Wonderful Life: The Burgess Shale and the Nature of History by Stephen Jay Gould

My rating: 3 of 5 stars


This book is a headache-inducing mixture of contrary qualities. My emotions swung wildly as I read, from pleasure to exasperation, from wonder to annoyance.

The subject of the book is fascinating. Gould here sets out to tell the story of the Burgess Shale, a site discovered in 1909 by Charles Walcott in the Canadian Rockies. Owing to unusual conditions (an underwater mudslide, 500 million years ago) fossils of thousands of ancient organisms were preserved in excellent condition—not only the typical hard parts that survive fossilization, but their soft tissue as well. This was an unprecedented and highly significant find, as it provided a kind of snapshot of life shortly after the famous Cambrian Explosion (when macroscopic animals suddenly become common in the fossil record).

Amazingly, though the importance of the find was widely understood, most of the specimens collected remained unanalyzed and poorly understood for decades. Their discoverer, Walcott, published a few preliminary studies but nothing of significant depth. It was not until the 1970s that a team of researchers—Harry Blackmore Whittington, Derek Briggs, and Simon Conway Morris—started reinvestigating these samplings and publishing long monographs on their findings. Their research and conclusions form the core of Gould’s book.

It is when Gould is describing the story of the shale’s discovery, and especially when he is describing how the scientists went about their work, that he is at his absolute best. And I think he deserves all the praise in the world for this. Few books for the general public, if any, take the reader so intimately into the process of paleontological research—collecting specimens, dissecting samples, debating taxonomies. What is more, Gould manages to make a book about fossilized worms into absolutely gripping reading. There were times when I felt my heart swell with admiration for these scientists hunched over a microscope, scratching through the exoskeleton of some three-inch creature in order to write a 100-page monograph to be read by a handful of experts. This is no small accomplishment for an author.

Where Gould falters, for me, is in the interpretation he gives to this story. Gould argues that the Burgess fossils completely overturn our notions of evolution and constitute a breakthrough comparable with Darwin’s theory. He bases this grandiose claim on the amount of taxonomic diversity encountered within the samples. Now, when Walcott first analyzed the fossils, he classified the animals as members of modern groups. But when Whittington, Briggs, and Conway Morris re-examined these fossils, they found a great many organisms that did not fit into any modern classification.

From this, Gould draws two conclusions. First, he asserts that this was a particularly fertile time for evolution; and he bases this claim on the large variation present in the fossils. Second, he makes the grander claim that the large number of viable organisms whose descendants did not survive to the present day proves that the history of life is highly contingent.

Now, let me explain why both of these claims are frustrating.

For one, there is a perfectly obvious reason why the Cambrian period could have involved more evolutionary “experimentation” than the present day. This is that, when organisms are first evolving to occupy previously unoccupied niches—as the first macroscopic animals were doing—there would be comparatively less environmental pressure to do so in the most efficient possible way. The first freely swimming animal, for example, did not need to swim faster than any other animal trying to eat it. Gould himself eventually gets around to this explanation, though he seems not entirely satisfied with it.

But the other—and even more obvious—explanation is that these animals are only more varied in a limited sense. As Gould himself admits, modern animals occupy more varied environments and exhibit a far large range of sizes and behaviors. What the Burgess animals possess is taxonomic variety—requiring the creation of new phyla, families, or orders to accommodate them into our filing cabinets. However, what he fails to mention is that this is exactly what one should expect from such ancient fossils. Our taxonomic categories were developed to classify modern animals, so why should they apply neatly to animals living half a billion years ago?

I think this point deserves further elaboration. Our modern categories were purposefully defined using the traits which remain invariable over the largest number of extant species. They would be useless otherwise. It is thus nonsensical when Gould wonders why there should arise so many new phyla in the Cambrian, but not today. This is like wondering why there are so many strange words in Chaucer but not the newspaper. In other words, the huge taxonomic variation Gould discusses is largely an artifact of our own categories and not an actual property of the animals themselves. A foreign language does not have an objectively more complex vocabulary just because there are more unfamiliar words in it.

Ironically, some of the taxonomic weirdness that Gould discusses has since been revised away, as the “radical” interpretations of Whittington, Briggs, and Conway Morris were themselves re-interpreted, placing many animals in more familiar categories. To take just one example, the Burgess animals hallucigenia is here presented as a truly nightmarish creature which walks on spikes and has tentacles sticking out of its back. Most concerning of all, it has a formless blob for a head! This animal was, understandably, at first given its own genus; but it has since been recognized as a lobopodian when a more accurate reconstruction was hit upon. (It turns out that the spikes are on its back and are not its legs; and that its head is, in fact, not a formless blob.)

Of course, Gould can hardly be blamed for basing his book on outdated science. After all, it was published over 30 years ago. But I do think that Gould, of all people, should have known that interpretations of fossil remains are frequently contested and revised, so it seems unwise of him to have based such grandiose conclusions on such a shaky foundation.

Now for Gould’s second claim, that the Burgess fossils prove that the history of based on chance—or, in his words, “decimation by lottery.” He repeatedly states that no modern scientist, if transported back in time, could have selected which lineages would survive to the present day or would end in extinction, since there is no obvious anatomical flaw in the extinct lineages. And he argues that this proves chance, not superiority, determined evolutionary survival.

This strikes me as a completely bizarre argument, since I am not sure who Gould is even arguing against. What scientist believes they could pick out evolutionary winners and losers 500 million years from now? Or even 500 years from now? Virtually every organism is well-adapted to its environment. If you traveled to the previous ice age, could you find an anatomical flaw in the woolly mammoth? I doubt it. But does this prove that extinction is totally random? Hardly.

Indeed, Gould eventually admits that even he himself does not believe that extinction is a truly random process (though this makes his use of the term “lottery” rather puzzling). Instead, he adopts the wholly conventional view that extinction occurs when the environment changes too quickly for a species to adapt, and that the species which do survive environmental shifts are able to do so because of traits they evolved under different circumstances.

In other words, yes, luck is a factor in evolutionary survival. Indeed, considering that evolution is a physical process, one could argue that it is completely a matter of luck which species survive or perish. In a general sense, luck determines whether an asteroid will hit, whether an invasive species will outcompete you, or whether hairless primates will destroy your habitat. But what biologist would deny that? It strikes me that the only people who seriously object to this do so for religious reasons, seeing the evolution of human life as something pre-ordained or divinely guided. Yet Gould acts as if he is arguing against the dominant view in his field.

To sum up, Gould is grandly making a non-controversial point using evidence that does not even prove his point. The burgess fauna were not necessarily more varied than today’s; and even if they were, that would not prove that evolution is random.

I am sorry to be writing such a critical review of this book. It just seemed such a shame that such a great story could be weighed down by so much unnecessary intellectual baggage. If this had simply been an exploration of the Burgess fauna it would have been delightful. Indeed, though perhaps outdated, the many illustrations of ancient creatures are still charming to contemplate. And though I found the book quite frustrating, if read in the right spirit, I still think it is wonderfully educational about the process of science—though, perhaps not in the manner it intended to be.

(Cover photo a reconstruction of hallucigenia by Martin R. Smith and Jean-Bernard Caron. Taken from Wikimedia Commons, CC BY-SA 4.0.)



View all my reviews