American Scientist

The Crucible of Creation: The Burgess Shale and the Rise of Animals. Simon Conway Morris. 242 pp. Oxford University Press, 1998. $30.

Stephen Jay Gould's Wonderful Life (1989) brought the beautifully preserved fossils of the mid-Cambrian Burgess Shale to the attention of the reading public. But Gould used the fossils to promote a highly contentious interpretation that stressed the contingency of the evolutionary process. He argued that the "Cambrian explosion," the comparatively rapid appearance of a multitude of animal phyla at the start of the Palaeozoic era, had generated a vast array of different types, many of which have subsequently been lost through extinction. According to Gould, if we could somehow "rewind the tape" of evolution and let it play again, chance would favor a different selection of that original multitude, and the world would be a very different place from the one we see around us. There is nothing "preordained" about the appearance of humanity or the human level of awareness.

Simon Conway Morris has produced another account of the Burgess Shale creatures, explaining to a wide audience the significance of new discoveries and new interpretations that have emerged since Gould wrote. As a piece of popular science writing the book is very well done, and many will read it for the latest information about the bizarre world of the earliest animals. But make no mistake: Despite the plethora of illustrations and a fictionalized account of what it would be like to explore these ancient seas, this is no coffee-table excursion through the details of an ancient ecosystem. It is a full-scale assault on Gould's interpretation of the Cambrian explosion and on the materialist philosophy of life embodied in that interpretation. Conway Morris wants to convince us that we (or thinking beings very like us) are the unique yet intended goal of evolution. The word "creation" in his title is not to be taken lightly.

Like Wonderful Life, Conway Morris's book takes us through the story of Charles Doolittle Walcott's discovery of the Burgess Shale and his efforts to describe the animals revealed by the high level of preservation in this very special (but now by no means unique) location in the Canadian Rockies. Walcott was Gould's anti-hero, the paleontologist who shoehorned a whole range of bizarre Cambrian types into a few known categories, mostly arthropods. I have often wondered if Gould's attitude was influenced by the fact (which I found out from his book) that Walcott had masterminded the plan to blacklist Franz Boas's anti-racist anthropology within the American scientific community. Be that as it may, Gould's real complaint was that Walcott was blind to the obvious strangeness of the Burgess Shale creatures because he was committed to the orthodox view that the cone of evolutionary diversity must expand through time. Conway Morris to some extent rehabilitates Walcott's reputation by showing that the diversity of Cambrian forms was by no means as extensive as Gould claims. To establish this point he gives us a fascinating tour through the research that has transformed our understanding of the Burgess Shale creatures, revealing that the strangeness is often only skin deep, concealing underlying features that confirm their position within, or between, known phyla.

More discoveries from similar sites in Greenland and China have also thrown light on the Cambrian fauna. The climax of the book is an imaginary tour of the Cambrian seas via submersible and time machine, with detailed descriptions of the structures and habits of the various species as now understood (including some fascinating color plates).

The concluding chapters survey the theoretical significance of the new interpretations. Conway Morris is keen to explain how and why the Cambrian explosion took place, constructing a theory that combines genetic triggers for structural innovations with an ecological pressure generated by the origin of predators. His real concern, though, is to refute the claim that the explosion requires the postulation of evolutionary forces that are no longer in operation. The main plank of the argument is the denial of Gould's alleged diversity of form. Using cladistic analysis, Conway Morris argues that the Burgess Shale creatures can all be fitted into known phyla, or show intermediate states that actually throw light on the process by which the known phyla diverged from one another. He notes that Harry Whittington and Derek Briggs, the Cambridge paleontologists who made the first modern studies of the Burgess Shale species, were influenced by Sidnie Manton's thesis that the arthropods are polyphyletic. According to Manton, there was no "arthropod Eve," no single ancestor from which all modern arthropods are descended. The chelicerates (spiders and scorpions), crustaceans (crabs and prawns), uniramians (insects and myriapods) and the extinct trilobites had each independently evolved the characteristic arthropod structure. On such a model it would not be surprising that some other, equally independent, arthropod types might have appeared in the Cambrian and then become extinct. Modern studies have now shown that all the Burgess Shale arthopods can be accommodated within a scheme that explains their origin in monophyletic terms—from a single common ancestor in which the basic arthropod structure was developed. Meanwhile, Wiwaxia (from the Burgess shale) and the halkieriids (from Greenland) show how the mollusks and brachiopods evolved from the annelid worms. Major transformations are involved, of course, but nothing that requires the postulation of evolutionary forces outside the range of what can be studied in more recent times.

