I would track my experiments, find what works, and discover the best way to cure a cold.

But a cold is too rare and brief for that methodology to work. And you never catch the same cold twice. But what about clinical trials? Clinical trials let you do science on complex things like the cold. They organize the efforts of thousands of individuals and create space for understanding a cure one variable at a time.

What is the effect of using this simple methodology on a complex phenomenon? It ends up that you can't really say much: a lot of things maybe work a little, or work well but just on symptoms, and there are a few things that don't work but taste good. Citing the Mayo Clinic, the Wikipedia article on the common cold says "There are currently no medications or herbal remedies which have been conclusively demonstrated to shorten the duration of illness."

But for the many many folk cures, the cost of adding an ineffectual herb to your nostrum is low enough to be worth the chance that it is doing something. And if it tastes good, all the better.

I'm just going to try everything at the same time with no control until I get better. When I get sick again, I'll try everything again. There is a lot of room here for superstition to find a home in ritual. In a psych lab, you might provide a subject with both prayer and a hammer. If they use both on a nail, and the nail goes in to the board, they will tend to conclude that both things worked. In the lab you will call that person irrational. But this failure in the lab might be perfect preparation for the real world. This clean reasoning makes assumptions that may not suit the problems people face in a complicated-but-forgiving world. Everyday problems are mundane, low risk, and incredibly complex.

The engineering methodology helps people achieve goals in simple environements: environments that are static, manipulable, and well understood. We need a different methodology to achieve goals in complex environments: those that change and are poorly-understood. My old boss at [necsi.org NECSI] crafted the ideas behind evolution into an engineering methodology for complex environments. It is like a genetic algorithm on social systems, and, for me, it is an important part of the complex systems toolkit.

For the cold, you can count on some other heuristics as well: trying everything every time, and the uncontrolled, partly ineffectual, net-useful rituals that come out of it.