Wonderful Life: The Burgess Shale and the Nature of History

by Stephen Jay Gould

  Walcott had long maintained a firm commitment to the artifact theory. It provided the keystone for his entire approach to Cambrian geology and life. He never doubted that Cambrian complexity and diversity required a long series of abundant Precambrian ancestors of similar anatomy. In an early article he wrote: “That the life in the pre-Olenellus sea was large and varied there can be little, if any, doubt.… It is only a question of search and favorable conditions to discover it” (1891). Olenellus, as then defined, was the oldest Cambrian trilobite, so pre-Olenellus meant Precambrian. And in one of his late papers: “When the advanced stage of development of some of the earliest-known forms is considered it seems almost certain that such existed far back in Precambrian time” (1916, p. 249).

  Walcott had long defended a particular approach to the artifact theory that a profusion of new Burgess phyla would have undermined. The artifact theory demanded long Precambrian histories for many modern groups, yet no fossils had been found. Therefore, the existence of Precambrian life would have to be inferred from some aspect of later, recorded history. Accordingly, Walcott sought support for the artifact theory in the concept of stability. If the number of basic anatomical designs had not changed throughout life’s recorded history, then such stability must surely guide our concept of what came before. Could a system so constant for hundreds of millions of years arise in a geological flash just a moment before? Protracted stability surely implied a very long and stately approach from a common ancestry deep in the distant Precambrian mists, not a gigantic burp of creativity from a starting point just below the Cambrian borderline.

  We can now understand why Walcott was virtually compelled to propose the Burgess shoehorn. He interpreted his new fauna in the light of thirty previous years spent (largely in frustration) trying to prove the artifact theory, as an ultimate tribute to Darwin from a Cambrian geologist. He could not grant Burgess organisms the uniqueness that seems so evident to us today because a raft of new phyla would have threatened his most cherished belief. If evolution could produce ten new Cambrian phyla and then wipe them out just as quickly, then what about the surviving Cambrian groups? Why should they have had a long and honorable Precambrian pedigree? Why should they not have originated just before the Cambrian, as the fossil record, read literally, seems to indicate, and as the fast-transition theory proposes? This argument, of course, is a death knell for the artifact theory.

  If, instead, he could shoehorn all Burgess creatures into modern groups, he would be giving the strongest possible boost to the artifact theory. For such a condensation of disparity increased the proportion of modern groups already represented right at the start of life’s recorded history—and greatly enhanced the apparent stability of major designs through time. Obviously, and with both vigor and delight, Walcott chose this alternative. What does any man do when faced with destruction or affirmation?

  Walcott approached the artifact theory from both geological directions—down from the Cambrian, as illustrated by the Burgess shoehorn, and up from the Precambrian. His argument about the Precambrian has become, in the typically perverse manner of textbook histories, his most enduring legacy. Most textbooks contain a traditional, almost mandatory, two-or three-page introductory section on the history of their discipline. These travesties of scholarship dismiss the fine thinkers of our past with two-liners about some error, usually misinterpreted, that shows how stupid they were and how enlightened we have become. Charles Doolittle Walcott was one of the most powerful men in the history of American science. Yet ask any student of geology about him, and if you get any response at all, you will probably hear: “Oh yeah, that doofus who invented the nonexistent Lipalian interval to explain the Cambrian explosion.” I first heard of Walcott in this context, long before I knew about the Burgess Shale. History can be either enlightening or cruel. However, with the preceding discussion of the Burgess and the artifact theory in mind, I think we can finally understand the story of the Lipalian interval properly—and recognize Walcott’s proposal as a reasonable, if outstandingly wrong, inference within his general commitments.

  The artifact theory was central to Walcott’s scientific perspective. His conclusions about the Burgess fauna supported this theory, but he needed a more direct argument from the Precambrian. Where had all the Precambrian animals gone? Other ideas included a universal metamorphism (alteration of rocks by heat and pressure) that had destroyed all Precambrian fossils, and an absence of fossilizable hard parts in Precambrian creatures. Walcott rejected the metamorphism theory because he had found many unaltered Precambrian rocks, and he argued that the hard-part theory, while probably true, could not explain the entire phenomenon.

