Review: Stephen Jay Gould, Full House: The Spread of Excellence from Plato to Darwin (Harmony, New York, 1996)
I had read a lot about Stephen Jay Gould, but had only read some of his essays, and never one of his book-length works, before a few weeks ago I noticed a copy of Full House at a sale of old stock from the local central library. The press on Gould has been particularly voluminous this year, after he died in May, aged 60, to the cancer he had evaded earlier in his life—and so he had been higher on my list (you know, the list, the neverending, constantly shifting ranked order of titles of works you want to read, most of which are destined to remain keys to unexplored worlds at your own death). Finding the book was particularly fortuitous for me, however, since one of the big problems I have been wrestling with during the year has been why evolution should drive towards complexity; and I didn’t even know, buying Full House, that this was the central question it addresses. As soon as I started, I realised how lucky the find was, and felt straight away that I was talking one-on-one with Gould about the problem; at every stage, he would anticipate my questions with his answers, as if I had been whispering doubts in his ear as he composed it. The book is really quite a bit longer than it needs to be to answer the question, in the end—and it isn’t long, a well-spaced 230 pages, with illustrations. The solution, in the grand theme of intellectual revolutions, is actually a dissolution—an argument that it is a mistake to see evolution as driving anywhere at all. Complexity turns out to be something that results given enough free play in a randomly varying system, rather than a pre-ordained endpoint towards which the system is aimed. We have been misled, Gould argues, by the same anthropocentrism that saw us at the centre of the Ptolemaic universe, to view ourselves as a kind of endpoint of evolution—when by any measure available, the humble bacteria remains the most successful form of life around. Bacteria is the most numerous form of life in terms of any number of measures: raw numbers of individuals, numbers of species, genetic range between different species, ubiquity in differing environments, and even, perhaps, biomass (the argument here relies on the empirically somewhat shaky speculation that bacteria prosper throughout at least the outer crust of the earth, fed by geothermal rather than solar energy—an argument needed to get their biomass over that of plant life, which otherwise easily dominates due to the weight of the worlds forests). However, our anthropocentrism leads us to draw those quaint evolutionary sequences from single-celled through multi-celled, through plant, animal and finally human life, as if evolution in total is pushing in our direction - as if it is progressing, when really what we have is simply variation away from the bacteria. Life cannot really get any simpler than the simplest bacteria, and so an increasing amount of variation could only result in an increasingly complex few forms of life, so long as they are adaptively successful - like the drunk man’s random walk between a wall (through which he cannot pass) and a gutter (into which he will fall), he’s going to end in the gutter eventually, even if he never aimed there. Progress towards complexity, then, is a by-product of random variation given enough time, and so can be gotten out of local adaptational principles without adding anything at all that dictates an overall trend towards complexity. Pretty simple really, and made all the more clear by Gould’s reverse example of the disappearance of 0.400 hitting in baseball, which is an example of variation shrinking over time, giving the illusion of a driving trend (the illusory trend being the decline of hitting skill). In addition to the substance of the book, I liked Gould’s often celebrated knack for quoting cultural sources in the midst of scientific argument—in Full House he drops Whitman, Shakespeare, the Bible, Peirce and many others like they are the air he breathes. The only thing that remains troubling for me is Gould’s conjecture that if a replay of evolution were played, we might not end up with conscious life, because of the many chance events required. He does say that he argues independently for this claim elsewhere (I think the book is called Wonderful Life, from memory), and so I might have to check that out—I recall seeing a headline the other day in one of the popular science journals that proclaimed precisely the reverse, however—a principle of convergence that entails a universe with physical laws the same as our own will produce life, as if inevitably, every replay. I’ll post back here once I’ve read a bit more widely on it; it’s a shame we don’t have Gould around any more, to see what his own reaction would be.