The Structure of Evolutionary Theory Read online

  Ariel's telling verse in Shakespeare's The Tempest proclaims in dense metaphor:

  Full fathom five thy father lies;

  Of his bones are coral made;

  Those are pearls that were his eyes:

  Nothing of him that doth fade

  But doth suffer a sea-change

  Into something rich and strange.

  With the exception of one possible (and originally unintended) modern reading of these images, this famous and haunting verse provides a beautiful description of both the priceless worth and intriguing modern transformation of Darwin's original theory. (For the exception, several connotations of deep burial in the sea — full fathom five — might be viewed negatively, as in “deep sixing” or going to Davy Jones's locker. But, for natural historians who read this book, and coming from an invertebrate paleontologist as author, the seafloor could not represent a more positive resting place or point of origin — and I intend to evoke only these upbeat images in citing Ariel's lines.) Other­wise, Darwin's original structure has only yielded greater treasure in cascad­ing implications and developments through the subsequent history of evolu­tionary thought — the conversion of the bones of an original outline into precious coral and pearls of current substance. Nothing of Darwin's central logic has faded or fully capsized, but his theory has been transformed, along his original lines, into something far different, far richer, and far more ade­quate to guide our understanding of nature.

  The last three lines of Shakespeare's verse also appear on the tombstone of the great poet Percy Bysshe Shelley (also the author of the preface to his wife's novella, Frankenstein, which cites Erasmus Darwin in its first line of text). I believe that these words would suit, and honor, Charles Darwin just as well and just as rightly.

  Apologia Pro Vita Sua

  A TIME TO KEEP

  The Preacher spoke ever so truly in writing his famous words (Ecclesiastes 3:1-7): “For every thing there is a season, and a time to every purpose ... A [Page 25] time to break down, and a time to build up ... A time to rend, and a time to sew: a time to keep silence, and a time to speak.” Evolutionary theory now stands in the happier second state of these genuine dichotomies (in part be­cause the first state had been mined to the limited extent of its utility): we live in a time for building up, for sewing together, and for speaking out.

  Not all times are such good times, and not all scientists win the good for­tune to live within these times of motion. For theories grow as organisms do, with periods of Sturm und Drang, long latencies of youth or ossifications of age, and some happy times of optimally productive motion in between (an­other Goldilockean phenomenon). I recently studied the life and career of E. Ray Lankester (Gould, 1999a), clearly the most talented evolutionary morphologist of the generation just after Darwin. He did “good” work and had a “good” career (see Chapter 10, pages 1069–1076 for his best theoretical efforts), but he never transcended the ordinary. Perhaps the limitation lay largely within his own abilities. However, I rather suspect that he did possess both the temperamental gumption and the requisite intellectual might — but that the tools of major empirical advance just didn't emerge in his generation, for he remained stuck in a relatively unproductive middle, as Darwin had seized the first-fruits from traditional data of natural history, and the second plucking required a resolution of genetic mechanisms.

  I felt a similar kind of frustration in 1977, after writing my first technical book, Ontogeny and Phylogeny (see Chapter 10, pages 1061–1063). I had spent the best years of a young career on a subject that I knew to be rele­vant (at a time when most of the profession had forgotten). But then defeat snatched my prize from the jaws of victory. I am proud of the book, and I do believe that it helped to focus interest on a subject that became doable soon thereafter. But I ran up against a wall right at the end — for the genetics of de­velopment clearly held the key to any rapprochement of embryology and evo­lution, and we knew effectively nothing about eukaryotic regulation. Indeed, as we could then only characterize structural genes by electrophoretic tech­niques, our major “arguments” for regulatory effects (if they even merited such a positive designation) invoked such negative evidence as the virtual identity in structural genes between chimps and humans, coupled with a fair visceral sense of extensive phenotypic disparity in anatomy and behavior — with the differences then attributed to regulatory genes that we could not, at the time, either study or even identify.