  Walcott was primarily a field geologist, specializing in Cambrian rocks. Following the proclivities of any field man, he approached his growing interest in the problem of the Cambrian explosion in the obvious way—he decided to search the latest Precambrian rocks for the elusive ancestors of Cambrian fossils. He worked for many years in the western United States, the Canadian Rockies (where he discovered the Burgess), and in China, but he found no Precambrian fossils. So he tried to reconstruct the geological and topographic history of the late Precambrian earth in a way that would explain this frustrating absence.

  Walcott eventually reached a conclusion opposite to Darwin’s speculation but in the same tradition—the rocks that might house abundant Precambrian fossils just aren’t accessible to us. Darwin had suggested vast Precambrian oceans with no continents nearby to serve as a source of sediments. Walcott argued that the late Precambrian was a time of uplift and mountain building, with continents far more extensive than today’s. Since life, according to Walcott and others, had evolved in the oceans and had not yet colonized land or fresh waters, these vast Precambrian continents permitted no marine sedimentation in areas now accessible to us. (Walcott wrote long before the era of continental drift and never doubted the permanent position of continents. Thus, he argued that places available for geological observation today were the centers of more extensive Precambrian continents, and were therefore devoid of late Precambrian marine sediments. Late Precambrian sediments might lie under miles of deep ocean, but no technology then existed to recover or even to sample such potential treasures.)

  The infamous “Lipalian interval” was Walcott’s name for this time of Precambrian nondeposition. Walcott proposed a world-wide break in accessible marine sedimentation, just during the critical interval of extensive Precambrian ancestry for modern groups. In a famous address to the Eleventh International Geological Congress, meeting in Stockholm on August 18,1910, he stated:

  I have for the past 18 years watched the geological and paleontological evidence that might aid in solving the problem of Precambrian life. The great series of Cambrian and Precambrian strata in eastern North America from Alabama to Labrador; in western North America from Nevada and California far into Alberta and British Columbia, and also in China, have been studied and searched for evidences of life until the conclusion had gradually been forced upon one that on the North American continent we have no known Precambrian marine deposits containing traces of organic remains, and that the abrupt appearance of the Cambrian fauna results from geological and not from biotic conditions.… In a word, the thought is that the Algonkian [late Precambrian] period … was a period of continental elevation and largely terrigenous sedimentation in non-marine bodies of water, also a period of deposition by aerial and stream processes over considerable areas (1910, pp. 2–4).

  And he added:

  Lipalian is proposed for the era of unknown marine sedimentation.… The apparently abrupt appearance of the lower Cambrian fauna is … to be explained by the absence near our present land area of the sediments, and hence the faunas of the Lipalian period (1910, p. 14).

  Walcott’s explanation may sound forced and ad hoc. It was surely born of frustration, rather than the pleasure of discovery. Yet the nonexistent Lipalian was not a fool’s rationalization, as usually presented in our textbooks, but a cr
edible synthesis of geological evidence in the context of a vexatious dilemma. If Walcott deserves any brickbats, direct them at his failure to consider any alternative to his favored way of thinking about the artifact theory—and at his false assumption, imposed by the old bias of gradualism, that equated evolution itself with a long sequence of ancestral continuity for any complex creature. For even if the Lipalian hypothesis made sense in the light of existing geological information, it rested, as Walcott knew only too well, upon the most treacherous kind of argument that a scientist can ever use—negative evidence. Walcott admitted: “I fully realize that the conclusions above outlined are based primarily on the absence of a marine fauna in Algonkian rocks” (1910, p. 6).

  And, as so often happens in the face of negative evidence, the earth eventually responded, offering to later geologists abundant late Precambrian marine sediments—still with no fossils of complex invertebrates. The Lipalian interval ended up on the trash heap of history.

