World of Wonders

I read this book by the late Stephen Jay Gould shortly after it was published in 1989, and more or less randomly grabbed it off the shelf and re-read it recently. I'm not a scientist, let alone an invertebrate paleontologist, so the impressions that follow are strictly those of a layperson.

To be as brief as possible (and with no pretense of doing justice to a rich and intricately argued book), Gould's immediate subject was the rare assemblage of soft-bodied fauna discovered, a century ago, by Charles Doolittle Walcott in the geological formation known as the Burgess Shale in the Canadian Rockies. Walcott, a geologist who was the Secretary of the Smithsonian Institution at the time, classified (Gould's word is "shoehorned") his new discoveries into existing taxa (mostly of arthropods), but subsequent work by Harry Whittington, Simon Conway Morris, Derek Briggs, and others revealed that many of the creatures properly belonged in previously unknown classes or even phylla that later become extinct. These Cambrian fossils, Gould reasoned, provided evidence that in the aftermath of the Cambrian explosion that produced the ancestors of the metazoan animals with which we are familiar (e.g. crustaceans, molluscs, vertebrates — and of course us) there was a wide diversity or disparity of body plans, many of which ultimately proved to be evolutionary dead ends. Instead of a steady progress of expanding diversity from a small number of original phylla, the analysis of the Burgess Shale fauna suggested a winnowing over time of possible avenues for evolutionary development. There may be more species now, but the roster of higher-level taxa was greater during the Cambrian than now. (Or so it seemed.)

Gould's broader argument, which he used the Burgess Shale reclassification to support, centered on the importance of contingency to the evolutionary history of life on earth. The lineages that became extinct were not necessarily essentially inferior to the lineages that survived; they perished because of circumstances that were equivalent, at least in part, to a roll of the dice. Climates changed, continents moved around, asteroids impacted the earth, etc. The fact that these lineages no longer exist (while others do) proves, in a certain sense, that they were less fitted to the challenges that followed, but since the challenges themselves were contingent and not predictable we can't say that a different roll of the dice might not have produced an entirely different outcome. A minor shift might well have led to the premature extinction of the chordates and hence precluded our own existence. (Gould's title alludes explicitly to Frank Capra's film It's a Wonderful Life, in which one small change holds the possibility of triggering a cascade of altered consequences.)

In the past thirty years continuing analysis of the Burgess fauna has, inevitably, complicated the picture considerably. It's no longer as clear as it seemed in 1989 that all of the re-classified species are true "weird wonders" that can't be contained within the familiar major phyla and classes. That process is highly technical and well beyond my ken, but although Gould, had he lived, might have been disappointed by the implications of more recent developments for his thesis, he was well aware of (and specifically notes) the fact that science proceeds by a grinding process of error and rectification. ("One cannot hope to do anything significant or original in science unless one accepts the inevitability of substantial error along the way.") His larger point, reading the role of contingency, remains open.

Ironically, one of Gould's main sources (fully credited and repeatedly praised in the book), Simon Conway Morris, has been one of Wonderful Life's most vigorous detractors. For Conway Morris, who has made a particular study of convergent evolution, the logic of evolution imposes such constraints that certain outcomes can be said to be predictable or even inevitable; one of them, in his opinion, is higher intelligence of the kind that we possess. Even if the chain of contingent events had been modified, for instance, so as to extinguish the primate lineage, "humanoid" intelligence would have been highly likely to evolve (or would eventually evolve) in another. Underlying Conway Morris's argument are his Christian beliefs and his philosophical antimaterialism (although criticism of Gould's book is not limited to those who hold such beliefs).

Though both scientists — Gould and Conway Morris — have used specific paleontological evidence to buttress their positions, the broader logical and philosophical arguments are fairly resistant to easy resolution by mere fact, and one suspects that if and when the classification of the Burgess fauna is finally settled, the debate will simply be continued in another venue. Regardless, Wonderful Life seems to me to hold up as a landmark of intelligent science writing for a general audience.