  By sheer good fortune (abetted in minuscule ways by my own pushes and those of my paleontological colleagues), the field moved fast and I lived long enough to witness a sea change (if I may cite Ariel yet again) towards potentiation on all three major intellectual and social substrates for con­verting a subject from great promise combined with even more frustrating inoperability, into a discipline bursting with new (and often utterly surpris­ing) data that led directly to testable hypotheses about basic issues in the structure of evolutionary theory.

  Empirics. During the last third of the 20th century, new techniques and conceptualizations opened up important sources of data that challenged [Page 26] orthodox formulations for all three branches of essential Darwinian logic. To cite just one relevant example for each branch, theoretical development and accumulating data on punctuated equilibrium allowed us to reconceptualize species as genuine Darwinian individuals, fully competent to participate in processes of selection at their own supraorganismic (and suprademal) level — and then to rethink macroevolution as the differential success of species rather than the extended anagenesis of organismal adaptation (see Chapter 9). This validation of the species-individual aided the transformation of what had begun as a particular argument about group (or interdemic) selection into a fully generalized hierarchical theory, with good cases then documented from the genic to the cladal level (see Chapter 8).

  On the second branch of full efficacy for natural selection as an externalist and functionalist process, the stunning discoveries of extensive deep homologies across phyla separated by more than 500 million years (particularly the vertebrate homologs of arthropod Hox genes) — against explicit statements by architects of the Modern Synthesis (see p. 539) that such homologies could not exist in principle, in a world dominated by their conception of nat­ural selection — forced a rebalancing or leavening of Darwinian functionalism with previously neglected, or even vilified, formalist perspectives based on the role of historical and structural constraints in channeling directions of evolu­tionary change, and causing the great dumpings and inhomogeneities of morphospace — phenomena that had previously been attributed almost exclu­sively to functionalist forces of natural selection.

  On the third branch of extrapolation, the discovery and relatively quick validation, beginning in 1980, of a truly catastrophic trigger for at least one great mass extinction (the K-T event of 65 million years ago), fractured the uniformitarian consensus, embraced by a century of paleontological compla­cency, that all apparent faunal overturns could be “spread out” into sufficient time for explanation by ordinary causes under plausible intensifications that would not alter conventional modes of evolution and extinction.

  Moreover, as we shall see, these three apparently rather different kinds of data and their attendant critiques cohere into a revised general structure for evolutionary theory — thus marking our age as a time for building up and not only as a time for breaking down.

  Concepts. Following the Kantian dictum that percepts without con­cepts are blind, but concepts without percepts empty, these two categories interpenetrate as “pure” data suggest novel ideas (how can one not rethink the causes of mass extinction when evidence surfaces for a bolide, 7-10 km in diameter, and packing 104 the megatonnage of all the earth's nuclear weapons combined), whereas “abstract” concepts then taxonomize the natural world in different ways, often “creating” data that had never been granted enough previous intellectual space even to be conceived (as when punctuated equilib­rium made stasis a theoretically meaningful and int
eresting phenomenon, and not just an embarrassing failure to detect “evolution,” in its traditional definition of gradual change — and paleontologists then began active stud­ies of a subject that had previously been ignored as uninteresting, if conceptualized [Page 27] at all). But, speaking parochially as a student of the fossil record, I can at least say that the conceptual revolution in macroevolutionary thinking re­vitalized the field of paleobiology (even creating the name as a subdiscipline of paleontological endeavor). Whatever the varied value of different individ­ual efforts in this burgeoning field, we may at least be confident that our pro­fession will no longer be humiliated as a synecdoche for ossified boredom among the natural sciences — as Nature did in 1969 when editorializing about the salutary value of plate tectonics in revitalizing the geological sciences: “Scientists in general might be excused for assuming that most geologists are paleontologists and most paleontologists have staked out a square mile as their life's work. A revamping of the geologist's image is badly needed” (Anonymous, 1969, p. 903).

