Complexity

The other day I finished reading Complexity: The Emerging Science at the Edge of Order and Chaos by M. Mitchell Waldrop (hilarious back-page plug: “if you liked chaos, you’ll love complexity!”). I’m not much of a bookworm usually, but very occasionally I will relentlessly read something from cover to cover in a short space of time, which I managed to do in this instance (this also happened recently with Robin Ince’s Bad Book Club [which is hilarious, and you should buy it immediately], so it’s not just a science book phenomenon). Complexity was written in the early 90s, shortly after the Santa Fe Institute for complexity research was founded. It serves partly as a documentary of the founding of the institute and its early years, partly as a biography of the key figures involved in its development, and partly as an introduction to the ideas of the new kind of science that the institute was trying to pursue. This works remarkably well – you get a good overview of the relevant scientific ideas, but almost feel like you’re reading a novel. The constant breathless excitement in the tone does begin to grate a little, but it manages to convey the passion of the researchers and the thrill of scientific discovery. It certainly rekindled my excitement with some of the ideas that made me interested in computational modelling and biological systems in the first place.

As the book was written at the height of enthusiasm of the scientists involved, it unfortunately lacks a critical view – no stories of the less successful avenues of research that were pursued, and little mention of criticisms by “mainstream science”. Interestingly, the Santa Fe Institute’s own history page mentions this kind of criticism in passing:

As the Institute’s research interests and reputation grew, so did its list of detractors. Exploring new scientific territory meant that some lines of inquiry failed to live up to expectations. Some researchers in mainstream science felt that complexity science was long on promise but short on results. The criticism culminated in a June 1995 Scientific American article by senior writer John Horgan that openly mocked not only the science of complexity but the scientists doing it. The article, today regarded at the Institute as a wakeup call, caused many in the complexity community to do some soul searching about their field.

I am definitely sympathetic to the approaches pioneered by the Santa Fe institute (amongst other places, of course), though. I particularly love the emergence of complexity in the behaviour of cellular automata – such simple rules can lead to such interesting behaviours. You can’t prove truths about the universe from such simulations, but I think they can give you very deep insights into problems that you just can’t get from other approaches. But that’s just, like, my opinion, man. Go and have a play with a game of life simulator, like this one, to see what I mean.

Anyway, I recommend it, but bear in mind it is quite un-critical, and quite out-of-date. If you want to have a hands-on play with some of the ideas of complexity theory and know a bit of programming, you could do a lot worse than pick up Alan Downey‘s book “Think Complexity” (available for free from his website).

I’m now onto an old software classic, The Cathedral and The Bazaar, which actually has some surprising parallels – complex emergent behaviour in software development processes, defying standard theory through the collective behaviour of individual programmers…