  Scientists have a favorite term for describing a phenomenon like Walcott’s allegiance to the Burgess shoehorn—overdetermined. The modern concept of maximal disparity and later decimation (perhaps by lottery) never had the ghost of a chance with Walcott because so many elements of his life and soul conspired to guarantee the opposite view of the shoehorn. Any one of these elements would have been enough in itself; together, they overwhelmed any alternative, and overdetermined Walcott’s interpretation of his greatest discovery.

  To begin, as we have seen, Walcott’s persona as an archtraditionalist in thought and practice did not lead him to favor unconventional interpretations in any area of life. His general attitude to life’s history and evolution implied stately unfolding along predictable pathways defined by the ladder of progress and cone of increasing diversity; this pattern also held moral meaning, as a display of God’s intention to imbue life with consciousness after a long history of upward striving. Walcott’s specific approach to the key problem that had focused his entire career—the riddle of the Cambrian explosion—favored a small set of stable and well-separated groups during Burgess times, so that a long history of Precambrian life might be affirmed, and the artifact theory of the Cambrian explosion supported. Finally, if Walcott had been at all inclined to abandon his ideological commitment to the shoehorn, in the light of contradictory data from the Burgess Shale, his administrative burdens would not have allowed him time to study the Burgess fossils with anything like the requisite care and attention.

  I have labored through the details of Walcott’s interpretation and its sources because I know no finer illustration of the most important message taught by the history of science: the subtle and inevitable hold that theory exerts upon data and observation. Reality does not speak to us objectively, and no scientist can be free from constraints of psyche and society. The greatest impediment to scientific innovation is usually a conceptual lock, not a factual lack.

  The transition from Walcott to Whittington is a premier example of this theme. The new view—as important an innovation as paleontology has ever contributed to our understanding of life and its history—was in no way closed to Walcott. Whittington and colleagues studied Walcott’s specimens, using techniques and tools fully available in Walcott’s time, in making their radical revision. They did not succeed as self-conscious revolutionaries, touting a new view in a priori assault. They began with Walcott’s basic interpretation, but forged ahead on both sides of the great dialectic between theory and data—because they took the time to converse adequately with the Burgess fossils, and because they were willing to listen.

  The transition from Walcott to Whittington marks a milestone that could hardly be exceeded in importance. The new view of the Burgess Shale is no more nor less than the triumph of history itself as a favored principle for reading the evolution of life.

  THE BURGESS SHALE AND THE NATURE OF HISTORY

  Our language is full of phrases that embody the worst and most restrictive stereotype about science. We exhort our frustrated friends to be “scientific”—meaning unemotional and analytic—in approaching a vexatious problem. We talk about the “scientific method,” and instruct schoolchildren in this supposedly monolithic and maximally effective path to natural knowledge, as if a single formula could unlock all the multifarious secrets of empirical reality.

  Beyond a platitudinous appeal to open-mindedness, the “scientific method” involves a set of concepts and procedures tailored to the image of a man in a white coat twirling dials in a laboratory—experiment, quantification, repetition, prediction, and restriction of complexity to a few variables that can be controlled and manipulated. These procedures are powerful, but they do not encompass all of nature’s variety. How should scientists operate when they must try to explain the results of history, those inordinately complex events that can occur but once in detailed glory? Many large domains of nature—cosmology, geology, and evolution among them—must be studied with the tools of history. The appropriate methods focus on narrative, not experiment as usually conceived.

  The stereotype of the “scientific method” has no place for irreducible history. Nature’s laws are defined by their invariance in space and time. The techniques of controlled experiment, and reduction of natural complexity to a minimal set of general causes, presuppose that all times can be treated alike and adequately simulated in a laboratory. Cambrian quartz is like modern quartz—tetrahedra of silicon and oxygen bound together at all corners. Determine the properties of modern quartz under controlled conditions in a laboratory, and you can interpret the beach sands of the Cambrian Potsdam Sandstone.