  The intricate and multifaceted concepts that have nuanced and altered the central logic on all three branches of Darwinism's essential postulates repre­sent ideas of broad ramification and often remarkably subtle complexity, as we vain scientists soon discovered in our fractured bubbles of burst pride — for we had been so accustomed to imagining that an evening in an armchair could conquer any merely conceptual issue, whereas we all acknowledge the substantial time and struggle that empirical problems, demanding collection and evaluation of data, often require. Yet, on these basic questions in formu­lating evolutionary theory, we often read and thought for months, and ended up more confused than when we began.

  The general solution to such procedural dilemmas lies in a social and intellectual activity that scientists do tend to understand and practice better than colleagues in most other academic disciplines — collaboration. Far more than most colleagues, I have tended to work alone in my professional life and pub­lication. But for each of the conceptually difficult and intellectually manifold issues of reevaluation for the central logic of the three essential Darwinian postulates, I desperately needed advice, different skills, and the give and take of argument, from colleagues who complemented my limited expertise with their equally centered specialties and aptitudes for other aspects of these large and various problems. Thus, on the first leg or branch of hierarchy theory, I worked with Niles Eldredge on punctuated equilibrium, and with Elisabeth Vrba on levels of selection and sorting. On the second leg of structuralist al­ternatives to adaptationist argument, I worked with Dick Lewontin on span­drels, Elisabeth Vrba on exaptation, David Woodruff on the functional and structural morphology of Cerion, and with “the gang of four” (increased to five with the later inclusion of Jack Sepkoski) — Dave Raup, Tom Schopf, Dan Simberloff, and me — on trying to specify how many aspects of apparently or­dered phyletic patterns, heretofore confidently attributed to selection for little reason beyond the visual appearance of order itself, could plausibly be gener­ated within purely random systems. And on the third leg of extrapolationism, my earliest interests in the logic and justification of uniformitarianism in phi­losophy, and of Lyellian perspectives in the history of science, could not have developed without advice and substantial aid (but not collaborative publica­tion this time) with historians Martin Rudwick, Reijer Hooykaas, and Cecil [Page 28] Schneer, and with philosophers Nelson Goodman, Bonnie Hubbard, and George Geiger. (Geiger, my mentor at Antioch College, was the last student of John Dewey and played with Lou Gehrig on the Columbia University base­ball team, thus embodying both my professional and avocational interests.)

  In fact, and as a comment within the sociology of science, I would venture that future historians might judge the numerous seminal (and published) collaborations between evolutionary biologists and professional philosophers of science as the most unusual and informative operational aspect of the reconstruction of evolutionary theory in the late 20th century. Research scientists tend to be a philistine lot, with organismic biologists perhaps at the head of this particular pack (for we work with “big things” that we can see and un­derstand at our own scale. Thus, we suppose that we can afford to be more purely empirical in our reliance on “direct” observation, and less worried about admittedly conceptual problems of evaluating things too small or too fast to see). Most of us would scoff at the prospect of working with a profes­sional philosopher, regarding such an enterprise as, at best, a pleasant waste of time and, at worst, an admission that our own clarity of thought had be­come addled (or at least as a fear that our colleagues would so regard our in­terdisciplinary collaboration).

  And yet, the conceptual problems presented by theories based on causes operating at several levels simultaneously, of effects propagated up and down, of properties emerging (or not) at higher levels, of the interaction of random and deterministic processes, and of predictable and contingent influ­ence, have proven to be so complex, and so unfamiliar to people trained in the simpler models of causal flow that have served us well for centuries (see the next section on Zeitgeist), that we have had to reach out to colleagues ex­plicitly trained in rigorous thinking about such issues. Thus, we learned, to our humbling benefit, that conceptual muddles do not necessarily resolve themselves “automatically” just because a smart person — namely one of us, trained as a scientist — finally decides to apply some raw, naive brain power to the problem. Professional training in philosophy does provide a set of tools, modes and approaches, not to mention a feeling for common dangers and fal­lacies, that few scientists (or few “smart folks” of any untrained persuasion) are likely to possess by the simple good fortune of superior raw brainpower. (We might analogize this silly and vainglorious, although regrettably com­mon, belief to the more popular idea that great athletes should be able to ex­cel at anything physical by reason of their general bodily virtue — a myth and chimaera that dramatically exploded several years ago when Michael Jordan discovered that he could not learn to hit a curve ball, just because he excelled so preeminently in basketball, and possessed the world's best athletic body in general — for he ended up barely hitting over 0.200 in a full season of minor league play. I do, however, honor and praise his persistence in staying the course and taking his lumps.)