Conway Morris thus claims that Gould's scenario for the origin of animals is disproved: There was no vast radiation and no winnowing out of many early phyla by extinction. But the disagreement between the two paleontologists is more fundamental than this, because Conway Morris thinks that Gould's whole rerunning-the-tape idea is misleading if it is meant to imply that the outcome could be significantly different from what we observe. He certainly does not want to imply that evolution is directed by mysterious goal-directed forces. But he appeals to the widespread existence of convergence to argue that at least in its broad outlines, the outcome of evolution is predetermined. Convergence occurs when two lines of evolution independently develop the same or very similar structures, as when ichthyosaurs (reptiles) and whales (mammals) independently evolved a fish-like body plan. This occurs because certain structures are simply the best for certain adaptive purposes—any vertebrate wanting to swim in the water is going to evolve in the same direction. Conway Morris believes that the combined limitations of the developmental pathways triggered by genetics and the demands of the environment mean that the possible outcomes of the evolutionary process are very limited. We can conceive of all sorts of alien creatures, but they could never exist in the real world—and what can exist is pretty much confined to what we actually see. So rerunning the tape would produce more or less the same results, although the details might be different. There would be something like whales swimming in the modern seas, although they might have evolved from different mammalian ancestors.

Curiously, Conway Morris has himself demolished the most effective case for the power of convergence—Manton's theory of the independent origin of the arthropod body plan by several different phyla. In fact, the possibility of a polyphyletic origin for the arthropods goes back to Walcott's time and may have influenced his original interpretations of the Burgess Shale creatures. But it has now been disproved and with it the best example of the power of convergence to predetermine the outcome of evolution. Instead, Conway Morris offers us the parallels between the marsupials and the placental mammals, his best example being the independent evolution of a marsupial very much like the saber-toothed tiger. The main problem with this argument (and it is not a new one) is the kangaroo. If convergence is so powerful, how was it possible for the kangaroos to proliferate into a major component of the Australian fauna whereas nothing like them ever became dominant among the placentals of the rest of the world? Gould's interpretation of the Cambrian explosion may have been demolished, but this reviewer, at least, remains unconvinced by Conway Morris's argument that the outcome of evolution is predetermined. If placental kangaroos had taken charge outside Australia, who knows what the world would look like now.

There is thus far more at stake here than the nature of the Cambrian explosion. Conway Morris is quite clear about how far he wants to extend the power of convergence: It guarantees the emergence of high intelligence (in mollusks like the octopus and in vertebrates) and of human spiritual faculties (in the Neandertals as well as our own ancestors). In the end, he wants us to believe that something very like human nature was bound to emerge sooner or later from the evolutionary process. This contrasts with Gould's position, which follows a materialist tradition pioneered by the founder of modern Darwinian paleontology, George Gaylord Simpson, who insisted that humans were a most unlikely product of so haphazard a process. Gould's Marxist leanings are well known, and we can see why he would favor a viewpoint that leaves the human race to figure out its own moral values with no hints provided by any transcendental source. Conway Morris's opposition to this is driven by a more traditional perception of the human situation. He tells us that our intelligence is a gift, that we shall be called into account and that the evils perpetrated by humanity make sense only if they can be redeemed. For him, we are not only the intended outcome of evolution—we may also be the unique embodiments of spiritual faculties in the universe. His last chapter is a brief but clear-cut rejection of the popular assumption that there are many life-bearing planets throughout the galaxy. Evolution is predetermined, but it has only happened once.

To a historian of science such as myself, the books by Gould and Conway Morris seem themselves like a rerunning of the tape of history, but in this case there is a loop that was first played in the late-19th century and is now repeating itself almost exactly. For all the new discoveries and the modern apparatus of cladistic analysis, the alternative visions of the nature of history are as clear today as they were to the biologists who tried to defend their belief in a purposeful universe against the assault of Darwin's Origin of Species. Gould himself once wrote about the "eternal metaphors" of paleontology, and on that point, Conway Morris has merely confirmed his claim that the rival visions of nature are still in play.