  But suppose you want to know why dinosaurs died, or why mollusks flourished while Wiwaxia perished? The laboratory is not irrelevant, and may yield important insights by analogy. (We might, for example, learn something interesting about the Cretaceous extinction by testing the physiological tolerances of modern organisms, or even of dinosaur “models,” under environmental changes proposed in various theories for this great dying.) But the restricted techniques of the “scientific method” cannot get to the heart of this singular event involving creatures long dead on an earth with climates and continental positions markedly different from today’s. The resolution of history must be rooted in the reconstruction of past events themselves—in their own terms—based on narrative evidence of their own unique phenomena. No law guaranteed the demise of Wiwaxia, but some complex set of events conspired to assure this result—and we may be able to recover the causes if, by good fortune, sufficient evidence lies recorded in our spotty geological record. (We did not, until ten years ago, for example, know that the Cretaceous extinction corresponded in time with the probable impact of one or several extraterrestrial bodies upon the earth—though the evidence, in chemical signatures, had always existed in rocks of the right age.)

  Historical explanations are distinct from conventional experimental results in many ways. The issue of verification by repetition does not arise because we are trying to account for uniqueness of detail that cannot, both by laws of probability and time’s arrow of irreversibility, occur together again. We do not attempt to interpret the complex events of narrative by reducing them to simple consequences of natural law; historical events do not, of course, violate any general principles of matter and motion, but their occurrence lies in a realm of contingent detail. (The law of gravity tells us how an apple falls, but not why that apple fell at that moment, and why Newton happened to be sitting there, ripe for inspiration.) And the issue of prediction, a central ingredient in the stereotype, does not enter into a historical narrative. We can explain an event after it occurs, but contingency precludes its repetition, even from an identical starting point. (Custer was doomed after a thousand events conspired to isolate his troops, but start again in 1850 and he might never see Montana, much less Sitting Bull and Crazy Horse.)

  These differences place historical, or narrative, explanations in an unfavorable light when judged by restrictive stereotypes of the “scientific method.” Th
e sciences of historical complexity have therefore been demoted in status and generally occupy a position of low esteem among professionals. In fact, the status ordering of the sciences has become so familiar a theme that the ranking from adamantine physics at the pinnacle down to such squishy and subjective subjects as psychology and sociology at the bottom has become stereotypical in itself. These distinctions have entered our language and our metaphors—the “hard” versus the “soft” sciences, the “rigorously experimental” versus the “merely descriptive.” Several years ago, Harvard University, in an uncharacteristic act of educational innovation, broke conceptual ground by organizing the sciences according to procedural style rather than conventional discipline within the core curriculum. We did not make the usual twofold division into physical versus biological, but recognized the two styles just discussed—the experimental-predictive and the historical. We designated each category by a letter rather than a name. Guess which division became Science A, and which Science B? My course on the history of earth and life is called Science B-16.

  Perhaps the saddest aspect of this linear ranking lies in the acceptance of inferiority by bottom dwellers, and their persistent attempt to ape inappropriate methods that may work higher up on the ladder. When the order itself should be vigorously challenged, and plurality with equality asserted in pride, too many historical scientists act like the prison trusty who, ever mindful of his tenuous advantages, outdoes the warden himself in zeal for preserving the status quo of power and subordination.

  Thus, historical scientists often import an oversimplified caricature of “hard” science, or simply bow to pronouncements of professions with higher status. Many geologists accepted Lord Kelvin’s last and most restrictive dates for a young earth, though the data of fossils and strata spoke clearly for more time. (Kelvin’s date bore the prestige of mathematical formulae and the weight of physics, though the discovery of radioactivity soon invalidated Kelvin’s premise that heat now rising from the earth’s interior records the cooling of our planet from an initially molten state not long past.) Even more geologists rejected continental drift, despite an impressive catalogue of data on previous connections among continents, because physicists had proclaimed the lateral motion of continents impossible. Charles Spearman misused the statistical technique of factor analysis to designate intelligence as a single, measurable, physical thing in the head, and then rejoiced for psychology because “this Cinderella among the sciences has made a bold bid for the level of triumphant physics itself” (quoted in Gould, 1981, p. 263).