  Indeed, I know of no other substantial conceptual advance in recent science so abetted by the active collaboration of working scientists and professional philosophers (thus obviating, for once, the perennial, and justified, complaint [Page 29] of philosophers of science that no scientists read their journals or even en­counter their analyses). Several key achievements in modern evolutionary theory, particularly the successful resolution of conceptual difficulties in for­mulating a workable theory of hierarchical selection (rooted in concepts like emergence and simultaneous selection at several levels that our minds, with their preferences for two-valued logics, don't handle either automatically, or well at all), have appeared as joint publications of biologists and philoso­phers, including the books of Sober and Wilson, 1998, and Eldredge and Grene, 1992; and articles of Sober and Lewontin, 1982, and Mayo and Gilinsky, 1987. My own understanding of how to formulate an operational theory of hierarchical selection, and my “rescue” from a crucial conceptual error that had stymied my previous thinking (see Chapter 8, pages 656–673), emerged from joint work with Elisabeth Lloyd, a professional philosopher of science. I take great pride in our two joint articles (Lloyd and Gould, 1993; Gould and Lloyd, 1999), which, in my partisan judgment, resolve what may have been the last important impediment to the codification of a conceptually coherent and truly operational theory of hierarchical selection.

  Zeitgeist. Although major revisions to the structure of evolutionary theory emerged mainly from the conventional substrates of novel data and clearer concepts, we should not neglect the admittedly fuzzier, but by no means unimportant, input from a distinctive social context, or intellectual “spirit of the times” (a literal meaning of Zeitgeist) that, at the dawn of a
calendrical millennium, has suffused our general academic culture with a set of loosely coherent themes and concerns far more congenial with the broad revisions here proposed within evolutionary theory than any previous set of guiding concepts or presuppositions had been. Needless to say, Zeitgeists are two edged swords of special sharpness — for either they encourage sheeplike conformity with transient ghosts of time (another literal meaning of Zeitgeist) that will soon fade into oblivion, or they open up new paths to insights that previous ages could not even have conceptualized. Any intellectual would therefore be a fool to argue that conformity with a Zeitgeist manifests any preferential correlation with scientific veracity ipso facto. Zeitgeists can only suggest or facilitate.

  Nonetheless, we would be equally foolish in our naive empiricism if we claimed that major advances in science must be entirely data driven, and that social contexts can only act as barriers to our vision of nature's factuality. Both the social and scientific world were “ready” for evolution in the mid 19th century. People of equal intelligence could neither have formulated nor owned such a concept in Newton's generation, even if some hypothetical Dar­win had then advanced such a claim (and probably ended up in Bedlam for his troubles). In Chapter 2, I shall document not only this general readiness of Western science within the Zeitgeist of Darwin's time, but also the specific so­cial impetus that Darwin gained from studying the distinctive theories (also a product of the earlier Enlightenment Zeitgeist, and not accessible before) of Adam Smith and the Scottish economists. Thus, and by analogy a century later, the altered Zeitgeist of our own time may also facilitate a fruitful [Page 30] reconsideration of major evolutionary concepts that still bear the originating stamp of a Victorian scientific context strongly committed to unidirectional, single-level and deterministic views of natural causality — subtly controlling con­cepts that many scientists would now label as limiting and outmoded.